Zainab Haseeb¹ (), Anna Kosovicheva^1,2; ¹University of Toronto, ²University of Toronto Mississauga

Human observers differ substantially in how effectively they can attend to locations across the visual field. Even when gaze is fixed, some regions within the functional visual field (FVF) are consistently identified more readily than others. Recent work shows that these “strong” and “weak” regions vary as a function of polar angle around fixation. These results show stable and idiosyncratic patterns of performance within individuals, suggesting that covert attention is distributed unevenly across space. This study tested whether individual differences in these attentional asymmetries can be modified through feedback, or if they are resistant to change, and whether metacognitive insight is associated with feedback-related improvements. Participants completed a visual search task in which a briefly presented rotated T shape appeared among distractor Ls at eight isoeccentric locations. Participants reported the orientation of the shape. A baseline session measured each participant’s error distribution and metacognitive judgments of performance across spatial locations. In a subsequent session, participants received trial- and block-level feedback of error rates by location. The baseline and feedback conditions both revealed reliable idiosyncratic asymmetries in covert attention, with clear “strong” and “weak” regions for each participant. Feedback improved performance overall but it did not reliably improve uniformity; a comparison of entropy in error rates between conditions showed no significant systematic reduction in spatial asymmetries. As expected, however, metacognition improved robustly with feedback, such that participants’ confidence ratings and end-of-task spatial rankings aligned closely with actual accuracy. However, these gains did not translate into more uniform performance. Moreover, individuals with higher baseline metacognitive accuracy did not show greater feedback-related increases in entropy, suggesting that even accurate insight does not readily compensate for attentional limitations. While awareness of accuracy can be modified, the underlying attentional asymmetries are seemingly resistant to change.

Acknowledgements: This work was supported by an NSERC Discovery Grant to AK