Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Poster Slams

Causal Representation of Abstract Phonological Properties in Temporal Cortex

Poster D22 in Poster Session D with Social Hour, Friday, October 7, 5:30 - 7:15 pm EDT, Millennium Hall

David Gow1,2,3,4, Enes Acvu1, Adriana Schoenhaut1, David Sorensen4, Skyla Lynch1, Seppo Ahlfors1,2; 1Massachusetts General Hospital, 2Athinoula A. Martinos Center for Biomedical Imaging, 3Salem State University, 4Harvard-MIT Division of Health Sciences and Technology

The generativity shown in language users ability to evaluate or generate novel, well-formed linguistic structures depends on abstract representations (variables). Linguistic variables enable generalization by allowing a single computation or structural constraint to apply to a potentially open set of specific instances independent of first order similarity to other members of the set. Variable-dependent models describing the patterning of classes of speech sounds in words or syllables, or the classes of words defined by grammatical and semantic roles are used to capture structural constraints on the acquisition, perception, and production of language, as well as its breakdown after pathology. Given the variable’s foundational role in the theory, it is striking that the neural basis of variable representation has been largely unexplored. Here, we present evidence for a neural basis for the representation of linguistic variables. Our test case was syllable repetition or reduplication, a morphophonological process in which words or parts of words are duplicated to change their meaning or grammatical properties. We aimed to identify localized patterns of brain activity associated with reduplication and determine whether they function as variable representations. We collected simultaneous magnetoencephalography (MEG) and electroencephalography (EEG) data in 12 subjects (native speakers of Standard American English) while they completed an artificial grammar learning task used to explore abstract rule learning in infants (Marcus et al., 1999). In each block, subjects first heard a series of trisyllabic CVCVCV nonsense words (exposure stimuli) with a common syllable reduplication pattern (e.g., AAB as in ba-ba-di or di-di-ba) and then were asked to indicate by button press whether subsequent nonsense words composed of different syllables (e.g., fu-fu-ni) “came from the same imaginary language”. We compared the neural responses in the three-syllable reduplication conditions (AAB, ABA, and ABB) using three complementary approaches: activation contrast, neural decoding, and effective connectivity analyses. Specifically, to determine whether there is an algebraic variable representation of reduplication, we examined whether previously unheard nonsense words evoked localized patterns of activation that support machine learning categorization (i.e., neural decoding) of syllable reduplication and explored potential causal downstream consequences of those localized activations using effective connectivity measures. Neural results showed that the within-region activity of a small set of temporal lobe regions known to be associated with the representation of phonetic and phonological structure supported the decoding of syllable reduplication. Control analyses demonstrated that decoding effects were separable from low-level repetition enhancement or task-specific processing demands. Effective connectivity analyses demonstrated that decodable signals from these early posterior middle temporal gyrus regions causally influenced downstream processes to influence sensitivity to reduplication patterns in more anterior middle temporal regions. Collectively, the results suggest that the localized activation patterns function as neural representations of a property, i.e., syllable reduplication, which appears to require variable representation. While these results do not resolve questions about the nature of the processes that rely on variable-based representations of stimulus properties to produce generative thought, future progress will benefit from developing independent, empirically derived characterizations of the representations upon which any mechanism must depend.

Topic Areas: Phonology and Phonological Working Memory, Methods

SNL Account Login

Forgot Password?
Create an Account