Pre-registration in Materials Science: A Practical Guide to Enhancing Research Credibility and Rigor

Abstract

This article provides a comprehensive guide to pre-registering materials science studies, a pivotal open science practice for improving research transparency and reproducibility. It covers the foundational principles of pre-registration, detailing how it distinguishes confirmatory from exploratory research to reduce biases like p-hacking and HARKing. The guide offers a step-by-step methodological framework for application, including template selection on platforms like the Open Science Framework (OSF). It addresses common troubleshooting concerns and optimization strategies and concludes with a comparative analysis of its benefits for validation, peer review, and funding acquisition, specifically tailored for researchers and professionals in materials science and drug development.

What is Pre-registration? Building a Foundation for Transparent Science

Pre-registration is the method of formally recording a research plan—including its hypotheses, methodologies, and analysis procedures—in a time-stamped, unchangeable repository before commencing a study [1]. This practice entails submitting a detailed research plan to a registry, which creates a permanent and specific blueprint for the upcoming study [2]. In the context of materials science and drug development, pre-registration serves to future-proof research by clearly separating hypothesis-generating exploratory work from rigorous hypothesis-testing confirmatory research, thereby significantly increasing the credibility and transparency of scientific findings [2] [3].

The central aim of pre-registration is to address critical issues contributing to the replication crisis in science, including publication bias, p-hacking (running multiple analyses until obtaining significant results), and HARKing (Hypothesizing After Results are Known) [3] [1]. By documenting research decisions before data collection and analysis, pre-registration provides a transparent record that helps researchers, reviewers, and readers distinguish between pre-planned confirmatory analyses and unplanned exploratory analyses, thus preserving the diagnostic value of statistical inferences [2] [1].

The Importance of Pre-registration in Materials Science and Drug Development

Distinguishing Between Planned and Unplanned Research

Pre-registration creates a crucial distinction between confirmatory and exploratory research, both of which are valuable but serve different scientific purposes [2]. The table below outlines the key differences:

Table 1: Characteristics of Confirmatory vs. Exploratory Research

Characteristic	Confirmatory Research	Exploratory Research
Primary Goal	Hypothesis testing	Hypothesis generation
Statistical Standards	Results held to highest standards; minimizes false positives	Results deserve replication and confirmation; minimizes false negatives
Data Relationship	Data-independent	Data-dependent
P-value Diagnostic Value	Retained	Loses diagnostic value
Role in Discovery	Rigorously tests predicted effects	Identifies unexpected discoveries and relationships

Combating Questionable Research Practices

In materials science and drug development, where research outcomes can have significant commercial, regulatory, and clinical implications, pre-registration provides a robust defense against questionable research practices (QRPs) [3]. These practices include cherry-picking results that align with hypotheses, p-hacking, and HARKing, which collectively increase the risk of false positives and support for false hypotheses [3]. One alarming estimate indicates that nearly 85% of research funding in biomedical sciences is avoidably wasted due to various QRPs [3]. Pre-registration helps address this problem by requiring researchers to specify their analysis plan in advance, making deviations from the planned methodology transparent and subject to scrutiny during peer review [2] [3].

Practical Implementation of Pre-registration

When to Pre-register

The timing of pre-registration is critical to its effectiveness in preserving the validity of research findings. The optimal points for pre-registration include [2]:

• Before data collection: When no data have been collected, created, or realized [2]
• Prior to data observation: When data exist but have not been quantified, constructed, or observed by anyone [2]
• Before data access: When using existing data that the researcher has not yet accessed [2]
• Before data analysis: When data exist and have been accessed but no analysis related to the research plan has been conducted [2]

For research involving existing datasets, special care must be taken to ensure that the confirmatory nature of the research plan remains uncompromised. Researchers must certify that they have not accessed the data or performed any analyses related to the research plan prior to pre-registration [2].

Pre-registration Templates and Registries

Various specialized templates are available for pre-registration across different research designs and methodologies. The table below summarizes common pre-registration templates and their applications:

Table 2: Pre-registration Templates and Their Applications

Template Name	Description	Primary Research Applications
OSF Preregistration	Standard, comprehensive, and general-purpose preregistration form; most commonly used [4]	General materials science research, drug development studies
OSF-Standard Pre-Data Collection	States whether data have been collected or viewed and includes other pertinent comments [4]	Experimental studies where data collection status must be documented
Secondary Data Preregistration	Designed for research projects using existing datasets [4]	Computational materials science, data mining studies, retrospective analyses
Registered Report Protocol Preregistration	For protocols after receiving "in-principle acceptance" from a Registered Report journal [4]	High-impact materials science and drug development research
Qualitative Preregistration	Template for registering primarily qualitative work [4]	Qualitative studies in materials science, such as expert interviews or case studies

Several registries are available for depositing pre-registration documents, with choice depending on research discipline and study type [3]:

• Open Science Framework (OSF) and AsPredicted.org: General registries welcoming pre-registrations from any discipline [3] [1]
• PROSPERO: For systematic reviews with health-related outcomes [3]
• ClinicalTrials.gov and ISRCTN Registry: For clinical trials and randomized controlled trials [3] [1]
• American Economic Association Registry for Randomized Controlled Trials: For RCTs in economic, political, and social sciences [3]

Essential Components of a Pre-registration

A rigorous pre-registration should function as a detailed roadmap for the entire research project. Effective practices for creating a comprehensive pre-registration include [4]:

• Precision and explicitness: Clearly list hypotheses and specify which variables will be used
• Pre-data decision making: Make all design and analysis decisions before viewing the data
• Contingency planning: Consider potential variations and write "if-then" decision trees for statistical tests and interpretation criteria
• Methodology demonstration: Describe how hypotheses will be tested, including exclusion rules, variable combinations, model forms, covariates, and outcome reporting
• Anticipation of deviations: Describe potential unplanned work and possible deviations from the primary plan

Experimental Protocol for Pre-registration

Workflow for Pre-registration in Materials Science

The following diagram illustrates the complete pre-registration workflow for materials science studies:

Protocol for Split-Sample Analysis in Exploratory Research

For researchers in early, exploratory stages of materials development, the following protocol enables rigorous hypothesis generation while maintaining methodological integrity:

Step-by-Step Protocol:

• Data Partitioning: Randomly divide available or incoming data into two subsets (A and B) using a reproducible method [2].
• Exploratory Analysis: Use Subset A for comprehensive exploration to identify potential relationships, effects, or differences. This phase minimizes false negatives to discover unexpected patterns [2].
• Hypothesis Formalization: Based on exploratory findings, formalize specific, testable hypotheses regarding material properties, structure-property relationships, or processing parameters.
• Pre-registration: Submit a pre-registration document specifying the formalized hypotheses, exact methodological approach for testing, and precise analysis plan [2].
• Confirmatory Testing: Using the untouched Subset B, conduct rigorous hypothesis testing following the pre-registered plan exactly [2].
• Validation Assessment: Evaluate whether the results from Subset B confirm the hypotheses generated from Subset A.

This approach, while reducing available sample size for confirmatory analysis, provides substantial benefits through increased credibility and provides a strong rationale for analytical decisions [2].

Research Reagent Solutions for Pre-registration

The following table details essential components for implementing pre-registration in materials science and drug development research:

Table 3: Essential Research Reagent Solutions for Pre-registration

Resource Category	Specific Examples	Function and Application
Pre-registration Registries	Open Science Framework (OSF), AsPredicted.org, ClinicalTrials.gov, PROSPERO [3]	Provide public, time-stamped repositories for depositing research plans; ensure transparency and establish precedence
Template Libraries	OSF Preregistration, Secondary Data Preregistration, Registered Report Protocol [4]	Standardized forms for documenting research plans; ensure comprehensive coverage of methodological details
Protocol Design Tools	CONSORT guidelines, PRP-QUANT template, specialized preregistration forms [3]	Assist in developing rigorous methodological protocols with appropriate statistical power and analysis plans
Data Management Platforms	OSF Projects, Electronic Lab Notebooks, Data Repositories	Enable organization of research materials, data, and analytical code linked to pre-registrations
Contrast Checking Tools	Acquia Color Contrast Checker, ColorZilla [5]	Ensure accessibility and readability of visual materials in publications and presentations

Addressing Common Challenges in Pre-registration

Managing Changes to Pre-registration Plans

Ideal research execution follows the pre-registered plan exactly, but practical research often requires adaptations. The protocol for handling changes depends on when modifications are needed [2]:

• For pre-registrations less than 48 hours old: Cancel the original registration and create a new one if no contributors have confirmed it yet [2].
• For serious errors or before data collection: Create a new pre-registration with updated information, withdraw the original, and explain the rationale with reference to the new registration [2].
• For changes after study commencement: Create a "Transparent Changes" document uploaded to the OSF project, explicitly detailing deviations from the original plan and their justifications when reporting results [2].

Reporting and Interpreting Pre-registered Analyses

A fundamental principle of pre-registration is the comprehensive reporting of all pre-planned analyses, regardless of their outcomes [2]. Selective reporting or interpretation based on results disrupts the diagnosticity of statistical inferences. For example, if a researcher plans 100 tests and only reports the 5 significant results while omitting the 95 non-significant findings, the interpretation would be misleading and likely reflect false positives [2]. Therefore, pre-registration requires complete reporting of all planned analyses and transparent interpretation that considers the full set of results [2].

Advanced Pre-registration Formats

Registered Reports

Registered Reports represent an advanced implementation of pre-registration that incorporates peer review before data collection [3] [1]. This format occurs in two stages:

Stage 1: Researchers submit the introduction, methods, and proposed analysis plan to a journal for peer review before conducting the study. Based on this review, journals may provide "in-principle acceptance" (IPA), guaranteeing publication regardless of the study outcomes provided the researchers follow the approved methodology [3].

Stage 2: After receiving IPA, researchers conduct the study, then submit the complete manuscript including results and discussion. Reviewers verify that the researchers followed the registered methodology and that the conclusions are supported by the data [3].

This approach significantly reduces publication bias by ensuring that studies are evaluated based on their methodological rigor rather than their results [3]. Over 200 journals now offer the Registered Reports format, with adoption approximately doubling each year [1].

Specialized Pre-registration Formats

Materials scientists and drug development researchers can utilize specialized pre-registration formats tailored to specific research designs [1]:

• Adaptive pre-registration: For studies incorporating adaptive designs that may modify parameters based on interim results [1]
• Pre-registration for existing data: For analyses of previously collected datasets with specific safeguards to maintain confirmatory status [1]
• Pre-registration of systematic reviews: For meta-analyses and systematic reviews to combat selective inclusion of studies [1]
• Pre-registration of exploratory research: For documenting hypothesis-generating research with clear distinction from confirmatory testing [1]

Pre-registration represents a transformative methodological innovation for materials science and drug development research. By requiring researchers to document their hypotheses, methodologies, and analysis plans before conducting studies, pre-registration enhances the transparency, credibility, and diagnostic value of research findings. The practice helps combat questionable research practices, distinguishes between exploratory and confirmatory research, and facilitates more accurate interpretation of statistical results.

As the scientific community continues to address challenges related to reproducibility and research quality, pre-registration and related approaches like Registered Reports provide powerful tools for strengthening research practices. For materials scientists and drug development professionals, adopting pre-registration represents a proactive step toward increasing the rigor, transparency, and impact of their research contributions.

Questionable Research Practices (QRPs), such as HARKing (Hypothesizing After the Results are Known) and p-hacking (collectively manipulating data analysis to obtain statistically significant results), undermine the credibility of scientific research by increasing the rate of false-positive findings [2]. These practices often arise from a flexible research environment where analytical decisions are made based on how the data appear, blurring the line between confirmatory (hypothesis-testing) and exploratory (hypothesis-generating) research [2].

Preregistration is a powerful methodological solution that involves specifying a research plan—including hypotheses, methodology, and analysis strategy—in a time-stamped, immutable document before data collection or observation begins [2] [4]. This practice distinguishes planned confirmatory research from unplanned exploratory work, thereby preserving the diagnostic value of statistical tests and safeguarding against both HARKing and p-hacking [2]. For materials scientists, preregistration enhances the rigor and credibility of empirical findings, whether in exploratory materials discovery or the confirmatory testing of a specific material property.

Core Protocol: Preregistering a Materials Science Study

This protocol provides a step-by-step guide for preregistering a research plan on the Open Science Framework, the most commonly used platform for this purpose [4].

Stage 1: Pre-Registration Planning

Objective: To draft a comprehensive and precise research plan.

• Define the Research Question and Hypotheses: Formulate a clear, specific, and testable primary research question. List all specific hypotheses, including the direction of expected effects for one-tailed tests [2].
• Design the Experimental Methodology:
  • Materials Synthesis & Specification: Detail the procedures for sample preparation, including precursor materials, synthesis conditions (e.g., temperature, pressure, time), and equipment used.
  • Sample Characteristics: Define the sample size and the unit of analysis (e.g., individual specimen, batch). Specify all inclusion/exclusion criteria for samples (e.g., purity thresholds, crystal structure verification) [6].
  • Measurement & Characterization: Describe all characterization techniques (e.g., XRD, SEM, tensile testing) and the specific metrics that will be recorded. Detail the calibration procedures for instruments.
  • Experimental Design: Specify the design (e.g., randomized control trial, case-control, comparison of groups) and all experimental groups or conditions [6].
• Devise the Data Analysis Plan:
  • Data Management: Outline procedures for data handling, coding, and verification [7].
  • Variable Definition: Define all variables, noting which are independent, dependent, and covariates.
  • Statistical Tests: Specify the exact statistical tests (e.g., t-test, ANOVA, regression model) that will be used to test each hypothesis [2] [4].
  • Decision Criteria: Define the criteria for interpreting results, including the alpha level (e.g., p < 0.05) and how effect sizes will be calculated and reported [7].
  • Contingency Plans: Anticipate potential issues, such as violations of statistical assumptions, and specify the alternative analyses that will be used (e.g., "if data are non-normal, a Mann-Whitney U test will be used instead of an independent samples t-test") [4].

Stage 2: Platform Submission on OSF

Objective: To formally submit the research plan to a registry.

• Access the OSF Platform: Navigate to the Open Science Framework website and log in.
• Initiate a New Registration: From your "My Registrations" page or a specific project, click "Add a Registration" [4].
• Select a Preregistration Template: OSF offers multiple templates. For most materials science studies, the OSF Preregistration template (a standard, comprehensive form) or the OSF-Standard Pre-Data Collection Registration is appropriate. Select the one that best fits your study [4].
• Complete the Registration Form: Fill in the template sections, which will align closely with the plan developed in Stage 1. Provide all required metadata, such as title, contributors, and abstract [4].
• Finalize and Submit: Review the completed form for accuracy and any potentially identifying information if a blinded review is anticipated. Submit the registration, which becomes a time-stamped, immutable public record [4].

The following workflow diagram summarizes the preregistration lifecycle, from initial planning to dealing with post-registration changes:

Quantitative Data Analysis Plan

A rigorous preregistration requires a precise plan for analyzing quantitative data. The analysis strategy must be specified before data collection begins to prevent p-hacking [2] [7].

Table 1: Common quantitative data analysis methods to specify in a preregistration plan.

Analysis Category	Purpose	Common Statistical Techniques	Application Example in Materials Science
Descriptive Statistics	To summarize and describe the basic features of the sample data [8].	Mean, Median, Mode, Standard Deviation, Skewness [8].	Reporting the average tensile strength, median grain size, and standard deviation of hardness measurements for a new alloy.
Inferential Statistics	To make predictions or inferences about a population based on sample data [8] [7].	T-tests, ANOVA, Correlation, Regression Analysis [8].	Using an independent samples t-test to compare the mean corrosion resistance between a new coating and a standard coating.
Data Management & Cleaning	To ensure data quality and prepare for analysis [7].	Procedures for handling missing data, identifying outliers, data transformation.	Pre-specifying that any measurement >3 standard deviations from the mean will be flagged as an outlier and subjected to a sensitivity analysis.

Protocol for Confirmatory Data Analysis

Objective: To execute and report the preregistered confirmatory analysis with transparency.

• Follow the Pre-Analysis Plan: Conduct the statistical tests exactly as specified in the preregistration [2].
• Report All Results: Report the outcomes of all preregistered analyses, regardless of whether they were statistically significant. Selective reporting invalidates the purpose of preregistration [2].
• Interpret Results Appropriately: Avoid selectively interpreting only the significant results. The interpretation must account for all planned tests, for example, by adjusting for multiple comparisons if this was part of the plan [2].
• Conduct Exploratory Analysis Separately: Clearly label any unplanned, post-hoc analyses as "exploratory." These results are hypothesis-generating and require confirmation in future studies [2].

The pathway below outlines the critical decision points in a preregistered data analysis workflow, highlighting the separation between confirmatory and exploratory work:

Research Reagent Solutions: The Preregistration Toolkit

For researchers, the "reagents" for implementing preregistration are not physical materials but conceptual tools and platforms that facilitate rigorous, transparent science.

Table 2: Essential tools and resources for the preregistration process.

Tool / Resource	Function	Usage Note
OSF Preregistration Template	A standardized form on the Open Science Framework that guides researchers through the key components of a research plan [4].	The most commonly used template; ideal for general materials science studies.
AsPredicted.org Template	An alternative, concise preregistration template consisting of eight key questions [4].	Useful for studies with a simpler design where a shorter form is sufficient.
Secondary Data Preregistration Template	A specialized OSF template for research projects that use an existing dataset [4].	Critical for materials informatics or studies re-analyzing public datasets. Must justify that outcomes have not been observed.
Transparent Changes Document	A living document used to record and justify any deviations from the original preregistered plan after the study has begun [2] [4].	Upload to the OSF project; essential for maintaining transparency when perfect adherence to the plan is not possible.
Data Visualization Software (e.g., R/ggplot2, Python/Matplotlib)	Software capable of producing complex, effective figures beyond default spreadsheet charts [9].	Necessary for creating publication-quality visuals that accurately represent the data and follow best practices.

Data Visualization for Preregistered Research

Effective data visualization is crucial for communicating the results of a preregistered study clearly and honestly.

Principles for Effective and Truthful Visuals

• Diagram First: Before using software, prioritize the information you want to share. Focus on the core message (e.g., a comparison, trend, or distribution) [9].
• Use an Effective Geometry: Match the visual representation (geometry) to your data and message. For example, use scatterplots or line plots for relationships, box plots or violin plots for distributions, and bar plots only for simple comparisons where distributional information is not critical [9].
• Show the Data: Maximize the data-ink ratio by removing non-data ink and, where possible, visualizing the underlying data points (e.g., using data points overlayed on a bar or box plot) rather than just summary statistics [9].
• Use Color Effectively:
  • Use qualitative palettes for categorical data.
  • Use sequential palettes for numeric data with a natural order.
  • Use diverging palettes for numeric data that diverges from a center value [10].
  • Ensure sufficient color contrast and consider colorblind-friendly palettes [11].

Protocol for Creating a Visualization

Objective: To generate a figure that truthfully represents the study findings.

• Determine the Message: Based on your confirmed results, state the single most important message the figure must convey.
• Select the Geometry: Choose the most appropriate plot type based on your data type and message (see [9] and [11]).
• Create the Plot: Using your software of choice, generate the initial figure.
• Refine and Simplify: Remove unnecessary gridlines, borders, or legends (chartjunk). Ensure axes are labeled clearly and the scale is appropriate to avoid misinterpretation [10] [11].
• Apply Color: Use color strategically to highlight key findings, not for decoration. Verify that the color palette is accessible [10] [11].

Within the rigorous field of materials science, the strategic planning of research activities is paramount for efficient resource allocation and generating reliable, translatable results. A critical first step in this process is the explicit classification of a study as either exploratory or confirmatory. These are two distinct, complementary modes of investigation, each with different overarching goals, designs, and interpretations [12].

Exploratory investigation aims to generate robust pathophysiological theories of disease or, in the context of materials science for drug development, to navigate a vast landscape of potential targets, drugs, doses, and treatment regimens. Its primary goal is to develop theories, measurement techniques, and evidence for selecting a manageable number of interventions to carry forward [12]. In contrast, confirmatory investigation aims to demonstrate strong and reproducible treatment effects in relevant models. It produces evidence that is sufficiently compelling to warrant the significant economic and moral costs of clinical development [12].

The following application notes and protocols are designed to guide materials scientists and drug development professionals in effectively distinguishing and implementing these two research modes, with a specific focus on integrating pre-registration practices into their workflow.

Comparative Analysis: Exploratory vs. Confirmatory Research

The table below summarizes the core distinctions between these two research approaches, providing a clear framework for study planning.

Table 1: Key Characteristics of Exploratory and Confirmatory Research

Characteristic	Exploratory Research	Confirmatory Research
Primary Goal	Generate hypotheses and theories; select promising candidates from a wide field [12].	Produce rigorous, reproducible evidence of clinical promise for a specific intervention [12].
Inferential Focus	Developing pathophysiological theories; sensitivity (detecting all potentially useful strategies) [12].	Demonstrating clinical utility; specificity (excluding useless strategies) [12].
Study Design	Flexible, evolving sequence of experiments; often a package of small studies [12].	Rigid, pre-specified design established before data collection [12].
Hypotheses	Loosely articulated or evolving; may not be pre-specified [12].	Clearly stated a priori (before data collection) [12].
Sample Size	Often smaller; effect sizes may be unknown [12].	Adequately powered; large sample sizes to minimize random variation [12].
Measurement Techniques	May be in development; uncertainty surrounds assays [12].	Well-established and clinically relevant [12].
Role of Pre-registration	Less common; can be used to document initial plans despite flexibility.	Essential for specifying hypotheses, methods, and analysis plan before data collection [4].

