Lab Manual

Version 0.1.0, updated 05-02-2026

Author

Insight Lab

This Lab Manual

This lab manual is a living document to transparently communicate and democratically update the expectations and procedures practiced in the INSIGHT lab. Clear norms reduce ambiguity, prevent misunderstandings, and help lab members work together productively towards commonly goals. A lab manual also promotes fairness, transparency, and psychological safety by making the lab culture explicit. As Kovacs and colleagues (2022) write:

Every lab has a culture and norms of research laboratories can greatly influence the quality of scientific outputs produced by the lab and the well-being of its members. [A] lab manual can ensure that lab members have a shared and mutual understanding of who they are and how they do things.

This manual takes inspiration from these main sources: the SAFE Labs Handbook, Kovacs and colleagues (2022), the FORRT Code of Conduct, and Merz, Strahmann, and Frank’s (2025) work on CRediT for students.

Note: university, departmental, and legal frameworks always take precedence over this manual.

How to update the manual

Lab members can propose additions to the lab manual via pull request at https://github.com/cjvanlissa/insight_lab/pulls (private repository, ask to join if you are not yet a collaborator)

Using pull requests keeps track of the date and author of the update.

When changes are made, also update the metadata to increment the version (use semantic versioning of release.major.minor).

Active lab members vote on pull requests, either online (thumbs up/down) or in a meeting. Pull requests are accepted if a majority of active members vote thumbs up.

Lab members may organize meetings to discuss and vote on urgent updates.

The INSIGHT Lab

Lab mission/focus

Theories reflect scientists’ understanding of phenomena. Ideally, they are unambiguous, and continuously updated based on new findings. However, many social scientific theories lack formalization and are flexible enough to accommodate contradictory findings. This limits scientific progress. The INSIGHT lab asks the question: How can we construct better theories? We seek to develop reproducible inductive (data-driven) methods to construct (formal) theories based on patterns in different kinds of data, including quantitative empirical data, text data from published papers, and qualitative data.

Lab philosophy/values

Our core values are:

  • Transparency
  • Reproducibility
  • Curiosity
  • Usefulness/contribution

Lab history

The INSIGHT lab was founded in 2026 by Caspar van Lissa, to unify and support the different projects resulting from his Vidi-funded research line “From Patterns to Principles: Machine Learning-Informed Theory Formation Methods for the Social Sciences”.

Contact information

The lab is physically embedded in the Department of Methodology and Statistics, Tilburg University.

Postal address

Tilburg University Department Methodology and Statistics, t.a.v. Caspar van Lissa PO Box 90153 5000 LE Tilburg The Netherlands

Email

To contact the lab, email specific lab members directly.

SAFE Labs Handbook

INSIGHT Lab subscribes to the SAFE Labs Handbook commitments. These can be found in the following sections:

SAFE Policies

The INSIGHT lab commits to publicly document:

The INSIGHT lab commits to internally document:

  • the procedure for reporting bullying and/or harassment
  • available resources to support mental health
    • In principle, mental health support should procede through the health care system. The general practitioner is the first (free) port of call.
    • The university describes procedures for mental health support for students. Some of these are relevant for employees too.
    • Use the My Employee Portal to report that you are sick, if you are suffering from mental health problems.
  • the procedure for raising lab or inter-personal issues: Section 10.3

SAFE Teams

The INSIGHT lab commits to publicly document:

The INSIGHT lab commits to internally document:

  • the onboarding procedure for new lab members: Section 6.3
  • how equal access to lab resources across lab members is maintained.
    • At present, there are no lab resources other than the standard laptop employees receive.

The INSIGHT lab commits to establish:

  • annual lab-wide feedback sessions.
    • The first lab meeting of the academic year devotes time to lab-wide feedback sessions.
  • annual bilateral feedback and appraisal sessions for each lab member.
    • This is embedded in policy at Tilburg University: every year there is a Performance and Talent Development meeting, which allows for bilateral feedback
  • annual lab-wide meetings to normalise failures.
    • The last lab meeting of the academic year devotes time to discussing successes and failures.
  • a mechanism to record key outcomes from each 1-on-1 meeting.
  • a “PhD steering committee” to annually monitor progress and mediate feedback.

