Perceptual and Neural Correlates of Expectation

Date of Award

Spring 5-15-2011

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Neurosciences)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



A central challenge of systems neuroscience is to understand how we make decisions based on information from our environment. Although incoming sensory information is a primary determinant of decisions, prior expectations about the world can also significantly influence our choices. Many recent studies have examined neural responses in primates during visual decision tasks. These studies have inspired a model which explains how incoming sensory information is accumulated over time until the subject commits to a decision. However, neither this model nor the available literature specify how prior expectations might bias the decision-making process. There are two possible mechanisms by which prior expectations could influence decisions: 1) Expectations might modify how the brain represents the sensory stimulus, or 2) Expectations might alter how much evidence is required to commit to a particular choice. The central goal of this thesis is to identify a neural correlate of the effect of expectations and to understand how it interacts with sensory information during decision-formation.

We explored these issues by training rhesus macaques to make challenging judgments about the direction of a visual motion stimulus. A cue that predicted the direction of the motion was presented before the motion began, and the monkeys’ reports of the motion direction were biased by this cue. We recorded extracellular activity from individual neurons in two areas of the macaque brain: 1) the middle temporal area (MT), which has a robust representation of motion direction, the feature of interest in our task; and 2) the lateral intraparietal area (LIP), which is thought to reflect the process of converting sensory information into a decision. We found that expectations did not alter the representation of motion direction in MT. However, LIP activity was altered by the cue, and we extended the decision-making model to show how the measured effects in LIP can quantitatively explain how prior expectations can bias decisions.


English (en)

Chair and Committee

Lawrence Snyder

Committee Members

Dora Angelaki, Maurizio Corbetta, Timothy Holy, Camillo Padoa-Schioppa, David Van Essen


Permanent URL:

This document is currently not available here.