Experimental Protocols for Materials Science Research

Protocol for Exploratory Research: Screening Novel Biomaterial Formulations

1. Objective: To identify promising novel polymer formulations for controlled drug release from a large set of candidate materials. 2. Pre-Study Registration (OSF Preregistration Template): While full pre-registration is not always feasible, document the initial research question, the library of candidate materials, and the primary high-throughput screening assay. 3. Materials Synthesis: * Prepare a library of 50 candidate polymer hydrogels with systematic variations in cross-linking density and monomer ratios. * Use combinatorial chemistry techniques for efficient synthesis. 4. High-Throughput Screening: * Load each polymer formulation with a model fluorescent drug compound. * Immerse formulations in a simulated physiological buffer (pH 7.4) at 37°C. * Use an automated plate reader to measure drug release kinetics at 1, 3, 6, 12, and 24 hours. 5. Data Analysis: * Calculate cumulative drug release profiles for each formulation. * Apply unsupervised machine learning (e.g., k-means clustering) to group formulations with similar release profiles. * Select the top 3-5 formulations for further, more detailed investigation based on desired release kinetics (e.g., sustained release over 24 hours). 6. Interpretation: Results are hypothesis-generating. The selected formulations become candidates for a confirmatory study. Negative findings are still valuable for understanding structure-property relationships.

Protocol for Confirmatory Research: Validating Drug Release from a Lead Formulation

1. Objective: To rigorously demonstrate that the lead polymer formulation (Formulation A23) provides sustained release of Therapeutic X over 72 hours in a physiologically relevant model. 2. Pre-Study Registration (OSF Preregistration Template): This step is mandatory. The pre-registration must include [4]: * Primary Hypothesis: "Formulation A23 will release 80% ± 5% of Therapeutic X over 72 hours in an in vitro model, a significant improvement over the current standard (60% release in 48 hours)." * Design: Pre-specified sample size (n=15 per group, justified by a power analysis), and detailed methodology. * Materials: Exact chemical specification of Formulation A23 and the control formulation. * Analysis Plan: Pre-specified statistical test (e.g., one-sided t-test comparing release at 72 hours) and criteria for success. 3. Experimental Workflow: * Prepare 15 identical samples of the pre-specified lead Formulation A23 and 15 control samples. * Load each with a precise dose of Therapeutic X. * Conduct drug release studies in triplicate under simulated physiological conditions, sampling at 0, 4, 8, 12, 24, 48, and 72 hours. * Use HPLC to quantify drug concentration in the release medium with high precision. 4. Data Analysis: * Adhere strictly to the pre-registered analysis plan. * Calculate mean cumulative release and standard deviation for each time point. * Perform the pre-specified t-test to compare the 72-hour release of Formulation A23 against the control. * Do not deviate from the plan or add unplanned post-hoc tests without clearly labeling them as exploratory. 5. Interpretation: The study provides a strong, falsifiable test of the pre-registered hypothesis. A positive result provides compelling evidence to advance the formulation to more complex (e.g., in vivo) testing.

Visualizing the Research Workflow

The following diagram illustrates the typical workflow and the critical decision points when navigating between exploratory and confirmatory research phases.

Research Workflow: Exploratory and Confirmatory Modes

The Scientist's Toolkit: Research Reagent Solutions

For materials scientists conducting research in drug development, several key resources and platforms are essential for planning and documenting both exploratory and confirmatory studies.

Table 2: Essential Research Reagents and Resources for Materials Science

Item / Resource	Function / Description	Applicability to Research Mode
Open Science Framework (OSF)	A multidisciplinary web application for collaborating, documenting, and registering research projects, materials, and data [4] [13].	Critical for Confirmatory, useful for Exploratory. Used for pre-registration and sharing protocols.
Springer Nature Experiments	A database of over 60,000 peer-reviewed protocols from sources like Nature Protocols and Methods in Molecular Biology [14].	Both Modes. Provides established methodologies for theory-building and validation.
Journal of Visualized Experiments (JoVE)	A peer-reviewed video journal publishing methods articles accompanied by videos of experiments [14].	Both Modes. Aids in learning complex techniques essential for both exploration and confirmation.
protocols.io	A platform for creating, organizing, and publishing reproducible research protocols; allows for dynamic version control [14].	Both Modes. Ideal for developing and sharing detailed, executable lab procedures.
Current Protocols Series	A subscription-based resource with over 20,000 updated, peer-reviewed lab methods across disciplines [14].	Both Modes. A key source for standardized, vetted laboratory techniques.
Pre-registration Template (OSF)	A standardized form to document study aims, hypotheses, methods, and analysis plan before data collection [4].	Essential for Confirmatory. Locks in the analytical pathway to prevent bias.

Embracing the distinction between exploratory and confirmatory research, and leveraging tools like pre-registration, is fundamental to enhancing the reliability and efficiency of materials science research in drug development. By intentionally designing studies for either theory generation or rigorous validation, and by transparently documenting their plans, researchers can build a more robust and reproducible foundation for translating novel materials from the lab to the clinic.

Preregistration is the practice of posting a time-stamped, read-only version of a study plan to a public registry before beginning data collection or analysis. This establishes a transparent record of research intentions [4]. For materials science researchers and drug development professionals, this represents a fundamental shift toward enhanced research credibility. The core value of preregistration lies in its ability to clearly distinguish between confirmatory research (planned, hypothesis-testing) and exploratory research (unplanned, hypothesis-generating), preserving the diagnostic value of statistical inferences [2]. In an era of increasing scrutiny of research reproducibility, particularly in fields involving advanced materials characterization and therapeutic development, preregistration provides a mechanism to future-proof research by staking claim to ideas earlier and planning for more rigorous experimental design.

Theoretical Framework: The "Essential Why" of Preregistration

Distinguishing Planned from Unplanned Research

The theoretical foundation of preregistration rests on clarifying the distinction between two equally important but fundamentally different modes of scientific inquiry:

• Confirmatory Research: Pre-specified hypothesis testing where results are held to the highest standards; this data-independent planning minimizes false positives and ensures p-values retain diagnostic value [2].
• Exploratory Research: Hypothesis-generating work that minimizes false negatives to discover unexpected relationships; results deserve replication and confirmation [2].

Preregistration does not prohibit unplanned analyses but rather creates transparency about which analyses were planned versus exploratory, allowing readers to appropriately evaluate the evidence presented [2].

Addressing Challenges in Materials Science Research

Materials science research faces particular challenges that preregistration can help address:

• High-throughput experimentation: Preregistration helps manage the risk of false discoveries when screening numerous material compositions simultaneously.
• Complex characterization data: By prespecifying analysis methods for XRD, SEM, TEM, and other characterization techniques, researchers reduce analytical flexibility.
• Reproducibility crises in advanced materials: Transparent methods and analysis plans facilitate replication across different laboratories and instrumentation.

Application Notes: Implementing Preregistration in Practice

Preregistration Templates for Materials Research

The Open Science Framework (OSF) provides multiple registration templates suitable for materials science studies [4]:

Table 1: Preregistration Templates for Materials Science Research

Template Name	Description	Best Use Cases
OSF Preregistration	Standard, comprehensive, and general purpose	Most materials science studies; most commonly used
Open-Ended Registration	Most flexible template	When no other template fits study design or registering completed work
Pre-registration in Social Psychology	Template emphasizing hypotheses, methods, and analysis plans	Research focusing on material properties with clear directional hypotheses
AsPredicted.org Template	Simplified 8-question format	Straightforward materials testing with minimal variables
Secondary Data Preregistration	For projects using existing datasets	Computational materials science using existing databases

Effective Practices for Rigorous Preregistration

Research indicates several effective practices for creating rigorous preregistrations [4]:

• Use preregistration as a draft methodology section: This helps systematically think through experimental procedures and how results will be reported.
• Precision and explicitness: Clearly list all hypotheses and specify which variables will be used for each test.
• Make design decisions before data collection: Finalize experimental designs, sample sizes, and characterization methods before viewing data.
• Anticipate contingencies: Describe "if-then" decision trees for potential outcomes (e.g., "if assumptions for ANOVA are violated, we will use Kruskal-Wallis test").
• Specify statistical approaches: Detail all statistical tests, decision criteria for interpretation, exclusion rules, and how variables will be combined.
• Plan for unplanned work: Explicitly describe what exploratory analyses might be conducted alongside the confirmatory tests.

Experimental Protocols: Preregistration Workflow

Protocol 1: Creating a Draft Registration from Scratch

For researchers beginning a new materials science study without existing OSF projects:

• Access registry platform: Navigate to the OSF Registry from the OSF dashboard or "My Registrations" page [4].
• Initiate new registration: Click "Add a Registration" and select "No" when asked if content exists in an existing OSF project [4].
• Template selection: Choose the registration template that best fits the research project (see Table 1 for guidance) [4].
• Complete draft registration: Fill in the registration metadata and provide responses to all template questions [4].
• Save and confirm: An email containing the link to the registration draft will be sent to the OSF-affiliated email address; save this link for future access [4].

Protocol 2: Registering from an Existing Project

For researchers with preliminary data or established experimental frameworks:

• Project selection: From "Add a Registration," select "Yes" when asked about existing content, then choose the relevant project or component from the dropdown list (only projects with admin permissions are visible) [4].
• Alternative access: Navigate to the project's "Registrations" tab and select "Add a Registration" – Steps 1 and 2 will be prefilled [4].
• Template selection: Choose the appropriate registration form (note: files associated with the project up to 5GB will be included) [4].
• Component-level registration: For complex materials science projects with multiple sub-studies, navigate to specific components and use their Registration tab to register discrete aspects separately [4].

Protocol 3: Preregistration with Existing Data

Materials science research often utilizes existing datasets; special considerations apply:

• Prior to data observation: Certify that data exist but have not been quantified, constructed, or observed by anyone [2].
• Prior to data access: Certify that data exist but have not been accessed by the researcher or collaborators [2].
• Prior to data analysis: Certify that data exist and have been accessed but no analysis related to the research plan has been conducted [2].
• Split-sample validation: For exploratory work, split data into exploration and validation sets; preregister findings from the exploratory set before confirming with the validation set [2].

Data Presentation and Visualization Standards

Effective Data Communication

Research communication in materials science benefits from strategic data presentation:

• Use non-textual elements judiciously: Incorporate approximately one table or figure per 1000 words to break textual monotony and enhance understanding [15].
• Select appropriate formats: Use tables for precise numerical values and figures for trends, patterns, or relationships [16].
• Ensure self-contained presentation: Each table and figure should be independently understandable with descriptive titles, clear legends, and necessary contextual information [15].

Table 2: Research Reagent Solutions for Materials Characterization

Reagent/Material	Function	Application Examples
Standard Reference Materials	Calibration and validation	Instrument calibration, method verification
High-Purity Solvents	Precursor synthesis and processing	Nanomaterial synthesis, thin film deposition
Etchants and Developers	Surface patterning and characterization	Microfabrication, metallographic analysis
Certified Matrix Matched Standards	Quantitative analysis calibration	Spectroscopy, chromatography calibration
Functionalization Reagents	Surface modification	Nanoparticle functionalization, interface engineering

Data Visualization Workflow

The following diagram illustrates the strategic decision process for selecting appropriate data visualization formats in materials science research:

Accessibility in Scientific Visualization

Creating accessible visualizations ensures research is available to the broadest possible audience:

• Color contrast requirements: Maintain a minimum contrast ratio of 4.5:1 for normal text and 3:1 for large text (approximately 18 point or 14 point bold) [17].
• Color deficiency considerations: Use patterns, textures, or direct labeling in addition to color coding [18].
• Text alternatives: Provide detailed descriptions for complex diagrams and flowcharts, explaining how you would describe the visual over the phone [18].
• High-contrast color palette: Utilize the specified color palette (#4285F4, #EA4335, #FBBC05, #34A853, #FFFFFF, #F1F3F4, #202124, #5F6368) with explicit text color settings to ensure readability [19].

Advanced Implementation Strategies

SPIRIT 2025 Framework for Experimental Protocols

The updated SPIRIT 2025 statement provides an evidence-based checklist of 34 minimum items for trial protocols, relevant for materials science studies involving experimental testing [20]:

• Open science emphasis: New sections address trial registration, sharing of full protocols, statistical analysis plans, and de-identified participant-level data [20].
• Harm assessment: Enhanced focus on adverse event documentation and safety monitoring in experimental trials [20].
• Patient and public involvement: Specification of how stakeholders will be involved in research design, conduct, and reporting [20].

Managing Preregistration Revisions

Research evolution necessitates protocol adjustments; transparent change management is essential:

• Immediate corrections: If the preregistration is less than 48 hours old and not yet confirmed, it can be canceled and recreated [2].
• Substantive changes post-data collection: Create a "Transparent Changes" document uploaded to the OSF project, explicitly documenting deviations from the original plan [2].
• Version control: Maintain all protocol versions with clear audit trails documenting dates and rationales for changes [20].

Preregistration represents a transformative practice for enhancing credibility, transparency, and reproducibility in materials science research. By creating a time-stamped research plan before experimentation, researchers can clearly distinguish confirmatory from exploratory findings, thereby increasing the diagnostic value of their statistical inferences. The protocols and application notes provided herein offer practical guidance for implementing preregistration within diverse materials research contexts—from fundamental materials characterization to applied drug development studies. As the research community increasingly values transparency and rigor, preregistration stands as an essential tool for establishing robust, defensible, and impactful materials science research.

The practice of preregistration—submitting a time-stamped, immutable research plan before beginning data collection or analysis—is a cornerstone of open science in fields like psychology and medicine [4] [2]. It establishes a transparent record of research intentions, distinguishing between hypothesis-generating (exploratory) and hypothesis-testing (confirmatory) research to enhance the credibility of findings [2]. This article argues for the urgent adoption of preregistration within materials science, a field where valuable data on composition, processing, characterization, and performance are often scattered across text, tables, and figures in research papers [21]. By formalizing the "story" of a study from its inception, preregistration provides a robust framework for combating subjective analytical flexibility and ensuring that materials research is both rigorous and reproducible [4].

Application Notes: Data Presentation in Materials Science

Effective data presentation is critical for communicating complex materials research. The tables below summarize quantitative findings on data distribution in scientific literature and provide guidelines for optimizing table design, tailored for materials science reporting.

Table 1: Distribution of Materials Data Types Across Article Sections An analysis of materials science papers reveals where key data types are typically located, highlighting the importance of structured data planning and presentation [21].

Data Type	Primarily Found In	Secondary Locations	Notes on Presentation
Composition	Text, Tables	Figures	Often detailed in experimental sections; requires precise specification.
Processing Conditions	Text	Tables, Figures	Parameters (e.g., temperature, pressure) are frequently text-based.
Characterization Data	Figures	Tables, Text	Includes microscopy, spectroscopy; visual presentation is dominant.
Performance Properties	Tables, Figures	Text	Quantitative results (e.g., efficiency, strength) are often tabled or graphed.

Table 2: Guidelines for Designing Effective Data Tables Adherence to design principles significantly enhances the clarity and interpretability of data tables in scientific publications [22].

Design Principle	Application	Rationale
Aid Comparisons	Right-flush align numbers and their headers; use a tabular font.	Facilitates vertical comparison of numerical values by aligning place values [22].
Reduce Visual Clutter	Avoid heavy grid lines; remove unit repetition within cells.	Reduces cognitive load and directs focus to the data itself [22].
Increase Readability	Ensure headers stand out; use active, concise titles; orient tables horizontally.	Makes the table self-explanatory and easier to scan [22].

Experimental Protocols

Protocol for Preregistering a Materials Science Study

This protocol provides a step-by-step methodology for preregistering a materials science study on the Open Science Framework (OSF), following rigorous open science standards [4] [20].

I. Preparation and Planning

• Define Study Type: Determine if the study involves new data collection, uses an existing dataset, or is a replication. This dictates the appropriate preregistration template [2].
• Select a Registration Template: On the OSF, choose a template that best fits the study design. For general materials science studies, the "OSF Preregistration" template is recommended. For studies using existing data, the "Secondary Data Preregistration" template is appropriate [4].
• Anonymization Check: If submitting for blinded peer-review, ensure all identifying information is removed from the planned metadata and documents. Plan to use an embargo during registration [4].

II. Drafting the Preregistration Fill out the selected template comprehensively, addressing the following key elements derived from SPIRIT 2013 guidance and OSF requirements [4] [20]:

• Hypotheses: State all primary and secondary research questions and hypotheses clearly and precisely.
• Variables: Define all variables, including independent (e.g., material processing parameters), dependent (e.g., measured properties), and covariates.
• Sample and Materials Preparation:
  • Specify the source, purity, and synthesis method of all starting materials.
  • Detail the exact steps for sample preparation, including all equipment models and environmental conditions (e.g., temperature, humidity).
  • Define the sample size (n) for each experimental group and the rationale for this size (e.g., based on power analysis or practical constraints).
  • Outline the randomization and blinding procedures for sample treatment and measurement, if applicable.
• Characterization and Data Collection Plan:
  • List all characterization techniques to be used (e.g., SEM, XRD, tensile testing).
  • Specify the exact settings and parameters for each instrument.
  • Describe the data to be collected from each technique and the file formats.
• Analysis Plan:
  • Describe all planned data processing steps (e.g., filtering, normalization).
  • Specify the exact statistical tests or numerical models that will be used to test each hypothesis.
  • Define the criteria for interpreting results (e.g., significance level α = 0.05).
  • Plan for any contingent analyses (e.g., "If assumptions for ANOVA are violated, a Kruskal-Wallis test will be used") [4].
• Data Management and Dissemination:
  • State whether and how the full protocol, raw data, and analysis code will be shared.
  • Declare the intended licensing for data and materials.
  • Describe the plan for disseminating results, regardless of outcome [20].

III. Submission and Registration

• Create a Draft: On the OSF, initiate a new registration from an existing project or from scratch.
• Complete the Form: Navigate through the template sections, providing all required information.
• Finalize and Submit: Once the draft is complete, submit the registration. The submitted version becomes a time-stamped, immutable record [4].

Protocol for Data Management and Analysis of Quantitative Materials Data

Robust data management is fundamental to the integrity of quantitative research. This protocol outlines the key stages for processing numerical data in materials science [7].

I. Data Management

• Data Entry and Cleaning:
  • Carefully check all entered data for errors and outliers.
  • Identify and document procedures for handling missing values.
• Variable Definition and Coding:
  • Define all variables clearly (e.g., units, measurement scale).
  • Code categorical variables appropriately for statistical software.

II. Data Analysis

• Descriptive Statistics: Summarize the variables using measures of central tendency (mean, median) and spread (standard deviation, range) to show what is typical for the sample [7].
• Inferential Statistics:
  • Use statistical tests to test predefined hypotheses.
  • Report the P-value, which indicates the probability that the observed effect is due to chance [7].
• Effect Size Calculation: Crucially, accompany P-values with a measure of effect size (e.g., Cohen's d, R²) to interpret the practical or scientific magnitude of the finding, which is essential for clinical or industrial decision-making [7].

Mandatory Visualization

Diagram: Preregistration Workflow for Materials Science

This diagram illustrates the end-to-end process of preregistering a materials science study.

Diagram: Materials Data Management and Analysis Pathway

This diagram outlines the logical flow from raw data collection to final interpretation, emphasizing the distinction between planned and unplanned analysis.

The Scientist's Toolkit: Key Research Reagent Solutions

This table details essential materials and resources for conducting and documenting preregistered materials science research.

Table 3: Essential Resources for Preregistered Materials Science

Item / Resource	Function / Purpose
Open Science Framework (OSF)	A free, open-source platform to create a project, collaborate, and submit preregistrations. It is the central hub for the preregistration process [4] [2].
OSF Preregistration Template	A standardized form that guides researchers in specifying their hypotheses, methods, and analysis plan, ensuring all critical elements are considered [4].
SPIRIT 2013 Guidelines	An evidence-based set of recommendations defining the minimum content of a clinical trial protocol. While for clinical trials, its principles of completeness and transparency are highly applicable to materials science study protocols [20].
ColorBrewer	An online tool designed to select effective, colorblind-safe color palettes for sequential, diverging, and qualitative data in figures and charts [23] [24].
Tabular Font (e.g., Lato, Roboto)	A monospace font where each character has the same width. Its use in numerical columns of tables ensures decimal points and place values align vertically, dramatically aiding number comparison [22].
Transparent Changes Document	A living document, uploaded to the OSF project, that records and justifies any deviations from the original preregistered plan after data collection has begun, maintaining transparency [2].

How to Pre-register: A Step-by-Step Guide for Materials Scientists

The Open Science Framework (OSF) is a free, open-source research management platform developed by the Center for Open Science (COS). It provides a centralized environment for managing the entire research lifecycle, integrating components like datasets, protocols, and analysis scripts into a structured repository to reduce fragmentation and streamline workflows [25]. For materials scientists and drug development professionals, OSF offers a powerful solution to enhance research transparency, reproducibility, and collaboration, which are critical in fields characterized by complex experimental procedures and data analysis.

A core functionality of OSF is its support for preregistration, the practice of publicly documenting a research plan—including hypotheses, methods, and analysis strategy—before conducting the study [2]. Preregistration distinguishes confirmatory (hypothesis-testing) research from exploratory (hypothesis-generating) research, thereby safeguarding the diagnostic value of statistical inferences and increasing the credibility of reported results [2]. Within the context of materials science, preregistering studies on novel polymer synthesis or catalyst efficiency creates an immutable, time-stamped record of the experimental intent, mitigating both conscious and unconscious bias in reporting outcomes.

Core OSF Features and Benefits for Researchers

The OSF platform is designed to support the unique needs of the research community through a suite of collaborative and transparent features.

