Event Information


RIKEN BSI Seminars & Forums


This event is free and open to the public.
こちらのイベントは理研BSI職員以外の方にも開放されています。参加費は無料です。
Contact BSI | お問い合わせはこちら

Forum: Representation of Confidence Associated with a Decision by Neurons in the Parietal Cortex
Speaker Dr. Roozbeh Kiani, Howard Hughes Medical Institute and Department of Neurobiology, Stanford University School of Medicine, Stanford, USA
Date/Time Thursday, April 22, 2010 11:00am - 12:30pm
Place BSI East Bldg. 1F Seminar Room
Abstract

The degree of confidence in a decision provides a graded and probabilistic assessment of expected outcome. Although neural mechanisms of perceptual decisions have been studied extensively in primates, little is known about the mechanisms underlying choice certainty. In my talk, I will show that the same neurons that represent formation of a decision encode certainty about the decision. We trained two rhesus monkeys to make decisions about the direction of moving random dots at different levels of difficulty. The monkeys were rewarded for correct decisions. On some trials, after viewing the stimulus, the monkeys could opt out of the direction decision for a small but certain reward. Monkeys exercised this option in a manner that revealed their degree of certainty. Neurons in parietal cortex represented formation of the direction decision and the degree of certainty underlying the decision to opt out.

Longer abstract:
The degree of confidence in a decision provides a graded and probabilistic assessment of expected outcome. Although neural mechanisms of perceptual decisions have been studied extensively in primates, little is known about the mechanisms underlying choice certainty, in part because it is difficult to measure. In a direction discrimination task, where monkeys report decisions about direction by saccadic eye movements, lateral intraparietal (LIP) neurons contribute to decision formation and the representation of the final choice. I will describe our recent experiments which demonstrate that the same LIP neurons also encode certainty about the decision. We trained two rhesus monkeys to discriminate opposing motion directions for random dot stimuli. Task difficulty was controlled by changing stimulus duration and strength. The monkey indicated its decision by making a saccade to a choice target, and was rewarded for correct decisions. On a random half of the trials, after the motion stimulus was extinguished, the monkey was given the option to opt out of the direction decision for a small but sure reward. The monkey exercised this sure option in a manner that revealed its degree of certainty. The sure option was chosen more frequently when the motion was weaker or shorter. Importantly, when the monkey did not bail out of the direction decision the probability of a correct response, for each motion duration and strength, was higher than on trials where the sure option was not presented, indicating that the monkey chose the sure option when it was more likely to make an error. LIP neurons represented formation of the direction decision and the degree of certainty underlying the decision to opt out. When the correct or error targets were in the cells' response field, trials with intermediate rate of rise in neural activity and intermediate persistent firing rates were more likely to lead to choosing the sure option. The monkey chose the correct or error target on trials with extreme levels of neural activity. A decision model based on bounded accumulation of evidence can explain both the choice and its associated certainty. In this model certainty is informed by the state of evidence at the time of decision and the time taken to reach the decision. Fundamentally, our results remove the need for metacognitive explanations for certainty monitoring, and advance the understanding of the neural mechanisms of decision making by coupling mechanisms leading to decision formation and the establishment of a degree of confidence.

This work was performed in collaboration with Mike Shadlen at the University of Washington, and was supported by the Howard Hughes Medical Institute, National Eye Institute grant EY11378, and National Center for Research Resources grant RR00166 to Mike Shadlen.

Host Keiji Tanaka, Laboratory for Cognitive Brain Mapping