Date of Award

Spring 2017

Degree Name

Bachelor of Science

Major

Computer Science and Psychology

First Advisor

Michael Grubb

Abstract

Attention is traditionally divided into two types: voluntary, goal-directed attention and involuntary, stimulus-driven attention (Corbetta & Shulman, 2002; Theeuwes, 2010). Seminal work on value-driven attentional capture (VDAC) has shown that stimuli associated with reward during a reward learning phase slowed reaction time (RT) in a test phase even when task-irrelevant and non-salient (Anderson, Laurent, & Yantis, 2011). However, performance-contingent reward and a response deadline impose additional constraints in the VDAC paradigm: responding too quickly decreases reward likelihood and responding too late drops the reward probability to zero. Thus, to maximize reward, participants must carefully decide when to respond, potentially altering the strategic balancing of speed and accuracy and confounding attentional effects with decisional ones. We replicated the VDAC paradigm to address the influence of different response strategies. Using maximum likelihood estimation, RT distributions were fitted with an exGaussian model. We found that RT variability (σ) was significantly greater in the experimental group (p<0.05), suggesting that reward learning produced a less stable strategy. Further, RT variability positively correlated with error rate (r=0.51, p<0.001), reflecting a behavioral cost with greater RT variability. These results call into question the validity of the baseline trials used in the VDAC paradigm, as reward learning altered the response strategy even after the reward was removed.

Comments

Senior project completed at Trinity College Hartford CT for the degree of Bachelor of Science in Computer Science and Psychology.

Download

Share

COinS