Date: Friday, September 25, 2020
Time: 12:00pm – 1:00pm
Location: Zoom Webinar – Registration Required
Register in advance for this webinar: click here
* After registering, you will receive a confirmation email containing information on how to join the webinar.
Speaker: Hiroki Sugihara, Ph.D.
Affiliation: Research Scientist, Mriganka Sur Laboratory, Picower Institute for Learning and Memory, Department of Brain and Cognitive Sciences, MIT
Talk title: Exploring Temporal Expectation in Marmosets
Abstract: Atypical temporal prediction or expectation has been suggested as one of phenotypes in autism spectrum disorder. Here, we used marmosets as a next generation model animal for exploring mechanisms of cognitive dysfunction in neurodevelopmental disorders. We implemented a simple timing task to investigate whether marmosets form an internal model of task timing and modify their behavior accordingly. Using an unrestrained home-cage setup, marmosets were trained to make a timed response prompted by a visual stimulus change with a uniform distribution of stimulus durations. We and others have shown that in macaques and humans, reaction time measurements follow the hazard rate of stimulus change whereby longer trials elicit a faster reaction time. By investigating reaction times for marmosets that were trained in our task, we found that like humans and macaques, marmoset reaction times followed the hazard rate. In addition, our experiments revealed an unexpected effect of trial history on performance. As marmosets learned the task, their reaction time was also modulated by the immediate temporal context, such that the reaction time on a given trial was modulated by the duration of the previous trial and reaction time. This effect emerged over time and with continued exposure to the task. The effects were well described by a multiple regression model, and computationally by Bayesian updating of the hazard function. These results, together with comparable results in humans, demonstrate the evolution of dynamic internal models of temporal expectation which are shaped by global as well as local temporal task structure. Our findings, and the quantitative measures we describe, can be used to assess atypical temporal prediction and its underlying neural processes in autism spectrum disorder.