High-capacity preconscious processing in concurrent groupings of colored dots

Abstract

Grouping is a perceptual process in which a subset of stimulus components (a group) is selected for a subsequent—typically implicit—perceptual computation. Grouping is a critical precursor to segmenting objects from the background and ultimately to object recognition. Here, we study grouping by color. We present subjects with 300-ms exposures of 12 dots colored with the same but unknown identical color interspersed among 14 dots of seven different colors. To indicate grouping, subjects point-click the remembered centroid (“center of gravity”) of the set of homogeneous dots, of heterogeneous dots, or of all dots. Subjects accurately judge all of these centroids. Furthermore, after a single stimulus exposure, subjects can judge both the heterogeneous and homogeneous centroids, that is, subjects simultaneously group by similarity and by dissimilarity. The centroid paradigm reveals the relative weight of each dot among targets and distractors to the underlying grouping process, offering a more detailed, quantitative description of grouping than was previously possible. A change detection experiment reveals that conscious memory contains less than two dots and their locations, whereas an ideal detector would have to perfectly process at least 15 of 26 dots to match the subjects’ centroid judgments—indicating an extraordinary capacity for preconscious grouping. A different color set yielded identical results. Grouping theories that rely on predefined feature maps would fail to explain these results. Rather, the results indicate that preconscious grouping is automatic, flexible, and rapid, and a far more complex process than previously believed.

Multiple concurrent centroid judgments imply multiple within-group salience maps

“…Centroid judgments are a high-capacity preconscious computation It is important to keep in mind that a centroid judgment is a statistical summary representation, that is, it is a preconscious computation that utilizes much more display information than can be subsequently recalled. For example, in stimuli quite similar to those used here, Sun, Chubb, Wright, and Sperling (2018) found that even with a very generous scoring procedure in a change-detection memory task, the best subject remembered the approximate locations of fewer than two of 26 dots. Yet, in the same stimuli, subjects were able to estimate centroids that minimally required the accurate processing of more than 15 of 26 dots.…”

“…Recent studies show that subjects can compute SSRs for selective subsets that are spatially segregated (Attarha, Moore, & Vecera, 2014;Im & Chong, 2014) or are defined by different features (Chong & Treisman, 2005;Drew, Chubb, & Sperling, 2010;Halberda, Sires, & Feigenson, 2006;Inverso, Sun, Chubb, Wright, & Sperling, 2016;Poltoratski & Xu, 2013;Sun, Chubb, Wright, & Sperling, 2016a, 2016b). More impressively, some studies show that subjects can even compute multiple concurrent SSRs (Attarha, Moore, & Vecera, 2014;Chong & Treisman, 2005;Halberda, Sires, & Feigenson, 2006;Im & Chong, 2014;Poltoratski & Xu, 2013;Sun, Chubb, Wright, & Sperling, 2018). The present study investigates multiple SSRs.…”

“…In the present experiments, when subjects make three centroid judgments, they remember primarily the statistical summary of the stimulus, the three centroids. We know from Sun et al (2018) and other prior work that they remember very little else about the stimulus. All the computations represented in Fig.…”

“…The more complex aspects of filter-group-salience processing are not exposed in the present experiments. Consider instead two complex examples: In a briefly flashed stimulus consisting of 12 dots, all of the same unknown color, and 14 dots of seven different colors, Sun et al (2018) demonstrated that subjects could concurrently compute both centroids, the centroid of the homogeneous group and the centroid of the heterogeneous group. There is no predefined map devoted to "heterogeneity versus homogeneity of colored dots" nor any pre-defined color filter that can make such a discrimination.…”

“…That two gratings moving in the same direction, each of which evokes a strong motion, can nevertheless cancel when they are of opposite phase indicates that the salience system is totally unable to distinguish these two inputs and simply adds their respective salience values. A consequence of salience being a scalar versus a vector is that it is no more difficult to compute salience, and therefore centroids, on any array of different objects than on an array of homogeneous objects (e.g., Sun et al, 2018). Obviously, other parts of the brain know the difference between a depth grating and a color grating and between heterogeneously and homogeneously colored dots, but not the salience map (or maps) on which motion and centroids are computed.…”