Abstract
Pupillary responses are known to differentiate younger and older adults (e.g., senile miosis) and have been recently linked to successful recognition in memory tasks. However, it is yet to be understood whether age-related changes in pupil dilation would show differential patterns during recognition memory tasks. In the present change-detection study, younger and healthy older adults viewed 192 original and modified versions of indoor photographs to determine whether a change had (or not) occurred with equal probability (i.e., 50% of the trials). In the modified version of the image, a change could be made to the identity of one of the objects in the scene (e.g., a toothbrush became a torchlight), to its location (e.g., a toothbrush moved from left to right), or to both features (e.g., a toothbrush moved and became a torchlight). On normalised (z-scored) pupil dilation collected during the recognition phase while the changed target object was fixated, we found that pupil size was significantly smaller in trials in which participants correctly than incorrectly detected a change. Crucially, this difference was greater for the younger adults than for the older adults. Further analyses on correct trials also revealed a significant interaction between group and type of change: older adults had smaller dilation when the target changed in identity compared to when it changed in location, whereas the opposite effect was observed in younger adults. We argue that the greater effort to identify a change in an object’s location compared to a change in identity may reflect the reduced useful field of view in older adults. Our results align with previous evidence of a negative relationship between pupil size and recognition memory and suggest that such pupillometry can help identifying the use of compensatory strategies in older adults during visual working memory tasks.