SAFE Careers

The INSIGHT lab commits to publicly document:

The INSIGHT lab commits to publicly document:

  • the procedure for requesting reference letters: Section 16.1
  • the procedure for leaving the lab, including an exit interview: Section 16

The INSIGHT lab commits to establish

  • annual lab-wide meetings to review training and outcomes for “core skills”.
  • a mechanism for sharing lab management updates: Section 10.2
  • an objective and equitable interview process: Section 6.2

Ethics

See the Code of Conduct.

Policies and Law

Studies of human participant data are subject to ethics review, see Tilburg University Ethics Review Boards.

All data should be handled in compliance with the GDPR law.

Personal data should be handled in compliance with the AVG law.

Code of Conduct

See the Code of Conduct.

Joining the Lab

Vacancies

In the Netherlands, a PhD position is a paid job, and thus must be publicly announced to ensure fair chances to applicants.

There is no point in contacting PIs directly to enquire about PhD positions, they are all posted publicly.

Tilburg University anounces vacancies here.

They are also automatically reposted on Academic Transfer.

All Dutch universities have the same labor agreements and salary. See here for more information.

Hiring Process

The INSIGHT lab has conducted the first pilot of blind selection at Tilburg University. While this approach was positively reviewed, it places an undue burden on HR due to the lack of suitable automated systems. We will continue to push for more equitable recruitment at University strategy meetings.

Meanwhile - we ask applicants to avoid mentioning information in their applications that could lead to biased evaluation, including demographic and cultural background, and photographs. Only mention information that is relevant to the job at hand (e.g., it’s fine to mention that you studied at the University of Amsterdam, but don’t mention that you are Dutch).

Selection panels typically consist of:

  • The hiring PI/promotor
  • Other collaborators in the project/supervision team
  • One junior lab member

The members of interview panels are anounced in the job vacancy.

Onboarding

When joining the lab:

  • Read this Lab Manual
  • Read the Code of Conduct
  • Talk to the MTO Secretariat (Anne-Marie van der Heijden) about
    • Your room
    • Your IT equipment
    • Your IT account
  • Request a key card from the Library
  • Install software described in Section 9
  • Subscribe to platforms mentioned in Section 9.2, or an open source alternative

Required skills

Basic skills:

  • Programming or scripting in R or Python
  • Writing in markdown
  • Version control with Git

Advanced skills:

  • Integration tests
  • Programming in C++
  • Writing in latex
  • Containerization

Completing Previous Work

In principle, starting a position at INSIGHT lab involves taking on specific responsibilities, which will fill the contractually agreed-upon hours.

In practice, it may be possible to negotiate the completion of previous work. Here are some possible approaches:

  • Starting with reduced hours
  • Connecting the previous work to your new tasks. Submit a proposal as for “starting a new study” (Section 12.1), but you may skip the validity check and ethical approval if it has already been given by another institution.

Roles and Duties

Roles

PI

  • Acquiring funding
  • Establishing research lines
  • Supervising lab members
  • Establishing collaborations
  • Giving feedback
  • Evaluating performance
  • Participating in scholarly communities, for example
    • Theory Methods Society
    • Paul Meehl Graduate School
    • IOPS Graduate School
    • Data Science Lab
    • Meta Lab

PhD student

  • Reading and synthesizing relevant literature
  • Conducting independent research under supervision and with feedback
  • Publishing approximately 1 paper per year, with the goal of 4 papers for a 4-year PhD project
  • Following agreed upon postgraduate education
  • Participating in scholarly communities, for example
    • Theory Methods Society
    • Paul Meehl Graduate School
    • IOPS Graduate School
    • Data Science Lab
    • Meta Lab
  • Networking with likeminded scholars, and exploring potential collaborations and future employment

Research Assistant

  • Performing well-defined research tasks
  • Reading assigned/agreed upon papers
  • Optionally contributing to papers as co-author (see Section 13.5)
  • Participating in scholarly communities, for example
    • Theory Methods Society
    • Paul Meehl Graduate School
    • IOPS Graduate School
    • Data Science Lab
    • Meta Lab

How we do Things

Lab Meetings

The lab meets monthly for a round table, to discuss ongoing research, present project proposals, and read relevant literature.

The first lab meeting of the academic year devotes time to lab-wide feedback sessions.