• Centralized Project Management: OSF provides a unified space to organize all research materials, reducing workflow fragmentation. Projects can be structured with components for different aspects of the study, such as literature review, experimental data, and statistical analysis [25].
• Seamless Collaboration: Multiple contributors can work on a project with customizable permission levels (administrator, read, write), facilitating teamwork across institutions and geographical distances [25].
• Enhanced Transparency and Reproducibility: Projects or specific components can be made publicly accessible to promote open inquiry and independent verification of results. The registration feature creates a permanent, time-stamped record of the research plan [25] [4].
• Integrated Workflows with Third-Party Tools: OSF supports numerous storage and reference management service integrations, or "add-ons," including GitHub, Dropbox, Google Drive, and Zotero [25] [26]. This allows researchers to incorporate their preferred tools directly into the OSF project environment.
• Version Control: The platform tracks changes to files and documents, allowing users to revert to previous versions. This is crucial for maintaining an accurate audit trail of the research process [25].

Table 1: Overview of OSF Integrations (Add-ons) Relevant to Materials Science

Category	Example Tools	Primary Research Function
Cloud Storage	Google Drive, Dropbox, Amazon S3	Centralized data backup and file management
Version Control	GitHub, GitLab	Management of analysis code and simulation scripts
Reference Management	Zotero, Mendeley	Organization of literature and citation collection
Preprint Servers	arXiv, bioRxiv	Dissemination of preliminary research findings

A Protocol for Pre-registering a Materials Science Study on OSF

This section provides a detailed, step-by-step methodology for preregistering a materials science research project using the OSF platform.

Pre-registration Preparation and Planning

• Define Your Research Question and Analysis Plan: Clearly articulate your primary research question, hypotheses, experimental design, materials to be used, and the exact statistical tests you will employ. For materials science, this could involve specifying the synthesis parameters, characterization techniques (e.g., SEM, XRD), and performance metrics you plan to measure.
• Select an Appropriate Preregistration Template: OSF offers multiple templates. The OSF Preregistration template is a general-purpose and commonly used form. For a systematic review of material properties, the Generalized Systematic Review template may be more suitable [4].
• Determine Anonymity Needs: If you plan to submit your registration for blinded peer review, ensure all identifying information is removed from the documents and metadata. You will need to embargo your registration [4].

Step-by-Step OSF Preregistration Workflow

• Initiate the Registration:
  • Navigate to your "My Registrations" page on OSF and click "Add a Registration" [4].
  • Choose whether to start from an existing OSF project (which will attach its files to the registration) or start from scratch [4].
• Choose and Complete the Template:
  • Select the template you identified in the preparation phase (e.g., OSF Preregistration) [4].
  • Click "Create Draft" to open the draft registration form.
  • Meticulously complete all required sections (marked in red or with an exclamation mark) [4]. Adhere to effective practices for rigor, such as being precise with variables and detailing statistical tests [4].
• Finalize and Submit:
  • Review all entered information and attached files for accuracy and completeness.
  • Once submitted, the registration becomes a permanent, time-stamped record. Edits are not possible after finalization, barring cancellation within the first 48 hours [2].

The workflow for this protocol, from preparation to the creation of a time-stamped research plan, is summarized in the diagram below.

Research Reagent Solutions for a Featured Experiment

Table 2: Essential Materials for a Hypothetical Study on "Novel Battery Anode Synthesis"

Research Reagent / Material	Specification / Purity	Primary Function in the Experiment
Graphite Powder	99.99%, -325 mesh	Primary precursor material for graphene oxide synthesis.
Potassium Permanganate (KMnO₄)	ACS Reagent, ≥99.0%	Strong oxidizing agent for the chemical exfoliation of graphite.
Hydrochloric Acid (HCl)	37% w/w, Analytical Grade	Washing and purification of the synthesized material.
N-Methyl-2-pyrrolidone (NMP)	Battery Grade, 99.9%	Solvent for slurry preparation during electrode casting.
Polyvinylidene Fluoride (PVDF)	MW ~534,000	Binder agent to adhere active materials to the current collector.
Silicon Nanoparticles	50-100 nm, 99.8%	Active anode material to enhance lithium-ion capacity.

Advanced OSF Capabilities: APIs and Institutional Use

For power users and institutions, OSF provides advanced capabilities through its Application Programming Interface (API) and customization options.

The OSF API is a RESTful interface that allows for the programmatic management of research projects and data. Researchers can automate workflows, such as creating projects, managing files, and integrating with other tools [27]. Key capabilities include:

• Project Management: Programmatically create nodes (projects), define their structure with components, and set contributor permissions [27].
• File Handling: Automate file uploads and downloads, manage metadata, and track versions across OSF Storage and connected add-ons like GitHub [27].
• Preregistration Automation: Facilitate the creation and management of study protocols and registered reports at scale [27].

The API uses JSON and supports authentication via OAuth 2.0 or personal access tokens. The following diagram illustrates a potential automated workflow for data collection and archiving enabled by the OSF API.

For institutions, OSF can be customized with branded solutions to align with specific organizational policies and create a cohesive research infrastructure [25]. The recent enhancement of OSF add-ons as a standalone service lowers technical barriers, allowing institutions and developers to build and maintain their own tool integrations, thereby extending the platform's functionality to meet unique community needs [26].

The Open Science Framework represents a paradigm shift towards more transparent, collaborative, and rigorous research practices. For researchers in materials science and drug development, where the integrity of experimental data is paramount, OSF provides a comprehensive platform to structure the entire research lifecycle. By leveraging its core features—particularly preregistration—scientists can fortify the credibility of their findings, streamline collaboration, and enhance the reproducibility of their work. The integration of familiar tools and the power of the OSF API make it an indispensable component of the modern researcher's toolkit, paving the way for more reliable and impactful scientific discovery.

Preregistration is the practice of publishing a time-stamped, read-only version of a research plan in a public repository before beginning data collection or analysis [2]. This process establishes a transparent record of research intentions, distinguishing clearly between hypothesis-generating exploratory work and hypothesis-testing confirmatory research [2]. For materials scientists and engineers, who frequently contribute to research proposals, technical reports, and patent applications, preregistration offers a structured approach to documenting study plans [28].

The Open Science Framework (OSF) provides a centralized platform for creating preregistrations, hosting multiple template options ranging from general-purpose forms to discipline-specific protocols [29]. This guide examines these template options through the specific lens of materials science research, providing detailed protocols for selecting and implementing the most appropriate framework for your study.

OSF does not recommend a specific template, acknowledging that researchers must select forms aligned with their study details, institutional policies, and community standards [29]. The platform offers numerous templates, each with distinct structures, sections, and questioning approaches.

Comprehensive Template Comparison

The table below summarizes the key OSF registration templates relevant to materials science research:

Template Name	Primary Purpose	Key Focus Areas	Example Use Cases in Materials Science
OSF Preregistration [29]	Standard, comprehensive, general-purpose preregistration	Hypotheses, methodology, analysis plan	General materials testing, characterization studies, comparative analysis of material properties
OSF-Standard Pre-Data Collection [29]	Document pre-analysis plans, state data collection status	Confirmation of no prior data collection/viewing, uploaded analysis plan	Experiments where preliminary data exists but formal analysis hasn't begun
Open-Ended Registration [29]	Maximum flexibility for non-standard designs	Minimal structured questions, broad documentation	Complex computational modeling, multi-stage materials synthesis projects
Secondary Data Preregistration [29]	Preregister analyses using existing datasets	Analysis plan for previously collected data	Utilizing public datasets (e.g., ASM Alloy Center, ASTM standards) [30]
Registered Report Protocol Preregistration [29]	Formal protocol after journal in-principle acceptance	Complete study methodology for peer review	Registered Reports for journals like Journal of Materials Science

Template Selection Workflow

The following diagram illustrates the decision process for selecting an appropriate preregistration template:

Materials Science Research and Preregistration Considerations

Materials science encompasses diverse methodologies ranging from fundamental materials characterization to applied engineering testing. Effective preregistration must account for this methodological diversity while maintaining rigorous documentation standards.

When preregistering studies, particularly those utilizing existing data or literature, materials scientists should consult these essential databases:

Resource Name	Content Focus	Utility in Preregistration
ASM Handbooks Online [30]	Comprehensive materials property data	Establishing baseline expected outcomes, reference standards
ASTM Compass [30]	Full-text standards and technical publications	Referencing standardized testing methodologies
Compendex [28] [30]	Engineering and technical literature (1884-present)	Comprehensive literature review documentation
Scopus [30]	Multidisciplinary research literature	Identifying knowledge gaps, supporting research rationale
Knovel [31]	Reference works, materials properties data	Accessing materials property data for hypothesis formation

Materials-Specific Template Selection Protocol

Objective: Select the optimal OSF preregistration template for a materials science study.

Procedure:

• Define Data Status: Determine your study's data collection status using the categories in [2]:
  • Category 1: No data collected or realized (use OSF Preregistration or OSF-Standard Pre-Data Collection)
  • Category 2: Data exists but not observed/measured (use OSF-Standard Pre-Data Collection with justification)
  • Category 3: Data accessed but not analyzed (use Secondary Data Preregistration with transparency about prior access)
  • Category 4: Analysis begun on portion of data (use Secondary Data Preregistration with split-sample justification)

• Identify Research Type: Classify your study following the confirmatory-exploratory framework [2]:

  • Confirmatory: Testing specific, pre-defined hypotheses about material behaviors or properties (use OSF Preregistration)
  • Exploratory: Investigating new material systems, unknown relationships, or generating hypotheses (use Open-Ended Registration)

• Methodology Alignment: Match your experimental approach to template strengths:

  • Standardized Testing: Studies using ASTM, ISO, or other standardized methods (use OSF Preregistration)
  • Computational/Modeling: Simulation studies, phase diagram calculations (use Open-Ended Registration or OSF Preregistration)
  • Materials Synthesis & Characterization: Development of new materials with multiple characterization stages (use OSF Preregistration with detailed analysis plan)

• Document Experimental Workflow: Create a detailed methodology section covering:

  • Materials specifications (composition, source, lot numbers)
  • Processing parameters (temperature, pressure, time, environment)
  • Characterization techniques (equipment, settings, standards)
  • Testing conditions (load rates, environments, replicates)

Detailed Template Application Protocols

Protocol for OSF Preregistration Template

Application: Ideal for confirmatory studies with well-defined hypotheses in materials science.

Required Elements:

• Hypotheses: State specific, testable predictions about material behavior or properties
• Variables: Define independent, dependent, and controlled variables with measurement units
• Sample Characteristics: Specify material sources, preparation methods, and inclusion/exclusion criteria
• Experimental Design: Describe experimental groups, controls, and replication strategy
• Analysis Plan: Pre-specify statistical tests, software, significance thresholds, and data transformation rules

Materials Science Specific Considerations:

• Document material sourcing and characterization history
• Predefine failure modes and outlier criteria for mechanical testing
• Specify environmental conditions and their monitoring during experiments
• Plan for microstructural characterization correlation with properties

Implementation Workflow:

Protocol for Secondary Data Preregistration

Application: Analyzing existing materials data from publications, databases, or previous experiments.

Key Considerations:

• Justify how confirmation is maintained despite data existence [2]
• Document data source, extraction methods, and any preprocessing
• Specify analysis plan without reference to observed outcomes
• Disclose all prior knowledge of the dataset

Procedure:

• Data Source Documentation: Record database name, version, access date, and extraction parameters
• Variable Definition: Map raw data to analysis variables with transformation rules
• Analysis Preservation: Define statistical models before conducting analysis
• Sensitivity Analysis: Plan alternative approaches to test robustness of findings

Quantitative Data Management Framework

Effective preregistration requires careful planning for quantitative data handling. The quality assurance process for quantitative data in materials science involves systematic procedures to ensure accuracy, consistency, and reliability throughout the research process [32].

Data Quality Assurance Protocol

Objective: Establish rigorous data management procedures before data collection.

Pre-Collection Planning:

• Define all variables, measurement units, and precision levels
• Specify data recording formats and storage protocols
• Plan for backup procedures and version control

Data Cleaning Procedures:

• Check for duplications: Identify and remove identical copies of data, leaving only unique participant data [32]
• Handle missing data: Establish thresholds for inclusion/exclusion (e.g., 50-100% completeness) and report removal criteria [32]
• Identify anomalies: Run descriptive statistics to detect values outside expected ranges (e.g., Likert scale boundaries) [32]

Quality Control Measures:

• Instrument calibration documentation
• Inter-rater reliability assessments for subjective measurements
• Batch effect controls for multi-session experiments

Statistical Analysis Planning

Protocol for Analysis Specification:

• Normality Assessment: Pre-specify tests for distribution (Kolmogorov-Smirnov, Shapiro-Wilk) and criteria for parametric vs. non-parametric tests [32]
• Descriptive Statistics: Plan frequency counts, measures of central tendency (mean, median, mode) and variability (standard deviation, range) [32]
• Inferential Statistics: Select appropriate tests based on study design and variable types [32]
• Psychometric Properties: For standardized instruments, plan reliability assessment (Cronbach's alpha > 0.7) and validity testing [32]

Essential Research Reagent Solutions

Materials science research requires careful documentation of materials and characterization tools. The following table details key resources and their functions in the preregistration context:

Resource Category	Specific Examples	Function in Research
Reference Databases	ASM Alloy Center [30], Phase Equilibria Diagrams Online [30]	Access materials property data, phase diagrams for hypothesis formation
Standards Repositories	ASTM Compass [30], ISO Standards	Reference standardized testing methodologies and protocols
Literature Databases	Compendex [28] [30], Scopus [30], Web of Science [28]	Conduct comprehensive literature reviews, identify knowledge gaps
Characterization Tools	JoVE Journal--Chemistry [30]	Access video protocols for experimental techniques
Data Analysis Software	Statistical packages (R, Python, specialized materials informatics tools)	Implement pre-specified analysis plans, computational models

Selecting the appropriate OSF preregistration template requires careful consideration of study design, data status, and disciplinary conventions. For materials scientists, the OSF Preregistration template typically provides the optimal balance of structure and flexibility for most experimental studies, while the Secondary Data Preregistration and Open-Ended Registration templates serve specialized needs. By following the protocols outlined in this guide and utilizing the provided decision framework, researchers can effectively document their study plans, enhance research transparency, and contribute to the growing movement toward open science in materials research.

Pre-registration, the practice of submitting a time-stamped, read-only research plan to a public registry before beginning a study, represents a cornerstone of open science and is increasingly critical for ensuring research credibility across scientific fields, including materials science [2] [33]. It establishes a transparent record of a researcher's intentions, creating a distinct, verifiable boundary between hypothesis-generating exploratory research and hypothesis-testing confirmatory research [2] [34]. This distinction is vital because it safeguards against problematic research practices such as p-hacking (selectively analyzing data to find statistically significant results) and HARKing (Hypothesizing After the Results are Known), thereby preserving the diagnostic value of statistical inferences [2] [34] [33]. For researchers in materials science and drug development, where experimental rigor and reproducibility are paramount, a well-executed pre-registration strengthens the validity of findings and provides a clear, defensible roadmap for the research journey. The central aim of this protocol is to delineate the essential components of a powerful pre-registration, with a focused examination of formulating hypotheses, justifying sample size, and detailing the analysis plan.

The Core Components of a Pre-registration

A robust pre-registration transforms a general research idea into a specific, testable, and transparent plan. The following three elements form the foundation of any strong pre-registration, ensuring that the research is held to the highest standards of rigor [2].

Hypotheses and Research Questions

The pre-registration must articulate clear, specific, and measurable hypotheses. This moves beyond a vague research idea to a precise prediction that can be rigorously tested.

• Confirmatory vs. Exploratory Research: Pre-registration primarily concerns confirmatory research, where a specific hypothesis is tested. This is distinct from exploratory research, which is aimed at generating new hypotheses [2] [34]. A pre-registration must specify at least one confirmatory test [2]. While exploratory analyses are valuable for discovery, the pre-registration ensures they are not later presented as confirmatory, a practice known as HARKing that inflates false positive rates [34].
• Specificity and Falsifiability: Each hypothesis should be stated in a falsifiable manner, clearly identifying the variables involved and the expected direction of the effect. For example, a pre-registration in materials science should not state "We will study the effect of doping on catalyst performance," but rather "We hypothesize that doping Catalyst A with Element B at 5 at.% will increase its catalytic efficiency for reaction C by at least 15% under standard conditions T and P."

Sample Size and Data Collection Plan

A transparent plan for data collection is crucial for assessing the statistical power and reliability of the study.

• Sample Size Justification: The pre-registration must detail the planned number of observations or experimental runs and provide a justification for this number [33] [35]. The gold standard for this is an a priori power analysis, which calculates the sample size required to detect a predetermined effect size with a given level of statistical power (typically 80%) [35]. This prevents the problematic practice of collecting data until a significant result is found.
• Stopping Rules and Exclusion Criteria: The plan should pre-specify rules for stopping data collection. Furthermore, it must explicitly state the criteria for excluding data points from analysis (e.g., outliers defined by a specific statistical test, failed experimental controls) before any data are collected or viewed [4] [33].

Analysis Plan

This is the most detailed component of the pre-registration, serving as a blueprint for how the data will be transformed into results.

• Statistical Tests and Models: The plan must specify the exact statistical tests or models that will be used to test each pre-registered hypothesis (e.g., "a two-tailed t-test," "a one-way ANOVA," "a linear regression model with covariates X, Y, and Z") [4] [33].
• Variable Definition and Data Handling: Precisely define how key variables will be measured, transformed, or combined. For instance, specify if a composite score will be created from multiple measurements and how that will be calculated [4].
• Decision Trees for Contingencies: A strong analysis plan anticipates potential scenarios and outlines "if-then" decision trees. For example, "if the data violate the assumption of normality, we will use a Mann-Whitney U test instead of a t-test" [4]. This demonstrates thorough planning and reduces analytical flexibility.

Table 1: Core Components of a Pre-registration and Their Key Elements

Component	Key Elements to Specify	Common Pitfalls to Avoid
Hypotheses	Specific, directional predictions; Primary and secondary hypotheses; Clear identification of variables.	Vague statements; Failing to distinguish confirmatory from exploratory hypotheses.
Sample Size	Total N; Justification (e.g., power analysis); Data collection stopping rule; Participant/sample exclusion criteria.	Arbitrary N; No justification; Flexible stopping rules.
Analysis Plan	Exact statistical tests for each hypothesis; Definition of outcome variables; Handling of outliers and missing data; Model specifications and covariates.	Stating "appropriate tests will be used"; Not defining analysis outcomes; No plan for protocol deviations.

Experimental Protocol for Pre-registration

The following workflow provides a step-by-step methodology for creating and submitting a pre-registration, adaptable for various research domains.

The diagram below outlines the logical sequence for developing a robust pre-registration.

Step-by-Step Procedure

• Step 1: Select a Pre-registration Template

  • Action: Choose a standardized template that fits your research design. The OSF Preregistration template is an excellent, comprehensive starting point for first-time users and general-purpose studies [33]. For specific study types, consider specialized templates such as the Preregistration for Studies with Existing Data or the Replication Recipe template [33].
  • Rationale: Using a template ensures you address all necessary components and improves the consistency and completeness of your registration [4] [33].

• Step 2: Draft Hypotheses and Research Questions

  • Action: Clearly articulate your primary and secondary hypotheses. Frame them as specific, testable statements. Use this section to define the scope of your confirmatory research and acknowledge areas you plan to explore Exploratory analyses are valid and important, but must be identified as such in the pre-registration [2] [34].
  • Rationale: This step formally stakes a claim to your research ideas and forces critical thinking about the study's theoretical foundation, reducing the temptation for HARKing [2] [34].

• Step 3: Plan Sample Size and Data Collection

  • Action: Perform an a priori power analysis to determine the required sample size. Document the effect size used, the desired power, and the alpha level. Pre-specify rules for when data collection will end and under what conditions data points will be excluded.
  • Rationale: Justifying the sample size mitigates bias from arbitrary data collection decisions and protects against p-hacking by eliminating flexibility in the sample [34] [35].

• Step 4: Detail the Analysis Plan

  • Action: For each hypothesis, specify the exact statistical model, including the dependent and independent variables, any covariates, and the type of test. Describe how you will handle data transformations, outliers, and missing data. Create "if-then" decision trees for potential analysis contingencies [4].
  • Rationale: A highly specific analysis plan minimizes "researcher degrees of freedom," where arbitrary analytical choices can inflate the false positive rate. It also demonstrates rigorous forethought [4] [34].

• Step 5: Finalize and Submit

  • Action: Before finalizing, review the entire pre-registration to ensure all sections are complete and precise. Once satisfied, submit the registration to a public repository like the Open Science Framework (OSF) to create a time-stamped, immutable record [2] [4].
  • Rationale: Submission creates a public, verifiable record of your research plan, which "future-proofs" your research by establishing the chronology of your ideas and analysis plan relative to data collection and access [2].

Successfully navigating the pre-registration process requires leveraging available tools and resources. The table below catalogs essential "research reagents" for this task.

Table 2: Key Resources for Effective Pre-registration

Resource Name	Type	Primary Function	Relevance to Materials Science/Research
OSF Preregistration	Template	Comprehensive form for general research; most commonly used.	Ideal for most experimental studies; ensures all critical components are addressed. [4] [33]
AsPredicted.org	Template	Streamlined template with 8 essential questions.	Good for simpler studies or when a concise plan is sufficient. [33]
Secondary Data Preregistration	Template	Template for studies using existing datasets.	Crucial for ensuring confirmatory analysis when using public or pre-existing data. [33]
SPIRIT 2025 Guidelines	Reporting Guideline	Evidence-based checklist for clinical trial protocols.	Provides a gold-standard reference for rigorous protocol design, especially in applied/drug development contexts. [36]
Open Science Framework (OSF)	Registry Platform	Public repository to submit and store pre-registrations.	Creates the time-stamped, public record essential for the pre-registration's validity. [2] [4] [13]

A strong pre-registration is not a restrictive straitjacket but a strategic foundation for credible and transparent science. By meticulously detailing the hypotheses, sample size plan, and analysis strategy before data collection or analysis, researchers in materials science and drug development can significantly enhance the rigor and persuasiveness of their findings. This practice distinguishes planned confirmatory tests from unplanned exploratory analyses, protects against pervasive cognitive biases, and ultimately increases the probability that research contributions will be recognized as robust and reliable [2] [34]. The components and protocols outlined herein provide a actionable framework for integrating this powerful open science practice into the research lifecycle, fostering a culture of reproducibility and trust.

