| Popis: |
To navigate our complex social world, it is crucial for people to deploy multiple learning strategies, such as learning from directly experiencing the outcomes of one’s actions – experiential learning (EL) – as well as learning from observing the behavior of other people – observational learning (OL). Despite the prevalence of EL and OL in humans and other social animals, two fundamental questions remain unaddressed: how is control over behavior assigned to one strategy over the other depending on the environment? And how do individuals vary in their strategy use? Here, we describe an arbitration mechanism in which the prediction errors associated with each learning strategy influence their weight over behavior. We designed an online behavioral task to test our computational model, in which the uncertainty of each strategy is manipulated, resulting in dynamic variations in prediction errors. Model comparisons revealed that a substantial proportion of participants rely on our proposed arbitration mechanism, but we also found some meaningful heterogeneity in how people solve this task, with four other groups identified: those who use a fixed mixture between the two strategies, those who rely on a single strategy (EL or OL) and non-learners who perform an irrelevant strategy. Furthermore, the groups were found to differ on key behavioral signatures, and on transdiagnostic symptom dimensions relevant to psychopathology, in particular autism traits and anxiety. Together, these results demonstrate the benefits of leveraging large heterogeneous datasets using computational methods to better characterize individual differences and pave the way toward individualized diagnoses and treatment. |
