Keynote speaker

Eran Eldar is a Research Associate at University College London, based at the Wellcome Trust Centre for Neuroimaging and the Max Planck UCL Centre for Computational Psychiatry and Ageing Research. Eran studies how people differ from one another in the algorithms their brain implements to solve fundamental decision-making problems, and how these differences manifest in normal and pathological real-life behavior. To do this, Eran develops computational models of cognitive and neural processes and tests these models against behavioral, neuroimaging and physiological data that he collects in and outside the lab. Eran received a PhD in Neuroscience from Princeton University in 2014, where he was funded by the Howard Hughes Medical Institute. Prior to that, Eran earned a Medical Doctorate from Tel Aviv University and a BSc in Creative Computing from the University of London.

Abstract
I propose that changes in pupil diameter track brain-wide levels of neural gain, which control the balance between broadly integrative information processing and narrowly focused selective attention. Neural gain, which is thought to be modulated throughout the brain by the locus coeruleus-norepinephrine system, determines how strongly neurons respond to input signals. When gain is high, both excitatory and inhibitory signals have greater impact, and as a result, competitive interactions between different neural representations increase, such that weak representations are further inhibited while strong representations become more dominant. In such a state, we expect processing to be more narrowly focused on the most strongly represented sources of information. In contrast, low gain may allow simultaneous representation, and thus integration, of a broader range of sources of information. In this talk, we will first examine the whole-brain effects of neural gain using functional connectivity and graph-theoretic analyses of neuroimaging data. The results reveal signs of brain-wide fluctuations in gain that are tracked by pupillometric indices, and show that high gain is coupled with a focusing, clustering effect on neural interactions throughout the brain. I will then present four behavioral experiments designed to investigate the effects of variations in gain on information processing. In the first experiment, I will show that pupillary and neuroimaging indices of high gain are associated with learning that is more narrowly focused on particular types of stimulus features, in accordance with individual predisposition. The second experiment will show that high gain has a similar effect on perception of an ambiguous stimulus, making it more focused and less integrative, and that the effects of gain do not have to be tied to individual predisposition, but rather, they can be flexibly manipulated by means of subliminal priming. The third experiment will show that with high gain, memories also become more specific to attended aspects of experimental stimuli. In the fourth experiment, I will show that the reduced integration that is associated with high gain comes with a benefit – weaker susceptibility to classical decision making biases. Finally, I will propose a Bayesian account of the presented results that highlights the effects of gain on inference.