Individual Differences in Attention Allocation During a 2-dimensional Inhibitory Control Task

It has been proposed that visual attention behaves as a spotlight or zoom lens that gradually focuses on goal-relevant features of a stimulus over the course of a trial. Several lines of evidence suggest that for spatially-contiguous stimuli, the spotlight naturally takes on the shape of a horizontally-biased ellipse. Analyses of group level behavior in the presence of horizontally- versus vertically-configured stimuli, however, potentially obfuscate an important source of between-subject variability in the early stages of attentional processing. In the current study, we used a 2-dimensional flanker task paradigm and nested variants of a model of within-trial attention and decision mechanisms to investigate individual differences in spotlight shapes. To account for the influence of distractor stimuli in both horizontal and vertical positions relative to the target, we operationalized the attentional spotlight as the density function for a bivariate normal distribution within our models. Horizontal and vertical shape parameters governing the spotlight were constrained to be equal in one model variant, and were allowed to vary in the other. Within-subject comparisons of Bayesian goodness of fit statistics revealed a general preference for an elliptical rather than a circular spotlight. Follow-up analyses, however, demonstrated substantial variability in spotlight shapes across subjects. Although data from most subjects were best-captured by a horizontally-biased elliptical spotlight, we observed individual differences in the extent of the bias, with some subjects even demonstrating a circular or vertically-biased elliptical spotlight.