Caricatures and Prototypes 1 Conceptual Interrelatedness and Caricatures

Concepts are interrelated to the extent that the characterization each concept is influenced by the other concepts, and isolated to the extent that the characterization of one concept is independent of other concepts. The relative categorization accuracy of the prototype and caricature of a concept can be used as a measure of concept interrelatedness. The prototype is the central tendency of a concept, whereas a caricature deviates from the concept’s central tendency in the direction opposite to the central tendency of other acquired concepts. The prototype is predicted to be relatively well categorized when a concept is relatively independent of other concepts, but the caricature is predicted to be relatively well categorized when a concept is highly related to other concepts. Support for these predictions comes from manipulations of the labels given to simultaneously acquired concepts (Experiment 1) and the order of categories during learning (Experiment 2). Caricatures and Prototypes 3 Conceptual Interrelatedness and Caricatures Concepts seem to be simultaneously connected to each other and to the external world. On the one hand, concepts seem to gain their meaning by the role that they play within a network of concepts (Collins & Quillian, 1969; Field, 1977). The notion of a “conceptual web” by which concepts all mutually define one another has been highly influential in all of the major fields that comprise cognitive science, including linguistics (Saussure, 1915/1959), computer science (Lenat & Feigenbaum, 1991), psychology (Landauer & Dumais, 1997), and philosophy (Block, 1999). However, there is also dissatisfaction in some quarters with the circularity of this conceptual web account. Researchers have argued that concepts must be grounded in the external world rather than merely related to each other (Harnad, 1990). The British empiricists argued that our conceptual ideas originate in recombinations of sensory impressions (Hume, 1740/1973). More recently, Barsalou (Barsalou, 1999; Goldstone & Barsalou, 1998; Solomon & Barsalou, 2001) has argued that concepts are not amodal, completely abstracted symbols, but rather are intrinsically perceptually based. In an attempt to reconcile arguments for a conceptual web and externally grounded concepts, Goldstone (1996) described a continuum between purely isolated and purely interrelated concepts, arguing that a concept is interrelated to the extent that its characterization is influenced by other concepts. For example, a Caucasian in a white majority culture may characterize black people in terms of their relation to white people (Linville & Jones, 1980). This Caucasian’s concept of black person would be dependent on and influenced by their white person concept, with the possible consequence that their representation of black people is distorted to accentuate perceived differences between the two races. Goldstone’s empirical basis for the continuum was the convergence of a set of experimental manipulations and measures of conceptual interrelatedness. A set of manipulations was designed to influence the degree of interrelatedness between simultaneously acquired concepts, and the influence of these manipulations was gauged by a set of measures of interrelatedness. These experiments gave support to the hypothesis that small experimental manipulations were capable of changing how influential one concept was on another concept’s representation and processing. The goal of the current experiments is to further test the claim for a continuum between isolated and interrelated concepts, using rich and naturalistic stimuli, and new manipulations and measures of interrelatedness. Caricatures and Prototypes 4 Isolated and Interrelated Concepts In evaluating the claim for a continuum between isolated and interrelated concepts, it is helpful to consider theories at the two poles of the continuum. We will consider representative models of isolated and interrelated concepts, leaving a fuller description to Goldstone (1996). Conceptual interrelatedness is a component of many linguistic treatments of concepts. Ferdinand de Saussure (1915/1959) argued that all concepts are completely “negatively defined,” that is, defined solely in terms of other concepts. He contended that “language is a system of interdependent terms in which the value of each term results solely from the simultaneous presence of the others” (p. 114) and that “concepts are purely differential and defined not in terms of their positive content but negatively by their relations with other terms in the system” (p. 117). This notion has evolved into the modern treatment of semantic networks (Collins & Quillian, 1969; Quillian, 1967). In these networks, concepts are represented by nodes in a network, and gain their functionality by their links to other concept nodes. Often times, these links are labeled, in which case different links refer to different kinds of relations between nodes. Dog would be connected to Animal by an Is-a link, to Bone by an Eats link, and to Paw by a Has-a link. Lenat and Feigenbaum have argued (1991) that interconceptual linkages are sufficient for establishing conceptual meanings even without any external grounding of the concepts. A computational approach to word meaning that has received considerable recent attention has been to base word meanings solely on the patterns of co-occurrence between a large number of words in an extremely large text corpus (Burgess, Livesay, & Lund, 1998; Burgess & Lund, 2000; Landauer & Dumais, 1997). Mathematical techniques are used to create vector encodings of words that efficiently capture their co-occurrences. If two words, such as “cocoon” and “butterfly” frequently co-occur in an encyclopedia or enter into similar patterns of cooccurrence with other words, then their vector representations will be highly similar. The meaning of a word, its vector in a high dimensional space, is completely based on the contextual similarity of the word to other words. Finally, researchers have argued that concepts are frequently characterized by their associative relations to other concepts. Barr and Caplan (1987) provide evidence, by having subjects list features associated with words, that many concepts are characterized by what they call “extrinsic features,” features that are “represented as the relationship between two or more entities” (p. 398). Caricatures and Prototypes 5 From the theories above, one might conclude that concepts cannot stand alone, and there could not be such a thing as a system with only one concept (Stich, 1983). However, if one looks at the field of pattern recognition rather than theories inspired by linguistics, then examples of isolated concepts become apparent. One way to conceive of an isolated concept is as a feature detector. A feature detector can become active when an input with a particular perceptual feature is present. Ascending in complexity, a concept can also be represented as a template in a physical or more abstract space (Edelman, 1999). Concepts are applied to external objects in order to determine the category membership based on the perceived input patterns of the objects. If patterns are categorized by comparing them with stored templates for categories, the representation of the categories do not depend on the other categories. A category’s representation is simply the image that best matches the members of the category. It is possible to have a feature detector or template for a concept without having any other concepts in the system. Categorizing an object may require comparing the relative degrees of match of the object to the representations for the candidate categories (Nosofsky, 1986), but if the categories themselves are represented by templates or feature detectors, then each can exist independently of the other categories. A comparison of some representative examples of interrelated and isolated concepts suggests a useful heuristic for assessing degree of interrelatedness. A Concept X is dependent on Concept Y to the extent that Concept X cannot exist without Concept Y. If the concept Vermicelli is represented as “thinner pasta than spaghetti” then no system could possess Vermicelli without also possessing Spaghetti. However, if the concept Vermicelli is represented by “a long pasta with a typical diameter of 6 mm” then possession of the concept does not depend upon possession of Spaghetti. Partial degrees of dependency owe to the multi-faceted nature of conceptual representations. A person’s concept of Vermicelli may incorporate both characterizations given above, and the relative importance of these characterizations determines how much Vermicelli’s representation is affected by the presence or absence of a Spaghetti concept. Prototypes and Caricatures Consider the example of two categories shown in Figure 1. Categories A and B each have 6 members, and each member has a unique combination of values on Dimensions 1 and 2. We will define the prototype of a category as the central tendency of the category along each of its dimensions (Posner & Caricatures and Prototypes 6 Keele, 1968; Reed, 1972). The prototype of Category A has a value of 2 on Dimension 1, while Category B’s prototype has a value of 4 on Dimension 1. In this usage, a prototype for a category would be the physical object that assumes average values along the dimensions that comprise the category’s members, rather than the internal representation of the category (Lakoff, 1987). In the experiments that follow, we use uniform distributions of dimension values in constructing categories, and consequently our description of a category prototype remains the same if we define central tendency as the average or median. -----------------------------INSERT FIGURE 1 ABOUT HERE ------------------------------We will define a caricature of a category as an object that assumes dimension values that depart from the central tendency of the category in the opposite direction of the central tendency of other simultaneously acquired concepts. In Figure 1, “X” represents the prototype of Category B, and “Y” represents a caricature of Category B. Y has a value of 5 on Dimension 1, and this value is a distortion of Category B’s central tendency in the direction opposite of Category A. This definition of caricature captures the intuitive notion of a caricature as an exaggeration. If Category B has a large value on Dimension 1 (relative to Category A), then the caricature of Category B exaggerates this large value, in the same way that a newspaper caricature exaggerates distinctive facial features of a politician. The term “caricature” is often reserved for exaggerations of a member of a category from the category’s prototype. A caricature of a face typically distorts the face away from a prototypical face (Rhodes, Brennan, & Carey, 1987). Although our use of the term “caricature” is consistent with this traditional use in that both refer to distortions of a representation away from contrasting objects, our use extends the traditional use to