Speakers
Overview
This session focused on decision-making research outside the lab and the practical tradeoffs that come with large, naturalistic datasets. Speakers discussed how scale and ecological validity can reveal robust effects and individual differences, while also magnifying noise and data-quality challenges. Two talks used Healthy Brain Study data to show how computational modelling can extract motivational parameters that relate to real-life wellbeing, beyond standard questionnaires and symptoms. A third talk presented a large-scale MuZIEum inattentional blindness study, emphasising design choices such as embedded internal controls, transparent exclusion pipelines, and appropriate statistical modelling.
Key Points
• Outside-the-lab data can deliver power and diversity that are difficult to achieve in typical lab samples, but requires careful attention to comprehension, exclusions, and data quality.
• Motivational parameters derived from decision tasks captured variance in real-life wellbeing, with punishment sensitivity predicting higher negative affect and lower positive affect, and reward-rate sensitivity predicting the opposite pattern once anxiety and depression were accounted for.
• Cross-task computational phenotyping was presented as a way to estimate shared motivational processes beyond task-specific noise, improving sensitivity and generalisability for mechanism-driven measurement.
• The MuZIEum inattentional blindness dataset illustrated both the promise of truly naïve participants at scale and the need for internal validity checks, preregistered cleaning decisions, and regression or hierarchical approaches.
• Open question: how can we design short field tasks that retain strong internal controls while still benefiting from ecological realism and scale?
Next Steps
Possible collaborations: combining field-style data collection with theory-driven modelling pipelines, and sharing design templates for short tasks with embedded control trials and preregistered exclusion criteria.
If you are interested in field-based decision research and would like to connect with the speakers or suggest a speaker for another session, contact Catalina.