Methods for a fully online automated cognitive training study on Prolific
Abstract
Intervention studies are the current gold standard when investigating the causal link between an intervention (e.g., physical activity, cognitive training, meditation, action video games) and its impact on cognitive functions. Such studies are resource intensive, especially when conducted to the latest standards in the field. Recently, it has been noted that the development of online tools to conduct such studies may significantly reduce resource demands, and thus allow more of these acutely necessary studies to be carried out. Here we present a series of tools to conduct intervention studies in a fully online fashion such that participants may go through the entire experimental pipeline without any contact with the experimenters. In particular this included Prolific for participant recruitment and management, the implementation of a pseudo-randomized group assignment procedure such that groups are matched at pre-test, and the development of various dashboards for experimenters and participants to follow their progression throughout the pipeline. These tools were implemented in a 12-h mechanistic cognitive training study where participants completed the training and pre- and post-test assessments remotely over multiple weeks. This new digital pipeline allowed us to limit the resource demands, implement strong masking practices, recruit a sample more diverse than the usual WEIRD in-laboratory samples, and complete the study in less time than usually needed.
Study specs
Participants were recruited via Prolific, assigned to groups using a pseudo-randomized procedure, and completed a 12-hour remote cognitive training study with pre- and post-test assessments monitored via custom dashboards.
- Authors
- F Joessel,S Denkinger,PE Joessel,CS Green
- Institution
- Max Planck Institute,University of Potsdam,University of Maryland,University of Zurich,University of Arizona
- Discipline
- Psychology
- Study Type
- methodology
- Year
- 2025
- Human Data Platform
- Prolific
- Source
- View Source Google Scholar
Measured Outcomes
Impact of a 12-hour cognitive training intervention on participants' cognitive functions, conducted in a remote and automated manner.
Peer Review & Critical Discussion
Potential Selection Bias in 2023 Cohort
The participant pool shows a concerning overrepresentation of users from high-income demographics. Looking at Table 3, we can see that 78% of respondents had annual incomes above $75k, which significantly limits the generalizability of these findings to broader populations.
Non-naive Participants Issue
I've noticed a methodological concern regarding participant naivety. Given that Prolific users often complete multiple studies, there's a real risk that participants had prior exposure to similar experimental paradigms, which could confound the results.
RLHF Applicability to This Study Design
The implications for RLHF training pipelines are understated. If we accept the authors' conclusions about preference stability, this has direct consequences for how we should structure reward model training. The temporal decay effect described in Section 4.2 is particularly relevant.
Verify your expertise to join discussion
Create an account and verify your credentials to participate in peer discussions.