cover situations where the contrasting objects comprise an alternative, rather than superordinate, category. If the prototype for a superordinate category such as face is established, then it can be used as the standard away from which all caricatures of specific faces are distorted. In the current research, we pursue the alternative strategy of creating two categories based on novel faces, and treating the caricature of one category as the distortion of the face away from the other simultaneously acquired category rather than a prototypical face. The current experiments investigate the conditions under which the prototype or caricature of a category is more easily categorized. On the one hand, one might predict better categorization for the prototype because it is, by definition, the item that is most similar to the members of its category (Posner & Caricatures and Prototypes 7 Keele, 1968; Rosch, 1975). On the other hand, one might predict better categorization accuracy for the caricature because it emphasizes a distinctive value for a category. In fact, there have been several experiments that have found a categorization advantage for caricatures relative to prototypes (Goldstone, 1996; Nosofsky, 1991; Palmeri & Nosofsky, 2001; Rhodes, Brennan, & Carey, 1987). Our current goal is not to argue that caricatures or prototypes are better categorized, but to identify experimental factors that modulate the benefit of one over the other. Specifically, experimental factors that promote relatively isolated concepts should tend to promote an advantage for prototypes over caricatures. In the absence of interconceptual influences, the representation that best exemplifies a concept will be its central tendency. If we try to represent Category B and do not know anything about Category A, then our best representation of Category B will be the point marked “X” in Figure 1. However, if a concept is characterized relative to other simultaneously acquired concepts, then characterizations of the form “Concept B members have relatively large Dimension 1 values compared to Concept A” will be formed. Caricatures fit these relational descriptions better than do prototypes. Nosofsky (1991) argued that classification of an object into a category depends on both its absolute similarity to members of the category, and its relative similarity to members of all of the candidate categories. The current experiments explore factors that affect the relative importance of these absolute and relative determinants of categorization. We instantiate caricatures and prototypes by face stimuli that are formed using automatic morphing software developed by Steyvers (1999). An example of caricaturization using this software is shown in Figure 2. A prototypical bald head was generated by combining together 62 bald heads taken from Kayser (1997). To create this prototype without the blurred quality typical of superimposed face photographs (Busey, 1998; Galton, 1878), 127 defining points were found for each of the 62 heads, and the average location for each point was assigned to the average face. The gray scale values of corresponding pixels across the 62 heads were blended to create the gray scale values for the average face. The caricature (shown on the right in Figure 2) of a particular face (shown in the middle of Figure 2) was generated by taking each of the 127 defining points on the face, and distorting their positions by 20% away from the defining points on the average face. In the experiments to be reported, the caricatures are distortions away from an alternative face category rather than the prototype for the superordinate category of all faces shown in Figure 2. Caricatures and Prototypes 8 -----------------------------INSERT FIGURE 2 ABOUT HERE ------------------------------Manipulations of Conceptual Interrelatedness The relative ease of categorizing prototypes and caricatures will be used as the measure of concept interrelatedness. The other main task is to develop experimental manipulations that are predicted to affect the interrelatedness of concepts. There already exists a literature suggesting such manipulations. Goldstone (1996) found that interrelated categories were promoted when 1) participants were encouraged to look for features that discriminated between the learned categories, 2), categories were alternated frequently, 3) participants were practiced categorizers, and 4) stimuli were relatively undistorted versions of the categories’ prototypes. In contrast, isolated categories were promoted when 1) participants were encouraged to form images of the categories, 2) categories were alternated rarely, 3) participants were relatively unpracticed, and 4) stimuli were highly distorted versions of the categories’ prototypes. Niedenthal and Beike (1997) explored whether people’s self-concepts were relatively independent of other people, or relationally defined. They found that self-concepts were relatively interrelated when participants were asked to consider their distinctive attributes, and when participants were younger siblings comparing themselves to older siblings rather than vice versa. McKenzie (1998, 1999) has explored the conditions under which finding out information about one hypothesis affects mutually exclusive alternative hypotheses. Even in situations where participants are told that a patient has one and only one of two candidate diseases, evidence that increases participants’ confidence that the patient has one disease does not always decrease confidence that the patient has the other