Mapping attractor fields in face space: the atypicality bias in face recognition

Face Familiarity, Distinctiveness, and Categorical Perception

“…For instance, distinctiveness may induce a stronger memory trace of a newly learned face and increase its memorability (Bartlett, Hurry, & Thorley, 1984). Alternatively, and argued here, a distinctive endpoint may correspond to a larger attractor field (Tanaka et al, 1998) in the representational face space and produce the observed shift of the boundaries on the typical -distinctive continua towards the distinctive end-point. Not least, distinctive faces are by definition located in a sparser region of the face space (Valentine, 1991); hence it is quite possible that it is easier to discriminate faces lying close to a distinctive end-point because they are distinctive themselves.…”

“…Valentine's model successfully explains the easy recognition of distinctive faces and how CP might emerge from distortions in that face space. For example, representations could be perceived as "attractors" tending to gather exemplars close together (Tanaka, Giles, Kremen, & Simon, 1998). Although such a mechanism might plausibly affect novel as well as familiar faces, Campanella et al instead suggest that morphing reduces the distinctiveness for the central items of the continuum and hence achieves the categorical effects found in Tanaka et al (1998).…”

Recognition, Classification and Inversion of Faces in the Multidimensional Space

“…Bruce et al, 1994;Byatt & Rhodes, 1998;Tanaka et al, 1998;Valentine & Endo, 1992;Wickham et al, 2000) have been explained in reference to an MDS metaphorical model (Valentine, 1991a(Valentine, , 2001. One of the problems with this model is its ability to explain different phenomena using opposite arguments.…”

“…As Levin (1996) pointed out, by changing the estimated weight of different dimensions or the relative density of the theoretical faces, one can explain different phenomena. Thus, although many studies have referred to the MDS metaphoric model (e.g., Bruce et al, 1994;Burton & Vokey, 1998;Busey, 1998;Byatt & Rhodes, 1998;Johnston, Kanazawa, Kato, & Oda, 1997;Lewis & Johnston, 1997, 1999aTanaka et al, 1998;Valentine, Chiroro, & Dixon, 1995;Valentine & Endo, 1992;Wickham et al, 2000), it is necessary to test its predictions on an empirically defined MDS. Recently, an attempt was made (Catz, Kampf, Nachson, & Babkoff, 2009) to construct an operational MDS that included 200 faces which were each rated on 21 dimensions.…”

“…The faces were ranked for distinctiveness in a previous study (Catz et al, 2009) by assessing the relative ease of spotting them in a crowd Tanaka et al, 1998;Valentine & Bruce, 1986a,c;Valentine et al, 1995;Valentine & Endo, 1992). On the basis of these rankings, faces above the median were considered typical, and those below the median were considered distinctive.…”