Integrating signals for perceptual decision

Kristine Krug (Physiology, Anatomy and Genetics, Oxford University, UK)

Abstract: Neural firing of single brain cells can directly be linked to visual perception or decision-making. This output firing pattern of a brain cell is shaped by a range of inputs at different stages of processing. Quantitative models have been employed to characterize the mechanisms by which such inputs affect brain activity and ultimately behaviour. One example is how expecting a reward affects decisions animals and humans make. When judging the appearance of a visual stimulus, the more rewarded interpretation is chosen more often. There is a long-standing debate about whether this is due to a bias in reporting or reflects a real change in the individual’s perception. We have investigated the interactions between reward, visual information and an artificial electrical signal in shaping perceptual decisions in the visual cortex of macaque monkeys. To discern which stages of brain processing are influenced by reward, a drift diffusion model was employed. Our results strongly suggest that expected reward does affect visual processing, not just the animal’s behavioural report or preparation for making it. Using this example from my own research, I will discuss some of the challenges we face when trying to build realistic models of brain mechanisms for cognitive function.