Within the context of pre-registering materials science studies, selecting the appropriate starting point for an Open Science Framework (OSF) project is a critical early-stage decision that influences collaborative efficiency, organizational clarity, and long-term project integrity. An OSF Project acts as a flexible, collaborative workspace for planning, managing, and sharing every part of the research lifecycle [37]. Researchers are primarily faced with two pathways: creating a project from an entirely blank slate or utilizing an existing project as a structural template. This application note delineates these two workflow options, providing a structured comparison and detailed protocols to guide researchers, scientists, and drug development professionals in making an informed choice that aligns with their pre-registration goals in materials science.

Comparative Analysis: Scratch vs. Existing Project

The decision between starting from scratch or from an existing template involves trade-offs between customization, effort, and standardization. The quantitative and qualitative differences are summarized in the table below.

Table 1: Comparative Analysis of OSF Project Starting Workflows

Feature	Starting from Scratch	Using an Existing OSF Project
Development Time	Higher (Baseline: 100%)	Significantly Lower (Approx. 30-40% of baseline effort for a good template fit) [38]
Customization Flexibility	Complete freedom to design structure	Constrained by the template's existing structure; requires customization
Best-Suited Use Case	Highly unique studies with no established workflows	Studies that follow established, repeatable patterns or lab workflows
Organizational Overhead	Researcher must design all components and structure	Inherits a pre-defined, logical structure
Collaboration Onboarding	Can be slower without a familiar structure	Faster, especially within a lab using standardized templates [38]
Maintenance & Technical Debt	Isolated code/workflows; harder to improve systematically	Lower if maintained properly; allows for systematic improvements across projects [38]
Ideal for Pre-registration	When no pre-existing template matches the study design	When using a standardized pre-registration template (e.g., OSF Preregistration, AsPredicted) [4]

Workflow Visualization

The following diagram illustrates the key decision points and subsequent steps for choosing between the two workflow options, from project initiation to the active research phase.

Experimental Protocol

This section provides the step-by-step methodology for implementing the two primary workflow options.

Protocol 1: Starting an OSF Project from Scratch

This protocol is designed for creating a new project without a pre-existing template, offering maximum flexibility [37].

4.1.1 Research Reagent Solutions

Table 2: Essential Materials for OSF Project Creation

Item Name	Function
OSF Dashboard	The primary interface for accessing and creating new projects [37].
Project Title & Description	Provides a unique identifier and context for collaborators and the public.
Storage Location Selector	Determines the geographical location (e.g., U.S., Germany) of stored data to comply with institutional or privacy requirements like GDPR [37].
Components Feature	Creates sub-projects to organize different phases of research (e.g., "Literature," "Synthesis," "Characterization," "Analysis") for modular management [37].
Wiki Tool	Serves as an Electronic Lab Notebook (ELN) for documenting hypotheses, protocols, and daily logs within the project [39].

4.1.2 Step-by-Step Procedure

• Navigate to OSF Dashboard: Log in to your OSF account and access your main dashboard.
• Initiate Project Creation: Click the "Create new project" button.
• Define Project Identity:
  • Enter a descriptive Title for your project.
  • (Optional) Add a short Description outlining the project's scope.
  • Select a Storage Location (United States, Canada, Germany, or Australia) based on your project's data governance requirements [37].
• Create Project: Click "Create Project". You will be redirected to the new project's overview page.
• Establish Project Structure using Components:
  • In the "Components" section of the project page, click "Add Component".
  • For each component (e.g., "Experimental Protocol," "Raw Data," "Analysis Scripts," "Manuscript"), provide a name and click "Create".
  • This creates a nested, organized workspace where each component can have its own contributors, files, and privacy settings [37].
• Document the Project: Use the Wiki tool to create a project overview, document research questions, and outline standard operating procedures. This acts as the central documentation hub [37].
• Initiate Pre-registration:
  • Navigate to the "Registrations" tab in your project's left sidebar.
  • Click "Add a registration".
  • When prompted "Do you have content for registration in an existing OSF project?" select "No" to start the pre-registration from scratch [4].
  • Select the pre-registration template that best fits your study design (e.g., "OSF Preregistration") and proceed with completing the form [4].

Protocol 2: Starting an OSF Project from an Existing Project

This protocol leverages a pre-existing project structure to save time and standardize workflows across a research group [37].

4.2.1 Research Reagent Solutions

Table 3: Essential Materials for Templating an OSF Project

Item Name	Function
Source Project (Template)	A well-structured, existing OSF project (from your lab or public) that serves as a model.
"Duplicate" Icon	The interface element used to copy the structure of a public project.
Template Library	A curated list of public OSF projects designed for reuse (e.g., Lab Manager Research Group, Electronic Lab Notebook, Data Management Template) [37].
Contributor Management	Interface to add team members with specific roles (Read, Write, Admin) and bibliographic credits [37].
Storage Add-ons	Integrations with third-party storage providers (e.g., Google Drive, GitHub, Dropbox) to centralize resources without creating data silos [39].

4.2.2 Step-by-Step Procedure

• Identify a Template Project: Locate a suitable project to use as a template. This can be:
  • A public template from the OSF community (e.g., "Electronic Lab Notebook" or "Lab Manager Research Group") [37].
  • One of your own existing projects that has a desirable structure.
• Duplicate the Project Structure:
  • Navigate to the overview page of the chosen template project.
  • In the top-right corner, click the "Duplicate" icon (two overlapping squares).
  • In the dialog box, select "Duplicate project (structure only)" [37]. This ensures you copy the component hierarchy and wiki structure without copying any of the actual files or data.
  • Click "Create" to generate your new project. The title will be prefixed with "Templated from...".
• Customize the New Project:
  • Rename the Project: Change the project title to reflect your new study.
  • Modify Components: Add, remove, or rename components within the inherited structure to perfectly fit your new research plan.
  • Update Documentation: Revise the Wiki pages to describe the new study's context, hypotheses, and protocols.
• Configure Collaborators and Storage:
  • Add Contributors via the "Contributors" section in the left navigation menu, assigning appropriate permissions [37].
  • Connect necessary Storage Add-ons (e.g., GitHub for code, Google Drive for large datasets) under the "Files" section for each component [39].
• Initiate Pre-registration from the Project:
  • From your new project's page, go to the "Registrations" tab and click "Add a registration".
  • Since you are starting from an existing project, "Yes" will be pre-selected for the question "Do you have content for registration in an existing OSF project?" [4].
  • Your current project will be pre-filled. Select your desired pre-registration template and click "Create Draft" to begin. Note that files associated with the project (up to 5 GB) will be attached to this registration [4].

In the rigorous field of materials science research, the timing of pre-registration is a critical determinant of a study's credibility. Pre-registration is the practice of publicly depositing a time-stamped, read-only version of a research plan before beginning data collection or analysis [4]. This establishes a transparent, unchangeable record of the researcher's initial intentions, creating a formal "story" of the study that describes planned procedures, any necessary updates, and eventual results [4]. The central purpose of this practice is to clearly distinguish between confirmatory (hypothesis-testing) and exploratory (hypothesis-generating) research, thereby safeguarding the diagnostic value of statistical inferences [2].

The strategic importance of pre-registration timing cannot be overstated. When executed at the appropriate research stage, it stakes a clear claim to research ideas, constrains researcher degrees of freedom that could inadvertently bias outcomes, and ultimately enhances the credibility of reported findings [40]. For materials scientists engaged in drug development and advanced materials characterization, proper timing ensures that analytical decisions remain independent of observed results, thus controlling for false positives and preserving the statistical integrity of confirmatory tests [2].

Optimal Timing Frameworks for Pre-registration

The optimal timing for pre-registration depends on the nature of the data source and the point in the research lifecycle when the plan is finalized. The following scenarios represent the most common and methodologically sound timelines for pre-registration relative to data collection and access.

Primary Data Collection Scenarios

For research involving new data generation, the pre-registration timeline must precede any data observation or analysis.

Table 1: Pre-registration Timeline for Primary Data Collection

Research Stage	Pre-registration Status	Key Considerations
Study Design Finalized	Draft pre-registration begins	Research questions, hypotheses, and methodological details are specified [40]
Before Data Collection	Pre-registration submitted and finalized	Certification that no data exist yet [2]
Data Collection	Pre-registration is immutable	Plan is locked; deviations must be transparently documented [4]
Data Analysis	Comparison with pre-registered plan	Distinction between confirmatory and exploratory analyses [2]

The most straightforward and methodologically robust scenario occurs when pre-registration is completed before any data have been collected, created, or realized [2]. In this situation, researchers must certify that the data do not exist at the time of pre-registration submission, ensuring complete independence of the research plan from knowledge of results. This approach is ideal for experimental materials science studies involving novel synthesis techniques, characterization of new compounds, or development of advanced materials for drug delivery systems.

Existing Data Set Scenarios

Using existing data for confirmatory research introduces specific timing considerations to maintain methodological rigor.

Table 2: Pre-registration Timeline for Existing Data

Data Access Status	Pre-registration Deadline	Required Certifications
Data unobserved by anyone	Before any human observation	Certification that data exist but haven't been quantified, constructed, or observed [2]
Data accessible to others	Before researcher access	Justification of how confirmatory nature is preserved despite others' access [2]
Data accessed but unanalyzed	Before analysis related to research plan	Explanation of prior analysis and justification for confirmatory status [2]
Split-sample analysis	Before analyzing held-out validation sample	Clear separation between exploratory and confirmatory datasets [2]

For existing datasets, the "gold standard" approach involves split-sample analysis, where incoming data is divided into exploration and confirmation sets [2]. Researchers can explore the first dataset for unexpected trends, pre-register the most tantalizing findings, then confirm these findings with the second dataset that had been held in reserve. This method is particularly valuable for materials researchers working with large, computationally-generated datasets or high-throughput experimental results.

Experimental Protocol for Timely Pre-registration

Workflow for Ideal Pre-registration Timing

The following workflow diagram outlines the sequential steps for pre-registering a materials science study at the optimal timeframe relative to data collection:

Protocol Details

The pre-registration protocol requires careful planning and execution to ensure methodological integrity:

• Study Design Finalization: Develop precise research questions and hypotheses. Specify all materials synthesis protocols, characterization methods, and analytical procedures. For drug development studies, this includes exact compound specifications, experimental conditions, and outcome measures [40].

• Pre-registration Drafting: Select an appropriate template from repositories like the Open Science Framework (OSF). The OSF Preregistration template is commonly used for general materials science studies, while specialized templates exist for secondary data analysis or qualitative research [4]. Complete all required sections with specific, detailed responses that constrain researcher degrees of freedom [40].

• Pre-registration Submission: Submit the finalized pre-registration to a public repository before data collection begins. For studies requiring blinding, utilize embargo features (up to 4 years maximum) while creating anonymized view-only links for peer review [4].

• Data Collection and Analysis: Execute the research exactly as pre-registered. Maintain clear documentation of any unforeseen circumstances or necessary deviations from the original plan [40].

• Transparent Reporting: When publishing results, include a link to the pre-registration and comprehensively document all deviations from the initial plan using a "Transparent Changes" document [2].

Research Reagent Solutions

Table 3: Essential Resources for Pre-registration in Materials Science

Resource	Function	Application Context
OSF Preregistration Template	Standard, comprehensive form for study planning	General materials science research; most commonly used template [4]
AsPredicted Template	Simplified 8-question pre-registration form	Rapid study registration with essential elements [4]
Secondary Data Preregistration Template	Specialized form for existing dataset analysis	Studies utilizing previously collected materials characterization data [4]
Registered Report Format	Two-stage peer review before data collection	High-impact confirmatory studies seeking results-blind review [40]
Open Science Framework (OSF)	Public repository for pre-registrations	Time-stamped, read-only storage of research plans [4]

Implementation Workflow for Different Scenarios

The appropriate workflow for pre-registration depends on whether researchers are working with new or existing data, as illustrated below:

Properly timed pre-registration represents a fundamental shift toward enhanced methodological rigor in materials science research. By creating an immutable record of research plans before data collection or access, scientists can preserve the diagnostic value of statistical tests, distinguish confirmatory from exploratory findings, and enhance the credibility of their conclusions. The protocols outlined provide a practical framework for implementing this practice across diverse research scenarios, ultimately contributing to more reproducible and reliable materials science innovation. As federal funding agencies increasingly mandate open science practices [41], mastering the timing of pre-registration becomes essential for researchers pursuing impactful work in drug development and advanced materials design.

For researchers in materials science and drug development, preregistration is a powerful tool to enhance research rigor and credibility by specifying the research plan, including hypotheses, methodology, and analysis strategy, before a study is conducted [2] [42]. A common misconception is that a preregistration is merely a bureaucratic hurdle. However, a well-constructed preregistration serves as a detailed blueprint, facilitating a more efficient, transparent, and defensible manuscript writing process. This protocol provides detailed guidance on leveraging your preregistration document to systematically draft key sections of your future manuscript, ensuring alignment with your initial plans and clearly distinguishing confirmatory from exploratory research [2] [43].

The table below summarizes the core quantitative and descriptive elements of a preregistration and their direct application in manuscript drafting.

Table 1: Translating Preregistration Elements into Manuscript Sections

Preregistration Element	Description & Data Type	Primary Manuscript Section	Application Notes for Drafting
Hypotheses	Specific, testable predictions (Qualitative Text)	Introduction, Results, Discussion	Use the exact wording from the preregistration in the introduction to state what you planned to test. In results and discussion, refer back to these specific statements.
Study Variables	Primary/Secondary Outcomes (Categorical/Nominal)	Methods, Results	Clearly define which variables were designated as primary and secondary in the methods. Report results for all preregistered variables.
Sample Size & Power	Target sample size (Integer), Justification (Text)	Methods	Detail the planned sample size and the power analysis (e.g., effect size, alpha) used to determine it [43]. Report final sample size against this target.
Experimental Protocol	Step-by-step methodology (Text/Flowchart)	Methods	The preregistered protocol forms the core of the methods section. Document any deviations transparently in a "Deviations from Protocol" subsection.
Analysis Plan	Statistical tests, software, alpha level (Text)	Methods, Results	Specify the pre-planned tests in the methods. Use the results section to report the outcomes of these tests, avoiding post-hoc justification of analytical choices.
Data Handling	Criteria for outlier exclusion, missing data treatment (Text)	Methods	Describe these pre-specified criteria in the methods to justify the data processing steps taken, reducing concerns about selective reporting.

Experimental Protocol: From Preregistration to Manuscript Drafting

This protocol outlines the systematic process of using a completed preregistration to draft a research manuscript in materials science and drug development.

Reagents and Materials Solutions

Table 2: Essential Research Reagent Solutions for Preregistered Studies

Item	Function/Explanation
Pre-registration Template (e.g., from OSF/COS)	Provides a structured framework to detail hypotheses, variables, and analysis plans, creating the foundational document for the future manuscript [2] [42].
Data Management Plan	Specifies protocols for data collection, storage, and sharing (e.g., on repositories like OSF), ensuring data used in the manuscript aligns with preregistered plans for transparency and verification [43].
Analysis Scripts (e.g., R, Python)	Pre-written or outlined code for planned statistical tests. Using these scripts on the collected data ensures the analysis reported in the manuscript faithfully follows the preregistered plan.
Transparent Changes Document	A living document to record any deviations from the preregistered protocol. This is critical for the manuscript's methods section to maintain rigor and honesty when reporting unavoidable changes [2].

Step-by-Step Methodology

Step 1: Initialize the Manuscript Shell Immediately after finalizing your preregistration, create a new document with the headings for a standard research article (Abstract, Introduction, Methods, Results, Discussion, References). Directly copy the relevant text from your preregistration into the respective sections:

• Introduction: Paste your a priori hypotheses and the theoretical framework that justified them [43].
• Methods: Populate the sections on experimental design, materials synthesis/preparation protocols, characterization methods, sample size justification, and data analysis plan.

Step 2: Draft the Methods Section with Fidelity Use the preregistered methods as the core narrative. The goal is to demonstrate that the conducted study adhered to the planned protocol.

• Detail Experimental Workflows: Describe materials synthesis, processing, and characterization steps (e.g., XRD, SEM, DSC protocols) as predefined.
• Specify Data Collection: List all equipment, settings, and measured variables exactly as planned.
• Document the Analysis Plan: State the statistical tests, software, and significance thresholds (e.g., p < 0.05) that were pre-specified for each hypothesis [2].

Step 3: Report Results with Direct Reference to the Plan This section should be a straightforward reporting of the outcomes from the pre-planned analyses.

• Structure by Hypothesis: For each preregistered hypothesis, report the corresponding statistical results (test statistic, p-value, effect size).
• Report All Pre-registered Outcomes: You must report the results of all analyses specified in the preregistration, not just the statistically significant ones [2]. This prevents cherry-picking and publication bias.
• Use Clear Labeling: Clearly distinguish between results from confirmatory tests (preregistered) and those from exploratory analyses (unplanned) [2] [35].

Step 4: Compose the Discussion with a Planned versus Executed Lens Frame the discussion around the initial predictions and the observed results.

• Interpret All Results: Discuss both the supported and unsupported preregistered hypotheses. If many tests were planned, avoid selectively interpreting only the significant results, as this can be misleading [2].
• Address Deviations Transparently: If changes to the protocol were necessary, describe them in the discussion or a dedicated "Transparent Changes" section and explain the rationale [2]. This does not invalidate the study but enhances its credibility.
• Contextualize Exploratory Findings: Clearly identify any unplanned, post-hoc analyses as exploratory and frame them as hypothesis-generating for future research, requiring confirmation [2] [43].

Workflow Visualization

The following diagram illustrates the logical workflow for transforming a preregistration into a complete manuscript.

Diagram 1: Manuscript Drafting Workflow

The diagram above outlines the sequential process. For a more granular view of handling analytical outcomes, the following decision tree is applied during the results and discussion writing phases.

Diagram 2: Results Interpretation Logic

Navigating Challenges and Optimizing Your Pre-registration Strategy

Application Note on Pre-Registration Concerns

Table 1: Common Preregistration Concerns and Evidence-Based Counterpoints

Common Concern	Empirical Evidence & Prevalence	Proposed Mitigation Strategy
Rigidity - Inability to adapt research plan	Distinguishes confirmatory/exploratory analyses; exploratory work remains vital [2].	Use "if-then" decision trees for anticipated contingencies [4]. Create a "Transparent Changes" document for deviations [2].
Scooping - Theft of research ideas	Timestamped preregistration stakes claim to ideas and methods early [2].	Utilize embargo features (up to 4 years) to maintain privacy during peer review [4].
Time Investment - Burden on researchers	Preregistration functions as a draft for journal method/results sections, streamlining later writing [4].	Leverage standardized templates (e.g., OSF Preregistration) to structure and speed up the process [4].

Conceptual Workflow for Addressing Concerns

Diagram: Strategic Response Pathway to Common Pre-Registration Concerns

Experimental Protocols

Protocol for Preregistration with Existing Data

Objective: To conduct a confirmatory analysis using an existing dataset without compromising the validity of statistical inferences.

Methodology:

• Eligibility Certification: The lead researcher must formally certify the level of prior interaction with the data [2]. Categories include:
  • Prior to Analysis: Data exists and has been accessed, but no analysis related to the current Research Plan has been conducted.
  • Prior to Access: Data exists but has not been accessed by the entrant or study collaborators.
• Justification & Disclosure: Document and justify how any prior observation, analysis, or reporting of the data avoids compromising the confirmatory nature of the Research Plan [2].
• Registration: Select and complete the appropriate preregistration template (e.g., "Secondary Data Preregistration" on OSF) [4].
• Transparent Reporting: In the final manuscript, disclose the data's prior status and all planned vs. unplanned analyses clearly [2].

Protocol for Split-Sample Validation

Objective: To combine exploratory hypothesis generation with rigorous confirmatory testing within a single dataset.

Methodology:

• Data Partitioning: Randomly split the incoming dataset into two distinct parts immediately upon collection [2]:
  • Exploration Set: Dedicated for model training and uncovering unexpected trends.
  • Validation Set: Held entirely off-limits for initial exploration.
• Exploratory Phase: Analyze the Exploration Set to generate tantalizing hypotheses or refine model structures.
• Preregistration: Formally preregister the specific hypotheses and analysis plan derived from the exploratory phase using the "OSF Preregistration" template [4].
• Confirmatory Testing: Test the preregistered hypotheses exclusively on the untouched Validation Set [2].
• Reporting: Clearly report the split-sample method, the preregistration, and which results emerged from each phase.

Workflow for Iterative Preregistration

Diagram: Managing Changes to a Preregistered Study Plan

The Scientist's Toolkit

Research Reagent Solutions for Preregistration

Table 2: Essential Resources for Effective Study Preregistration

Tool / Reagent	Function	Key Features / Considerations
OSF Preregistration Template	Standard, comprehensive form for drafting the research plan [4].	Most commonly used; helps think through methods and results reporting [4].
Secondary Data Preregistration Template	Form tailored for projects using existing datasets [4].	Requires explicit certification about prior data access and observation [2].
Embargo Feature	Keeps a registration private for a defined period (max 4 years) [4].	Mitigates scooping concerns; allows for blinded peer review [4].
Transparent Changes Document	A living document to record deviations from the original plan [2].	Upload to OSF project; cited in final manuscript to maintain transparency [2].
Generalized Systematic Review Template	For registering protocols for systematic reviews, scoping reviews, and meta-analyses [13].	Designed for cross-disciplinary use in evidence synthesis [13].
Split-Sample Dataset	A dataset randomly divided for exploration and confirmation [2].	"Model training" and "validation" are analogous terms; preserves confirmatory power [2].

