Synopsis

Navigation is one of the most intensively studied topics in systems neuroscience. Indeed, there have been groundbreaking results about the neural bases of this challenging and ecologically highly relevant computation that essentially all animals need to perform. However, despite this rich tradition, there has been a notable paucity of studies focussing on a key, but under-appreciated aspect of navigation: uncertainty. 

It is obvious to anyone who has ever tried to find their way in a foreign city that navigation is plagued by uncertainty. Furthermore, engineered systems for efficient navigation, from mobile phone apps to autonomous vehicles and Mars rovers, all make strategic use of uncertainty when tracking the relevant variables (such as location, head direction, or velocity). Thus, it is all the more surprising that while we have detailed knowledge about how key navigational variables are represented and computed in the brain, we know next to nothing about how uncertainty about these variables manifests in neural computations. Moreover, most if not all canonical neural-level models for how our brains track and compute with navigational variables lack any sense of uncertainty. Conversely, while studies of neural computations under uncertainty have been very successful in explaining a plethora of behavioral and neural phenomena in the cognitive domains of perception and decision making, they have barely addressed the domain of navigation. The goal of our workshop is to articulate these issues, and provide the necessary momentum and shared understanding with which researchers in the field will be able to begin to fill this gap at the intersection of navigation and uncertainty.

Our workshop will bring together experimentalists working with different species (humans, non-human primates, and rodents), and theoreticians using different approaches (Bayesian, reinforcement learning, neural networks), interested in navigation, and specifically the role of uncertainty in navigation. In particular, we will integrate recent, convergent results in the field that speak to this topic. We will put particular emphasis on discussing experimental studies whose original framing did not consider navigational uncertainty explicitly but whose results may still reveal important signatures of its neural underpinning when interpreted within an appropriate theoretical framework. We will also discuss combined experimental-theoretical strategies that could more directly test these signatures going forward. 

Venue