Name: Jiwon Yeon
Ph.D. Dissertation Proposal Meeting
Date: Thursday, September 24th, 2020
Time: 10:00am-12:00pm (EDT)
Location: meet.google.com/xjw-ekvm-wro

Advisor: Dobromir Rahnev, Ph.D. (Georgia Tech)

Dissertation Committee members:
Dobromir Rahnev, Ph.D. (Georgia Tech)
Thackery Brown, Ph.D. (Georgia Tech)
Audrey Duarte, Ph.D. (Georgia Tech)
Kobe Desender, Ph.D. (Ghent University)
Brian Odegaard, Ph.D. (University of Florida)

Title: Dissociating the brain regions involved in processing objective and subjective performance

Abstract: When making perceptual decisions, easier tasks produce higher task accuracy and, naturally, higher confidence levels. Although we recognize the two distinctive cognitive processes, and sometimes even we experience inconsistency between objective and subjective performance, it is challenging to judge exactly how objective and subjective performance affect different brain regions due to the foundational low of covarying human performance. Hence, previous studies that looked into the association between different brain areas with objective and subjective performance, have been unable to adjudicate which type of performance was responsible for the activations in which subregions of the brain. In the proposed fMRI study, I aim to reveal which brain regions are activated by objective and subjective performance, respectively. The experiment will present clouds of red and blue dots with different numbers. The task for subjects will be to determine which color of the dots was dominant and then to evaluate subjective confidence in their decision. The experiment is a 2 x 2 factorial design, with the two factors being task difficulty (i.e., easy and difficult conditions) and the total number of presented dots (i.e., high and low conditions). Similar visual stimuli were examined in our lab’s previous studies, where we found that the low number of dots condition reliably generated lower confidence level than the high number of dots condition. Using this phenomenon, it will be possible to find certain difficulty levels in which the low condition will produce lower subjective, but higher objective performance, compared to the high condition. Meanwhile, the easy condition will result in lower performance in both objective task accuracy and subjective confidence level compared to the difficult condition. Contrast test of the easy condition compared to the difficult condition will reveal brain regions that are associated with low subjective and objective performance. Similarly, contrast test of the low condition compared to the high condition will uncover the area related with low subjective, but high objective performance. Combinations of different contrast tests will demonstrate how the subjective and objective performance affects different brain regions, respectively. In addition to the contrast tests using simple general linear models, I will also analyze the data using multi-voxel pattern analysis to confirm the findings in the contrast tests. Decoding accuracy for distinguishing the performance level of the objective task accuracy and subjective confidence level in each activated brain region will be assessed. Finally, connectivity analyses will be used to examine how the flow of information between different parts of the brain relates to objective and subjective performance. I expect to find several subregions in the frontal cortex, including frontal eye field, dorsolateral prefrontal cortex, and anterior prefrontal cortex, showing dissociated activation for objective and subjective performance. Specifically, the anterior part of the frontal cortex will be mostly driven by subjective performance. The proposed study not only demonstrate separate brain regions associated specifically with objective and subjective performance, but will also establish a new experimental method to reliably distinguishing objective task accuracy and subjective confidence level.