Achromatic color categories

Sternheim and Boynton’s (1966) continuous judgmental color-naming technique was used to assess the elemental nature of achromatic colors. Results from three subjects indicate that the names “black” and “white” are necessary and sufficient to describe achromatic test lights associated with the name “gray.” On the basis of the criteria that establishes a color as elemental, black and white were confirmed as elemental, whereas gray was not. These findings support the assumption of various opponent processing models that there are only two elemental achromatic sensations. In addition, the results support Hering’s (1920) view that black and white differ from the paired-chromatic opponent colors in that they are not mutually exclusive.     

Achromatic surface color and the direction of illumination

Hochberg and Beck (1954) found that an objectively upright trapezoid, when illuminated from above, appeared darker if it was viewed monocularly and lighter if it was viewed binocularly. Illuminated from in front, the same trapezoid then appeared lighter under monocular and darker under binocular viewing. Since the target appeared slanted under monocular but upright under binocular viewing, these changes in apparent lightness could be attributed, wholly or in part, to the apparent angle of incidence of the illumination on the surface. In two experiments, when 8-min periods of dark adaptation were introduced between monocular and binocular viewing, but when the arrangements were otherwise approximately the same as those of Hochberg and Beck, their results could not be observed. A third experiment demonstrated that the monocularly observed trapezoids did appear slanted.     

Color categories and color appearance

We examined categorical effects in color appearance in two tasks, which in part differed in the extent to which color naming was explicitly required for the response. In one, we measured the effects of color differences on perceptual grouping for hues that spanned the blue-green boundary, to test whether chromatic differences across the boundary were perceptually exaggerated. This task did not require overt judgments of the perceived colors, and the tendency to group showed only a weak and inconsistent categorical bias. In a second case, we analyzed results from two prior studies of hue scaling of chromatic stimuli (De Valois, De Valois, Switkes, & Mahon, 1997; Malkoc, Kay, & Webster, 2005), to test whether color appearance changed more rapidly around the blue-green boundary. In this task observers directly judge the perceived color of the stimuli and these judgments tended to show much stronger categorical effects. The differences between these tasks could arise either because different signals mediate color grouping and color appearance, or because linguistic categories might differentially intrude on the response to color and/or on the perception of color. Our results suggest that the interaction between language and color processing may be highly dependent on the specific task and cognitive demands and strategies of the observer, and also highlight pronounced individual differences in the tendency to exhibit categorical responses.     

Recalibrating color categories using world knowledge

ABSTRACT— When the perceptual system uses color to facilitate object recognition, it must solve the color-constancy problem: The light an object reflects to an observer's eyes confounds properties of the source of the illumination with the surface reflectance of the object. Information from the visual scene ( bottom-up information) is insufficient to solve this problem. We show that observers use world knowledge about objects and their prototypical colors as a source of top-down information to improve color constancy. Specifically, observers use world knowledge to recalibrate their color categories. Our results also suggest that similar effects previously observed in language perception are the consequence of a general perceptual process.     

Language is not necessary for color categories

The origin of color categories is under debate. Some researchers argue that color categories are linguistically constructed, while others claim they have a pre‐linguistic, and possibly even innate, basis. Although there is some evidence that 4–6‐month‐old infants respond categorically to color, these empirical results have been challenged in recent years. First, it has been claimed that previous demonstrations of color categories in infants may reflect color preferences instead. Second, and more seriously, other labs have reported failing to replicate the basic findings at all. In the current study we used eye‐tracking to test 8‐month‐old infants’ categorical perception of a previously attested color boundary (green–blue) and an additional color boundary (blue–purple). Our results show that infants are faster and more accurate at fixating targets when they come from a different color category than when from the same category (even though the chromatic separation sizes were equated). This is the case for both blue–green and blue–purple. Our findings provide independent evidence for the existence of color categories in pre‐linguistic infants, and suggest that categorical perception of color can occur without color language.     

The development of color categories in two languages: a longitudinal study

This study unites investigations into the linguistic relativity of color categories with research on children's category acquisition. Naming, comprehension, and memory for colors were tracked in 2 populations over a 3-year period. Children from a seminomadic equatorial African culture, whose language contains 5 color terms, were compared with a group of English children. Despite differences in visual environment, language, and education, they showed similar patterns of term acquisition. Both groups acquired color vocabulary slowly and with great individual variation. Those knowing no color terms made recognition errors based on perceptual distance, and the influence of naming on memory increased with age. An initial perceptually driven color continuum appears to be progressively organized into sets appropriate to each culture and language.     

