The development of perceptual grouping biases in infancy: A Japanese-English cross-linguistic study

Perceptual grouping has traditionally been thought to be governed by innate, universal principles. However, recent work has found differences in Japanese and English speakers’ non-linguistic perceptual grouping, implicating language in non-linguistic perceptual processes (Iversen, Patel, & Ohgushi, 2008). Two experiments test Japanese- and English-learning infants of 5–6 and 7–8 months of age to explore the development of grouping preferences. At 5–6 months, neither the Japanese nor the English infants revealed any systematic perceptual biases. However, by 7–8 months, the same age as when linguistic phrasal grouping develops, infants developed non-linguistic grouping preferences consistent with their language’s structure (and the grouping biases found in adulthood). These results reveal an early difference in non-linguistic perception between infants growing up in different language environments. The possibility that infants’ linguistic phrasal grouping is bootstrapped by abstract perceptual principles is discussed.     

Time course of perceptual grouping

An investigation was made of the time course of perceptual grouping that is based on two qualitatively different spatial relationships: proximity and alignment. An index of grouping capacity was used to assess the processing time required before a backward pattern mask interfered with grouping. Stimuli consisted of bistable arrays of disjunct dots that were followed by a mask. Grouping cues, either proximity or alignment, were randomly assigned to either the horizontal or vertical orientation, and subjects indicated whether the dots appeared grouped as a series of horizontal or vertical lines. Spatial metrics of the cues were systematically altered until they no longer served as a cue for grouping, thereby determining the grouping threshold. The stimulus onset asynchrony (SOA) of the mask, relative to the test stimulus, ranged from 33.3 to 150 msec. The SOA at which grouping thresholds first became elevated identified the point at which the mask first interfered with the grouping process, thereby identifying the processing time required for grouping by the specified cue. The processing time for grouping by proximity and alignment differed significantly, requiring means of 87.6 and 118.8 msec, respectively, for processing to be completed. These measurements serve to identify the processing time necessary for spatially integrating stimulus elements into unified forms, thereby delineating temporal constraints at this stage of visual processing.     

Unifying sequential effects in perceptual grouping

Temporally-extended perception involves a delicate balance of constancy and change. 'This can be seen, for instance, when viewing bistable figures such as the Necker cube. A recent study by Gepshtein and Kubovy of sequential effects in multistable dot lattices demonstrates constancy and change within the same set of data. They propose that these opposing trends might be explained by the same single factor: a persistent random orientation bias that is intrinsic to brain activity. This proposal could form the basis for a new account of multistability.     

Learning perceptual organization in infancy

It has been demonstrated that older infants (6- to 7-month-olds), but not younger infants (3- to 4-month-olds), use form similarity to organize stimuli consisting of X and O elements. We investigated whether utilization of form similarity is governed by maturation or experience by contrasting how infants perform when familiarized with a single exemplar versus multiple exemplars depicting a particular organization. In Experiment 1, 3- to 4-month-olds failed to organize alternating columns or rows of squares and diamonds or Hs and Is, respectively. In Experiment 2, same-aged infants familiarized with all three patterns (X-O, square-diamond, H-I) displayed evidence of organization. The results suggest that 3- to 4-month-olds can use form similarity to organize visual patterns in a concept-formation task. The findings imply that perceptual organization based on form similarity is learned through experience with multiple patterns depicting a common arrangement, rather than immediately apprehended in an individual pattern.     

Development of perceptual completion in infancy

Perceptual completion consists of bridging the gaps imposed by occlusion, such as perceiving the unity of center-occluded objects. It is unknown at present what developmental mechanisms underlie the emergence of functional perceptual completion in infancy. One current debate centers on the role of visible surface motion. According to a core-principles account, perceptual completion emerges simultaneously with the onset of motion discrimination, the sole determinant of unity percepts in infants. According to a contrasting constructivist account, motion discrimination is but one of several independent inputs to perceptual completion. In the present study, 2-month-old infants were tested for both unity perception and motion discrimination in partial-occlusion displays. Motion discrimination obtained under all conditions, even under circumstances in which infants were unable to perceive completion. Four-month-olds showed marked improvements in perceptual completion, most likely because of improvements in information integration. Taken together, these findings support a constructivist view of early perceptual and cognitive development.     

