Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Neural markers of developmental processing of grammaticality from childhood to adolescence
Poster A29 in Poster Session A, Tuesday, October 24, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
Juliana Ronderos1, Zhenghan Qi2, Kenneth Wexler3, Helen Tager-Flusberg1, John Gabrieli3, Tyler Perrachione1; 1Boston University, 2Northeastern University, 3Massachusetts Institute of Technology
Errors in the acquisition of grammatical morphology characterize developmental stages of children’s acquisition of overt tense and aspect marking. In English, a common developmental error is the use of infinitive verb forms where finite forms are required. For example, omissions of the past tense “-ed,” the 3rd person singular “-s,” and copula “to be” are typical errors in the course of English language acquisition in early childhood. For example, in the sentence, “Last year, Bob play football,” the nonfinite form “play” has been used instead of the correct finite form “played” to indicate the past tense marking. Other kinds of overt tense and aspect marking errors, such as the subject and auxiliary form mismatch seen in the sentence, “We am seeing a movie,” are not attested in development. Little is known about the development of the neural bases of grammatical error processing, including whether there is differential sensitivity to the kinds of grammatical errors that children do vs. do not make during development. The present study examined developmental trajectories in the neural processing of grammatical vs. ungrammatical sentences in a cross-sectional sample of N=87 neurotypical children from 5-18 years old. A grammaticality judgment task, modeled after the stimuli in the Test of Early Grammatical Impairment (TEGI; Rice & Wexler, 2001), was used during functional magnetic resonance imaging (fMRI) to compare brain activation for sentences that were grammatically correct vs. those with either developmental or non-developmental tense/aspect agreement errors. In an event-related fMRI task, children heard 108 auditory sentences while making a grammaticality judgment on each trial via button press. We examined fMRI response magnitude to the three types of sentences in select frontal and temporal regions of interest based on a probabilistic atlas of the adult language network (Lipkin et al., 2022). We found that activation in left inferior frontal gyrus (IFG) pars opercularis and pars triangularis decreased with age in response to hearing grammatically correct sentences, but increased in response to ungrammatical sentences (both developmental and non-developmental errors). However, no significant differences in activation were associated with age when contrasting non-developmental versus developmental grammatical errors to each other. In addition, we used multi-voxel pattern analysis (MVPA) to compare how the neural response profiles to developmental vs. non-developmental errors in each parcel changed as a function of age. In left anterior temporal lobe (including superior and middle temporal gyrus), right posterior temporal lobe, and bilateral supplementary motor area, the neural responses to developmental and nondevelopmental errors became less similar as children got older. These results provide new evidence into the neural maturation of grammaticality processing across development. As children get older, left IFG shows greater dissociation of response magnitude to grammatical vs. ungrammatical sentences. Furthermore, the neural patterns that reflect processing of developmental vs. nondevelopmental grammatical errors become more distinct across development, suggesting more mature encoding of nuanced grammatical information.
Topic Areas: Morphology, Language Development/Acquisition