Abstract: We have the capacity to (1) understand and represent the value of the actions available to us and (2) to perform the computations needed to identify the most rewarding ones. However, we do not always make these best choices, even when there is no conceivable benefit to doing otherwise. Why do we make mistakes? In this talk, I will describe ongoing work that suggests that at least some mistakes emerge from fundamental constraints in neural hardware. We find that there are limits on both the stability and accuracy of neural representations that can explain at least some of our mistakes. While mistakes are certainly costly, the constraints that cause them may confer surprising advantages: they seem to increase the efficiency and flexibility of decision-making over longer time scales, even as they produce the occasional misstep in the moment.
ZOOM Passcode: Cogs595
Noise stands for Neural Motifs, Internal States, and Evolution. This backronym describes our goal (discovering the neural motifs underlying cognition), our approach (studying variability across internal states), and one of our philosophical commitments (that the brain is the product of evolution and must be understood in its ecological context). Noise itself is just a powerful mechanism for discovery and learning. We're interested in how goals, beliefs, expectations and even arousal shape how we see and interact with the world. We study (1) how these internal states change the way we transform sensation into action, and (2) how we adjust our internal states in order to achieve different goals. Our work shows that we can generate the same sensorimotor transformation in very different ways, depending on why we're doing it (e.g. 1, 2).