Specifying the brain bases of syntax: Distinct fMRI effects for syntactic complexity and syntactic violations

Christian J. Fiebach (1), Matthias Schlesewsky (2), Ina D. Bornkessel (1) and Angela D. Friederici (1)
(1) Max Planck Institute of Cognitive Neuroscience, Leipzig, Germany
(2) Department of Linguistics, University of Potsdam, Germany

Several previous functional neuroimaging studies have investigated the neural bases of syntactic processing by examining either the effects of structural complexity of stimulus sentences on neural activity (e.g., Just et al., 1996; Stromswold et al., 1996) or, alternatively, by mapping brain activation changes due to the processing of sentences with syntactic violations (e.g., Meyer et al., 2000; Ni et al., 2000; Newman et al., 2001). The results, however, are very heterogeneous and thus it is difficult to derive a unified model of syntax and the brain.

In the present study, we investigated the neural correlates of syntactic complexity and syntactic violations within one experiment. Syntactic complexity was increased by scrambling one or two object noun phrases in German declarative sentences before the subject NP. Thus, syntactic complexity was varied parametrically in three steps (i.e., S-IO-DO; IO-S-DO; IO-DO-S; cf. also Rösler et al., 1998). A syntactic violation condition was constructed by moving the verb from its clause-final position to a position before the object NPs, a transformation that is illegal in German (e.g., Lenerz, 1977).

Event-related fMRI (20 axial slices measured at 3 Tesla with a TR of 2 sec.) indicates that increasing syntactic complexity leads to an activation increase in the inferior-posterior portion of Brodmann?s area 44, i.e., in the classical area of Broca. In addition, a sensitivity to syntactic complexity was also observed in the right hemisphere homologue of BA 44, as well as in the pre-supplementary motor area and in the cerebellum.

Syntactic violations also activated Broca?s area, albeit to a lesser degree than complex grammatical sentences. Specific activity for syntactic violations, which was not found for syntactic complexity, was identified about 2 cm posterior to Broca?s area in the ventral premotor cortex. Therefore, it can be concluded that syntactic complexity and syntactic violations draw upon distinguishable aspects of syntactic processing, represented separately in the brain. These results will be discussed in the context of a neurocognitive model of sentence processing (Friederici, 2002). Whereas phrase structure building processes might be localized in posterior inferior frontal areas (i.e., in ventral premotor cortex/BA 6) more complex aspects of hierarchical structuring of the input might rely on areas located immediately anterior to classical premotor cortex, namely in inferior BA 44.

References

Friederici, A. F. (2002). Towards a neural basis of auditory sentence processing. Trends in Cognitive Science, 6, 78-84.

Just, M. A., et al. (1996). Brain activation modulated by sentence comprehension. Science, 274, 114-116.

Meyer, M., Friederici, A. D., & von Cramon, D. Y. (2000). Neurocognition of auditory sentence processing: Event-related fMRI reveals sensitivity to syntactic violations and task demands. Cognitive Brain Research, 9, 19-33.

Newman, A. J., et al. (2001). An event-related fMRI study of syntactic and semantic violations. Journal of Psycholinguistic Research, 30, 339-364.

Ni, W. et al., (2000). An event-related neuroimaging study distinguishing form and content in sentence processing. Journal of Cognitive Neuroscience, 12, 120-133.

Rösler, F., et al. (1998). Parsing of sentences in a language with varying word order: Word-by-word variations of processing demands are revealed by event-related brain potentials. Journal of Memory and Language, 38, 150-176.

Stromswold, K., Caplan, D., Alpert, N., & Rauch, C. (1996). Localization of syntactic comprehension by Positron Emission Tomography. Brain and Language, 52, 452-473.