Color-function categories that prime infants to use color information in an object individuation task

There is evidence for developmental hierarchies in the type of information to which infants attend when reasoning about objects. Investigators have questioned the origin of these hierarchies and how infants come to identify new sources of information when reasoning about objects. The goal of the present experiments was to shed light on this debate by identifying conditions under which infants’ sensitivity to color information, which is slow to emerge, could be enhanced in an object individuation task. The outcome of Experiment 1 confirmed and extended previous reports that 9.5-month-olds can be primed, through exposure to events in which the color of an object predicts its function, to attend to color differences in a subsequent individuation task. The outcomes of Experiments 2-4 revealed age-related changes in the nature of the representations that support color priming. This is exemplified by three main findings. First, the representations that are formed during the color-function events are relatively specific. That is, infants are primed to use the color difference seen in the color-function events to individuate objects in the test events, but not other color differences. Second, 9.5-month-olds can be led to form more abstract event representations, and then generalize to other colors in the test events if they are shown multiple pairs of colors in the color-function events. Third, slightly younger 9-month-olds also can be led to form more inclusive categories with multiple color pairs, but only when they are allowed to directly compare the exemplars in each color pair during the present events. These results shed light on the development of categorization abilities, cognitive mechanisms that support color-function priming, and the kinds of experiences that can increase infants’ sensitivity to color information.     

Achromatic color matching as a function of apparent target orientation,target and background luminance,and lightness or brightness instructions

In two experiments, achromatic color matches for a fixed target, under constant illumination, were compared under conditions where the target appeared perpendicular to illumination direction and coplanar with the background (monocular viewing) and where the target appeared nonperpendicular to illumination direction and noncoplanar with the background (binocular viewing). Contrary to the coplanar ratio hypothesis, which predicts a “lightening” of the target seen coplanar with a darker background, a general “darkening” of the target occurred for both white (N9.5/) and black (N3/) backgrounds and for both dark (N5.5/) and light (N6.5/) targets. This darkening effect was greatest for the darker target and black background, and approximately equal for other combinations of target value and background. The direction of the darkening effect is consistent with the albedo hypothesis, which assumes an inferential correction for changes in conditions of illumination. However, variation in the magnitude of the darkening effect is problematical, and cannot be easily explained by any existing theory. In both experiments, instructions to judge “lightness” or “brightness” failed to produce any substantial differences in performance, although postexperimental questioning suggested that subjects had a verbal understanding of these concepts. Apparently, under reduction conditions, subjects lack cues to illumination and make only lightness matches, regardless of instructions.     

Achromatic transparency and the role of local contours

In this paper we investigate the role of contours and junctions in the perception of single-plane achromatic transparency. In order to measure the accuracy with which observers encode transparency, a six-luminance stimulus was employed in which the figural properties could be easily manipulated. Accuracy was measured by requiring subjects to select (either by the method of adjustment or by using a forced-choice procedure) the luminance that best completed a simulated transparent filter. The X junctions in the stimulus were destroyed or perturbed in three experiments. Simple occlusion of the junction (experiment 1), and perturbation of the orientation of the contours of the filter as they pass through the junction (experiment 3) resulted in small but significant reductions in performance. On the other hand, a sudden change in orientation of the background (material) contours (experiment 2) resulted in a small but significant enhancement of overall performance compared with the control stimulus. In the forced-choice task, reversals in the polarity of contours (as defined by the brightness order of flanking regions) around the junction were shown to effect large changes in subjects' accuracy in processing transparency. The overall results show that X and Psi junctions are indeed salient properties of transparent stimuli. The findings suggest that jagged contours with sudden changes in direction are more likely to be attributed to reflectance (material) changes than to changes due to a transparent filter (or to illumination).     

Relational categories are more mutable than entity categories

Across three experiments, we explore differences between relational categories—whose members share common relational patterns—and entity categories, whose members share common intrinsic properties. Specifically, we test the claim that relational concepts are more semantically mutable in context, and therefore less stable in memory, than entity concepts. We compared memory for entity nouns and relational nouns, tested either in the same context as at encoding or in a different context. We found that (a) participants show better recognition accuracy for entity nouns than for relational nouns, and (b) recognition of relational nouns is more impaired by a change in context than is recognition of entity nouns. We replicated these findings even when controlling for factors highly correlated with relationality, such as abstractness–concreteness. This suggests that the contextual mutability of relational concepts is due to the core semantic property of conveying relational structure and not simply to accompanying characteristics such as abstractness. We note parallels with the distinction between nouns and verbs and suggest implications for lexical and conceptual structure. Finally, we relate these patterns to proposals that a deep distinction exists between words with an essentially referential function and those with a predicate function.     

