Purchase this article with an account.
Carmela Gottesman; Disambiguating the Effect of clutter on Boundary Extension. Journal of Vision 2014;14(10):871. doi: 10.1167/14.10.871.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Perceptual processes involved in processing the spatial layout of scenes often result in memory distortions called Boundary Extension (BE); viewers tend to remember more of the scenes than they actually saw, as if they had seen more wide-angle views than they actually did. The influence of different scene characteristics on BE is still to a large degree unknown. Prior research looking at the effect of clutter on BE has led to contradictory findings. Natural scenes obtained from the database developed by Oliva and Torralba (2001) induced more BE for cluttered than uncluttered scenes (Gottesman, 2010). However, the pictures in these categories varied on other dimensions as well as clutter. Gottesman (2011) used computer modeled scenes with cluttered and uncluttered versions of each scene to control for any variables other than the number of objects presented in a given space. This study produced the opposite result: uncluttered images induced more BE. The current study sought to determine which result is the real effect of clutter and whether the discrepancy was caused by the variability of the scenes in the first study or the artificiality of the stimuli in the second study. Clusters of objects in real world environments were photographed (e.g., books and papers in a bookcase). In the cluttered version many more objects than in the uncluttered version were crowded into the same space. Participants viewed eight cluttered and eight uncluttered scenes in random order; picture version for each scene was counterbalanced across participants. As in the previous study using the uncontrolled natural scenes, the cluttered versions induced more BE than the uncluttered versions of the same scenes. These results reaffirm the probability that the more information present in a scene the more viewers need to expand their representation of the spatial context.
Meeting abstract presented at VSS 2014
This PDF is available to Subscribers Only