Parametric Approaches to Syntax in the Brain: Spatial and Temporal Signatures
Linguistic utterances are not just mere strings of words but rather manifest intricate syntactic structures where the semantic interpretation of an utterance depends not just on the constituents (e.g. words) but how each constituent relates to other constituents (constituent structures) [1,2]. Neural research on syntax has focused mostly on the left inferior frontal gyrus (IFG; Brodmann's areas 44, 45, and 47) , the left anterior temporal lobe (ATL) , and the left posterior superior temporal sulcus (pSTS)  as regions important for syntactic processing. However, in light of the granularity mismatch and ontological incommensurability problems [6,7] inherent to neurolinguistic research the complex nature of the stimuli used in most studies raise an important question -- are these aforementioned regions activated by basic combinatorial mechanisms required for syntax (e.g. MERGE), or by other predictive mechanisms that underpin sentence comprehension [8-10]? Two related findings further complicate a straightforward localization effort to these brain regions. First, if said regions are truly core to structure building then one would expect them to consistently activate in response to complex syntactic structures and increased complexity to lead to increased activation. Yet, a number of studies failed to report such activation patterns [11-13]. Secondly, one would expect that patients with lesions in the IFG and/or pSTS would exhibit significant and systematic comprehension deficits. Similarly, it would be logical to expect that patients with Broca’s Aphasia, associated with damage to the left IFG, would suffer in acceptability judgment tasks which require syntactic combinatorics. However, studies have failed to find any systematic comprehension deficits associated with damage to the IFG and pSTS [14,15], and in fact found that mere damage to the left IFG is not sufficient to cause Broca’s Aphasia [16-18]. At a superficial level, then, it would seem that the two sets of studies -- those reporting activation of IFG, ATL and pSTS to syntactic structures [4-7], and those that did not [11-18]-- are at odds.
One suggestion for the reconciliation of such findings comes in the form of structural prediction hypotheses [8-10], i.e. increased IFG and pSTS activity reflects top-down prediction of structural nodes (e.g. NP, VP) and syntactic features (tense, number etc.). That is, given that there is good evidence that users make structural prediction in a variety of syntactic contexts -- filler-gaps , ellipsis  -- a logical hypothesis is that said brain regions support such top-down predictions but are not necessarily core structure-building circuits . Overall prediction effects have been studied in four seminal fMRI experiments. Santy and Grodzinsky  report that the IFG is activated by predictable filler-gaps, while increased activation of anterior IFG is observed for stimuli where prediction of filler-gaps or antecedent dependencies have to be maintained over longer distances [20,21]. Likewise, an fMRI experiment in German reported  that predictions of the syntactic category of the final word of jabberwocky sentences (i.e. real words replaced by pseudowords) resulted in increased activation in left IFG and pSTS compared to unstructured lists where syntactic prediction is not possible.
In light of the evidence reviewed above, a converging empirical design to probe the temporal and spatial neural signatures over the broader IFG, ATL and pSTS, indeed the so-called syntax network , in both typical (control) populations, and patients with Broca’s Aphasia (Condition-1) and damages/lesions to IFG (Condition-2) is likely to be informative. In particular, we propose a parametric approach to increasing syntactic complexity in our experimental stimuli. Precisely, we propose to use three levels of complexity -- unstructured word lists, simple two-word phrases, and short simple sentences ranging between five and eight words. Further, replicating the German study the short sentences will be divided into two broad groups -- typical (control) and nonce (sentences using pseudowords). Participants are required to recall specific target words. Regarding the choice of stimuli design two important factors are worth highlighting. First, the use of two-word phrases and short sentences is meant to allow for a selective probing of the structural prediction hypotheses. We predict that given the cost-ineffectiveness of predictive processes, participants are less likely to engage them in two-word phrases (even though, technically such phrases do manifest syntactic structures), especially since the following word arrives immediately after the first word. On the other hand, with sentences if candidates were to use the entire structure, as opposed to predictive mechanisms, then they would need to wait until the entire sentence has been heard. On the other hand the extensive literature on word-recall tasks [24-26] allow us to reliably predict that participants would make use of their implicit knowledge of syntactic dependencies to actively make predictions about target words. Our task will further ensure detection of prediction mechanisms being employed by shortening the time between the end of the sentence and the response deadline, thus rewarding active predictions. Second, the use of nonse sentences with pseudowords allow us to disentangle syntax from semantics. Finally, we propose to employ a combined fMRI + EEG recording in order to collect both spatial and temporal activation patterns during the task. While we predict increased activation of the IFG and pSTS for structural prediction acts in control participants, the neural responses of Condition-1 and Condition-2 subjects is likely to highlight both the role of concerned areas regarding processing and prediction of syntactic structures, as well as possible neuroplasticity related workaround in cases where the IFG, for example, is unable to perform its functions due to pathology.
 In spite of using nonce materials, however, it is likely that some aspects of sementics, such as abstract event structures, will remain in the nonse sentences. (cf. Chomsky & Keyser, 1982) [Chomsky, N., & Keyser, S. J. (1982). Some concepts and consequences of the theory of government and binding. MIT press.]