disease. Presenting the two diseases concurrently rather than successively makes the diagnoses of the diseases more (negatively) dependent on one another, as does mentioning both diseases when participants make confidence judgments about each disease. The current research explores two experimental manipulations that might be expected to affect concept interrelatedness. Experiment 1 manipulates the labels given to categories being acquired. One category is given a positive label (Club A) and the other category is given a negation label (Not Club A). Although this manipulation was used by Goldstone (1996), it has never been used with the caricature versus prototype measure of interrelatedness, and never used with naturalistic stimuli. Experiment 2 manipulates Caricatures and Prototypes 9 the order of learning categories, testing the hypothesis that the first learned category will be relatively isolated, while the second category tends to be characterized relative to the first category. Faces as Categories One of the goals of the experiments is to explore interrelated and isolated categories using materials that are natural and familiar to participants. A particularly apt domain for exploring the categorization of caricatured representations is faces, from which the notion of caricatures originally developed. An interesting empirical question is whether the prior results indicating caricature-based representations for interrelated categories will generalize to categories defined by faces. One reason for questioning the applicability of the previous results to faces is that faces are often thought to be processed by special-purpose mechanisms that are functionally and physiologically distinct from those involved in recognizing other objects (Bentin et al., 2002; McCarthy et al., 1997; Kanwisher, McDermott, & Chun, 1997). Researchers have argued that faces are typically processed holistically, in terms of entire face patterns rather than analyzed into separate components (Maurer, Le Grand, & Mondloch, 2002; Tanaka & Farah, 1993). If faces are processed holistically, one might predict that a face category would be represented by its prototypical image, rather than by a caricature that selectively exaggerates certain components of the face. However, previous research has indicated better categorization for caricatures rather than prototypes of faces (McKone, Martini, & Nakayama, 2001; Rhodes, Brennan, & Carey, 1987), consistent with previous results with unfamiliar objects. When choosing faces as materials, several levels of categorization are possible. The categories to be learned can be defined by particular individuals; by groups of individuals based on gender, race, emotion, age, or another category; or by the superordinate category of faces (versus non-faces). The framework of isolated and interrelated concepts is general enough to be applicable at all of these levels of face categorization. The current experiments focus on categories defined by particular individuals, and the variability within a category is obtained by gradually transforming the identity of a face using morphing software (Steyvers, 1999). Several other researchers have created morph-based continua between faces, and tested people’s ability to discriminate or categorize the faces (Beale & Keil, 1995; McKone et al., 2001; Levin & Beale, 2000). A commonly obtained result from these studies is that categorization accuracy for a face increases monotonically as a function of the distance of the face from the category boundary (see also Ashby & Maddox, 1993). However, in these studies, the endpoints of the continua are presented as the Caricatures and Prototypes 10 representatives of the categories, and this may explain why these endpoints are well categorized rather than their status as caricatures. The current experiments explore whether a caricature advantage for categorization is still found even when prototypes are presented as often as caricatures during category learning. The experiments also explore factors that modulate the degree of the caricature advantage. Experiment 1 Numerous studies have shown that the labels given to categories of objects influence the characterization of those categories (Harnad, 1987; Malt et al., 1999; Waxman, 1990; Wisniewski & Medin, 1994). One example that is particularly related to concept interrelatedness is the mutual exclusivity hypothesis (Markman, 1990; Waxman, Chambers, Yntema, & Gelman, 1989). Children adopting this hypothesis determine the referent of a noun by assuming that nouns are mutually exclusive, and consequently, if a new term is applied to one of two objects and one object already has a name, children will tend to assume that the term refers to the other object. Similar to Saussure’s (1915/1959) notion of competition between concepts, the mutual exclusivity hypothesis assumes that as one concept gains control of a conceptual region, its competitor concepts lose control of the region. This is the same competition that is predicted to make interrelated concepts increasingly characterized by their caricatures rather than prototypes. In both cases, a category is displaced away from another category. Labeling may make pairs of concepts asymmetrically dependent on one another. One concept can be labeled “Category A” while another concept is labeled “ not Category A.” In this case, the concept labeled “Not Category A” is predicted to be more influenced by “Category A” than vice versa (Clark, 1990). The concept that has a label that refers to another concept is predicted to be