The signer and the sign: Cortical correlates of person identity and language processing from point-light displays

Highlights ► First cortical imaging study of point-light signed language (SL). ► Distinguishes carrier and content in SL. ► Describes cortical circuitry for both carrier and content in SL.
In this study, the first to explore the cortical correlates of signed language (SL) processing under point-light display conditions, the observer identified either a signer or a lexical sign from a display in which different signers were seen producing a number of different individual signs. Many of the regions activated by point-light under these conditions replicated those previously reported for full-image displays, including regions within the inferior temporal cortex that are specialised for face and body-part identification, although such body parts were invisible in the display. Right frontal regions were also recruited – a pattern not usually seen in full-image SL processing. This activation may reflect the recruitment of information about person identity from the reduced display. A direct comparison of identify-signer and identify-sign conditions showed these tasks relied to a different extent on the posterior inferior regions. Signer identification elicited greater activation than sign identification in (bilateral) inferior temporal gyri (BA 37/19), fusiform gyri (BA 37), middle and posterior portions of the middle temporal gyri (BAs 37 and 19), and superior temporal gyri (BA 22 and 42). Right inferior frontal cortex was a further focus of differential activation (signer > sign). These findings suggest that the neural systems supporting point-light displays for the processing of SL rely on a cortical network including areas of the inferior temporal cortex specialized for face and body identification. While this might be predicted from other studies of whole body point-light actions (Vaina, Solomon, Chowdhury, Sinha, & Belliveau, 2001) it is not predicted from the perspective of spoken language processing, where voice characteristics and speech content recruit distinct cortical regions (Stevens, 2004) in addition to a common network. In this respect, our findings contrast with studies of voice/speech recognition (Von Kriegstein, Kleinschmidt, Sterzer, & Giraud, 2005). Inferior temporal regions associated with the visual recognition of a person appear to be required during SL processing, for both carrier and content information.