Seminar SCALab

Lucie Charles (Queen Mary University of London): 

How does explicit reliability guide choices? Effect of trustworthiness on choice and metacognition 

Can we successfully ignore information we deem irrelevant or unreliable? And can we become aware of how such information can bias our choices? The ability to introspect and evaluate the factors influencing our decisions constitutes a key metacognitive function but remains poorly understood. In this talk, I will present studies on how people integrate explicit reliability of information to make choices, testing their ability to disregard unreliable information. Computational modelling revealed that participants distorted the stated reliabilities of information sources rather than using them at face value:  even unreliable or irrelevant sources continued to influence choices, while reliably wrong sources impaired performance. By comparing participants’ influence reports to model-derived estimates of each piece of information’s causal contribution to choice, we show that participants had partial metacognitive access to these biases. However, this awareness did not translate into successful compensation: participants could detect some unwanted influences but remained unable to fully correct for them. These findings reveal a dissociation between metacognitive sensitivity and metacognitive control, and suggest that computational measures of influence provide a useful framework for studying introspection in decision-making.