Preregistration is the practice of specifying a research plan in advance of a study and submitting it to a registry, creating a timestamped blueprint that distinguishes pre-existing hypotheses from exploratory findings [2]. In materials science research, where experimental conditions often require real-time adjustment, it is crucial to frame preregistration not as an inflexible constraint, but as a dynamic plan that documents initial intent while accommodating the iterative nature of scientific discovery. This approach enhances the credibility of your results by clearly separating confirmatory hypothesis testing from exploratory data analysis, ultimately future-proofing your research against claims of data dredging or post-hoc narrative building [2] [44].

The concept of a "Transparent Changes" document is central to this flexible approach. It is a mechanism that allows researchers to openly document and justify any deviations from the original preregistered plan that occur after a study has begun [2]. This practice maintains the transparency and rigor that preregistration is designed to introduce, while acknowledging that research, particularly in experimental fields like materials science, is rarely a linear process. Properly handled, deviations do not weaken your study; instead, a transparent record of changes strengthens the validity of your conclusions and provides a true account of the research pathway.

The Protocol for Handling Deviations

When your materials science research deviates from its preregistered path—whether due to equipment limitations, unforeseen material properties, or optimized synthesis conditions—a systematic protocol ensures maintained rigor.

Decision Workflow for Managing Deviations

The following workflow outlines the critical decision points when a deviation from the preregistration occurs:

When to Use a Transparent Changes Document

Initiate a Transparent Changes document when modifications to the preregistered plan occur after data collection has begun [2]. This is the most common scenario in active research and includes situations such as:

• Adaptations in material synthesis: Changes to reaction times, temperatures, or precursor concentrations that yield improved or necessary results.
• Analytical technique substitutions: Replacing a characterization method (e.g., switching from SEM to TEM for superior resolution) due to availability or technical constraints.
• Sample size adjustments: Modifying the number of experimental replicates based on interim results or resource availability.
• Statistical analysis modifications: Altering data analysis plans to better address the research question with the collected data.

The Transparent Changes document should be uploaded to the same Open Science Framework (OSF) project from which the original registration was created [2]. This maintains a direct and public link between the original plan, the documented changes, and the final research outcomes.

Creating and Implementing a Transparent Changes Document

A Transparent Changes document is a living record that accompanies your research from the first deviation through to publication. Its effective implementation relies on both comprehensive content and a consistent workflow.

Essential Components of the Document

A robust Transparent Changes document should systematically capture the following information for every deviation:

• Description of the Change: A clear, precise explanation of what aspect of the preregistered plan was altered.
• Rationale for the Deviation: The scientific justification for the change, referencing unexpected experimental results, technical limitations, or opportunities for optimization.
• Date of Implementation: When the change was introduced in the research timeline.
• Impact Assessment: A statement on how the change affects the original hypotheses, analytical approach, or interpretation potential.

Table: Template for Documenting Individual Changes

Preregistered Plan Element	Deviation Implemented	Scientific Justification	Date
Original material synthesis parameter	Actual parameter used	Reason for change (e.g., safety, yield, availability)	YYYY-MM-DD
Planned characterization technique	Technique actually employed	Justification (e.g., resolution, data quality, access)	YYYY-MM-DD
Initial statistical model	Final model used	Reason (e.g., model fit, data structure)	YYYY-MM-DD

Workflow for Maintaining the Document

Integrating the Transparent Changes document into your research lifecycle ensures it remains an accurate and useful record.

Practical Solutions for Materials Science Research

Materials science research involves specific challenges and reagents that often require deviations from initial plans. The following table outlines common scenarios and solutions for maintaining transparency.

Research Reagent Solutions and Common Deviations

Table: Common Materials Science Scenarios and Transparency Solutions

Research Context	Potential Deviation	Transparent Solution	Key Reagents/Techniques
Novel Material Synthesis	Altering precursor ratios or reaction conditions from planned parameters.	Document the optimized conditions in the Transparent Changes document, justifying how they improved yield, purity, or material properties.	Precursors, Solvents, Catalysts, Autoclaves
Process Optimization	Changing sintering temperatures/times or deposition parameters from the original plan.	Justify changes based on interim characterization data (e.g., XRD, SEM) that indicated the need for parameter adjustment.	Furnaces, Sputtering/ALD Systems, Process Gases
Characterization & Analysis	Substituting or adding analytical techniques (e.g., using TEM instead of planned SEM).	Explain why the alternative technique was necessary or provided superior data, referencing initial results.	SEM, TEM, XRD, XPS, AFM
Data Processing	Applying different data smoothing algorithms or model fitting approaches than preregistered.	Detail the original and final methods, providing a statistical or visual rationale for the change (e.g., improved fit).	Analysis Software (e.g., ImageJ, Origin, MATLAB)

Quantitative Data Presentation on Preregistration

The movement toward preregistration and research transparency is driven by empirical evidence of its benefits. The following table summarizes key quantitative findings from meta-research.

Table: Impact of Preregistration and Transparency on Research Practices

Metric	Without Preregistration/Transparency	With Preregistration/Transparency	Source/Context
Reported Significant Results	96% of biomedical papers using P-values claimed significant results (1990-2015) [44]	Increased tolerance for "negative" results, especially in clinical trials [44]	Biomedical Literature
Reproducibility of Landmark Studies	Only 6 of 53 (∼11%) landmark oncology studies were reproducible [44]	N/A - Highlights the crisis preregistration aims to address	Amgen Replication Effort
False Positive Economics Results	~70% of significant results would not be significant in a bias-free world [44]	Preregistration minimizes false positives by distinguishing confirmatory/exploratory work [2]	Economics Literature
Data Availability	Raw data and algorithms often not shared; "stealth research" mode exists [44]	Public preregistration timestamps ideas; OSF links plans, changes, and data [2] [4]	Scientific Ecosystem Analysis

In materials science, where discovery is often nonlinear and iterative, preregistration must be embraced as a planning tool, not a punitive constraint. The use of a Transparent Changes document transforms unavoidable deviations from potential liabilities into documented, justified, and scientifically valid components of the research pathway. By systematically implementing the protocols outlined in this article—using the decision workflow, maintaining a detailed changes document, and leveraging discipline-specific solutions—researchers can enhance the credibility, reproducibility, and overall impact of their work, fostering a culture of openness and rigor in materials science.

In the pursuit of scientific discovery, researchers in materials science and drug development face intense pressure to produce novel, statistically significant results. This environment can inadvertently encourage two detrimental research practices: Hypothesizing After the Results are Known (HARKing) and selective reporting. HARKing occurs when researchers present post-hoc hypotheses as if they were developed a priori, while selective reporting involves cherry-picking statistically significant results from a larger dataset without acknowledging the full analysis [34]. These practices, often termed "questionable research practices," undermine the credibility of scientific findings by dramatically increasing the probability of false-positive results [34]. Within materials science, where accurate data extraction and reporting are fundamental to developing reliable structure-property relationships, these practices can lead to misplaced confidence in material performance, inefficient allocation of research resources, and ultimately, failures in downstream applications including drug development.

Pre-registration emerges as a powerful, low-cost intervention to mitigate these issues. Pre-registration involves specifying a research plan—including hypotheses, experimental design, sample size, and planned statistical analyses—in a time-stamped, immutable document before data collection begins [2] [34]. This simple act creates a clear distinction between confirmatory research (rigorous hypothesis testing) and exploratory research (hypothesis generation), allowing each to be valued appropriately while preventing the conflation that leads to false positives [2]. For materials scientists, adopting pre-registration enhances the credibility and reproducibility of research, which is paramount when research outcomes inform critical decisions in drug development pipelines.

Understanding HARKing and Selective Reporting

Definitions and Impact

• HARKing (Hypothesizing After the Results are Known): This practice involves presenting a post-hoc hypothesis (developed after seeing the data) as an a priori prediction (made before data collection) [34]. It misrepresents the exploratory process of data analysis as a confirmatory test, misleading readers about the true diagnostic value of the statistical evidence.
• Selective Reporting: Also known as "p-hacking" or "cherry-picking," this entails selectively reporting statistically significant analyses while omitting non-significant or contradictory results [34]. This creates a biased representation of the evidence, making effects appear more robust and reliable than they truly are.

Both practices stem from a research culture that has historically prioritized novel, positive findings over rigorous, reproducible science [34]. The consequences are severe: they distort the scientific literature, lead to false positives that fail to replicate, and ultimately impede cumulative scientific progress. In fields like materials science and drug development, where research findings guide significant financial investment and clinical decisions, the real-world costs of these practices can be extraordinarily high.

Quantitative Evidence of the Problem

Table 1: Survey Findings on Researcher Perceptions and Practices

Aspect	Finding	Source
Open Science Adoption	Pre-registration lags behind data and material sharing in psychology/OBHDP journals (2% vs 29% and 15%) [34].	Survey of published literature
Generational Adoption	Early-career researchers create more pre-registrations than senior colleagues (59% vs 30%) [34].	Researcher survey
Barrier: Freedom	41% of researchers believe pre-registration limits freedom to explore data [34].	Researcher survey
Barrier: Efficiency	60% of researchers believe pre-registration decreases research efficiency [34].	Researcher survey

The Pre-registration Solution

Core Principles and Benefits

Pre-registration functions as a safeguard for scientific integrity by enforcing transparency and pre-specification. Its core benefit is the clear demarcation it creates between confirmatory and exploratory analyses [2]. This distinction is vital because:

• Confirmatory research tests specific, pre-defined hypotheses. The results are held to the highest standards of rigor, and the diagnostic value of statistical tests (like p-values) is preserved [2].
• Exploratory research is for generating new hypotheses from data. Its results are inherently more tentative and must be confirmed in future studies [2].

By registering a research plan, researchers commit to a specific analytical path before encountering the data, thereby reducing the temptation to engage in HARKing or p-hacking [45]. This increases the credibility of the reported confirmatory results [34]. Furthermore, pre-registration maximizes the impact of data by communicating the reliability of the research to the wider scientific community, fostering trust and facilitating collaboration [45].

Evidence of Efficacy

Empirical studies demonstrate the tangible benefits of pre-registration and related transparency practices. The implementation of selective reporting (SR) of antimicrobial susceptibility testing (AST) results in a clinical setting serves as a powerful, analogous case study. This intervention guided prescribers toward optimal antibiotics by deliberately withholding susceptibilities for certain drugs from the initial report.

Table 2: Impact of Selective Reporting on Antibiotic Prescribing

Outcome Measure	Pre-Implementation (N=50)	Post-Implementation (N=50)	P-value
Appropriate antibiotic within 24h of AST report	62%	86%	0.01 [46]
30-day mortality	Similar	Similar	Not Significant [46]
Clinical success	Similar	Similar	Not Significant [46]
Microbiological failure	Similar	Similar	Not Significant [46]

This study shows that a simple nudge toward better practices through structured reporting significantly improved appropriate prescribing without compromising patient outcomes [46]. This parallels the goal of pre-registration in materials science: to guide researchers toward more rigorous and truthful reporting without stifling scientific inquiry.

Protocols for Pre-registration in Materials Science

Workflow for Pre-registration and Data Extraction

The following diagram illustrates a robust workflow that integrates pre-registration with modern data extraction techniques, such as the ChatExtract method, which is particularly relevant for systematic literature reviews and database construction in materials science [47].

Pre-registration Protocol

This protocol provides a step-by-step guide for pre-registering a materials science study.

• Step 1: Develop the Research Plan. Before any experimentation or data analysis, document the core components of your study. Essential elements include:

  • Hypotheses: State clear, specific, and testable primary and secondary hypotheses.
  • Materials & Synthesis: Describe all materials to be used, their sources, purity, and detailed synthesis protocols to ensure reproducibility.
  • Experimental Design: Specify all experimental conditions, control groups, measurement techniques, and instruments.
  • Sample Size: Justify the number of samples or data points, including power analysis if applicable.
  • Dependent & Independent Variables: Define all variables and how they will be quantified.
  • Analysis Plan: Pre-specify all statistical tests, software, data normalization procedures, and criteria for handling outliers or data exclusions [2] [34].

• Step 2: Select a Registry and Submit. Choose a public, time-stamped registry like the Open Science Framework (OSF) or AsPredicted. Upload your research plan to create an immutable record. This stakes a claim to your idea and plan, protecting against idea "scooping" as registries provide timestamps and potential privacy options [45] [2].

• Step 3: Conduct Research and Analyze. Execute the study as pre-registered. Adhere to the planned analysis for all confirmatory tests. It is acceptable and valuable to conduct unplanned, exploratory analyses, but these must be clearly identified as such in the final report [2].

• Step 4: Report with Transparency. When publishing, report all pre-registered analyses, regardless of the outcome. Clearly distinguish between confirmatory tests of your pre-registered hypotheses and any exploratory analyses conducted. If deviations from the pre-registered plan were necessary, explain them transparently using a "transparent changes" document [2].

ChatExtract Protocol for Automated Data Extraction

For literature reviews and meta-analyses, the ChatExtract method provides a protocol for accurate, automated data extraction using conversational Large Language Models (LLMs) like GPT-4, achieving precision and recall rates near 90% for materials data [47]. This is crucial for building reliable databases free from selective reporting biases.

• Application: Building structured databases (Material, Value, Unit triplets) from unstructured text in research papers.
• Workflow:
  • Preparation: Gather PDFs of research papers, remove XML/html syntax, and divide text into sentences [47].
  • Initial Classification (Stage A): Use a simple LLM prompt to classify sentences as "relevant" (containing a value and unit for the target property) or "irrelevant" [47].
  • Context Expansion: For each relevant sentence, create a short passage including the paper's title, the preceding sentence, and the target sentence to capture material names [47].
  • Data Extraction (Stage B):
    • Single/Multiple Value Determination: Prompt the LLM to determine if the passage contains a single data point or multiple values. This dictates the subsequent path [47].
    • Single-Value Extraction: Use direct prompts to extract the Material, Value, and Unit. The prompt should explicitly allow for a "not specified" answer to discourage hallucination [47].
    • Multi-Value Extraction & Verification: This is higher risk. Use a series of engineered prompts that introduce redundancy and suggest uncertainty. Ask follow-up questions that force the model to re-analyze the text and confirm or deny its initial extractions. Enforce a strict Yes/No format for answers to simplify automation [47].

The Scientist's Toolkit

Table 3: Essential Resources for Pre-registration and Transparent Research

Resource / Tool	Type	Function & Application
Open Science Framework (OSF)	Registry Platform	A free, open-source platform for pre-registering study plans, sharing data, materials, and code [2].
AsPredicted	Registry Platform	A dedicated service for creating and storing pre-registrations with a standardized template.
ChatExtract Code	Data Extraction Tool	An automated workflow (e.g., Python script) using conversational LLMs for high-accuracy data extraction from literature [47].
Registered Reports	Publishing Format	A journal format where peer review of the introduction and methods occurs before data collection, mitigating publication bias [45].
Pre-registration Template	Documentation	A standardized form (available on OSF/AsPredicted) guiding researchers on what to include in a pre-registration [2].
Transparent Changes Document	Documentation	A living document used to track and explain any deviations from the pre-registered analysis plan after data collection has begun [2].

The adoption of pre-registration represents a fundamental shift toward a more rigorous, transparent, and cumulative materials science. By proactively implementing the protocols outlined here—from drafting a detailed pre-registration to utilizing advanced tools like ChatExtract for literature mining—researchers and drug development professionals can effectively mitigate the misuse of HARKing and selective reporting. This commitment to integrity enhances the credibility of individual studies and accelerates scientific progress by building a literature foundation that is robust, reliable, and truly reproducible.

Within the broader framework of pre-registering materials science studies, researchers often face a fundamental dilemma: the need for flexible, hypothesis-generating exploratory analysis while upholding the highest standards of confirmatory research. Exploratory research is crucial for discovery, aiming to minimize false negatives and find unexpected patterns, whereas confirmatory research rigorously tests specific predictions to minimize false positives [2]. In materials science, where data acquisition through experiments or high-fidelity computations is often costly and time-consuming, this dilemma is particularly acute, frequently resulting in limited "small data" sets [48].

The data-splitting strategy provides a powerful methodological solution to this challenge. It allows researchers to legitimately use the same dataset for both exploration and subsequent confirmation, thereby increasing the credibility of discovered findings. This approach involves partitioning a single dataset into distinct subsets dedicated to different phases of the research workflow [2]. By creating a clear separation between the data used to generate hypotheses and the data used to test them, this method mitigates the risk of capitalizing on chance associations and provides an iron-clad rationale for statistical tests, making it an essential tool for rigorous materials science research [2].

Conceptual Framework and Key Principles

The data-splitting strategy is underpinned by a clear distinction between exploratory and confirmatory research. Exploratory research is inherently data-dependent, serving the primary purpose of hypothesis generation. In this phase, researchers delve into the data to identify potential relationships, effects, or differences, prioritizing the avoidance of false negatives to ensure promising discoveries are not overlooked [2].

In contrast, confirmatory research is data-independent, with its plan being finalized before any data inspection occurs. It focuses on rigorously testing a specific, pre-defined hypothesis and is held to the highest statistical standards to minimize false positives, ensuring that any declared findings are diagnostically meaningful [2].

The core principle of data-splitting is to create a formal firewall between these two modes of inquiry. By randomly splitting the data, researchers can use one portion for unrestricted exploration and the other for pristine confirmation. This process, also referred to as dividing data into "model training" and "validation" sets, provides a clear path from uncharted exploration to a statistically sound confirmatory test [2]. While this process reduces the sample size available for the final confirmatory analysis, the substantial benefit gained through the increased credibility of the results more than compensates for this drawback [2].

Table 1: Core Concepts of Exploratory and Confirmatory Research

Aspect	Exploratory Research	Confirmatory Research
Primary Goal	Hypothesis generation	Hypothesis testing
Error Minimization	False negatives (Type II)	False positives (Type I)
Data Relationship	Data-dependent	Data-independent
Statistical Diagnostic Value	P-values lose diagnostic value	P-values retain diagnostic value
Standards	Results deserve replication and confirmation	Results held to the highest standards

Application Protocol: Implementing Data-Splitting in Materials Science

This protocol outlines the step-by-step procedure for implementing the data-splitting strategy in a materials science research context, for instance, when investigating the relationship between processing parameters and the resultant yield strength of a novel alloy.

Pre-Splitting Phase: Data Preparation

• Data Collection and Curation: Assemble the complete, cleaned dataset. For a materials study, this may include data on composition, synthesis conditions (e.g., temperature, pressure, time), characterization data (e.g., from XRD, SEM), and the target property (e.g., yield strength, band gap, conductivity) [48].
• Feature Preprocessing: Perform necessary data normalization or standardization to remove the influence of different units and make data processing more agile. Address any missing values through appropriate imputation methods or deletion [48].
• Define the Split Ratio: Determine the proportion of data to allocate to the exploratory and confirmatory subsets. A common practice is a 50/50 or 70/30 split, but the ratio should be adjusted based on the total dataset size and the minimal sample required for a powerful confirmatory test. The split must be performed randomly to ensure both subsets are representative of the underlying data distribution.

Data-Splitting and Analysis Workflow

The following diagram illustrates the logical workflow and the distinct roles of the two data subsets.

Post-Splitting Phase: From Exploration to Confirmation

• Exploratory Analysis on the Exploratory Subset: Use this subset to conduct unsupervised learning, calculate correlation matrices, generate descriptive statistics, and build preliminary models. The goal is to identify tantalizing patterns, trends, or potential relationships between descriptors and the target material property [2].
• Hypothesis Formulation and Preregistration: Based on the insights gained, formulate a specific, testable hypothesis. For example: "For the studied alloy system, an increase in annealing temperature from 300°C to 400°C will lead to a statistically significant increase (p < 0.05) in yield strength." Formally preregister this hypothesis, along with the exact statistical test (e.g., one-tailed t-test), exclusion rules, and variable definitions, before any analysis of the confirmatory subset begins [2] [4].
• Confirmatory Analysis on the Confirmatory Subset: Execute the preregistered analysis plan exactly on the untouched confirmatory subset. No deviations from the plan are allowed without transparently reporting them as exploratory follow-ups.
• Interpretation and Reporting: Report the results of the confirmatory test as the primary finding. The results from the initial exploratory analysis can be reported as supporting, hypothesis-generating observations. Transparently disclose the entire data-splitting process in the final research publication.

Essential Reagents and Computational Tools

The following table details key resources for implementing this strategy in materials science research.

Table 2: Research Reagent Solutions for Data-Splitting and Analysis

Category / Item	Function / Description	Relevance to Data-Splitting Protocol
Data Management & Preregistration
Open Science Framework (OSF)	A free, open-source platform for managing and sharing the entire research lifecycle.	Used to preregister the analysis plan after hypothesis generation from the exploratory subset, creating a time-stamped, immutable record [2] [4].
Data Analysis & Machine Learning
Python (with scikit-learn)	A programming language with a powerful library for machine learning.	The train_test_split function in scikit-learn is used to perform the random split of the dataset into exploratory and confirmatory subsets.
R	A programming language for statistical computing and graphics.	Provides functions like sample() and createDataPartition() (from the caret package) to randomly partition the data.
Descriptor Generation
Dragon	Software for calculating a wide range of molecular descriptors from chemical structures.	Used in the data preparation phase to generate structural descriptors for materials, which become the features in the dataset [48].
RDkit	An open-source toolkit for cheminformatics.	Used to generate molecular descriptors that serve as inputs for machine learning models predicting material properties [48].
Feature Engineering
SISSO (Sure Independence Screening Sparsifying Operator)	A compressed sensing-based method for feature selection and combination.	Can be applied to the exploratory subset to identify the most relevant descriptors and create optimal feature combinations before confirmatory testing [48].