The last lab meeting of the academic year devotes time to discussing successes and failures.

Working Hours

Formally, every employee signs a contract for a specific number of hours - 38 hours/week for full time employment.

Informally, time management should be guided by predefined performance criteria, not by number of hours made.

It is a known problem that university employees work overtime; more than half of employees do so, on average 7.5 hours/week.

Some managers explicitly communicate the expectation of overwork to employees. At INSIGHT lab, we do not do this.

Remote Work and Attendance

Lab members are expected to work on location at least 2 days a week. The most important reason to do so is to nurture connections within the department and university. Therefore, try to prioritize days on which departmental social events (other lab meetings, guest lectures, MTO drinks) take place and attend the lunch meeting if possible.

Lab members are expected to be present in person for events organized by other lab members, including the lab meetings.

Establishing Collaborations

Lab members should discuss potential collaborations with their line manager/PI before committing.

In principle, the PI should be supportive. Reasons for advising against the collaboration include:

  • Concerns that the lab member is behind schedule on their prior agreed activities.
  • Concerns about the feasibility or quality of the proposed work
  • Reputation of the collaborating partner, their institution, or government

Supervisiory Meetings

  • Daily supervision: PhD’s should meet with their “daily” supervisor once a week (ironically).
  • Meetings with the full supervisory team take place once a month.
  • For a meeting to take place, a meeting agenda should be shared at least 24 hours before the scheduled meeting time.
  • Aim for at least one meeting a month for other collaborations.
  • If no agenda is shared, the meeting is assumed to be canceled.

Dealing with deadlines

Plan Backward

Step 1 - Determine the “external” deadline (e.g., journal submission, funder deadline, symposium submission).

Step 2 - Propose an internal deadline 1-2 weeks before the external deadline (longer for grants). The internal deadline is the “real” deadline.

Step 3 - Break the task up into distinct deliverables, e.g.:

  • Literature review
  • Study design
  • Analysis script
  • Data collection/cleaning
  • Results section
  • First draft

Step 4 - Block time in your calendar for the work. Schedule deadlines in calendar, add co-authors/supervisors

At each deliverable, and at least halfway between start and deadline, check if the project is on track.

If the project is not realistically on track at the halfway mark, discuss it with supervisors/co-authors.

Delays are normal. Silence is not.

If you anticipate missing a deadline:

  • Signal early - not at / after the deadline
  • Report the current status, remaining work, reason for delay, a revised timeline, and propose a solution if there is a bottleneck.

No Surprises

Do not:

  • Miss deadlines without prior communication
  • Explain after the fact
  • Be incommunicative

Unmanageable Deadlines

If work expands beyond what is feasible, these are valid solutions:

  • Reduce the scope of the project
  • Defaulting to a shorter paper format (e.g., brief report)
  • Splitting the work into two papers
  • Involving co-authors to do part of the work

Using “AI”

When most people say “AI” nowadays, they mean generative pretrained transformer models, e.g. ChatGPT. In our lab, we strive to use proper definitions instead; lab members are expected to specify the exact methods they mean (e.g., supervised machine learning, unsupervised machine learning, natural language processing, large language models). Certain classes of machine learning models (LLMs) that are used in generative pretrained transformer models are also objects of study in our work. Studying e.g. unsupervised clustering of theoretical concepts using vector embeddings derived from an LLM is conceptually distinct from using generative pretrained transformer models to assist with scientific activities like writing, coding, or literature review.

When such tools are used to support research activities, their role must be transparent, appropriate to the task, and aligned with open science and reproducibility standards. Clear distinctions between studying algorithms, applying algorithms, and using generative tools are essential to maintaining conceptual clarity and methodological integrity.

This section governs the use of “AI” tools in scientific activities. While there is no blanket ban on the use of AI tools in our lab, consider the following:

  • Is it clear and transparent how the tool works? In order to justify using the tool for scientific work, we must know how it operates.
  • What is the goal of the work you’re doing? In many cases, the goal is not to get the work done - the goal is learning how to do the work. For example, a PhD is expected to graduate with an academic level of competence in writing, coding, data analysis, and research skills. Outsourcing these tasks to an algorithm does not serve the purpose of learning those skills. If you are not able to complete your work without outsourcing it, the timeline and scope of your project should be adjusted accordingly, not the method by which you achieve it.
  • Is the AI tool ethically trained? Just like we would not resort to sweat shop labor or copyright infringement to obtain research data, we should not use tools that use these methods to train an algorithm.
  • Is the tool validated for the task you wish to assign it? Every tool that is used in scientific work must be validated; this includes measurement instruments, statistical methods, and of course, “AI tools”.

