Abstract
The contextual cueing effect describes the benefit of repeated contexts on visual search performance relative to non-repeated contexts. Larger contextual cueing effects, defined by a larger benefit of repeated contexts, have been found in real-world stimuli relative to simplistic stimulus arrays. However, it is unclear how much of this difference is due to increased complexity of the display versus other factors such as semantic information. The current work examined the impact of stimuli rich in complexity but with little semantic information on facilitating contextual cueing. We assessed participants’ visual search performance using highly complex aeronautical charts across five blocks in which half of the charts in each block were shown either once in the experiment (Non-Repeated) or repeated once each block (Repeated). Participants also completed a recognition post-test consisting of repeated charts from the search task, a subset of non-repeated charts from the search task, and novel charts which were not present in the search task to assess their memory for the charts. Non-linear modelling revealed no significant differences in search response time across blocks of trials between repeated and non-repeated charts. However, post-test results demonstrated strong recognition performance for repeated, followed by non-repeated charts compared to novel charts. These findings suggest that high complexity and search difficulty do not necessarily lead to a reliable contextual cueing effect. This lack of a contextual cueing effect was not due to a lack of memory for the arrays: Strong explicit memory for the charts was confirmed. Thus, the larger contextual cueing effects typically found in real-world stimuli may be due to semantic information more so than stimuli complexity. Furthermore, there may be an optimal level of complexity for contextual cueing that may have been surpassed in the current experiment.