Purchase this article with an account.
Barbara Hidalgo-Sotelo, Aude Oliva; Decomposing the effect of contextual priors in visual search: Where does the time go?. Journal of Vision 2006;6(6):835. doi: 10.1167/6.6.835.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Among the many visual searches we perform daily, often the same environment is searched for the same item multiple times. Contextual cueing research has shown that repeated configurations facilitate attentional guidance toward the target (Chun & Jiang 1998). We investigated the influence of contextual priors in guiding visual search by monitoring eye movements as participants search familiar real-world scenes. The level of search difficulty was either easy (target visible within a glance) or difficult (target requiring foveation). Additionally, we manipulated the expectation of target presence in a given scene. One group of subjects established strong contextual priors: the target would always be present or always absent. A second group established weaker priors: a given scene was presented with the target present only 50% of the time. After twenty epochs of search, reaction times in the strong contextual prior group improved by 280 ms, while the weak prior group improved by only 190 ms. A decomposition of reaction times shows this improvement is found in two stages. First, the scan time to fixate the target is decreased, implying a faster exploration stage. Second, the time spent fixating the target before response (gaze duration) also decreases. Interestingly, the magnitude of scan time improvement was the same for both groups. The remaining RT improvement in the strong prior group is entirely manifested in decreased gaze duration. These results demonstrate experience-dependent influences at different stages of visual search and have implications for models of context-dependent scene processing.
This PDF is available to Subscribers Only