Purchase this article with an account.
Qi Li, Kazuhiko Yokosawa; A task-irrelevant high memorability image can impair or enhance visual search performanc. Journal of Vision 2016;16(12):1440. doi: https://doi.org/10.1167/16.12.1440.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Some images are remembered at first glance, while others are forgotten quickly. A recent study has demonstrated that images carry the property of memorability (Isola, Xiao, Parikh, Torralba, & Oliva, 2014). An important question that arises from their study is what factors interact with memorability. Isola et al. (2014) have investigated factors that are intrinsic to an image such as aesthetics, simple features and semantic attributes. Our study focused on a possible factor that is extrinsic to an image — visual attention. Specifically, we used the concurrent visual search task (Wolfe, Horowitz & Michod, 2007) to investigate the interplay between image memorability and visual attention. Experiment 1 tested whether high and low memorability images capture visual attention in a different way. Participants viewed an L versus T search array that was superimposed over a task-irrelevant image on each trial. The task was to count the number of Ts presented on each trial (There could be 0, 1 or 2 Ts). Search performance was significantly impaired by high memorability images compared to low memorability images, suggesting that task-irrelevant high memorability images are difficult to ignore and capture visual attention automatically. Experiment 2 further examined whether presenting a task-irrelevant image prior to the search display can produce similar effects as those observed in Experiment 1. Interestingly, the results revealed that search performance was significantly enhanced by high memorability images that preceded the search display. A plausible explanation is that the participants were more alert and active after viewing a high memorability image, leading to improvements in search performance.
Meeting abstract presented at VSS 2016
This PDF is available to Subscribers Only