Ensemble perception includes information from multiple spatial scales

Abstract

Most visual scenes contain information at different spatial scales, including the local and global, or the detail and gist. Global processes have become increasingly implicated in research examining summary statistical perception, initially as the output of ensemble coding, and more recently as a gating mechanism for selecting which information is included in the averaging process itself. Yet local and global processing are known to be rapidly integrated by the visual system, and it is plausible that global-level information, like spatial organization, may be included as an input during ensemble coding. We tested this hypothesis using an ensemble shape-perception task in which observers evaluated the mean aspect ratios of sets of ellipses. In addition to varying the aspect ratios of the individual shapes, we independently varied the spatial arrangements of the sets so that they had either flat or tall organizations at the global level. We found that observers made precise summary judgments about the average aspect ratios of the sets by integrating information from multiple shapes. More importantly, global flat and tall organizations were incorporated into ensemble judgments about the sets; summary judgments were biased in the directions of the global spatial arrangements on each trial. This global-to-local integration even occurred when the global organizations were masked. Our results demonstrate that the process of summary representation can include information from both the local and global scales. The gist is not just an output of ensemble representationit can be included as an input to the mechanism itself.

Introduction to the special issue on ensemble perception

“…Much of the new work reported in this special issue explores higher level ensemble representations, the interactions among levels, and the underlying mechanisms. For example, ensemble perception of spatial properties may not require magnocellular or coarse spatial processing (Lee & Chong, 2020), but it does involve information drawn from multiple spatial scales (Sweeny et al, 2020).…”

Ensemble Representation of Animacy Is Based on Mid-level Visual Features

“…Ensemble summary statistics studies demonstrated that observers can extract the mean shape which could vary from very boxy to very curvy (Robinson & Brady, 2022) as well as mean texture (Koenderink, Doorn, & Pont, 2004) or other types of shape information (Sweeny, Bates, & Elias, 2021;Khayat, Fusi, & Hochstein, 2021). Thus, curvature is a feature that may be strongly related to object and ensemble animacy perception; however, additional studies are required to explore the origin features behind animacy perception.…”

“…Early works in this area were focused primarily on the representation of low-level features such as visual motion (Watamaniuk & McKee, 1998;Watamaniuk et al, 1989), brightness (Bauer, 2009), orientation (Dakin & Watt, 1997;Parkes et al, 2001;Attarha & Moore, 2015), hue (Gardelle & Summerfield, 2011;Maule & Franklin, 2015) and spatial location (Alvarez & Oliva, 2008). Recent studies have shown that observers can easily extract summary statistics for mid-level features such as object size (Ariely, 2001;Chong & Treisman, 2003;Tiurina & Utochkin, 2019;Markov & Tiurina, 2021;Haberman & Suresh, 2021), shape (Sweeny, Bates, & Elias, 2021;Khayat, Fusi, & Hochstein, 2021) and texture (Koenderink, Doorn, & Pont, 2004;Cant & Xu, 2012, 2017. Observers could also estimate summary statistics for high-level visual features: emotion, gender, gaze direction, and head rotation (Haberman & Whitney, 2007, 2009Sweeny & Whitney 2014;Yamanashi Leib et al, 2014;Florey et al, 2016;Han et al, 2020); categorical characteristics (Khayat & Hochstein, 2019); economic value (Yamanashi , and animacy (Yamanashi Leib et al, 2016).…”

A general ability for judging simple and complex ensembles.

“…At an ice cream shop, we estimate the flavors of ice cream by the diversity of colors in the freezer. People can make such judgments about ensembles of objects for a host of visual features, ranging from lower level features like size (Ariely, 2001), aspect ratio (Elias & Sweeny, 2020;Sweeny et al, 2021), lightness (Bauer, 2017;Takano & Kimura, 2020), and orientation (Alvarez & Oliva, 2009), to higher level features such as car models (Cha et al, 2021;Chang & Gauthier, 2022) and facial expression and identity (De Fockert & Wolfenstein, 2009;Haberman & Whitney, 2007). People can also evaluate groups of objects to consider summaries of abstract features, such as the lifelikeness or economic value of objects (Yamanashi Leib et al, 2016Leib et al, , 2020 and the attractiveness of faces (Luo & Zhou, 2018).…”