We used event-related potentials (ERPs) to investigate age differences in executive control. A cue-based task-switching paradigm with standard Stroop stimuli was employed, in which the participants were instructed either to name colors or to read words (single blocks) or to alternate between both (mixed blocks). Task-set selection was examined by comparing ERPs for single versus mixed blocks, interference control by comparing ERPs for compatible versus incompatible targets. Results of 14 younger (mean age = 22 years) and 14 older adults (mean age = 63 years) indicated a greater P3 component in the cue interval for mixed compared to single blocks for both age groups, showing a parietal-maximum distribution in younger and a more widespread distribution in older adults. This finding suggests that older adults have impairments in updating task-relevant information. Furthermore, older adults showed a substantially greater contingent negative variation (CNV) in mixed compared to single blocks, whereas no such difference was observed in younger adults. We assume that it reflects a stronger engagement in response preparation processing in the elderly. Finally, an enhanced negativity for incompatible targets (IN) was found, that was substantially smaller for older adults. The smaller IN in older adults presumably reflects age-related deficits in interference control.
In a follow-up study with a subset of the subjects from the first experiment we manipulated control demands by varying the probability of compatible/incompatible Stroop trials in order to examine whether this affects age-related changes in the three ERP components.