Writing

Generative AI may never be used for original writing, including papers, presentations, applications, and peer review.

Generative AI may be used to suggest minor edits, like reducing word counts or simplifying language. The results should be thoroughly proof-read and edited.

Code/Software

The research suggests that AI coding assistants do not save time and introduces bugs and vulnerabilities. At this point, you should not expect to save time or get acceptable results from AI coding assistants:

Fu, Y., Liang, P., Tahir, A., Li, Z., Shahin, M., Yu, J., & Chen, J. (2024). Security Weaknesses of Copilot Generated Code in GitHub (arXiv:2310.02059). arXiv. https://doi.org/10.48550/arXiv.2310.02059

Moradi Dakhel, A., Majdinasab, V., Nikanjam, A., Khomh, F., Desmarais, M. C., & Jiang, Z. M. (Jack). (2023). GitHub Copilot AI pair programmer: Asset or Liability? Journal of Systems and Software, 203, 111734. https://doi.org/10.1016/j.jss.2023.111734

Zhong, L., & Wang, Z. (2024). Can ChatGPT replace StackOverflow? A Study on Robustness and Reliability of Large Language Model Code Generation (arXiv:2308.10335). arXiv. https://doi.org/10.48550/arXiv.2308.10335

Code written in our lab must conform to best practices and style guides, see Section 12.6.

Suggested alternatives:

  • Read the documentation
  • Copy-paste a solution from StackOverflow
  • Ask a colleague
  • Involve a co-author with the required programming skill

Transcription

Speech to text is at a high risk of GDPR violations, especially for tools that join online meetings and stream the audio/video to servers that might be based outside the EU.

Suggested alternatives:

  • Ask someone to keep minutes
  • Use a local speech-to-text model

Literature Review

Note that there is a key distinction between search and summarization.

A search should typically be unbiased, see:

Staaks, J. (2019). Systematic Review Search Support. https://doi.org/10.17605/OSF.IO/49T8X

AI tools are not unbiased, and are thus a poor starting point for a literature search.

Other methods may be used to identify overlooked sources, this is where AI-based tools may come in.

Specific models may be useful for summarizing text; there is a distinction between extractive summarization (retain only the most important sentences in a text based on a ranking algorithm) and abstractive summarization (identify the meaning of text and reproduce it in shorter form).

The former approach stays true to the source material, although context may be lost.

The latter approach, being generative, can introduce errors.

Fully generative approaches, like ChatGPT, are not bound to stay true to the source material.

Suggested alternatives:

  • Skimming/targeted reading: Search the paper for specific information, using general knowledge of how papers are structured as guidance, and of course, searching the text for keywords
  • Extract highlighted text from papers in Zotero to make your own summary

Summarization

Using generative pretrained transformer models for summarizing text is not a good idea, because of knowledge bleed: the resulting summary is a mix of the original text in the prompt, and other information encoded into the model itself. Here is an excellent article from TU Eindhoven’s library explaining why this is problematic: https://www.tue.nl/en/our-university/library/library-news/24-02-2026-are-ai-generated-summaries-suitable-for-studying-and-research

Tools and Platforms

Tools

These are important tools:

  • A note taking app that continuously stores work, e.g.:
    • Notepad++
    • Obsidian
  • An IDE, like
    • RStudio
    • Visual Studio
  • Python
  • Anaconda, Miniconda or Mambo
  • Git (command line)

Online accounts

There are important online services:

Communication

Internal Communication

Please use university communication channels (email and teams). In the future, we may switch to open source/European alternatives, feel free to propose using these.

Check your university Outlook and Teams accounts at least once per working day.

To avoid creating an expectation that lab members are “always working”, please don’t use non-work related communication channels outside of official work hours unless previously agreed upon. For example: send an email instead of a WhatsApp message if it’s after 17:00 or on the weekend.

