Seeing Caricatures

The representation of the vision is present at the age of one month when is possible to recognize the prototype face. Three month olds relate information from one face to another and begin to form a category of face (de Haan, 2002). Studies on how adults learn perceptual categories indicate that recognition of individual examples is mediated by substrates other than categorization of the prototype (Knowlton, 1993): the former is mediated by medial temporal lobe structures while the latter involves visual cortex (Aizenstein, 2000). In adults both of these regions are believed to be part of the network involved in storing knowledge about object categories (Riesenhuber, 2000; Martin, 2001). Infants’ early ability to recognize individual faces may rely on medial temporal lobe structures whereas the subsequent ability to form mental prototypes and use these to facilitate new face encoding may rely on further development of areas of the ventral visual cortical pathway. Research indicates that adult like characteristics of face processing are present by 12 months of age (de Haan, 2002). Williams syndrome, a pathological disease that involves the visual pathways or object recognition, is characterized by mild to moderate mental retardation together with an uneven cognitive profile, with poor visuo-spatial skills but relatively intact verbal and face-processing skills (Donnai, 2000). In adults, damage to the occipito-temporal cortex can result in prosopognosia, a selective impairment in face processing (Nunn, 2001).

Fig. 1

enlarge image

+

Artists draw facial caricature using expressions to reflect a personal impression of a face. Facial caricatures reflecting subjective impression can be drawn by using expressions and caricatures can be recognized objectively even with features exaggerated [Fig. 1]. Iwashita (Iwashita, 2001) proposes that a caricature resembling the model is drawn better by the addition of appropriate exaggeration.

Indeed, caricatures of famous faces are recognized more quickly and accurately than life-like drawings. Studies on caricature recognition indicate that hair is not necessarily a consistent feature in media pictures of the famous and one or both ears were not always visible and therefore accurately delineated. Mode of representation (curve or line) was not found to have a significant influence on the level of exaggeration (Benson, 1994). Eyes are a very salient feature for recognition of identity, because the eye area is relatively invariant during development (Dal Martello, 2006). The eyes were recognized significantly better on caricatures than on anti caricatures. The effectiveness of caricatures varies from faces to face and according to experimental conditions, but these differences did not relate to differences in initial distinctiveness (Rhodes, 1996). The caricaturing age ranges of older faces does not seem to substantially alter the principal finding relating to ageing, distinctiveness, attractiveness and memorability (Defferenbacher, 1998). The validity of individual physical face spaces as model of human perception can be tested by comparing the properties of physical and perceptual face spaces (Lee, 2000).

The advantage of caricature in face recognition tends to be limited to an improvement in reaction time rather than recognition accuracy. Exaggeration aids recognition of face. Interaction between recognition advantage for image type and duration suggests that brief exposure is a form of image degradation and favours right-hemisphere configural processing (Blanz, 2000). Component and configural processes underlying face perception have also been investigated (Sergent, 1984). Caricatures appear advantageous at short image durations when the image is particularly degraded. Such degradation may favour configural processing, allowing amplification of distinctive benefit recognition, while at the same time preventing the subject from realizing that caricatures are in fact large distortions of reality. Caricatures have been found to improve recognition accuracy or to reduce reaction time. The presence of colour, incorporating textual information, has already been shown to be an important aid to face recognition and may be more pertinent when subjects are required to identify a best likeness (Lee, 1997; Zeki, 2005). It appears that subjects also select images with higher contrast as best likeness. In colour-caricatured and contrast-enhanced pictures, images appear to stand out more from the background and seem more vivid to the viewer. Contrast exaggeration does not exaggerate information with respect to the face, so a level of caricature near the veridical need not necessarily be predicted. The level of exaggeration required to bring out facial structure and wrinkles is lower when caricaturing against a face norm than against a neutral-grey image. At the shortest exposure, caricatures were found more accurately identified than veridical images. No differences were identified between veridical and shape caricatured representations at other exposure duration. Caricatured representation rather than veridical representation are stored in memory. In this case, caricatured images seem to be more efficiently recognized because they are closer to the stored representations. Caricaturing aids the process of matching the veridical representation of the input image. Enhancing contrast produces a better likeness but does not aid recognition (Lee, 1997).