1. Chomsky, N. (1965). Aspects of the theory of syntax. Cambridge, MA:MIT Press.
2. Everaert, M. B., Huybregts, M. A., Chomsky, N., Berwick, R. C., &
Bolhuis, J. J. (2015). Structures, not Strings: Linguistics as part of the cognitive sciences. Trends in Cognitive Sciences, 19(12), 729e743.
3. Embick, D., Marantz, A., Miyashita, Y., O'Neil, W., & Sakai, K. L. (2000). A syntactic specialization for Broca's area. Proceedings of
the National Academy of Sciences USA, 97(11), 6150e6154.
4. Brennan, J., Nir, Y., Hasson, U., Malach, R., Heeger, D. J., & Pylkkanen, L. (2012). Syntactic structure building in the anterior temporal lobe during natural story listening. Brain and Language, 120(2)
5. Stowe, L. A., Broere, C. A., Paans, A. M., Wijers, A. A., Mulder, G., Vaalburg, W., et al. (1998). Localizing components of a complex task: Sentence processing and working memory. NeuroReport, 9(13), 2995e2999.
6. Poeppel, D., & Embick, D. (2005). The relation between linguistics and neuroscience. In A. Cutler (Ed.), Twenty-first century psycholinguistics: Four cornerstones (pp. 103-120). Lawrence Erlbaum Associates.
7. Embick, D., & Poeppel, D. (2015). Towards a computational (ist) neurobiology of language: correlational, integrated and explanatory neurolinguistics. Language, cognition and neuroscience, 30(4), 357-366.
8. Crain, S., & Fodor, J. D. (1985). How can grammars help parsers. Natural Language Parsing: Psychological, Computational, and Theoretical Perspectives, 94e128
9. Stowe, L. A. (1986). Parsing WH-constructions: Evidence for on-line gap location. Language and Cognitive Processes, 1(3), 227e245
10. Friederici, A. D. (2002). Towards a neural basis of auditory sentence processing. Trends in Cognitive Sciences, 6(2), 78e84
11. Bemis, D. K., & Pylkkanen, L. (2012). Basic linguistic composition recruits the left anterior temporal lobe and left angular gyrus during both listening and reading. Cerebral Cortex, 23, 1859e1873.
12. Mazoyer, B. M., Tzourio, N., Frak, V., Syrota, A., Murayama, N.,Levrier, O., et al. (1993). The cortical representation of speech. Journal of Cognitive Neuroscience, 5(4), 467e479
13. Rogalsky, C., & Hickok, G. (2009). Selective attention to semantic and syntactic features modulates sentence processing networks in anterior temporal cortex. Cerebral Cortex, 19(4), 786e796.
14. Dronkers, N. F., Wilkins, D. P., Van Valin, R. D., Redfern, B. B., & Jaeger, J. J. (2004). Lesion analysis of the brain areas involved in language comprehension. Cognition, 92(1), 145e177.
15. Magnusdottir, S., Fillmore, P., Den Ouden, D. B., Hjaltason, H., Rorden, C., Kjartansson, O., et al. (2013). Damage to left anterior temporal cortex predicts impairment of complex syntactic processing: A lesion-symptom mapping study.Human Brain Mapping, 34(10), 2715e2723
16. Mohr, J. P., Pessin, M. S., Finkelstein, S., Funkenstein, H. H., Duncan, G. W., & Davis, K. R. (1978). Broca aphasia Pathologic and clinical. Neurology, 28(4), 311e311.
17. Thothathiri, M., Kimberg, D. Y., & Schwartz, M. F. (2012). The neural basis of reversible sentence comprehension: Evidence from voxel-based lesion symptom mapping in aphasia. Journal of Cognitive Neuroscience, 24(1), 212e222.
18. Wilson, S. M., Dronkers, N. F., Ogar, J. M., Jang, J., Growdon, M. E., Agosta, F., et al. (2010). Neural correlates of syntactic processing in the nonfluent variant of primary progressive aphasia. The Journal of Neuroscience, 30(50), 16845e16854.
19. Santi, A., & Grodzinsky, Y. (2012). Broca's area and sentence comprehension: A relationship parasitic on dependency, displacement or predictability? Neuropsychologia, 50(5), 821e832.
20. Matchin, W., Sprouse, J., & Hickok, G. (2014). A structural distance effect for backward anaphora in Broca's area: An fMRI study. Brain and Language, 138, 1e11.
21. Santi, A., & Grodzinsky, Y. (2007). Working memory and syntax interact in Broca's area. NeuroImage, 37(1), 8e17.
22. Bonhage, C. E., Mueller, J. L., Friederici, A. D., & Fiebach, C. J. (2015). Combined eye tracking and fMRI reveals neural basis of linguistic predictions during sentence comprehension. Cortex, 68, 33e47
23. Friederici, A. D. (2016). The neuroanatomical pathway model of Language: Syntactic and semantic networks. In G. Hickok, & A. Small (Eds.), The Neurobiology of language. Elsevier
24. Brener, R. (1940). An experimental investigation of memory span. Journal of Experimental Psychology, 26(5), 467.
25. Miller, G. A., Heise, G. A., & Lichten, W. (1951). The intelligibility of speech as a function of the context of the test materials. Journal of Experimental Psychology, 41(5), 329.
Marks, L. E., & Miller, G. A. (1964). The role of semantic and syntactic constraints in the memorization of English sentences. Journal of Verbal Learning and Verbal Behavior, 3(1), 1e5.