If you receive work-related communication outside of work hours, there is no obligation to respond.

To manage expectations about response times, please use an autoresponder when you are out of office for e.g. a holiday.

Lab members may opt in to communication outside of work channels (e.g., Signal, WhatsApp) for urgent matters. Do not communicate sensitive content via these channels.

Management Updates

Lab members will be kept informed about lab developments information via email. Relevant information includes spending, grant applications, and outlook for the future. Lab members will be involved to provide feedback on grant applications and funding considerations that affect the future of the lab.

Settling Disagreements

Horizontal disagreements between lab members may be brought to the PI for mediation. If this does not lead to resolution, HR will be requested to mediate.

In disagreements involve the PI, HR will be requested to mediate.

Communication with collaborators

Lab members should take the initiative to communicate with active collaborators at least once a month.

If the PI is involved in the collaboration, then CC them in email communication (even if no reply is expected).

Public Relations

Lab members are encouraged to communicate about their area of expertise, including about recent publications/releases, via social media, regular media, invited talks, et cetera.

Lab members are encouraged to boost (share) each other’s releases.

It is recommended to share content on multiple platforms.

It is recommended to tag the lab and involved lab members in posts.

Preferred outlets are:

  • linkedin.com
  • bsky.app
  • mastodon

Health & safety

See Tilburg University Intranet.

You can use My Employee Portal to report when you are sick (physical/mental).

The university offers trainings on work/life balance.

Research Workflows

Starting a New Study

  1. Propose a study by submitting a proposal to the supervisory team (for PhD students) or to your line manager and intended co-authors (for postdocs+), who can approve the project
  • A study proposal comprises:
    • Intended co-authors
    • 250 word introduction (including 3+ references)
    • Research question
    • (for confirmatory studies) hypothesis
    • Planned method for addressing research question/hypothesis
  1. Checking feasibility and validity
  • Conduct an informal literature review, look for similar works and works that support a knowledge gap
  • Present the study proposal at a lab meeting (local or another lab)
  • Ask two independent colleagues for feedback
  1. If the project involves analysis of human participant data, obtain ethical approval submit a request to the Ethics Review Board via the G.E.D. Started app: https://www.tilburguniversity.edu/research/ethics-review-boards/tsb
  2. Draft a timeline for the planned work.

Data Management Plan

Creating a data management plan:

Reproducibility

The lab’s approach to reproducibility is documented in these papers:

Van Lissa, C. J., Brandmaier, A. M., Brinkman, L., Lamprecht, A.-L., Peikert, A., Struiksma, M. E., & Vreede, B. M. I. (2021). WORCS: A workflow for open reproducible code in science. Data Science, 4(1), 29–49. https://doi.org/10.3233/DS-210031

Peikert, A., Van Lissa, C. J., & Brandmaier, A. M. (2021). Reproducible Research in R: A Tutorial on How to Do the Same Thing More Than Once. Psych, 3(4), 836–867. https://doi.org/10.3390/psych3040053

Theory Specification

The lab’s approach to theory specification is documented here:

Van Lissa, C. J., Peikert, A., Ernst, M. S., van Dongen, N. N. N., Schönbrodt, F. D., & Brandmaier, A. M. (2026). To Be FAIR: Theory Specification Needs an Update. Perspectives on Psychological Science, 17456916251401850. https://doi.org/10.1177/17456916251401850

Developing Research Materials

  • At least one other lab member checks these materials for mistakes, and functionality issues (e.g., randomization for any test materials).
  • Optionally, attend a code café or request a code review to obtain feedback

Developing Software

  • First, check if you can contribute to existing software from within the lab
  • Then, check if you can contribute to other existing open source software
  • If not, then propose the new software to your line manager in a similar way as Starting a New Study.
  • Follow the FAIR software principles
  • Follow the requirements of the intended repository
  • Adopt a standard for best practices in software development, e.g. OpenSSF Best Practices

Lamprecht, A.-L., Garcia, L., Kuzak, M., Martinez, C., Arcila, R., Martin Del Pico, E., Dominguez Del Angel, V., van de Sandt, S., Ison, J., Martinez, P. A., McQuilton, P., Valencia, A., Harrow, J., Psomopoulos, F., Gelpi, J. Ll., Chue Hong, N., Goble, C., & Capella-Gutierrez, S. (2019). Towards FAIR principles for research software. Data Science, 1–23. https://doi.org/10.3233/DS-190026

Confirmatory Research

All confirmatory studies should be preregistered as a “Preregistration-As-Code”, that is to say: a reproducible document comprising a draft manuscript, with results generated from synthetic data.

