In this project through a series of large-scale online and smartphone-based experiments, we investigated how imprecisions during learning (i.e. learning noise) and stochasticity in the choice process accounted for compulsivity and impulsivity. We showed that impulsivity was tied to imprecision in learning across reward and punishment learning domains, while choice stochasticity was more linked to compulsivity. In other words, when making a decision, a compulsive individual changed their mind and explored new options, while an impulsive behaviour was rather driven by erroneous updates of the received information. These findings demonstrate that distinct neurocomputational mechanisms can drive seemingly similar behavioural deficits, only dissociable using targeted computational approaches.
In this project, we pioneered new approaches to gather large scale cognitive data. In particular, we relied on a citizen science approach, which encourages the general population to partake in research. By using gamification, we further strengthened this new methodology and pioneered it in computational psychiatry. In forthcoming articles, we further discuss the advantages and drawbacks of such approaches and their implications. We believe that such approaches will be critical for the field.
Finally, we investigated how such processes were affected by the Covid-19 pandemic. In a longitudinal study, we investigated how compulsivity was linked to Covid-19 related behaviours. We found that in contrast to depression symptoms that seem to diminish during the ease of Covid-19 restrictions, the obsessive-compulsive symptoms OC further increased and were contributing to better adherence to governmental guidelines.
Together these results advance our understanding of psychiatric traits and their behavioural and underlying computational origins. By pioneering novel approaches, our research opened up new avenues for future research.
