skip to primary navigationskip to content

Brown bag lunch and poster session

When May 08, 2013
from 01:15 PM to 02:15 PM
Where Room 7, Mill Lane Lecture Rooms
Contact Name
Add event to calendar vCal

This was the last brown bag lunch of the 2012/13 academic year. A new season, at different venues around the University, will start next term. Details to be announced!



Morphological processing in a minimally inflected language

Huang, Y-H.1,2, Fonteneau, E. 1,2, Whiting, C. M. 1,2, Cai, Q. 3,4, Marslen-Wilson, W. 1,2

1 Department of Psychology, University of Cambridge

2 MRC Cognition and Brain Sciences Unit, Cambridge, U.K.

3 East China Normal University, China

4 INSERM U992, Cognitive Neuroimaging Unit, France

Previous research in English and Polish provides evidence that two distinct but partially overlapping networks underlie spoken word processing.  Morphological inflections, an assumed hallmark of human language, mainly engage a left-lateralized frontal-temporal network linking inferior frontal and posterior temporal regions; non-inflected simple words and derived words are supported by a more widely distributed bilateral fronto-temporal network (Bozic et al. 2010; Szlachta et al, 2012). The aim of this study is to examine the possible role of these two networks in a minimally inflected language, Mandarin Chinese. Native speakers of Mandarin passively listened to three types of disyllabic Mandarin words, where the second syllable contrasted different types of morphological complexity (simple compound words (hu shi ‘to breath’), derived words (jing hua ‘purify’) , and inflected words (chang zhe ‘singing’), while combined electro- and magneto-encephalography (EMEG) data were recorded. This allows us to track the transient real-time morphological computations underlying the three types of complexity. The alignment point was set to the onset of the second syllable, where information about word type and complexity becomes available. Minimum Norm Estimate (MNE) techniques were used to compute whole-brain source estimates. Early results suggest that derived words elicit more activation than inflected words in the bilateral fronto-temporal network at early time-windows, while inflected words preferentially activate left hemisphere frontal and temporal ROIs at a later time-window. This is preliminary evidence that, despite the reduced role of grammatical morphological combination in Mandarin, a similar functional differentiation can be observed in the underlying neurobiological systems.

Click here for the pdf.


Grammatical computations in the fronto-temporal language network

M Bozic, E Fonteneau, LK Tyler, B Devereux, P Buttery, W Marslen-Wilson

Converging evidence shows that language comprehension engages two distinct processing systems: a distributed bihemispheric system, involved in semantic and pragmatic interpretation of auditory inputs, and a left-lateralised fronto-temporal system, activated by core grammatical computations (Bozic et al, 2010; Marslen-Wilson & Tyler, 2007). Recent neuropsychological data suggest, however, that the bihemispheric system may also have a significant role in supporting aspects of sentence processing (Tyler et al, 2010), raising the question of whether grammatical computations are distributed over these two systems. We used event-related fMRI to test their grammatical processing capacities, by comparing the processing signatures of inflectional grammatical complexity and minimal phrasal complexity. We contrasted these effects in verbs, which have rich morpho-syntactic paradigms, with parallel ones in nouns. Participants heard unique or dominant verbs and nouns, presented as bare stems (e.g. sing, rug), inflected forms (e.g. sings, rugs) or short phrases (e.g. I sing, a rug). The results show that inflectional complexity and minimal phrasal complexity engage overlapping, yet distinct processing mechanisms. Inflectional complexity engages left-lateralised perisylvian regions, while the processing of minimal phrases engages bilateral temporal regions. Verb dominance modulated the processing of inflected forms and phrases, but not the processing of bare stems. The results represent a significant shift away from current views on the distribution of grammatical processing capacities within the language processing network. 

Click here for the pdf.

Music, language, and syntactic integration

Barry Ross, Centre for Music and Science

[abstract to follow]