Using the attention cascade model to probe cognitive aging

Young and older adults searched for 2 digit targets among black letter distractors in rapid serial visual presentation. Unsurprisingly, relative to the young, the older adults performed worse on both targets and exhibited greater and longer attentional blink. The data of each group were computationally accounted for by the attention cascade model (Shih, 2008) with 7 parameters; the optimum values and 95% confidence intervals of the parameters were based on 10,000 bootstrap samples. There was no age effect on the width of the attention window or on the capacity of the consolidation processor. However, relative to the young, the older adults suffered more masking effects of the salient (and brighter) stimulus, required longer consolidation duration, and had greater and more widespread decision noise. The processing rate prior to working memory was numerically slower in older adults. Both age groups adopted inefficient strategy during the task-engaging the consolidation processor for an unnecessarily long period. Further simulations suggest that varying the duration can emulate strong or weak blinkers or nonblinkers. The attention cascade model appears a useful tool for the investigation of cognitive aging and other comparative studies.