After preregistration, the data is collected/accessed/unscrambled, and the reproducible document is recompiled. Any necessary changes are made and logged (e.g., using Git commits). Then, the Discussion is written.

Registered reports are preferred for preregistered studies, to increase the probability of successful publication.

Exploratory Research

Define the scope of exploration before starting.

Reporting Results

General guidelines:

  • Report the results comprehensively in a reproducible dynamic document (e.g., Rmarkdown, Quarto, Python notebook)
  • Reference the comprehensive results in the paper, if applicable
  • Check for relevant reporting guidelines and follow these, if applicable

Preparing Plots

  • Use vector graphics, not raster graphics
  • If publishing plots in print, make sure that the publisher has a vector graph. They usually accept only EPS format
  • Consider Wilkinson’s Grammar of Graphics as a general philosophy on how to compose plots https://en.wikipedia.org/wiki/Wilkinson%27s_Grammar_of_Graphics
  • Common sense principles:
    • Consider whether a plot is needed; if a table is simpler, use a table
    • Consider the measurement level of variables; nominal/ordinal variables should not be plotted in a way that implies continuity, e.g., by connecting points with lines.
    • Consider overplotting
    • Make plots print-friendly (avoid reliance on color is it is not required), make plots accessible, e.g., consider color-blind friendly fonts

Co-Writing

Avoid emailing files back and forth; instead, maintain a version of record in a plain text format, in a repository where access can be controlled (e.g., GitHub).

The preferred mode of collaborative writing is via non-destructive edits:

  • Pull requests on GitHub
  • Suggested changes on Google Docs

The preferred reference manager is Zotero.

Proof-reading

All lab members share responsibility for manuscript they are co-authors on.

At minimum, they are required to proofread any manuscript they are a co-author on at least once.

During this proofreading, they should raise any remaining concerns (factual, ethical, stylistic, et cetera).

The lead author is not required to address all issues, but should make a good-faith effort to do so. Disagreement should be documented (e.g., in email history, commit history).

Authors may ask for specific kinds of feedback desired, but this does not override the requirement to raise all relevant issues (see above).

Lab members may ask each other for proofreading without this resulting in an automatic co-authorship. Lab members should reciprocate proofreading each other’s manuscripts.

Output

When is Work Done?

Work is done when its final incarnation has been approved by all involved co-authors and is archived in a FAIR-compliant repository.

All output that is not an academic paper should include a README file, describing how others can use it.

Licensing

Include a license with all resources created.

For guidance on what license to use, see https://choosealicense.com/

Recommendations:

  • CC-BY for papers
  • MIT for code, optionally GPL (>= 3) for R-packages
  • CC-0 for theories, but with request for citation

Preferred Journals

We developed a list of potential journals to publish in, based on considerations such as the relevance of the manuscript to the journal, journals that use the registered reports publishing format, publisher copyright and self-archiving policies (prioritised by the freedom to provide public access to pre-reviewed manuscripts).

  • Research Synthesis Methods
  • AMPPS
  • Psychological Methods
  • Behavioral Research Methods

Crediting Contributions

Be comprehensive and generous in crediting contributions. Nothing is lost by giving people fair credit.

For published work, report each author’s contribution according to the CRediT taxonomy: https://credit.niso.org/

When planning the work, fill out a CRediT taxonomy to define intended roles at the start

If project roles and levels of contribution change, update the CRediT taxonomy accordingly.

Before publishing a version of the manuscript (e.g., as preprint or in a journal), check that all authors recognize the CRediT taxonomy as accurate representation of the work done.

Student Contributions

Several students collaborating with the PI of the INSIGHT lab have been credited as co-authors on papers they made substantial contributions to.

Expectations regarding credit for student contributions will be discussed with students at any level and in any capacity (be it thesis students, interns, research assistants, etc.).