Quantitative Considerations and Best Practices

Table 3: Data-Splitting Guidelines and Statistical Considerations

Parameter	Recommendation	Rationale and Trade-offs
Split Ratio	50/50 to 70/30 (Exploratory/Confirmatory). Adjust based on N.	A 50/50 split maximizes the statistical power of the confirmatory test. A 70/30 split provides more data for robust exploration but reduces confirmatory power.
Sample Size (N)	The smaller the total N, the more impactful the split on power.	Splitting a small dataset can lead to underpowered confirmatory analysis. In such cases, alternative strategies like cross-validation followed by a full-dataset preregistration may be considered.
Randomization	Mandatory. Use a random number generator with a set seed for reproducibility.	Ensures both subsets are representative of the same underlying data distribution, preventing selection bias.
Preregistration Content	Hypothesis, primary test, variables, exclusion rules.	Distinguishes the planned confirmatory analysis from unplanned exploration, preserving the diagnosticity of statistical inferences [2].
Statistical Test Justification	Use of one-tailed tests can be justified.	The confirmatory hypothesis is derived from the exploratory subset, providing a strong, a priori direction for the test, making one-tailed tests appropriate [2].

The data-splitting strategy is a powerful, accessible, and rigorous method for optimizing exploratory research within a pre-registered framework. By formally separating hypothesis generation from hypothesis testing, it directly addresses the pervasive challenge of "small data" in fields like materials science [48]. This approach allows researchers to navigate the uncharted territory of exploratory analysis while building a verifiable path to credible, confirmatory findings. Integrating this strategy into the research workflow empowers scientists to make the most of precious data, ensuring that discoveries are not only intriguing but also statistically sound and reliable.

Leveraging Pre-registration for High-Risk, High-Impact Research Proposals

Pre-registration, the practice of publicly documenting a research plan before a study is conducted, is a powerful tool to enhance the credibility and transparency of scientific inquiry [2]. In materials science and drug development, where research often involves high-risk, high-impact investigations with substantial resource investments, pre-registration provides a structured framework to distinguish confirmatory (hypothesis-testing) from exploratory (hypothesis-generating) research [2] [49]. This distinction is critical for maintaining the diagnostic value of statistical inferences and reducing the proliferation of false positive findings [2]. By committing to a predefined analytical pathway, researchers can mitigate issues of selective reporting and data dredging, thereby strengthening the evidentiary value of their findings and building trust within the scientific community and among public stakeholders [49].

Understanding Pre-registration and Its Applicability to Materials Science

Core Concepts and Definitions

Pre-registration involves submitting a time-stamped research plan to a permanent, secure registry before beginning data collection or, in some cases, data analysis [2]. This plan specifies the study's hypotheses, primary outcome measures, methodology, and analytical strategy. The practice creates a clear distinction between confirmatory research, which tests specific, pre-defined hypotheses and is held to the highest evidential standards, and exploratory research, which generates hypotheses and identifies unexpected discoveries [2]. It is crucial to recognize that pre-registration does not prohibit unplanned analyses; rather, it ensures that the distinction between planned and unplanned work is transparent, allowing consumers of research to appropriately evaluate the evidence presented [2].

Advantages for High-Stakes Research

For fields like materials science and pharmacoepidemiology, pre-registration offers several compelling benefits that are particularly relevant for high-risk, high-impact proposals:

• Enhanced Credibility of Results: Pre-registration assures reviewers and future readers that the analytical approach was not shaped by the observed data, which protects against both conscious and unconscious analytic flexibility [2] [49]. This is paramount when research findings could influence significant policy or clinical decisions.
• Reduction of Publication Bias: By staking a claim to an idea earlier in the research process, pre-registration creates a record of all undertaken studies, including those that may yield null or negative results [49]. This helps combat the "file drawer problem," where statistically non-significant findings are less likely to be published, thus providing a more complete picture of the scientific evidence.
• Improved Research Rigor and Team Alignment: A detailed pre-registration fosters team alignment by clarifying roles, responsibilities, and the core analytical plan upfront [49]. It provides a structured foundation from which any necessary midstream changes can be transparently documented and justified.

Table 1: Comparison of Confirmatory vs. Exploratory Research in Pre-registration

Feature	Confirmatory Research	Exploratory Research
Primary Goal	Hypothesis testing	Hypothesis generation
Statistical Standard	Minimizes false positives (Type I errors)	Minimizes false negatives (Type II errors)
Data Dependence	Data-independent analysis plan	Data-dependent analysis
Diagnostic Value of P-values	P-values retain diagnostic value	P-values lose diagnostic value
Role in Science	Rigorous testing of predicted effects	Identifying unexpected discoveries and new relationships

Quantitative Data on Pre-registration Platforms and Practices

Researchers have several platform options for pre-registering their study protocols. The choice depends on the specific goals of the study, disciplinary norms, and any regulatory requirements [49]. The following table summarizes key platforms relevant to materials science and drug development research.

Table 2: Comparison of Protocol Pre-registration Platforms for Scientific Research

Platform	Primary Purpose & Focus	Leading Organization(s)	Key Features
Open Science Framework (OSF) Registries [2]	General platform for enhancing transparency/reproducibility across all sciences; includes specialized registries (e.g., RWE Registry).	Center for Open Science (COS)	Embargo option; flexible templates; citable DOI; integrated with OSF project workspace.
Real-World Evidence (RWE) Registry [49]	Pre-registration of RWE studies (e.g., comparative effectiveness studies using real-world data).	ISPOR, ISPE, Duke-Margolis	Requires protocol upload and data handling attestation; embargo feature; citable DOI.
HMA-EMA Catalogue [49]	Cataloging RWD sources and studies to facilitate discoverability and support regulatory decision-making.	Heads of Medicines Agencies (HMA), European Medicines Agency (EMA)	Links studies to institutions and data sources; protocol upload encouraged but not required.
ClinicalTrials.gov [49]	Public database of clinical studies, primarily for prospectively collected data from enrolled participants.	US National Library of Medicine	Widely recognized; required for many clinical trials; some fields may be irrelevant for RWE.

Quantitative data from the pharmaceutical literature indicates a positive trend towards adopting pre-registration, though it remains underutilized in observational research. For instance, registrations on ClinicalTrials.gov grew from approximately 12,825 in 2009 to over 22,131 in 2020, demonstrating the normalization of registration in clinical research [49]. This trajectory is beginning to extend into real-world evidence (RWE) studies, driven by guidance from global regulatory bodies and health technology assessment (HTA) agencies [49].

Experimental Protocols for Pre-registration

Core Pre-registration Protocol

This protocol outlines the standard workflow for pre-registering a research plan, applicable to most experimental materials science studies.

Step 1: Define Research Question and Hypotheses

• Clearly articulate the primary research question in a specific, measurable, and falsifiable manner.
• Formulate precise null and alternative hypotheses. For materials science, this could be a hypothesis about a material's property (e.g., "The novel ceramic composite will exhibit a fracture toughness ≥15 MPa·m¹/²").

Step 2: Specify Primary Outcome Measures

• Identify the key dependent variable(s) that will answer the primary research question. This prevents later ambiguity about what constitutes the main result.
• Example: "The primary outcome is fracture toughness (K_Ic) as measured by the single-edge V-notched beam (SEVNB) method according to ASTM C1421."

Step 3: Detail Methodology

• Describe the experimental design, including materials (sources, purity), synthesis or processing protocols (e.g., sintering temperature profile), and characterization methods (e.g., SEM, XRD, mechanical testing standards).
• Specify all experimental groups, controls, and the sample size per group, including the rationale for the chosen sample size.

Step 4: Define Analysis Plan

• Pre-specify the statistical tests (e.g., t-test, ANOVA, regression) that will be used to evaluate each hypothesis.
• Define the criteria for success (e.g., "A p-value < 0.05 will be considered statistically significant") and any correction methods for multiple comparisons.
• State pre-determined data exclusion criteria (e.g., "Samples with visible porosity >5% will be excluded from mechanical property analysis").

Step 5: Select Platform and Submit

• Choose a suitable registry (see Table 2). For most academic materials science research, the OSF Registries are a versatile option [2].
• Complete the platform's template and submit. The submission becomes a time-stamped, immutable public record (potentially under embargo).

Protocol for Pre-registration with Existing or Exploratory Data

This protocol is critical for research that involves high-risk, complex data analysis, such as optimizing synthesis parameters or analyzing text data from scientific literature using machine learning [50].

Step 1: Data Splitting

• Before any analysis, randomly split the existing dataset or incoming data stream into two parts: an exploratory set and a confirmatory set [2]. A common split is 50/50 or 70/30, depending on the total sample size.

Step 2: Exploratory Analysis

• Use the exploratory set for all initial, unplanned analyses. This includes training machine learning models, testing different data representation techniques, optimizing processing parameters, and searching for unexpected trends [2] [50].
• The goal here is to minimize false negatives to find all potential discoveries.

Step 3: Pre-registration of Confirmatory Plan

• Based on insights from the exploratory set, formulate one or more specific, testable hypotheses.
• Pre-register the exact analytical model and success criteria that will be applied to the untouched confirmatory set [2].

Step 4: Confirmatory Analysis and Reporting

• Execute the pre-registered analysis on the confirmatory set. This analysis is now a rigorous, hypothesis-testing endeavor.
• When reporting, clearly distinguish between results from the exploratory phase (which are tentative) and results from the confirmatory phase (which are held to a higher standard of evidence) [2].

The Scientist's Toolkit: Essential Materials and Reagents

For a materials science study focusing on the development and testing of a high-performance ceramic composite, the following key materials and analytical tools are critical to specify in a pre-registration.

Table 3: Research Reagent Solutions for a Ceramic Composite Study

Item Name	Function/Justification	Specifications / Standards
Alumina Powder (Al₂O₃)	Primary matrix phase for the composite, selected for its high hardness and chemical inertness.	Supplier: [e.g., Baikowski]; Purity: ≥99.99%; Average Particle Size: 0.3 µm.
Zirconia Powder (ZrO₂)	Toughening phase to enhance fracture resistance via transformation toughening mechanism.	Supplier: [e.g., Tosoh Corp.]; Purity: ≥99.9%; Stabilizer: 3 mol% Yttria (3Y-TZP).
Polyvinyl Alcohol (PVA) Binder	Binds powder particles for green body formation, enabling handling before sintering.	Concentration: 5 wt% in deionized water; Molecular Weight: ~89,000-98,000 g/mol.
Sintering Furnace	High-temperature consolidation of powdered compact into a dense, solid body.	Model: [e.g., High Temp Tube Furnace]; Max Temperature: ≥1600°C; Atmosphere: Air or controlled.
Single-Edge V-Notched Beam (SEVNB) Fixture	Standardized method for measuring fracture toughness (K_Ic) of advanced ceramics.	Test Standard: ASTM C1421; Notching method: Diamond saw followed by razor blade polishing.
Scanning Electron Microscope (SEM)	Microstructural characterization to analyze phase distribution, grain size, and fracture surfaces.	Model: [e.g., FE-SEM]; Required Capabilities: Secondary Electron (SE) and Backscattered Electron (BSE) imaging.

Protocol for Handling Post-Preregistration Changes

Inevitably, research may require deviations from the initial plan. Transparently documenting these changes is essential to maintain the credibility gained from pre-registration.

Process Overview:

• Before Data Collection: If a serious error is found or a significant change is needed before data collection has begun, the best practice is to create a new, corrected pre-registration and withdraw the original, providing a rationale for the change [2].
• After Data Collection Has Begun: If changes occur during the study, researchers should create a "Transparent Changes" document [2]. This document should be uploaded to the project's repository (e.g., the associated OSF project) and must:
  • Clearly describe each deviation from the original plan.
  • Explain the rationale for the change (e.g., equipment failure, new methodological insight).
  • Justify that the change was not motivated by the observed results.
• This document must be referenced and summarized in the final research manuscript to ensure readers can distinguish between the pre-registered analysis and any exploratory deviations.

Measuring Impact: How Pre-registration Validates and Strengthens Research

Registered Reports represent a transformative publishing format in empirical science, designed to validate research through peer review before results are known. This model shifts peer review to focus on the question and methodology rather than the outcome, aligning scientific practice with the hypothetico-deductive model and mitigating a range of questionable research practices [51]. For materials science researchers, this format offers a robust framework for pre-registering studies, ensuring methodological rigor from the earliest stages of experimental design.

The fundamental principle of Registered Reports is the provisional acceptance of research before data collection begins. This "in-principle acceptance" (IPA) guarantees publication regardless of the eventual results, provided authors adhere to their peer-reviewed protocol [52]. This model addresses publication bias by ensuring that null results and negative findings undergo the same rigorous peer review and dissemination as hypothesis-confirming outcomes, creating a more complete and accurate scientific record.

The Registered Reports Workflow: A Two-Stage Process

The following workflow illustrates the complete Registered Reports process from initial submission to final publication:

Stage 1: Pre-Data Collection Peer Review

The Stage 1 submission comprises a detailed research protocol containing: an Introduction establishing the research context and rationale; explicitly stated Hypotheses; comprehensive Methods describing proposed experimental procedures; and a Statistical Analysis Plan outlining all intended analyses [51]. For materials science studies, this should include detailed specifications of materials synthesis protocols, characterization methods, and performance testing methodologies.

During Stage 1 review, editors and peer reviewers evaluate the theoretical foundation, methodological rigor, and analytical soundness of the proposed research. High-quality protocols that pass this stage receive an "in-principle acceptance" (IPA), constituting a commitment to publish the final article regardless of the experimental outcomes [52]. This stage is crucial for materials science research where methodological transparency and reproducibility are paramount.

Stage 2: Complete Manuscript Review

Following data collection, authors submit the Stage 2 manuscript, which includes the original Introduction and Methods from Stage 1, plus Results and Discussion sections. The Results must include both the pre-registered analyses and any additional unregistered analyses clearly labeled as "Exploratory Analyses" [51]. Authors are typically required to share their data on public archives such as OSF or Figshare to enhance transparency and reproducibility [51].

Stage 2 peer review verifies that the authors conducted the research as described in the Stage 1 protocol and that the conclusions are supported by the data. While deviations from the registered protocol are permitted if justified and documented, significant unapproved changes may void the in-principle acceptance [52]. The final published article integrates both stages, providing readers with confidence that the hypotheses and primary analyses are free from questionable research practices.

Comparative Analysis: Registered Reports vs. Traditional Publishing

Table 1: Quantitative Comparison of Publishing Models

Feature	Traditional Publishing	Registered Reports
Review Timing	After data collection and analysis	Before data collection [51]
Result Dependency	Highly dependent on results (96% positive findings in psychology) [52]	Result-independent (44% positive findings in psychology) [52]
Publication Bias	High against null results	Eliminated through IPA guarantee [51]
Methodological Rigor	Assessed post-hoc	Ensured through pre-study review [52]
Analytical Flexibility	Often undisclosed flexibility in analyses	Distinction between confirmatory and exploratory analyses [51]
Researcher Incentives	Emphasis on novel, positive findings	Emphasis on methodological soundness [51]

Table 2: Peer Review Focus Across Publishing Models

Review Aspect	Traditional Publishing	Registered Reports Stage 1	Registered Reports Stage 2
Hypotheses	Evaluated in context of results	Evaluated for logical foundation and justification [51]	Verified for consistency with protocol
Methods	Assessed for appropriateness to obtained results	Rigorously reviewed for adequacy to test hypotheses [52]	Verified for adherence to approved protocol
Analysis Plan	Often reconstructed post-hoc	Pre-specified and reviewed for appropriateness [51]	Checked for adherence with deviations documented
Results	Primary focus of review	Not available for review	Reviewed for accuracy and completeness
Interpretation	Evaluated for plausibility	Not available for review	Assessed for support by data

Experimental Protocol for Materials Science Preregistration

Comprehensive Preregistration Framework

Effective preregistration for materials science studies requires meticulous documentation of experimental designs, materials specifications, and analytical procedures. The following protocol adapts the OSF Preregistration template for materials science applications [4]:

I. Research Question and Hypotheses

• Precisely define the primary research question and specific hypotheses
• Clearly state the materials system under investigation
• Identify the fundamental scientific or engineering problem being addressed

II. Experimental Design

• Specify all experimental variables (independent, dependent, controlled)
• Detail sample size and number of experimental replicates with justification
• Describe randomization procedures and blinding methods where applicable
• Outline complete experimental workflow from materials preparation to characterization

III. Materials Synthesis and Processing Protocol

• Document all starting materials with specifications (purity, supplier, lot numbers)
• Describe synthesis methods with precise parameters (temperature, time, atmosphere)
• Specify processing conditions (annealing treatments, mechanical processing)
• Detail environmental controls (humidity, temperature, cleanroom conditions)

IV. Characterization and Testing Methods

• Identify all characterization techniques (e.g., XRD, SEM, TEM, XPS, Raman)
• Specify instrument models and measurement parameters
• Describe standard calibration procedures and reference materials
• Document testing conditions and protocols for functional properties

V. Data Analysis Plan

• Predefine data processing methods and algorithms
• Specify statistical tests and significance thresholds
• Outline criteria for data inclusion/exclusion
• Define primary and secondary outcome measures
• Plan for handling missing data and outliers

Materials Science Research Reagent Solutions

Table 3: Essential Research Materials and Their Functions

Material/Reagent	Specification Requirements	Primary Function	Critical Parameters
Starting Precursors	Purity grade (>99.9%), particle size distribution, crystallographic phase	Base materials for synthesis of target compounds	Lot-to-lot consistency, impurity profiles, moisture content
Solvents & Dispersants	Anhydrous conditions, purity level, water content specification	Reaction medium, particle dispersion, processing aid	Boiling point, polarity, toxicity, environmental impact
Standard Reference Materials	Certified reference materials with traceable documentation	Instrument calibration, method validation, quantitative analysis	Certified values, uncertainty limits, stability conditions
Characterization Standards	Well-defined properties (size, composition, structure)	Measurement validation, comparative analysis	Property uniformity, stability, representative nature
Substrates & Supports	Surface roughness, crystallographic orientation, cleanliness	Support for thin films, catalysts, and nanostructures	Surface energy, thermal stability, chemical compatibility

Implementation Toolkit for Registered Reports

Color Selection Protocol for Data Visualization

The following workflow ensures accessible color contrast in scientific visualizations, a critical requirement for clear data communication:

WCAG Color Contrast Standards for Scientific Figures

Table 4: Minimum Color Contrast Requirements for Scientific Visualization

Element Type	Definition	Minimum Ratio (AA)	Enhanced Ratio (AAA)	Application Examples
Body Text	Standard text in figures and tables	4.5:1 [53]	7:1 [53]	Axis labels, legend text, data point labels
Large-Scale Text	120-150% larger than body text	3:1 [53]	4.5:1 [53]	Figure titles, section headings, highlighted annotations
Graphical Elements	User interface components and informational graphics	3:1 [53]	Not defined	Chart elements, icons, buttons, diagram components
Data Series	Distinct data traces in plots	3:1 (recommended)	4.5:1 (recommended)	Line graphs, bar charts, scatter plot categories

Practical Implementation Guide

Selecting Appropriate Journals: Over 300 journals currently offer Registered Reports as a regular submission option or through special issues [51]. In materials science, researchers should identify journals that have adopted this format or propose Registered Reports special issues to editors. Elsevier alone has numerous participating journals across disciplines [52].

Preregistration Platform Options: The Open Science Framework (OSF) provides multiple preregistration templates, including a general OSF Preregistration form and specialized templates for different research designs [4]. For materials science studies, the standard OSF Preregistration template typically provides sufficient structure, while the Open-Ended Registration offers flexibility for complex experimental designs.

Protocol Development Strategy: Effective Stage 1 protocols should be precise and explicit, detailing hypotheses, variables, methodological procedures, and analysis plans before data collection [4]. Materials scientists should pay particular attention to specifying materials characterization protocols, equipment parameters, and data processing algorithms to facilitate reproducibility.

Benefits and Validation of the Registered Reports Model

Advantages Across Stakeholder Groups

The Registered Reports model provides distinct benefits for different research community stakeholders:

For Researchers: Authors receive early feedback on their experimental design, increasing the likelihood of obtaining reliable and reproducible results [52]. The in-principle acceptance eliminates publication bias against null results and reduces the incentive for questionable research practices. For materials scientists, this is particularly valuable for replication studies and methodological validations that may not produce novel positive findings but are essential for scientific progress.

For the Scientific Community: The model increases the stock of research that is methodologically rigorous, transparent, and reproducible [51]. By publishing studies based on their methodological soundness rather than their results, the scientific literature becomes more representative of actual research outcomes. This is especially important in materials science where failed syntheses or non-performing materials provide valuable information for the community.

For Research Funders: Funding agencies benefit from supporting projects that are guaranteed publication in respected journals, eliminating publication bias and maximizing transparency [51]. Some funders have established partnerships with journals for simultaneous review of grant applications and Stage 1 protocols, increasing efficiency in research funding allocation.

Empirical Validation

Evidence supporting the validity of the Registered Reports model continues to accumulate. A comparative analysis revealed that only 44% of results in Registered Reports were positive (hypothesis-confirming), compared to 96% in traditional psychology articles [52]. This demonstrates the model's effectiveness in reducing publication bias. Additionally, studies examining the relationship between peer review scores and research impact have found moderate correlations between preliminary merit assessment and subsequent citation output, supporting the predictive validity of pre-study review [54].

The Registered Reports format represents a significant advancement in scientific publishing that aligns particularly well with the methodological needs of materials science research. By emphasizing methodological rigor over novel outcomes and ensuring transparency through preregistration, this model addresses key challenges in reproducibility and research quality that affect contemporary scientific practice.

