Abstract
The contextual cueing effect was initially thought to be the product of memory for the repeated context guiding attention to the target location. Recent work has disputed this explanation pointing to the absence of decreased search slopes (derived from response time × set size function) in contextual cueing, a criterion used for establishing the presence of attentional guidance. However, we argue that the candidate source of guidance in contextual cueing – memory for the repeated displays – is fundamentally different from previously established “guiding attributes” which are derived from features of the display items themselves. It is this difference, rather than a lack of attentional guidance, that explains the failure to observe decreased search slopes in contextual cueing. While increasing the number of items in the display will necessarily increase the amount of guiding information in typical feature-based guidance, the resulting changes in the perceptual characteristics of the display may attenuate memory-based attentional guidance. In particular, visual crowding may impair peripheral recognition of the target at larger set sizes due to the increased density of items in the local target region. This could reduce the efficacy of the guidance as set size increases thereby yielding steep search slopes. We investigated the influence of crowding on contextual cueing by varying the density of items in the local target region. When crowding was eliminated, we observed results characteristic of attentional guidance – a significant reduction in search slopes between the novel and repeated conditions. No contextual cueing effect was observed when crowding of the local target region was enhanced. These findings indicate that the absence of decreased search slopes in contextual cueing is the product of increased crowding at larger set sizes rendering the memory-based attentional guidance less useful.