Cody's categories

Of the two main interpretations of Spinoza's theory of the identity of the attributes, in particular those of Thought and Extension, the objective interpretation is now almost universally preferred to the subjective. Rejection of the subjective interpretation, according to which the attributes are merely our ways of cognizing a reality whose real essence remains unknown, is certainly justified, but the objective theory comes too near to replacing the identity by a mere correlation of différents to be quite satisfactory. Is it not better to say that Thought and Extension represent two complementary conceptions of reality which are both correct? Yes, but in saying so some commentators ascribe to mind, as Spinoza conceives it, an unplausibly abstract status. An alternative proposal is made as to a way in which Spinoza might be right in essentials, though it requires that a certain tension in Spinozism as to the nature of body be resolved in a particular direction.     

Natural categories

The hypothesis of the study was that the domains of color and form are structured into nonarbitrary, semantic categories which develop around perceptually salient “natural prototypes.” Categories which reflected such an organization (where the presumed natural prototypes were central tendencies of the categories) and categories which violated the organization (natural prototypes peripheral) were taught to a total of 162 members of a Stone Age culture which did not initially have hue or geometric-form concepts. In both domains, the presumed “natural” categories were consistently easier to learn than the “distorted” categories. Even when not central, natural prototype stimuli tended to be more rapidly learned and more often chosen as the most typical example of the category than were other stimuli. Implications for general differences between natural categories and the artificial categories of concept formation research were discussed.     

Ad hoc categories

People construct ad hoc categories to achieve goals. For example, constructing the category of “things to sell at a garage sale” can be instrumental to achieving the goal of selling unwanted possessions. These categories differ from common categories (e.g., “fruit,” “furniture”) in that ad hoc categories violate the correlational structure of the environment and are not well established in memory. Regarding the latter property, the category concepts, concept-to-instance associations, and instance-to-concept associations structuring ad hoc categories are shown to be much less established in memory than those of common categories. Regardless of these differences, however, ad hoc categories possess graded structures (i.e., typicality gradients) as salient as those structuring common categories. This appears to be the result of a similarity comparison process that imposes graded structure on any category regardless of type.     

Variation within categories

Two aspects of variation within categories, relating to different models of categorization, were investigated — frequency of dimensional values and typicality differences within values. The influence of range of typicality experienced during learning and of informational value of feedback was also studied. Finally, differential forgetting of values was examined.In the experiment subjects learned to categorize faces, and then performed a classification test task and pairwise comparisons of faces. A variety of dependent variables was employed, including the galvanic skin response (GSR).Typicality and frequency of values appeared to influence categorization performance independent of each other. It was concluded that both prototype distance models and frequency models explain different aspects of variation within the same categories, and that models of classification should account for frequency of values in contrasting categories. Results showed furthermore (1) the influence of typicality range on the extension of a category; (2) no influence of specific feedback regarding representativeness of a face; (3) less decay with more important values; and (4) a positive relationship between uncertainty reduction and GSR.     

Inference using categories

How do people use category membership and similarity for making inductive inferences? The authors addressed this question by examining the impact of category labels and category features on inference and classification tasks that were designed to be comparable. In the inference task, participants predicted the value of a missing feature of an item given its category label and other feature values. In the classification task, participants predicted the category label of an item given its feature values. The results from 4 experiments suggest that category membership influences inference even when similarity information contradicts the category label. This tendency was stronger when the category label conveyed class inclusion information than when the label reflected a feature of the category. These findings suggest that category membership affects inference beyond similarity and that category labels and category features are 2 different things.     

Abstract coherent categories

Many studies have demonstrated the importance of the knowledge that interrelates features in people's mental representation of categories and that makes our conception of categories coherent. This article focuses on abstract coherent categories, coherent categories that are also abstract because they are defined by relations independently of any features. Four experiments demonstrate that abstract coherent categories are learned more easily than control categories with identical features and statistical structure, and also that participants induced an abstract representation of the category by granting category membership to exemplars with completely novel features. The authors argue that the human conceptual system is heavily populated with abstract coherent concepts, including conceptions of social groups, societal institutions, legal, political, and military scenarios, and many superordinate categories, such as classes of natural kinds.     

Learning time-varying categories

Many kinds of objects and events in our world have a strongly time-dependent quality. However, most theories about concepts and categories either are insensitive to variation over time or treat it as a nuisance factor that produces irrational order effects during learning. In this article, we present two category learning experiments in which we explored peoples’ ability to learn categories whose structure is strongly time-dependent. We suggest that order effects in categorization may in part reflect a sensitivity to changing environments, and that understanding dynamically changing concepts is an important part of developing a full account of human categorization.