Development of associative and perceptual interference

In a choice reaction-time letter-identification task, the temporal development of perceptual and associative interference was studied with visual displays containing identical, irrelevant, or response-incompatible noise letters. Using the methods of variable-criterion theory, it was determined that perceptual interference is complete very early and has no effect on the shape of the function for the growth of associative strength. Associative interference begins later, gradually increases to a maximum, and then declines to zero at long latencies. In experiments with speeded performance, it was also found that the growth of excitatory strength for errors contains the same nonmonotonic process as for correct responses, but in reverse form. In addition, it was observed that the discriminability of the target stimuli affects the rate of growth of associative strength for the correct response. Analyses of individual differences in theoretical parameters indicate information-processing patterns similar to those reported for simple auditory tasks.     

The role of objects in perceptual grouping.

Perceptual organization can be viewed as the selection of the best or "most reasonable" parse of a given scene. However, the principles that determine which interpretation is most reasonable have resisted most attempts to define them formally. This paper summarizes a formal theory of human perceptual organization, called minimal model theory, in which the best interpretation of a given scene is expressed as the formally minimal interpretation in a well-defined space of possible interpretations. We then focus specifically on the role of types of grouping units, in particular the difficult notion of "object". Although grouping is often thought of as the process of dividing the image into objects, most research in perceptual grouping actually focuses on simpler types of units, such as contours and surfaces. Minimal model theory characterizes grouping units at a logical level, demonstrating how formal assumptions about units induce the observer to place a certain preference ranking on interpretations. The theory is then applied to the more subtle problem of objects, culminating in a definition for objects that is formally rigorous but at the same time captures some of the flexibility of human intuitions about objects.     

Time course of perceptual grouping by color

Does perceptual grouping operate early or late in visual processing? One position is that the elements in perceptual layouts are grouped early in vision, by properties of the retinal image, before perceptual constancies have been determined. A second position is that perceptual grouping operates on a postconstancy representation, one that is available only after stereoscopic depth perception, lightness constancy, and amodal completion have occurred. The present experiments indicate that grouping can operate on both a preconstancy representation and a postconstancy representation. Perceptual grouping was based on retinal color similarity at short exposure durations and based on surface color similarity at long durations. These results permit an integration of the preconstancy and postconstancy positions with regard to grouping by color.     

Influence of past experience on perceptual grouping

We used primed matching to examine the microgenesis of perceptual organization for familiar (upright letters) and unfamiliar (inverted letters) visual configurations that varied in the connectedness between their line components. The configurations of upright letters were available for priming as early as 40 ms, irrespective of connectedness between their line components. The configurations of connected inverted-letter primes were also available this early, but the configurations of disconnected inverted letters were not available until later. These results show that past experience contributes to the early grouping of disconnected line segments into configurations. These findings suggest an interactive model of perceptual organization in which both image-based properties (e.g., connectedness) and input from object memories contribute to perceptual organization.     

Temporal synchrony in perceptual grouping: a critique

It has been hypothesized that the human visual system can use temporal synchrony for the perceptual grouping of image regions into unified objects, as proposed in some neural models. It is argued here, however, that previous psychophysical evidence for this hypothesis is due to stimulus artifacts, and that earlier studies do not, therefore, support the claims of synchrony-sensitive grouping mechanisms or processes.     

Contour grouping inside and outside of facial contexts

We examine how contours group in isolation compared with when they are embedded in face-like contexts. As previously shown, contours that seem to group by phenomenological observation also show powerful effects on task performance: with contours that group, selective attention to one while ignoring another is poor (as indexed by Garner Interference (GI), but not Stroop Interference), whereas divided attention across contours is good. With contours that do not group, however, the reverse happens. Here we test pairs of curved lines (parentheses) displayed either in isolation or within contexts including cartoon faces, where these curves may serve as mouths or eyebrows. The results with isolated contours replicate previous findings of poor selective attention, but within face-like contexts the same contours showed nearly perfect selective attention (i.e., zero GI). Thus, contour grouping was weaker inside than outside of faces, a finding that contrasts with the widely-held belief that faces are processed configurally, not by local features.     

The construction of perceptual grouping displays using GERT

To study perceptual grouping processes, vision scientists often use stimuli consisting of spatially separated local elements that, together, elicit the percept of a global structure. We developed a set of methods for constructing such displays and implemented them in an open-source MATLAB toolbox, GERT (Grouping Elements Rendering Toolbox). The main purpose of GERT is to embed a contour in a field of randomly positioned elements, while avoiding the introduction of a local density cue. However, GERT's modular implementation enables the user to create a far greater variety of perceptual grouping displays, using these same methods. A generic rendering engine grants access to a variety of element-drawing functions, including Gabors, Gaussians, letters, and polygons.     

