Name: Jenny Autumn Walker

Master’s Thesis Proposal Meeting
Date: Friday, April 13th, 2018
Time: 11:00am
Location: J.S. Coon Building, room 150
 
Advisor:
Professor Mark Wheeler, Ph.D. (Georgia Tech)
 
Thesis Committee Members:
Professor Mark Wheeler, Ph.D. (Georgia Tech)
Professor Audrey Duarte, Ph.D. (Georgia Tech)

Professor Christopher Hertzog, Ph.D. (Georgia Tech)
 
Title: The Impact of Criterion Shifts on Evidence Accumulation in the Inferior Temporal Cortex

Abstract

Informative prior knowledge is a critical component of decision-making because, as sequential sampling models suggest, shifts in criterion may influence the quantity of evidence that is accumulated and processed at the neural level before a choice can be made with sufficient certainty. However, the majority of work in this area has focused almost exclusively on behavioral outcomes alone, so it is unclear if or how such criterion-based shifts are represented in brain activity. Therefore, this study aims to address if individuals who are likely to be either relatively liberal (younger adults) or relatively conservative (older adults) differ in how they integrate prior knowledge into their decision-making processes, make choices, and gather evidence at the neural level. A face/house discrimination task with probabilistic cues is used to examine evidence accumulation in the inferior temporal cortex (ITC) for both younger and older adults. The cues represent prior knowledge. The face/house stimuli are also overlaid with a certain amount of visual noise; the amount of which is controlled relative to each person’s visual noise threshold given a pre-defined accuracy. Additional goals include accounting for the difference between cue-related activation and decision-related activation, as well as examining possible relationships between performance, age, task-related cognitive load, and visual contrast sensitivity. Behavioral pilot data pertaining to the efficiency of an implemented visual noise thresholding task (relative to other methods) is first presented. Then, the conceptual framework of decision-making, sequential sampling models, and prior knowledge utilization across the lifespan is explained. Finally, pilot work aimed at establishing visual noise thresholds and controlling for accuracy is described, preliminary behavioral data is presented, and current hypotheses are proposed. fMRI data collection is ongoing.