ABSTRACT
Until recently, the processing of faces was believed to be carried out more efficiently in the right than the left hemisphere. This belief was based on evidence from numerous studies of facial recognition involving patients with unilateral cortical lesions (e.g., De Renzi, Faglioni, & Spinnler, 1968; Milner, 1968; Yin, 1970), commissurotomy patients (Levy, Trevarthen, & Sperry, 1972), and normal adults and children (e.g., Geffen, Bradshaw, & Wallace, 1971; Rizzolatti, Umiltà, & Berlucchi, 1971). However, more recent evidence indicates that processing facial information may require bilateral involvement. Thus, Hamsher, Levin, and Benton (1979) found that both right-posterior brain-damaged patients and left-hemisphere-damaged aphasic patients with an impairment in language comprehension showed defective performance on a facial recognition task. In addition, autopsy findings from patients with prosopagnosia, who show an inability to identify familiar faces, indicated the presence of bilateral lesions (Meadows, 1974). Furthermore, in studies of normal individuals it has been shown that, under certain conditions, the processing of faces may be superior in the left than the right hemisphere (Patterson & Bradshaw, 1975; Sergent, 1982a). Thus, the evidence from both normal subjects and brain-damaged patients makes it clear that facial recognition may predominantly involve either the right or left hemisphere, or both hemispheres equally, depending on task constraints.