Effects of extinction in multiple contexts on the renewal of associative tolerance to ethanol

The Pavlovian analysis of drug tolerance describes how the contexts of drug administration participate in the elicitation of compensatory conditioned responses that are, at least partly, responsible for tolerance. Findings indicate that if an association is acquired in one context and extinguished in another one, when exposing the subjects to the acquisition context, renewal of the conditioned response will take place. Ambiguous evidence exists about whether or not this effect diminishes when extinguishing the association in multiple contexts. This investigation evaluates the existence of the phenomenon of renewal of tolerance to ethanol, and whether it diminishes by the extinction in multiple contexts. Evidence of renewal of tolerance to ethanol in rats was found, but no diminution was observed when extinguishing in multiple contexts.     

Parts function as perceptual organizational entities in infancy

Both objects and parts function as organizational entities in adult perception. Prior research has indicated that objects affect organization early in life: Infants grouped elements located within object boundaries and segregated them from those located on different objects. Here, we examined whether parts also induce grouping in infancy. Five- and 6.5-month-olds were habituated to two-part objects containing element pairs. In a subsequent test, infants treated groupings of elements that crossed part boundaries as novel, in comparison with groupings that had shared a common part during habituation. In contrast, the same arrangement of elements failed to elicit evidence of grouping in control conditions in which the elements were not surrounded by closed part boundaries. Thus, infants grouped and segregated elements on the basis of part structure. Part-based processing is a key aspect of many theories of perception. The present research adds to this literature by indicating that parts function as organizational entities early in life.     

Evaluating a computational model of perceptual grouping by proximity

A formal approach to the phenomenon of perceptual grouping by proximity was investigated. Grouping judgments of random dot patterns were made by the CODE algorithm (van Oeffelen & Vos, 1982) and several related algorithms, and these judgments were compared with subjects’ grouping judgments for the same stimuli. Each algorithm predicted significantly more subject judgments than would be expected by chance. The more subjects agreed on how a given dot pattern should be grouped, the more successful was the algorithms’ ability to match the judgments for that pattern. CODE predicted significantly fewer subject judgments than did some of the other algorithms, largely because of its overemphasis on the extent of interactivity among dots as they are being grouped.     

Effects of perceptual grouping on positive and negative priming

The effects of perceptual grouping/segregation of targets and distractors by means of colour on positive and negative priming were examined in two experiments. In Experiments 1 and 2 we examined whether grouping of target and distractors by means of common colour in the prime display affected positive and negative priming, or whether these effects depend on prime–probe contextual similarity in colour. In addition, we examined the effects of the predictability of target colour in the prime and the probe displays across the experiments using mixed (Experiment 1) or blocked procedures (Experiment 2). The pattern of results was similar in both experiments, indicating that the positive priming effect was determined by target repetition and enhanced by perceptual segregation of target from distractors in the prime display. The negative priming effect was determined by grouping the target and distractors by common colour in the prime display. The results of the present experiments are consistent with inhibition-based models of negative priming.     

Labels can override perceptual categories in early infancy

An extensive body of research claims that labels facilitate categorisation, highlight the commonalities between objects and act as invitations to form categories for young infants before their first birthday. While this may indeed be a reasonable claim, we argue that it is not justified by the experiments described in the research. We report on a series of experiments that demonstrate that labels can play a causal role in category formation during infancy. Ten-month-old infants were taught to group computer-displayed, novel cartoon drawings into two categories under tightly controlled experimental conditions. Infants were given the opportunity to learn the two categories under four conditions: Without any labels, with two labels that correlated with category membership, with two labels assigned randomly to objects, and with one label assigned to all objects. Category formation was assessed identically in all conditions using a novelty preference procedure conducted in the absence of any labels. The labelling condition had a decisive impact on the way infants formed categories: When two labels correlated with the visual category information, infants learned two categories, just as if there had been no labels presented. However, uncorrelated labels completely disrupted the formation of any categories. Finally, consistent use of a single label across objects led infants to learn one broad category that included all the objects. These findings demonstrate that even before infants start to produce their first words, the labels they hear can override the manner in which they categorise objects.