Pre-registration is a methodological practice that involves specifying a research plan—including hypotheses, study design, and analysis strategy—in a time-stamped, immutable registry before data collection or analysis begins [2]. This practice aims to distinguish confirmatory (hypothesis-testing) from exploratory (hypothesis-generating) research, thereby safeguarding the diagnostic value of statistical tests and reducing the risk of false-positive results [2]. Within materials science and drug development, where reproducibility and reliability are paramount, preregistration offers a framework to enhance the credibility of research findings by mitigating questionable research practices such as p-hacking and HARKing (Hypothesizing After the Results are Known) [55].

Comparative Quantitative Analysis: Preregistered vs. Non-Prereregistered Studies

A comprehensive empirical comparison of 193 preregistered and 193 non-preregistered studies in psychology provides robust quantitative insights into the impact of preregistration [55]. The findings are summarized in the table below.

Table 1: Comparative Analysis of Preregistered and Non-Prereregistered Studies

Analytical Dimension	Preregistered Studies	Non-Prereregistered Studies	Comparative Findings
Proportion of Positive Results	Not significantly lower [55]	Not significantly higher [55]	No robust evidence of difference (Hypothesis 1 not supported) [55]
Effect Sizes	Not significantly smaller [55]	Not significantly larger [55]	No robust evidence of difference (Hypothesis 2 not supported) [55]
Statistical Error Rates	Not significantly fewer errors [55]	Not significantly more errors [55]	No robust evidence of difference (Hypothesis 3 not supported) [55]
Use of Power Analysis	More frequent [55]	Less frequent [55]	Statistically significant increase (Hypothesis 4 supported) [55]
Sample Sizes	Typically larger [55]	Typically smaller [55]	Statistically significant increase (Hypothesis 5 supported) [55]
Publication Impact	Better on several impact measures [55]	Lower on several impact measures [55]	Preregistered studies scored higher [55]
Time to Publication	No longer [55]	Similar [55]	No significant difference found [55]

Experimental Protocols for Pre-registration

Protocol 1: Preregistering a Novel Material Synthesis Study

This protocol provides a step-by-step guide for preregistering a study focused on synthesizing and characterizing a new material.

Workflow Overview:

Detailed Methodology:

• Step 1: Select a Registration Template. Utilize a standardized template such as the OSF Preregistration, which is a comprehensive, general-purpose form [4].
• Step 2: Draft the Research Plan. Formulate and document the primary research question and all specific hypotheses. This plan should be precise and explicit, detailing the variables involved [4].
• Step 3: Specify the Synthesis Protocol. Describe the experimental procedure in detail, including:
  • Precursors and Materials: Identity, purity, and supplier of all starting materials.
  • Reaction/Synthesis Conditions: Precise temperatures, pressures, durations, and atmospheric controls.
  • Equipment: Make and model of key instruments (e.g., furnaces, reactors).
• Step 4: Define Characterization Methods. Pre-specify all materials characterization techniques to be employed (e.g., XRD, SEM, DSC). For each technique, define the key operational parameters (e.g., scan rates, accelerating voltages) and the number of replicates.
• Step 5: Pre-specify the Analysis Plan. Outline the data processing steps and statistical tests. Anticipate potential contingencies; for example, "if the data violate the assumption of normality, a non-parametric Mann-Whitney U test will be used instead of a t-test" [4]. Detail exclusion rules for outliers and how results will be reported.
• Step 6: Submit the Preregistration. Finalize and submit the completed form to a registry such as the Open Science Framework (OSF), creating a time-stamped, immutable record [4].

Protocol 2: Preregistering a Drug Formulation and Stability Study

This protocol is designed for pre-clinical research on the efficacy and stability of a new drug formulation.

Workflow Overview:

Detailed Methodology:

• Step 1: Define Formulation and Metrics. Clearly describe the drug formulation's composition and the key performance indicators (e.g., dissolution rate, bioavailability, IC50).
• Step 2: Plan Bioassays. Detail the design of in-vitro or in-vivo assays. This includes specifying control groups, dosage levels, sample sizes (justified by a power analysis, if possible [55]), and primary endpoints.
• Step 3: Design Stability Studies. Outline the conditions for stability testing (e.g., temperature, humidity, light exposure), the time points for analysis, and the specific metrics for assessing degradation (e.g., concentration, impurity profiles).
• Step 4: Pre-define Outcomes and Analysis. Declare which outcomes are primary and which are secondary. Specify the exact statistical models (e.g., ANOVA for dose-response, regression for stability data) and any adjustments for multiple comparisons.
• Step 5: Document Experimental Rigor. Describe procedures for blinding during data analysis and the method of randomizing sample allocation to minimize bias.
• Step 6: Register and Begin. Submit the complete plan to a registry before commencing the experiments [2].

The Scientist's Toolkit: Research Reagent Solutions

Table 2: Essential Materials and Reagents for Pre-registered Materials Science and Drug Development Research

Item/Tool	Function & Application	Considerations for Pre-registration
OSF Preregistration Template	Standardized form for detailing hypotheses, methods, and analysis plan [4].	Serves as the foundational document for the preregistration, ensuring all critical elements are considered.
Reference Materials (e.g., NIST Standards)	Certified materials used to calibrate instruments and validate analytical methods.	Specify the specific standard(s) to be used for each characterization technique to ensure reproducibility.
High-Purity Precursors	Starting materials for synthesizing novel compounds or materials with defined properties.	Document supplier, catalog number, and lot number to control for variability between material batches.
Cell Lines/Animal Models	Biological systems for testing drug efficacy and toxicity in pre-clinical research.	Pre-define the source, species, strain, passage number, and specific model to be used. Justify sample size.
Statistical Analysis Software (R, Python)	Tools for executing the pre-specified data analysis plan.	The planned analysis script can be included as part of the preregistration to enhance computational reproducibility.

Logical Framework for Implementing Pre-registration

The following diagram illustrates the decision-making process and workflow for integrating preregistration into a research lifecycle, from initial planning to manuscript preparation.

In the competitive landscape of scientific research funding, demonstrating methodological rigor has become paramount for securing support from agencies like the U.S. National Science Foundation (NSF). The materials science community faces increasing pressure to produce reproducible, transparent research that aligns with funder priorities while accelerating the discovery-to-deployment timeline. Pre-registration of research plans represents a transformative approach to addressing these demands, providing a structured mechanism for establishing research credibility before data collection begins.

NSF's commitment to rigorous, impactful research is embedded throughout its strategic framework. The agency explicitly values "scientific leadership" and "integrity and excellence" as core principles [56]. Furthermore, specific programs like Designing Materials to Revolutionize and Engineer our Future (DMREF) actively promote "a deep integration of experiments, computation, and theory" and "the use of accessible digital data" - principles inherently supported by pre-registration practices [57]. This alignment between methodological rigor and funder priorities creates a compelling case for adopting pre-registration in materials science.

Aligning with Funding Body Priorities

NSF Strategic Goals and Research Funding

The NSF's strategic plan outlines four overarching goals that directly inform funding decisions and represent key alignment opportunities for researchers [56]:

• Empower STEM Talent to Fully Participate: NSF seeks to "increase the involvement of communities underrepresented in STEM" through programs like Research Experiences for Undergraduates (REU) [56].
• Create New Knowledge: The agency supports "exploratory research" and "smart risk-taking" that can disrupt traditional understanding [56].
• Benefit Society by Translating Knowledge into Solutions: NSF aims to "advance research and accelerate innovation that addresses societal challenges" [56].
• Excel at NSF Operations and Management: This focuses on internal agency effectiveness [56].

The DMREF program exemplifies how these strategic goals translate into specific funding priorities, emphasizing "a collaborative and iterative 'closed-loop' process wherein theory guides computational simulation, computational simulation guides experiments, and experimental observation further guides theory" [57]. This methodology naturally complements pre-registration approaches.

Table: NSF Strategic Goals and Research Alignment Opportunities

Strategic Goal	Research Alignment Opportunity	Relevant NSF Programs
Empower STEM Talent	Include broader impact statements addressing workforce development	Research Experiences for Undergraduates (REU)
Create New Knowledge	Propose high-risk, high-reward fundamental research	DMREF, Condensed Matter and Materials Theory
Benefit Society	Demonstrate clear translation path to applications	America's Seed Fund (SBIR/STTR)
Cross-cutting Priorities	Integrate computation, experiment, and theory	DMREF, Division of Materials Research

Materials Genome Initiative and DMREF Program

The DMREF program serves as NSF's primary implementation vehicle for the Materials Genome Initiative (MGI), representing a "transformational paradigm shift in the philosophy of how materials research is performed" [57]. DMREF specifically supports activities that "significantly accelerate the materials discovery-to-use timeline by building the fundamental knowledge base needed to advance the design, development, or manufacturability of materials with desirable properties or functionality" [57].

Key DMREF requirements that align with pre-registration include:

• Team Science: Proposals "must be directed by a team of at least two Senior/Key Personnel with complementary expertise" [57].
• Closed-Loop Research: The proposed research "must involve a collaborative and iterative 'closed-loop' process" [57].
• Data Accessibility: Emphasis on "the use of accessible digital data across the materials development continuum" [57].

The program supports awards ranging from "$1,500,000 - $2,000,000 over a duration of four years" with biennial competitions in odd-numbered years [57].

Pre-registration Protocols for Materials Science

Fundamental Principles of Pre-registration

Pre-registration involves "specifying your research plan in advance of your study and submitting it to a registry" to create "a specific plan for the upcoming study" [2]. This practice helps "distinguish planned from unplanned work," addressing the critical problem where "the same data cannot be used to generate and test a hypothesis, which can happen unintentionally and reduce the credibility of your results" [2].

The distinction between confirmatory and exploratory research is fundamental to pre-registration:

• Confirmatory Research: "Hypothesis testing" where "results are held to the highest standards," "minimizes false positives," and "P-values retain diagnostic value" [2].
• Exploratory Research: "Hypothesis generating" where "results deserve to be replicated and confirmed," "minimizes false negatives," and "P-values lose diagnostic value" [2].

Pre-registration is particularly valuable for the iterative, closed-loop research approach mandated by DMREF, as it creates a transparent record of initial hypotheses and analytical plans before the iterative optimization process begins.

When to Pre-register Materials Science Studies

Optimal timing for pre-registration depends on research context [2]:

• Ideal Scenario: "Right before your next round of data collection"
• Peer Review Response: "After you are asked to collect more data in peer review"
• Existing Data Analysis: "Before you begin analysis of an existing data set"

For research using existing data, special considerations apply, requiring researchers to "describe the steps that will ensure that the data or reported outcomes do not influence the analytical decisions" [2].

Table: Pre-registration Timing and Specifications for Different Research Contexts

Research Context	Pre-registration Timing	Key Specifications Required
New Data Collection	Before any data collection	Full experimental design, sample size justification, primary outcomes, analysis plan
Additional Data for Revision	Before collecting new data in response to reviews	Clear distinction between original and new data, analysis plan for combined dataset
Existing Dataset	Before any analysis related to research question	Detailed documentation of data access status, blinding procedures, analytical choices
Split-sample Analysis	Before exploring first data subset	Rationale for sample division, explicit hypothesis to test on holdout sample

Pre-registration Workflow Protocol

The pre-registration process follows a structured workflow that can be visualized and implemented systematically:

Protocol: Pre-registration Document Creation

• Research Question Development

  • Formulate primary research question with PICO elements (Population, Intervention, Comparison, Outcome) where applicable
  • Define secondary or exploratory questions clearly distinguished from primary question
  • Justify research question based on literature review and theoretical framework

• Hypothesis Specification

  • State primary hypothesis in testable form with clear directionality where appropriate
  • Define secondary hypotheses with clear prioritization
  • Specify exact statistical null and alternative hypotheses

• Methodology Description

  • Experimental design with detailed materials synthesis and characterization protocols
  • Sample size justification with power analysis where applicable
  • Materials specification including sources, purity, and processing history
  • Measurement techniques with precision and accuracy estimates
  • Control conditions and standardization procedures

• Analysis Plan

  • Primary and secondary outcome measures with operational definitions
  • Statistical analysis methods with software implementation details
  • Covariate adjustment strategy and model specification
  • Handling of missing data and outliers
  • Significance thresholds and multiple testing corrections

• Data Management Plan

  • Data storage and backup procedures
  • Documentation standards for materials characterization data
  • Data sharing commitments and timeline

Demonstrating Methodological Rigor Through Quantitative Data Presentation

Data Comparison and Visualization Strategies

Effective data presentation is crucial for demonstrating research rigor to funding bodies. Different comparison charts serve specific purposes in materials science research [58]:

• Boxplots: Ideal for "comparing quantitative data between individuals" and showing "parallel boxplots or side-by-side boxplots" that display "the minimum value; the first quartile (Q1); the median (Q2); the third quartile (Q3); and the maximum value" [58].
• 2-D Dot Charts: Effective for "small to moderate amounts of data" to show individual data points while comparing groups [58].
• Bar Charts: Suitable for "comparing different categorical data" and "monitoring changes over time" [59].

Table: Appropriate Data Visualization Techniques for Materials Research Data

Data Type	Recommended Visualization	Key Strengths	Common Applications in Materials Science
Group Comparisons	Boxplots	Shows distribution shape, central tendency, and variability	Comparing material properties across synthesis conditions
Time Series Data	Line Charts	Displays trends and patterns over time	Tracking material degradation, performance over cycles
Composition Data	Pie or Doughnut Charts	Illustrates part-to-whole relationships	Elemental composition, phase distribution
Multivariate Data	Combo Charts	Combines different data types in one view	Correlating structural properties with performance metrics
Small Datasets	2-D Dot Charts	Preserves individual data points	Preliminary studies, high-cost experiments

Statistical Analysis and Reporting Standards

When comparing quantitative data between groups, researchers should provide complete numerical summaries. For two-group comparisons, this includes "the difference between the means and/or medians of the two groups" [58]. The summary should include measures of central tendency, variability, and sample sizes for each group, with the difference between groups clearly indicated.

For example, in a study comparing material properties, a complete summary table would include [58]:

• Sample size (n) for each group
• Mean and/or median for each group
• Standard deviation or interquartile range (IQR)
• Difference between group means/medians

Funding bodies particularly value transparent reporting of all analyses, including non-significant results, to avoid selective reporting biases.

Experimental Protocols for Materials Characterization

Workflow for Closed-Loop Materials Research

The DMREF program mandates a "collaborative and iterative 'closed-loop' process" that integrates computation and experimentation [57]. This workflow can be systematically structured as follows:

Protocol: Integrated Materials Research Workflow

• Theoretical Framework Establishment

  • Define theoretical basis for material behavior predictions
  • Identify key parameters for computational modeling
  • Establish success criteria for material performance

• Computational Simulation Protocol

  • Select appropriate modeling approaches (DFT, MD, phase-field, etc.)
  • Define simulation parameters and boundary conditions
  • Establish convergence criteria and validation metrics
  • Protocol duration: Variable based on complexity (1 week - 3 months)

• Materials Synthesis Protocol

  • Specify precursor materials with purity and source information
  • Detail synthesis conditions (temperature, pressure, atmosphere)
  • Define processing parameters (time, heating/cooling rates)
  • Quality control checkpoints during synthesis

• Materials Characterization Protocol

  • Structural characterization (XRD, SEM, TEM)
  • Chemical analysis (EDS, XPS, FTIR)
  • Physical property measurement (mechanical, thermal, electrical)
  • Performance testing under application conditions

• Data Integration and Model Refinement

  • Compare experimental results with computational predictions
  • Identify discrepancies and refine theoretical models
  • Update simulation parameters for next iteration
  • Document all changes from initial predictions

Research Reagent Solutions and Essential Materials

Table: Essential Research Reagents and Materials for Advanced Materials Research

Reagent/Material	Function/Purpose	Key Specifications	Example Applications
High-Purity Metal Precursors	Source materials for synthesis	≥99.99% purity, certified trace metals	Thin film deposition, alloy development
Solvents (Anhydrous)	Reaction medium for synthesis	Water content <10 ppm, inhibitor-free	Polymer synthesis, nanoparticle preparation
Single Crystal Substrates	Epitaxial growth templates	Specific orientation (±0.5°), surface roughness <0.5 nm	Electronic materials, quantum structures
Sputtering Targets	Physical vapor deposition sources	99.95% purity, controlled density and grain size	Metallic thin films, multilayer structures
Polymer Monomers	Building blocks for macromolecules	Controlled functionality, purity >99%	Functional polymers, biomaterials
Characterization Standards	Instrument calibration	Certified reference materials	Quantitative microanalysis, property measurement
Lithography Resists	Pattern definition	Specific sensitivity, contrast, and resolution	Nanofabrication, device patterning
Gaseous Precursors	Vapor phase reactions	Specific purity levels, moisture/oxygen controls	CVD, surface functionalization

Implementation Framework for Funding Applications

Integrating Pre-registration into Proposal Development

Successful funding applications should explicitly incorporate pre-registration principles throughout the research plan:

• Introduction and Specific Aims

  • State commitment to rigorous methodology through pre-registration
  • Clearly distinguish confirmatory and exploratory aims
  • Reference agency priorities for transparent, reproducible research

• Research Strategy

  • Include pre-registration as a specific step in the research timeline
  • Detail which studies will be pre-registered and when
  • Specify the registry that will be used (e.g., OSF, discipline-specific registry)

• Broader Impacts

  • Highlight how pre-registration contributes to research transparency
  • Describe how pre-registered research facilitates data sharing and collaboration
  • Connect to NSF's goal of "strengthening connections among theorists, computational scientists, data scientists, mathematicians, statisticians, and experimentalists" [57]

Addressing Common Concerns and Implementation Barriers

Researchers often express concerns about pre-registration limiting scientific creativity and flexibility. However, pre-registration "does not mean that you can't do any unplanned analyses" but rather "distinguishes confirmatory and exploratory analyses" [2]. This distinction actually enhances scientific discovery by providing a clear framework for both hypothesis testing and exploration.

For materials researchers concerned about the iterative nature of materials development, the split-sample approach provides an excellent compromise: "It may be difficult to fully prespecify your model until you have a chance to explore through a real data-set... By randomly splitting off some 'real' data, you can build the model through exploration and then confirm it with the portion of the data that has not yet been analyzed" [2].

Pre-registration represents a powerful methodology for aligning materials science research with NSF and other funding agency priorities. By adopting pre-registration practices, researchers can demonstrate methodological rigor, enhance research transparency, and accelerate the materials development continuum that programs like DMREF explicitly champion. The structured protocols and workflows presented in this article provide practical implementation guidance that can strengthen funding applications while advancing the scientific quality of materials research.

Pre-registration is the practice of registering the hypotheses, methods, and analysis plans of a scientific study before it is conducted [1]. This practice creates a time-stamped, read-only research plan that distinguishes between confirmatory (planned) and exploratory (unplanned) research, thereby increasing transparency and rigor [2]. While more established in clinical trials and psychology, pre-registration offers significant benefits for materials science by mitigating questionable research practices, reducing publication bias, and improving the credibility of results [1] [35].

The core value of pre-registration lies in its ability to clearly separate hypothesis-generating exploration from hypothesis-testing confirmation. In materials science research, where experimental complexity often leads to unexpected observations, this distinction is particularly valuable. Pre-registration establishes a transparent record of your initial research intentions, allowing for a more rigorous evaluation of subsequent findings [4]. This practice is especially relevant for case studies in materials science, which often involve complex, multi-stage experimental processes where the path from hypothesis to conclusion can benefit from increased methodological transparency.

Theoretical Foundation and Framework

The Rationale for Pre-registration

Pre-registration addresses several systemic challenges in scientific research. It directly confronts the "replication crisis" by combating publication bias and questionable research practices such as p-hacking (trying various analytical approaches until a significant result is found) and HARKing (Hypothesizing After Results are Known) [35] [1]. By requiring researchers to specify their analysis plan before data collection, pre-registration ensures that results are evaluated based on the severity of the planned test rather than their statistical significance [1].

For materials science, this framework provides a structured approach to documenting research decisions that are often omitted from final publications, such as material selection criteria, processing parameter boundaries, and characterization protocols. The pre-registration process creates a specific plan for upcoming studies, which helps researchers clearly distinguish planned from unplanned work [2]. This distinction is crucial for maintaining diagnostic value in statistical testing while still allowing for serendipitous discovery through properly documented exploratory analysis.

Types of Pre-registration

Researchers can choose from several pre-registration formats depending on their needs:

• Standard Pre-registration: Researchers prepare and publicly deposit a research protocol before conducting their study, detailing hypotheses, sampling procedures, measures, and statistical analyses [1].
• Registered Reports: This format involves peer review of the introduction, methods, and analysis plan before data collection, with provisional acceptance of the study regardless of the eventual results [1].
• Specialized Pre-registration: Adaptations are available for various research contexts, including qualitative research, pre-existing data, and exploratory research [1].

For materials science case studies, the standard pre-registration or registered reports formats are most applicable, with adaptations to address field-specific requirements such as material sourcing, processing parameters, and characterization techniques.

Pre-registration Workflow for Materials Science

The process of pre-registering a materials science case study follows a structured pathway from initial planning through protocol development to registration and execution. The following workflow diagram illustrates the key stages and decision points in this process:

Workflow Description

The pre-registration workflow begins with comprehensive research planning and hypothesis generation based on thorough literature review. Researchers then select an appropriate pre-registration template, such as the OSF Preregistration template or the AsPredicted template [4]. The core of the process involves developing a detailed research protocol that specifies all key methodological elements before any experimental work begins. This protocol is then submitted to a registry such as the Open Science Framework (OSF), creating a time-stamped, immutable record [2]. Research is conducted according to the pre-registered plan, with any necessary deviations transparently documented in the final research report, which must link to the original pre-registration.

