Academic journal article Attention, Perception and Psychophysics

Characterizing Ensemble Statistics: Mean Size Is Represented across Multiple Frames of Reference

Academic journal article Attention, Perception and Psychophysics

Characterizing Ensemble Statistics: Mean Size Is Represented across Multiple Frames of Reference

Article excerpt

Published online: 18 December 2013

# Psychonomic Society, Inc. 2013

Abstract The visual system represents the overall statistical, not individual, properties of sets. Here we tested the spatial nature of ensemble statistics. We used a mean-size adaptation paradigm (Corbett et al. in Visual Cognition, 20 ,211-231, 2012) to examine whether average size is encoded in multiple reference frames. We adapted observers to patches of small-and large-sized dots in opposite regions of the display (left/ right or top/bottom) and then tested their perceptions of the sizes of single test dots presented in regions that corresponded to retinotopic, spatiotopic, and hemispheric coordinates within the adapting displays. We observed retinotopic, spatiotopic, and hemispheric adaptation aftereffects, such that participants perceived a test dot as being larger when it was presented in the area adapted to the patch of small dots than when it was presented in the area adapted to large dots. This aftereffect also transferred between eyes. Our results demonstrate that mean size is represented across multiple spatial frames of reference, supporting the proposal that ensemble statistics play a fundamental role in maintaining perceptual stability.

Keywords Adaptation . Aftereffects . Cognitive eye movements . Grouping . Segmentation

How does the visual system mediate between the needs to parse incoming sensory input into updated representations of objects, events, and scenes and to maintain our perception of a stable and continuous visual environment? Given that objects are of central importance for cognition and action, much vision research has focused on the problem of how objects are processed and recognized. However, object individuation is limited to, at most, only a few items in each glance. Not surprisingly, in addition to this form of detailed individual object processing, the visual system also relies on more global representations of the surrounding environment based on the average statistics of scenes and sets, or ensembles of objects (Ariely, 2001; Oliva & Torralba, 2001). Our naïve impression of a complete, detailed, continuously updated visual experience is likely based on a combination of these two types of visual processing. Together, the rich, foveal representation of one or two objects, memory for a handful of previously fixated items, and rapid impressions of the ensemble statistics of the rest of the scene may underlie the illusion of stable and complete visual perception (Ariely, 2001;Bar,2004; Hochberg, 1978; Hock & Schmelzkopf, 1980; Melcher & Colby, 2008; Rensink, 2000). This feeling of perceiving everything in detail would be heightened by the ability to quickly fixate items in the periphery via saccadic eye movements.

Despite renewed interest in ensemble statistics, the spatiotemporal nature of these representations and their actual purpose in guiding cognition and action remain largely unknown. Given that ensemble representations do not depend on any single element being in view at a particular time, this raises the question of how we combine information about elements across space and over time during realistic viewing conditions. Previous studies have briefly flashed groups of elements on the screen, but natural viewing involves moving our eyes and bodies. The spatial and temporal attributes of ensemble statistics are critical, as it has been suggested that the relative stability of ensemble statistics across separate views might play an important role in our sense of visual stability (Corbett & Melcher, 2013; Melcher & Colby, 2008).

Beyond object-based representations: Ensemble statistics

As we described above, growing evidence indicates that our perception of the world combines two separate mechanismsone involved in individuating a small number of objects, and the other representing the statistical properties of many objects. Visual input is highly redundant, allowing the brain to efficiently encode these regularities. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.