Below are guidelines for student co-authorship; these are derived from Merz, Strahmann, & Frank (2025). CRediTs for Students? - Promoting Student Authorships in Psychology. https://doi.org/10.17605/OSF.IO/TYGSP

A thesis is a scientific work and scientific results should, whenever possible, be made accessible to the public. In principle, you can be involved in a publication as a (co-)author. An essential prerequisite for this is your willingness to invest additional work in the publication in addition to writing the thesis. A thesis is usually not yet ready to be submitted directly to a journal, and there is a substantial amount of work that needs to be done between submission and publication. As is the case with collaborations between scientists: co-authors are expected to take responsibility for the work throughout this process, and are ethically responsible for the final result.

The co-authors of a publication will be (at least) the supervisor and student, along with potential collaborators. All co-authors are jointly responsible for the publication. The order of authors is determined by their relative contribution, unless otherwise agreed upon.

In the event of disagreement about the order of authors, a third person should be consulted for mediation. It is recommended to involve the Student Council. First authorships by students are required if they have clearly done the main work on an academic paper.

Any thesis may only be published in consultation with the supervisor(s), as they also bear responsibility for the work and its quality.

Deviations from the procedures described here require written agreements.

Student Authorship Guidelines

Authorship is required if at least one of the following points applies:

Example: Independent and complete conceptualization, implementation, analysis and documentation of a publication (e.g. as part of a thesis or the research-orientedinternship)

Authorship must be discussed case-by-case, but at least a mentioning in the Acknowledgements is expected if:

Authorship does not have to be discussed, Acknowledgements can be offered if:

In cases of disagreement: If there are disagreements regarding authorship, the student should approach the Student Council and initiate a discussion with reference to this document. This discussion then consists of the student concerned, the supervisor, a member of the Student Council and an independent person from the ranks of the academic staff, who is sought by the Student Council.

Open access

All lab output must be openly accessible. This requirement can be met by publishing a preprint, e.g. on https://psyarxiv.com/, or by making use of the Taverne Amendment to archive the paper in the institutional repository 6 months post publication, see https://www.openaccess.nl/en/policies/open-access-in-dutch-copyright-law-taverne-amendment, or by publishing in an open access journal or Peer Community In.

The priority for publication is:

  • Diamond open access journal
  • Gold open access via a journal covered by an institutional agreement, see openaccess.nl
  • Archival of a preprint or post-publication archival under the Taverne Amendment

Conference attendance

Each lab member has a conference budget of at least XXX/year.

It is expected that lab members attend at least one conference per year.

Please take climate concerns into account when planning conferences. If the choice is between two equally relevant conferences, and one can be reached by train whereas the other requires flying - consider prioritizing the one that can be reached by train.

To attend a conference, the lab member must present at the conference.

Making Mistakes

How to deal with mistakes

Making mistakes is an inherent part of research work (indeed, of being human). Our approach to reproducibility and open science intends to make it easier to audit one another’s work, to hopefully catch mistakes early and work in a way that makes it easy to propose and implement fixes.

Nevertheless, mistakes do occur.

If you detect a mistake, please take the following:

  1. Inform all involved co-authors
  2. If you can fix the mistake, propose/implement a fix
  3. If you cannot fix the mistake, ask for help to your co-authors or the responsible PI
  4. If the mistake credibly has implications beyond the lab (e.g., it’s in published code or other research output, or a published paper), take appropriate steps:
    • For code, prepare an update, document the mistake in the release notes
    • For other research output, prepare an update (e.g., on Zenodo), document the mistake in the release notes
    • For a paper: update the preprint, discuss potential retraction with the co-authors and/or editorial team, mention the mistake when citing the paper in future work

Caspar’s philosophy is: “I am guaranteed to make mistakes. The best I can strive for is to not make the same mistake twice.”

Data Leaks / Confidentiality Breaches

As required by law, Tilburg University has policy regarding data leaks and confidentiality breaches. See this website for more information.

Leaving the Lab

  • Please archive all your ongoing work as described in Section 13
  • We will schedule an exit interview on your last day (or before, if necessary)

Reference Letters

Lab members may request reference letters from their supervisor/line manager/PI at least 2 weeks in advance of the deadline. In principle, supervisors should comply with such requests - time permitting. Lab members should expect fair but honest evaluation of their performance and skills in the reference letters provided.