Abstract
Successful goal-directed behavior depends on efficient disengagement of attention. Certain brain lesions (Posner et al., 1984) and aging can cause a "disengagement deficit" (Cosman et al., 2011). The patterns of decline can provide insights on how attention demanding real-world behaviors (e.g. automobile driving) change over the lifespan. We used a cross-sectional experimental design to characterize how efficiency of attentional disengagement influences oculomotor behavior. We compared 15 college-aged and 15 older adults (mean age: 78 [4.13]) on a variant of the "gap" paradigm. Participants were asked to saccade to a target onset in the periphery. In gap trials, a central fixation spot was extinguished before target onset (eliminating the need to disengage attention). In no gap trials, fixation remained, but its color changed before target onset, controlling for general alerting effects that might accompany fixation offset. Saccadic reaction times (SRTs) are typically lower in gap trials (Saslow, 1967), as visuospatial attention is reflexively disengaged when the fixation target is removed, facilitating the ensuing saccade (Fischer & Weber, 1993). Disengagement deficits should be more evident in no gap trials (when the fixation target remains), as higher-level control processes required for disengagement of attention become less efficient with advancing age (Cashdollar et al., 2013). In this vein, we observed a significant interaction between gap condition and age group. Older adults showed a significantly (p <.0001) greater gap effect (95.6 [42.8] ms) compared to young adults (41.3 [24.5] ms), with the greatest slowing occurring in the no gap condition. Two follow-up experiments, investigated how cognitive load manipulations (VSTM or executive working memory) affect oculomotor disengagement. Results showed executive working memory loading selectively slowed disengagement. The pattern of findings suggest that variations in disengagement efficiency associated with aging can influence oculomotor behavior, and that individual differences in executive control processes mediate disengagement functions.
Meeting abstract presented at VSS 2014