Don't stop believin'––in the link between efficient scanning and working memory capacity

Poster Presentation: Tuesday, May 20, 2025, 8:30 am – 12:30 pm, Banyan Breezeway
Session: Eye Movements: Social, individual differences, visual preferences

McKenzie Sheets1 (), Micalee Segers2, Tori Jones3, Chloe Mitchell4, Shannon Ross-Sheehy5; 1University of Tennessee

There is a well-established relationship between working memory (WM) and visual attention. Previous studies have shown that the contents of WM can influence looking behavior (Soto et al., 2008; Woodman & Luck, 2007). Additionally, there has been a substantial amount of work indicating that individual differences in WM capacity (WMC) are related to performance on a variety of cognitive tasks (Unsworth & Engle, 2007). However, there is little work investigating the relationship between visual WMC and scanning patterns. Individual differences in WMC may be related to scanning efficiency (SE), such that individuals with high WMCs may employ more efficient scanning strategies than those with lower WMCs. To investigate this relationship, we used data collected from adults (n=55) who completed both a scanning task and a change detection task. Gaze was sampled at a rate of 500Hz using EyeLink 1000+ in both tasks. The scanning task consisted of four blocks of five 10s trials, in which participants were presented with two static faces (one male and one female), counterbalanced across blocks. For the change detection task, visual arrays consisted of colored circles (ss3,ss5,ss7,ss9), and each trial included a 100ms sample array, 900ms retention interval, and 3000ms test array. SE scores were calculated for each participant using fixation durations and subsequent saccade amplitudes (see Ross-Sheehy et al., 2022) during the scanning task. Preliminary results indicate a positive relationship between current fixation durations and subsequent saccade amplitudes, t(54)=11.99,p<0.001. WMC (k) was calculated for each participant based on their change detection performance. Participants were categorized as having high or low SE, and preliminary findings suggest a relationship between SE and WMC, as participants with low SE have higher WMC(M=4.77) than those with high SE(M= 4.30), t(51)=-1.98,p=0.05. While these findings do not support our hypothesis, further analyses will explore the relationship between WMC and SE.