2018 Friday Session B 1000

Friday, November 2, 2018 | Session B, Conference Auditorium | 10am

Syntactic Development and Neurolinguistic Processing Go Hand in Hand: Evidence from Early vs Late L1 Acquisition of ASL
R. Mayberry, M. Hatrak, D. Ilkbasaran, W. Matchin, A. Villwock, A. Roth, E. Halgren

Much theorizing about language acquisition posits that a successful outcome depends upon the age when it occurs for reasons related to brain maturation. Testing this hypothesis requires studying healthy individuals who first experienced language during infancy or after childhood. The interaction of infant deafness and social factors means that such individuals can be found within the ASL population. Previous research has shown that a delayed onset of L1 acquisition affects the outcome of syntactic acquisition.1 Other research has found that delayed L1 acquisition affects the neurolinguistic processing of ASL in ways uncharacteristic of late L2 learning.2 In two studies we asked whether delayed L1 acquisition affects the acquisition of syntax and neurolinguistic processing in a systematic and related fashion.

We first measured the acquisition of ASL syntax using a sentence-to-picture matching task. The 56 stimuli represented three levels of syntactic complexity: monoclausal and bi-clausal structures, and inter-sentential structures. Deaf native signers (n=10, AoA=birth) performed accurately across the complexity levels. The hearing L2 signers (n=8; AoA>9yrs) made errors on the inter-sentential structures, but their performance was not significantly different from that of the native signers. As predicted, the deaf late L1 signers (n=5; AoA >9yrs) performed well on the mono-clausal structures but showed limited comprehension of the bi-clausal and inter-sentential structures (p<.001). Thus post-childhood L1 acquisition affects the level of syntactic complexity that can be acquired. Next we asked whether this pattern of limited syntactic acquisition corresponds with neurolinguistic processing.

We used fMRI to neuroimage syntactic processing in the same 25 signers in two conditions. 3 Sentences were 60, 4-sign, ASL mono-clausal SVO stimuli with an adjective added to the subject or object noun phrase (which the late L1 learners could understand). Half the sentences were implausible (e.g., The table ate the black dog) with a plausibility judgment scanner task. Words were 60, 4-sign, unrelated lists of nouns, verbs, or adjectives from the sentences with the task being identification of an animal name. The data were collected using a 3T GE MR750 MRI scanner using echo-planar imaging. We examined activation for the sentences and word lists using a voxelwise p-value of <.05, and a 20 voxel cluster threshold for each subject within two main language regions, STS, superior temporal sulcus and MTG, middle temporal gyrus, and IFG, inferior frontal gyrus. These areas are activated by syntax in spoken (French, English) and signed (ASL, BSL, JSL) languages.4 The native signers showed significant activation in IFG (Broca’s area) for sentences, as did the L2 signers. Both groups also showed significant activation for sentences in posterior temporal lobe (STS/MTG). By contrast, the late L1 learners did not show significant activation in either major language area with no clear activation differences for sentences vs words, despite performing the scanner tasks well.

These results show that language experience during post-natal brain growth is essential for the brain language system to develop functionality. In turn, this may serve to facilitate or inhibit the acquisition of syntactic structure, and form the basis of the critical period for language.