Abstract
Aging is associated with impairments in functions that rely heavily on the prefrontal cortex, such as selective attention. To deal with potential information overload, older adults must rely on preserved functions. But which functions are preserved, and how can they help compensate for reduced capacity in selective attention? This study investigates one candidate mechanism – implicit learning. Previous research suggests that implicit learning is preserved in only simple, not complex, tasks in older adults. Here we tested implicitly learned attention in healthy older adults (60-80 years) and young adults (18-30 years). In Study 1 participants searched for a letter T among letter Ls. We introduced a learning component by placing the target T in one visual quadrant more frequently than in any of the other quadrants. Consistent with the idea that top-down attention declines with age, older adults showed longer response time (RT) and steeper RT-set size slope, relative to young adults. However, both groups demonstrated faster RT in the high-probability quadrant than the low-probability quadrants, even when they were unaware of the target's location probability. This learned attentional preference was durable, persisting over 200 trials of extinction training in both groups. Study 2 examined the spatial reference frame in which the frequently attended locations were coded. Participants were trained to prioritize one quadrant of the monitor placed flat on a stand. After training they changed their viewpoint by 90°. Older adults, like younger adults, showed a strong attentional bias toward locations in the same visual field as the previously trained high-probability locations. This finding suggests that learning induced a search habit that was viewer-centered. Study 3 showed that an explicit instruction to prioritize environment-centered locations can partially override the viewer-centered search habit in older adults. We conclude that implicitly learned spatial attention is preserved in healthy aging.
Meeting abstract presented at VSS 2016