Inferring Latent Features in Noisy, Dynamic Environments

The world changes constantly in ways that shape our decisions, but how do we learn the latent dynamics of our environment and use them to guide choices? Tahra Eissa presents recent evidence on how humans use latent environmental dynamics to modulate decision-making behaviour. Humans show systematic biases in their inference that can be tied both to the structure of the environment and to how relevant information is generalised, suggesting that people's ability to learn environmental dynamics is not uniform. The talk then extends this theoretical framework to how latent environmental inference may be shaped by social interactions, giving a basis for understanding when latent-feature inference is most feasible and valuable. Part of the NITMB Seminar Series, which brings mathematicians and biologists together to foster discussion and collaboration between the two fields; talks are held in person at NITMB with the ability to join remotely via the Zoom webinar link on the event page.