Neural substrates of sign language processing differ partially between Swedish and British signers
2013 (English)Conference paper, Abstract (Other academic)
The lexicons of British Sign Language (BSL) and Swedish Sign Language (SSL) only partially overlap. This offers the opportunity of studying the neurocognitive systems that support the processing of familiar and non-familiar lexical signs as well as non-signs under controlled conditions. In particular, effects relating either to a particular sign language or to the users of a particular sign language can be identified. Fifteen deaf native users of BSL and fifteen deaf native users of SSL took part in an fMRI study. All were right-handed and neurologically healthy. The two groups were matched on age, gender, level of education and non-verbal intelligence. In the scanner participants were presented with video recorded sign-based material that consisted of 1) BSL: signs lexicalized in BSL but not SSL, 2) SSL: signs lexicalized in SSL but not BSL, 3) Cognates: signs lexicalized in both BSL and SSL, 4) Non-signs made up of handshapes, movements, locations and orientations put together in a phonotactically illegal manner. The recordings were modeled by a deaf native signer of a third sign language (German Sign Language) to avoid bias to BSL or SSL. All stimuli were matched for complexity. Stimulus types 1, 2 and 3 were matched for familiarity and age of acquisition. Stimulus types 1 and 2 were matched for iconicity. Participants performed two tasks in the scanner. Both tasks involved monitoring stimuli for specific phonological components according to a cue presented before each block of 12 stimuli. One task involved monitoring for handshape and the other task involved monitoring for location. Whole brain analysis of the fMRI data revealed more activation for non-signs than unfamiliar signs (SSL for British signers and BSL for Swedish signers) in superior posterior parietal regions bilaterally across tasks. No differences in activation could be attributed to semantic content (familiar versus unfamiliar signs) or iconicity (cognates versus familiar signs). Interestingly, there was more activation in British compared to Swedish signers across conditions in a region anterior to the left cytoarchitectonic definition of area V5/MT+ (Malikovic et al., 2007) and right middle occipital gyrus. There were no interactions between group and either task or type of material. This pattern of results shows that the neural substrates of sign language processing partially differ between Swedish and British signers. This difference cannot be attributed to differences in processing the phonological parameters of handshape and location or to differences in processing semantics and iconicity. One possible explanation is a difference in reliance on non-manual components of sign language in British and Swedish signers.
Place, publisher, year, edition, pages
IdentifiersURN: urn:nbn:se:liu:diva-91772OAI: oai:DiVA.org:liu-91772DiVA: diva2:619040
Theoretical Issues in Sign Language Research (TISLR) Conference 11. University College London, 10th - 13th July 2013