hierarchical structure learning & metacognition

• How do we learn multiple spatiotemporal patterns and structures in our environment without external feedback? 

• What role does metacognition (evaluating one's own cognitive processes) play in their learning?

• How do structure learning and metacognition evolve across time? 

We built a novel behavioral paradigm to investigate the extent to which the temporal dynamics of hierarchical structure learning under uncertainty rely on metacognition. We also look at the role of stimulus familiarity in these processes - sneak peak: stimuli types really matter!  Check out my preprint with Dr. Megan Peters! 

Using Successor Representations (a type of reinforcement learning framework), we ask: 

• How do we represent information when having to learn multiple structures simultaneously? 

• How do representations and usage of our representations in decisions interact with metacognitive processes? 

This recent talk summarizes our behavioral and computational findings! Stay tuned for more details about ongoing work.