Detailed Experimental Protocol

Pre-registration Template Selection and Adaptation

For materials science case studies, the OSF Preregistration template provides a comprehensive foundation that can be adapted to field-specific requirements [4]. This template should be expanded to include materials-specific sections:

• Material Specifications: Include source, purity, lot numbers, and certification details for all raw materials
• Processing Parameters: Document all synthesis, fabrication, or processing conditions with acceptable ranges
• Characterization Methods: Specify all structural, chemical, and property measurement techniques with instrument parameters
• Data Analysis Protocols: Define how experimental data will be processed, normalized, and analyzed

When selecting a template, researchers should choose one that aligns with their study design and community standards, considering factors such as whether data have been collected or viewed already [4]. The OSF provides numerous specialized templates, and researchers should select the one that best fits their specific research design, with the OSF Preregistration template being the most commonly used for general purposes [4].

Protocol Development Guidelines

Effective pre-registration requires precise and explicit planning. Researchers should:

• Use the pre-registration as a draft for the methods and results sections of a future journal article [4]
• Make all design and analysis decisions before viewing the data [4]
• Describe statistical tests and decision criteria for interpreting results [4]
• Specify exclusion rules for experimental data and criteria for combining variables [4]
• Anticipate potential deviations from the plan and include contingency analyses [4]
• List all covariates and model characteristics that will be reported [4]

For materials science, this specifically includes defining criteria for material batch acceptance, establishing thresholds for measurement quality, and specifying how structural data will be quantified and correlated with properties.

Research Reagent Solutions and Materials

The following table details essential materials and research reagents commonly used in materials science case studies, with specifications critical for pre-registration:

Table 1: Essential Research Reagents and Materials for Materials Science Studies

Item Category	Specific Function	Pre-registration Specifications Required
Precursor Materials	Source compounds for material synthesis	Supplier, purity grade, lot number, chemical composition certificates, impurity profiles
Characterization Standards	Calibration and validation of instruments	Certified reference materials, standard provenance, uncertainty values, expiration dates
Processing Reagents	Solvents, additives, catalysts for material fabrication	Manufacturer, grade, concentration, purification methods, storage conditions
Substrate Materials	Support structures for deposition or testing	Material grade, surface finish, orientation, crystallographic parameters, cleaning protocols
Analytical Consumables	Electrodes, holders, containers for testing	Material composition, dimensions, surface treatments, single-use vs. reusable designation

Quantitative Data Standards and Compliance

Pre-registration in materials science requires careful specification of quantitative standards and acceptance criteria for experimental data. The following table outlines key parameters that should be defined during protocol development:

Table 2: Quantitative Data Standards for Materials Science Pre-registration

Data Category	Minimum Standards	Acceptance Criteria	Reporting Requirements
Structural Characterization	Crystallographic phase purity ≥95% by XRD; Microstructural images at minimum 3 representative areas	Defined peak intensity ratios; Statistical significance of representative areas	Full pattern/data files; Scale bars and magnification for all images
Compositional Analysis	Detection limits for all relevant elements; Minimum 3 measurements per sample	Acceptable variance between replicates (<5% RSD)	Complete elemental quantification with uncertainty estimates
Property Measurements	Minimum sample size n≥5 for statistical power; Standard testing protocols cited	Coefficient of variation thresholds; Outlier identification method	Raw data curves; Complete statistical summary with effect sizes
Process Parameters	Temperature control ±1°C; Pressure monitoring intervals; Atmosphere composition	Ranges for acceptable experimental conditions	Time-series data for critical parameters; Documentation of any excursions

Implementation and Reporting Framework

Pre-registration Submission Process

The technical process of submitting a pre-registration involves:

• Creating a Registration: Researchers can start from scratch or from an existing OSF project, with the registration pulling metadata and files from the project [4]
• Selecting a Template: Choosing the appropriate registration template that best fits the research project [4]
• Completing the Protocol: Filling in all required sections of the research registration template [4]
• Finalizing Submission: Submitting the completed registration, which becomes an immutable record [4]

When registering from an existing project, researchers should note that any files associated with the project will be attached to the registration up to 5 GB [4]. For blinded peer review, all identifying information must be removed from metadata and attached documents, and the registration should be placed under embargo [4].

Reporting and Deviation Management

When reporting pre-registered research, authors must:

• Report all results from pre-analysis plans, not just significant findings [2]
• Interpret all pre-planned analyses, not just those with statistically significant results [2]
• Clearly distinguish between confirmatory and exploratory analyses in the final manuscript [2]
• Transparently document any deviations from the pre-registered plan using a "Transparent Changes" document [2] [4]

For materials science case studies, this particularly includes documenting any adjustments to experimental protocols, material substitutions, or characterization method modifications necessitated by practical research constraints.

Visual Documentation Standards

Effective visual communication in materials science requires adherence to accessibility standards, particularly for color contrast in figures and diagrams. The following diagram illustrates the relationship between different elements of experimental documentation and the corresponding contrast requirements:

Visual Accessibility Requirements

All visual elements in pre-registered research must meet minimum color contrast ratio thresholds to ensure accessibility [60] [61] [62]. The specific requirements include:

• Normal Text: Minimum contrast ratio of 4.5:1 between text and background [62] [63]
• Large Text: Minimum contrast ratio of 3:1 for text at least 18pt (24 CSS pixels) or 14pt bold (19 CSS pixels) [62] [63]
• Enhanced Contrast: For higher accessibility standards, a contrast ratio of at least 7:1 for normal text and 4.5:1 for large text [60]

These requirements apply to all text elements in figures, diagrams, and presentations associated with the pre-registered research. Note that WCAG thresholds are absolute - for example, 4.49:1 fails the 4.5:1 requirement [61].

Publication bias, the systematic failure to publish research findings based on the direction or strength of the results, represents a significant challenge to scientific progress. This bias disproportionately affects null results—findings that do not support the experimental hypothesis—creating a distorted literature that overrepresents positive effects. In materials science and drug development, this "file drawer problem" [45] can lead to misallocated resources, repeated dead ends, and an inaccurate understanding of material properties and drug efficacy.

Preregistration has emerged as a powerful methodological solution to this challenge. By specifying research plans in advance of data collection and analysis, preregistration creates a transparent record of all studies conducted, regardless of their outcomes [2]. This practice is particularly valuable in fields with high resource commitments, such as materials synthesis characterization and preclinical drug development, where understanding what does not work is equally as important as understanding what does.

Understanding Publication Bias and Its Consequences

Publication bias stems from multiple sources, including researcher motivation, perceived journal preferences, and corporate interests. Studies with statistically significant results are more likely to be submitted and accepted for publication, while null findings often remain inaccessible. Analysis of NHLBI clinical trials reveals that the likelihood of null effects has increased over time, particularly after the adoption of transparent reporting standards [2].

The consequences of this bias are particularly severe in materials science and drug development:

• Resource Misallocation: Academic labs and companies may pursue research avenues based on selectively published positive results while remaining unaware of unsuccessful attempts.
• Safety Risks: In drug development, failure to publish null results or adverse findings can compromise safety assessments.
• Scientific Integrity: The literature becomes skewed toward positive findings, undermining meta-analyses and systematic reviews.

Preregistration as a Solution

What is Preregistration?

Preregistration involves formally documenting a research plan before beginning a study and submitting this plan to a timestamped registry. This process creates a clear distinction between confirmatory (hypothesis-testing) and exploratory (hypothesis-generating) research [2]. For materials science researchers, this might include specifying synthesis parameters, characterization methods, and performance metrics in advance.

The preregistration document becomes a permanent part of the research record, establishing the researcher's prior intentions regardless of the eventual outcomes. This practice "future-proofs" research by creating a specific plan for the upcoming study [2].

How Preregistration Combats Publication Bias

Preregistration addresses publication bias through several mechanisms:

• Reduces Pressure for Positive Results: Registered Reports format alleviates the pressure for positive results, directly countering the "file drawer problem" and promoting a more balanced literature [45].
• Ensures Result Visibility: By creating a public record of all undertaken studies, preregistration ensures that null results remain part of the scientific discourse.
• Detaches Value from Outcome: The methodological rigor of preregistered research provides inherent value independent of the results' direction.

Table 1: Comparison of Traditional vs. Preregistered Research Approaches

Aspect	Traditional Approach	Preregistered Approach
Hypothesis Development	Can be developed after seeing data	Must be specified before data collection
Analysis Plan	Flexible and adaptable	Fixed before data analysis begins
Result Interpretation	Potentially influenced by outcome	Based on predetermined criteria
Publication Likelihood	Heavily favors significant results	Equal opportunity for all outcomes
Transparency	Limited insight into analytical choices	Complete transparency of planned vs. unplanned analyses

Quantitative Evidence Supporting Preregistration

Empirical studies demonstrate the tangible benefits of preregistration and transparent research practices:

Table 2: Documented Benefits of Preregistration and Transparent Practices

Documented Effect	Impact	Evidence
Reduction in False Positives	Minimizes Type I errors through predefined analysis plans	P-values retain diagnostic value in confirmatory research [2]
Increased Credibility	Enhances trust in research findings	Establishes transparency and methodological rigor [45]
Improved Reproducibility	Facilitates study replication	Contributes to cumulative knowledge advancement [45]
Literature Balance	Counters selective publication	Registered Reports reduce file drawer problem [45]

Analysis of clinical trials has shown that "the number of NHLBI trials reporting positive results declined after the year 2000," with researchers attributing this trend to "prospective declaration of outcomes in RCTs, and the adoption of transparent reporting standards" [2]. This suggests that as transparent practices like preregistration become more common, the scientific literature becomes more representative of all research conducted.

Experimental Protocols for Preregistration

Protocol 1: Basic Preregistration Workflow

The following diagram illustrates the core workflow for implementing preregistration in materials science research:

Protocol 2: Advanced Preregistration with Data Splitting

For exploratory research in novel materials domains, a data-splitting approach provides both flexibility and rigor:

This approach is particularly valuable when "it may be difficult to fully prespecify your model until you have a chance to explore through a real data-set" [2]. Though it "reduces the sample size available for confirmatory analysis, the benefit gained through increased credibility more than makes up for it" [2].

Protocol 3: Preregistration with Existing Data

Using existing datasets requires special considerations to maintain confirmatory value:

Table 3: Guidelines for Preregistration with Existing Data

Data Status	Eligibility Requirements	Justification Needed
No data collected	Certify that data do not exist	Not required
Data exists, unobserved	Certify no human observation	Explain how data remains unobserved
Data exists, not accessed	Certify no access by research team	Explain who has accessed data
Data exists, not analyzed	Certify no analysis related to research plan	Justify how prior reporting doesn't compromise confirmatory nature

For existing data, researchers must describe "steps that will ensure that the data or reported outcomes do not influence the analytical decisions" [2]. Standards for effective preregistration using existing data are still evolving.

Implementation Framework

Template Selection Guide

The Open Science Framework provides multiple preregistration templates tailored to different research needs:

Table 4: OSF Preregistration Templates for Materials Science

Template Name	Best Use Cases	Key Features
OSF Preregistration	Standard materials research	Comprehensive, general purpose, most commonly used
Open-Ended Registration	Unconventional study designs	Maximum flexibility
Secondary Data Preregistration	Analysis of existing datasets	Specific questions about data access and prior observation
AsPredicted.org Template	Simple experimental studies	Streamlined, 8-question format
Registered Report Protocol	Journal-integrated preregistration	For use after "in-principle acceptance"

The OSF Preregistration template is "most commonly used" and provides a strong foundation for most materials science studies [4]. Researchers should "use your preregistration as a way to draft your methods and results section of a journal article" to ensure thorough planning [4].

The Scientist's Toolkit: Essential Research Reagent Solutions

Table 5: Essential Materials and Tools for Preregistered Research

Tool/Resource	Function	Implementation Example
OSF Registry	Timestamped preregistration submission	Creates immutable record of research plan
Preregistration Templates	Structured planning documents	Ensures comprehensive methodology description
Data Splitting Protocols	Separation of exploratory/confirmatory data	Enables hypothesis generation followed by rigorous testing
Transparent Changes Document	Records deviations from original plan	Maintains integrity while acknowledging real-world adaptations
Color Contrast Analyzers	Ensures accessibility of visual data	Verifies sufficient contrast in graphs and charts [60]

Practical Implementation Steps

• Select Appropriate Template: Choose based on research design and data status [4]
• Develop Detailed Plan: "Be precise and explicit with your plan, e.g. list out your hypotheses and which variables you'll use" [4]
• Anticipate Contingencies: "Consider writing out 'if then' decision trees, e.g., 'if the assumptions for this test are violated, we will use this alternate test'" [4]
• Submit Preregistration: Create timestamped, immutable record in OSF registry
• Conduct Research: Follow planned methodology while documenting deviations
• Report Transparently: Disclose all analyses, including exploratory work

Preregistration represents a paradigm shift in how researchers approach scientific inquiry, particularly in fields like materials science and drug development where the cost of unpublished null results is high. By creating a transparent record of research intentions before data collection begins, preregistration directly addresses the systemic problem of publication bias while enhancing the credibility and reproducibility of all findings—whether positive, negative, or null.

The ultimate benefit of this approach extends beyond individual studies to the entire scientific ecosystem. As more researchers adopt preregistration, the literature will become more representative of all research conducted, providing a more accurate foundation for future scientific and resource allocation decisions. For materials scientists and drug development professionals, this transparency accelerates progress by ensuring that both successful and unsuccessful experiments contribute to our collective knowledge.

Conclusion

Pre-registration is a powerful, practical tool that moves materials science toward a more robust, collaborative, and transparent future. By formally distinguishing between hypothesis-generating and hypothesis-testing research, it directly combats the replication crisis and enhances the credibility of published findings. For individual researchers, it streamlines the research process, strengthens study design, and can improve the reception of their work by journals and funders. For the field at large, it ensures that the scientific record—including null results—is complete, accelerating cumulative knowledge. The future of materials science will be shaped by these open practices, fostering greater innovation and trust in research outcomes that can reliably inform future drug development and clinical applications.

References

• 1. Preregistration (science) [https://en.wikipedia.org/wiki/Preregistration_(science)]
• 2. Preregistration [https://www.cos.io/initiatives/prereg]
• 3. Pre-registration vs. Registered Reports: What's the ... [https://www.aje.com/arc/pre-registration-vs-registered-reports]
• 4. Welcome to Registrations & Preregistrations! [https://help.osf.io/article/330-welcome-to-registrations]
• 5. Text color must provide high contrast with its background [https://docs.acquia.com/web-governance/text-color-must-provide-high-contrast-its-background]
• 6. Recommended format for a 'research protocol' [https://www.who.int/groups/research-ethics-review-committee/recommended-format-for-a-research-protocol]
• 7. An Overview of the Fundamentals of Data Management ... [https://www.sciencedirect.com/science/article/pii/S0749208123000293]
• 8. Quantitative Data Analysis Methods & Techniques 101 [https://gradcoach.com/quantitative-data-analysis-methods/]
• 9. Principles of Effective Data Visualization - PMC [https://pmc.ncbi.nlm.nih.gov/articles/PMC7733875/]
• 10. Designing Effective Data Visualizations [https://guides.library.jhu.edu/datavisualization/design]
• 11. Best Practices - Data Visualization [https://libguides.gwu.edu/dataviz/best_practices]
• 12. Distinguishing between Exploratory and Confirmatory ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4028181/]
• 13. A Guide to Evidence Synthesis: 5. Register a Protocol [https://guides.library.cornell.edu/evidence-synthesis/protocol]
• 14. Protocols and Methods - Materials Science - Research Guides [https://guides.library.ucdavis.edu/materials-science/protocols-methods]
• 15. Utilizing tables, figures, charts and graphs to enhance the ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10394528/]
• 16. Figures and Charts - UNC Writing Center [https://writingcenter.unc.edu/tips-and-tools/figures-and-charts/]
• 17. Understanding Success Criterion 1.4.3: Contrast (Minimum) [https://www.w3.org/WAI/WCAG21/Understanding/contrast-minimum.html]
• 18. Key Considerations for Flowchart Accessibility - MN.gov [https://mn.gov/mnit/about-mnit/accessibility/news/?id=38-436349]
• 19. Changing graph nodes text color · Issue #487 · mermaid-js ... [https://github.com/mermaid-js/mermaid/issues/487]
• 20. SPIRIT 2025 statement: updated guideline for protocols of ... [https://www.nature.com/articles/s41591-025-03668-w]
• 21. Quantifying the distribution of materials data types in ... [https://chemrxiv.org/engage/chemrxiv/article-details/655533d3dbd7c8b54b56ada6]
• 22. Design of tables for the presentation and ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10346464/]
• 23. What to consider when choosing colors for data visualization [https://academy.datawrapper.de/article/140-what-to-consider-when-choosing-colors-for-data-visualization]
• 24. How to choose colors for data visualizations [https://www.atlassian.com/data/charts/how-to-choose-colors-data-visualization]
• 25. Open Science Framework Review 2025: Pricing, Features, ... [https://research.com/software/reviews/open-science-framework]
• 26. Introducing Enhanced OSF Add-ons [https://www.cos.io/blog/cultivating-open-source-community-osf-add-ons]
• 27. OSF API: The Complete Guide [https://zuplo.com/learning-center/osf-api]
• 28. Materials Science & Engineering - Library Guides - Penn State [https://guides.libraries.psu.edu/c.php?g=821826&p=5865068]
• 29. Welcome to Registrations & Preregistrations! - OSF Support [https://help.osf.io/article/229-select-a-registration-template]
• 30. Materials Science: Home - Research Guides - Missouri S&T [https://libguides.mst.edu/materials]
• 31. Cross-Disciplinary Resources - Materials Science & Engineering [https://lib.guides.umd.edu/materialsscience/cross-eng]
• 32. Quantitative Data Quality Assurance, Analysis and Presentation [https://pmc.ncbi.nlm.nih.gov/articles/PMC12056448/]
• 33. Choosing the Right Preregistration Template: A Guide ... [https://www.cos.io/blog/choosing-preregistration-template-guide-for-researchers]
• 34. Pre-registration: Weighing costs and benefits for researchers [https://www.sciencedirect.com/science/article/abs/pii/S0749597821000649]
• 35. Adapting Open Science and Pre-registration to ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC10904029/]
• 36. SPIRIT 2025 statement: Updated guideline for protocols of ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12037212/]
• 37. Welcome to OSF Projects [https://help.osf.io/article/353-welcome-to-projects]
• 38. What's the actual cost difference between building ... [https://community.latenode.com/t/whats-the-actual-cost-difference-between-building-workflows-from-scratch-versus-starting-with-templates/56598]
• 39. 5 ways to optimize your research workflow with OSF [https://www.cos.io/blog/5-ways-to-optimize-your-research-workflow-with-osf]
• 40. Preregistration and Registered Reports [https://www.surrey.ac.uk/library/open-research/preregistration-and-registered-reports]
• 41. Public Access to Research Policy [https://nces.ed.gov/about/public-access-research]
• 42. Pre-Registering Studies – What Is It, How Do You Do It, ... [https://acf.gov/opre/blog/2022/08/pre-registering-studies-what-it-how-do-you-do-it-and-why]
• 43. Open Science practices in Quantitative research [https://www.sv.uio.no/psi/english/research/research-support/open-science-portal/quantitative/]
• 44. Transparency, bias, and reproducibility across science: a meta ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC11563668/]
• 45. Benefits and Concerns of Registration [https://help.osf.io/article/603-benefits-of-registration]
• 46. Selective reporting of antibiotic susceptibility testing results [https://pmc.ncbi.nlm.nih.gov/articles/PMC12181692/]
• 47. Extracting accurate materials data from research papers ... [https://www.nature.com/articles/s41467-024-45914-8]
• 48. Small data machine learning in materials science [https://www.nature.com/articles/s41524-023-01000-z]
• 49. Preregistration: A Key to Credible Real‐World Evidence ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC12397443/]
• 50. How to pre-register experimental studies that involve ... [https://www.sciencedirect.com/science/article/pii/S2214804325000783]
• 51. Registered Reports [https://www.cos.io/initiatives/registered-reports]
• 52. Registered Reports [https://www.elsevier.com/researcher/author/policies-and-guidelines/registered-reports]
• 53. Color contrast - Accessibility - MDN Web Docs - Mozilla [https://developer.mozilla.org/en-US/docs/Web/Accessibility/Guides/Understanding_WCAG/Perceivable/Color_contrast]
• 54. The Validation of Peer Review through Research Impact ... [https://pmc.ncbi.nlm.nih.gov/articles/PMC4153641/]
• 55. Preregistration in practice: A comparison of preregistered ... [https://link.springer.com/article/10.3758/s13428-023-02277-0]
• 56. Strategic Plan - About NSF | NSF - U.S. National Science ... [https://www.nsf.gov/about/performance/strategic-plan]
• 57. Designing Materials to Revolutionize and Engineer our ... [https://www.nsf.gov/funding/opportunities/dmref-designing-materials-revolutionize-engineer-our-future]
• 58. 14 Comparing quantitative data between individuals [https://bookdown.org/pkaldunn/SRM-Textbook/BetweenQuantData.html]
• 59. 7 Best Comparison Charts for Effective Data Visualization [https://ninjatables.com/types-of-comparison-charts/?srsltid=AfmBOor-YgF9klfiXz242GyrEuPAnR4R9_7VNMquZuhiYddnW_OWxhS6]
• 60. Text has enhanced contrast | ACT Rule | WAI [https://www.w3.org/WAI/standards-guidelines/act/rules/09o5cg/]
• 61. Contrast requirements for WCAG 2.2 Level AA [https://www.makethingsaccessible.com/guides/contrast-requirements-for-wcag-2-2-level-aa/]
• 62. Audit color-contrast [https://siteboard.io/en/resources/accessibility-audits/color-contrast]
• 63. Elements must meet minimum color contrast ratio thresholds [https://dequeuniversity.com/rules/axe/4.8/color-contrast]