Cartoons may present an essential minimum of information for building up an adequate structural description, and also a particularly ‘good’ minimum because they so closely resemble the kind of structural description that would be built up anyway. We are capable of extracting the essential skeleton-like structure from a normal natural image; we can also do this for a blurred natural image. Each ‘blob’ in a blurred image could easily be explicitly described with a separate symbol, and associated with this symbol, could be a statement of the blob’s position, and its rough orientation and extent, just as for the ‘blob’ of oriented line elements in the feature description of the image or figure (Frisby, 1980).

References

• Aizenstein HJ, MacDonald AW, Stenger VA, Nebes RD, Larson JK, Ursu S and Carter CS. Complementary category learning system identified using event-related functional MRI. J Cogn Neurosci 12: 977-987; 2000
• Benson PJ, Perrett DI. Visual processing of facial distinctiveness. Perception 23: 75-93; 1994
• Berger A. An equilibrium of idiocies: Cartoons as instant communication. Publ Commun Review 1: 42–46; 1981
• Blanz V, O’Toole AJ, Vetter T, Wild HA. On the other side of the mean: the perception of dissimilarity in human faces. Perception 29:885-891; 2000
• Dal Martello MF and Maloney LT. Where are kin recognition signals in the human face? J Vis 6: 1356-1366; 2006
• de Haan M, Pascalis O, Johnson MH. Specialization of neural mechanism underlying face recognition in human infants. J Cogn Neurosci 14: 199-209; 2002
• de Haan M, Humphreys K and Johnson MH. Developing a brain specialized for face perception: a converging method approach. Dev Psychobiol 40:200-212; 2002
• Deffenbacher KA, Vetter T, Johanson J, O’Toole AJ. Facial aging, attractiveness, and distinctiveness. Perception 27:1233-1243; 1998
• Donnai D, Karmiloff-Smith A. Williams syndrome: from genotype through the cognitive phenotype. Am J Med Genet 97: 164-171; 2000
• Frisby J P (1980) Seeing. Illusion, Brain and Mind. Oxford: Oxford University Press; 1980
• Iwashita S and Onisawa T. Facial caricature drawing based on subjective impression. Systems and Computers in Japan 23: 891-900; 2001
• Hartley L. Physiognomy and the meaning of expression in nineteenth-century culture. Cambridge, Cambridge University Press; 2001
• Kennedy JM, Silver J. The surrogate functions of lines in visual perception: evidence form antipodal rock and cave artwork sources. Perception 3:313-322; 1974
• Knowlton BJ and Squire LR. The learning of categories: parallel brain systems for item memory and category knowledge. Science 262:1747-1749; 1993
• Lee JK, Perrett DI. Presentation-time measures of the effects of manipulation in colour space on discrimination of famous faces. Perception 26:733-752; 1997
• Lee JK, Perrett DI. Manipulation of colour and shape information and its consequence upon recognition and best-likeness judgment. Perception 29: 1291-1312; 2000
• Martin A and Chao LL. Semantic memory and the brain: structure and processes. Curr Opin Neurobiol 11: 194-201; 2001
• Nunn JA, Postma P, Pearson R. Developmental prosopognosia: should it be taken at face value? Neurocase 7:15-27; 2001
• Peck S and Peck L. Selected aspects of the art and science of facial esthetics. Sem Orthod 2: 105-126; 1995
• Riesenhuber M and Poggio T. Models of object recognition. Nat Neurosci 3 Sup: 1199-1204; 2000
• Rhodes G, Byatt G, Tremewan T, Kennedy A. Facial distinctiveness and the power of caricatures. Perception 26: 207-223; 1997
• Zeki S. The Ferrier Lecture 1995 behind the seen: the functional specialization of the brain in space and time. Phil. Trans. R. Soc. B 360: 1145-1183; 2005