Co-pilot system

To reduce potential errors and increase the reliability of our findings, the lab has implemented a ‘co-pilot’ system12:

  • By default, there is a lead researcher (or ‘pilot’) and a ‘co-pilot’, who is involved in most parts of the workflow.

  • The co-pilot is typically assigned at the beginning of the project.

  • The ‘co-pilot’ researcher not only double-checks the data analysis scripts of the ‘pilot’ researcher, but is can also be involved in other tasks, from the conception of a research idea until the publication of the manuscript (e.g,. double-coding a certain percentage of videos).

  • The co-pilot does not necessarily have to be a lab member; for example, when no one of the lab members has the skills that are required for the lead researcher’s project, the lead researcher may look for an appropriate co-pilot in other labs.

The co-pilot system thus serves two goals: it is a quality control system and facilitates mutual learning. The system is described in more detail in Reimer et al. (2019), and more detailed guidelines can be found on the lab’s Github repo.

Sometimes, lab members might have reasons for opting out, but they should discuss this first with Dr. Verbruggen.

Footnotes

  1. Wicherts, J. M. (2011). Psychology must learn a lesson from fraud case. Nature, 480(7375), 7–7. https://doi.org/10.1038/480007a↩︎

  2. Reimer, C. B., Chen, Z., Bundt, C., Eben, C., London, R. E., & Vardanian, S. (2019). Open Up – the Mission Statement of the Control of Impulsive Action (Ctrl-ImpAct) Lab on Open Science. Psychologica Belgica, 59(1), 321. https://doi.org/10.5334/pb.494↩︎