highly influenced by the referenced concept. Even though the category structures are symmetric, and the labels are randomly assigned to two categories, the “Not Category A” concept is predicted to be characterized more in terms of a caricature than the “Category A” concept. More specifically, there should be a tendency to associate the “Not Category A” concept with a stimulus that is more of a caricature than the stimulus associated with the “Category A” concept. There may be a bias to associate both categories with caricatures rather than prototypes (Goldstone, 1996; Palmeri & Nosofsky, 2001; Rhodes, Brennan, & Carey, 1987), but the extent of caricaturization is predicted to be greater for the “Not Category A” concept. If we compare the relative Caricatures and Prototypes 11 categorization accuracies for prototypes and caricatures, we predict the advantage of caricatures over prototypes to be larger for the concept labeled “Not Category A” than for the “Category A” concept. The basis for this prediction comes from a combination of two assumptions. First, the more dependent Concept A is on a mutually exclusive Concept B, the more Concept A’s characterization will be caricatured away from Concept B. This assumption is supported by previous research showing that mutually exclusive concepts that are highly interrelated tend to be characterized by features that distinguish each from the other (Goldstone, 1996; Kruschke, 1996; Niedenthal & Beike, 1997). Exaggerating distinctive properties of a concept through caricature allows a concept to be differentiated from its close conceptual neighbors. This exaggeration will be the most extensive along attributes that distinguish a concept from others concepts that are most likely to be brought to mind by the concept – that is, by other concepts that most influence the concept in question. The second assumption is that explicitly labeling Concept B as not being Concept A makes Concept B dependent on Concept A. This assumption is supported by previous work showing that when a concept’s label explicitly makes reference to another concept, then this referenced concept exerts more influence on the referencing concept’s use and representation than vice versa (Goldstone, 1996; Van Wallendeal & Hastie, 1990). Putting these two assumptions together, we predict that the extent of the categorization advantage for caricatures over prototypes will be greater for a category labeled “Not club members” than for a category labeled “club members.” Method Participants. Sixty-two undergraduate students from Indiana University served as participants in order to fulfill a course requirement. The students were split evenly into the two labeling conditions. Materials . The stimuli were faces that were generated by morphing between photographs of two bald heads selected from Kayser (1997). Previous research has suggested that morphs generated from the two selected faces did not introduce conspicuous non-linearities between physical and psychological scalings (Goldstone & Steyvers, 2001). The morph sequence of 10 faces used is shown in Figure 3. Each of the morphs was automatically generated using a morphing technique described by Steyvers (1999). Applying this technique, the main contours in the face images were delineated by 127 control lines. These control lines served to align the features of the two faces. In the warping phase of this morphing algorithm, correspondences were calculated between the pixels of all the images to be morphed. Then, in the crossdissolving phase, the gray scale values of corresponding pixels were blended to create the gray scale values Caricatures and Prototypes 12 of the resulting morph image. The faces on the left and right ends of Figure 3 are actual faces, and the 8 intermediate faces are blends of the two actual faces, with the proportion of the left face shifting along the series in equal 11.11% increments. -----------------------------INSERT FIGURE 3 ABOUT HERE ------------------------------The prototype for a category was defined as the center face within the category’s set of five faces. This is the face that is the most similar, on average, to the other faces within its category. The caricature of a category is defined as the face that is least like the faces from the other category. The face between the caricature and the prototype is also a systematically caricatured face relative to the prototype, but to a lesser extent. Each face was displayed in grayscale with 256 possible brightness values per pixel (one pixel = .034 cm), and measured 14.48 cm tall by 11.68 cm wide. Each face was photographed against a dark background and displayed on a white Apple Imac computer screen. The average viewing distance was 46 cm. Procedure . On each trial, participants saw a face and categorized it by pressing either “Y” or “N” on the keyboard, with feedback on each trial from the computer indicating with a check or an “X” whether or not the participant was correct, and also indicating the correct category assignment for the face. Participants were instructed: "You will see faces appear on the screen. Half of them belong to certain club, while the remaining half do not. If you think that a face belongs to the club, press the ‘Y’ key for ‘Yes.’ If you think that it does not belong to the club, press the ‘N’ key for ‘NO.’” The dividing line between club members and non-club members is shown by the vertical line in Figure 3. For half of the participants, those in Group 1, the first five faces were club members, and the last five faces were not club members. For Group 2, the first five faces were not club members, and the last five faces were club members. The experiment consisted of 60 repetitions of the 10 faces shown in Figure 3, for a total of 600 trials. The order of the 600 trials was randomized. The placement of a face’s center was also randomized within a 6 X 6 cm square in the center of the screen. Each trial began with a face appearing on the screen. The face remained on the screen until the participant pressed the “Y” or “N” key. Immediately after pressing one of the keys, feedback was given to the participant, and after 1.5 seconds, the screen was erased. Written feedback was in the form of “Yes, this face is a club member,” “No, this face is not a club member,” “Yes, this face is not a club member,” or “No, this face is a club member.” The blank interval between trials was 1 second. Caricatures and Prototypes 13 Participants were given breaks every 100 trials. During these breaks, participants were informed of their accuracy and speed during the preceding block. Results The primary data of interest was participants’ accuracy at categorizing particular faces into the two categories. Accuracy averaged over the 600 trials was variable enough that it was a sensitive dependent measure, and was less noisy than response times. However, response times closely mirrored the accuracy data. The categorization accuracies for each face and each group of participants are shown in Figure 4. The data reveal a bias to respond “Club” rather than “Not club.” This is shown by the horizontal offset between the two lines. Overall, participants responded “Club” and “Not club” on 54% and 46% of trials respectively, paired t-test t(61)=6.24, p< .01. The accuracy results were submitted to a repeated measures ANOVA with within-subject factors for the stimulus presented (10 levels) and the block (2 levels, for the first and seconds halves of the experiment), and a between-subject factor for the labeling condition (Faces 1-5 “Club A” or Faces 1-5 “Not Club A”). There was a main effect of stimulus presented, interpretable as the stimuli closer to the category boundary being harder to categorize than stimuli near the endpoints, F(9, 540)=34.59, MSE=4.20, p<.001. There was also a main effects of block, F(1, 60)=23.49, MSE=3.86, p<.001, but not of labeling. There were interactions between stimulus presented and labeling, F(9, 540)= 3.45, MSE=3.55, p<.01, and between stimulus presented and block, F(9, 540)=2.38, MSE=3.18, p<.05, but not between labeling and block. To directly test hypotheses that compare categorization accuracy of caricatures and prototypes we conducted a restricted analysis in which only Faces 1, 3, 8, and 10 were considered. Faces 1 and 10 were coded as caricatures and Faces 3 and 8 were coded as prototypes. The labeling variable was recoded to take into account the interaction between labeling and presented stimulus. Given that the results for the 10 faces were highly symmetric across the categorization boundary, we created a new labeling variable that combined responses from Faces 1-5 with those from Faces 6-10. Labels were recoded as “negation labels” when Faces 1 and 3 were presented and Faces 1-5 were “Not Club A” or when Faces 8 and 10 were presented and Faces 1-5 were “Club A.” Labels were recoded as “standard labels” when Faces 1 and 3 were presented and Faces 1-5 were “Club A” or when Faces 8 and 10 were presented and Faces 1-5 were “Not Club A.” With this recoding, the computed ANOVA was now a repeated measure ANOVA with three within-subject factors of Stimulus (prototype or caricature), Labeling (standard or negation), and Block (first block or second). This restricted analysis yielded a significant main effect of stimulus presented, with caricatures being more Caricatures and Prototypes 14 accurately categorized than prototypes, with respective accuracies of 78.3% and 72.6%, F(1,61)=8.85, MSE=3.13, p < .01. There was no main effect of labeling, but there was a significant interaction between stimulus presented (prototype or caricature) and labeling, F(1, 61) = 8.25, MSE=3.18, p<.01. For negation labeled faces, caricatures and prototypes were categorized with respective accuracies of 78.6% and 68.6%. For standard labeled faces these respective accuracies are much more similar, at 78.0% and 76.6%. The faces between the caricature and the prototype (Faces 2 and 9) produced intermediate results to the caricature and prototype for both labeling conditions. Still using the ANOVA restricted to Faces 1, 3, 8, and 10, exploratory analyses revealed a main effect of blocks, with accuracy lower for the first 300 trials (70.7%) than for the final 300 trials (80.3%), F(1, 61)=13.76, MSE=2.92 p < .001. However, there was also an interaction between stimulus presented and block, F(1, 61)=9.14, MSE=4.14, p<.01. For early trials, caricatures and prototypes were categorized with respective accuracies of 72.0% and 69.4%, and for late trials, these percentages increased to 84.6% and 76.0%. This interaction indicates that the categorization advantage for caricatures over prototypes increases with practice. This is consistent with earlier results (Goldstone, 1996) and is predicted by the hypothesis that in a categorization task where two categories are frequently alternated, the categories will become increasingly interrelated. As categories become more interrelated, the degree of caricaturization of the category representations is expected to increase. -----------------------------INSERT FIGURE 4 ABOUT HERE -------------------------------