Presentation

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

Morphology of the posterior superior temporal plane and language abilities in autism

Poster E58 in Poster Session E, Saturday, October 8, 3:15 - 5:00 pm EDT, Millennium Hall

Kendrick Tak¹, Juliet Henderson¹, Gabrielle-Ann Torre¹, Alexandra Kapadia¹, Zhenghan Qi², John Gabrieli³, Helen Tager-Flusberg¹, Tyler Perrachione¹; ¹Boston University, ²Northeastern University, ³Massachusetts Institute of Technology

The posterior superior temporal plane exhibits remarkable structural variability across individuals. Morphological variability, particularly of Heschl’s gyrus (HG) and planum temporale (PT), have been implicated in language and communication impairments. Such impairments often occur in autism spectrum disorder (ASD), as do structural differences in the superior temporal plane. However, the aspects of morphological variability in autism that are related to language skills in particular remain poorly understood. We investigated differences in superior temporal morphology between neurotypical and autistic children and their relationship to language skills in autism. Methods: We obtained T1-weighted whole-brain structural MRI scans from neurotypical (NT: N=69; age 5-18, mean=10.2 years; 34 female, 35 male) and autistic children (ASD: N=39, age 5-18, mean=11.3 years, 7 female, 32 male), which underwent the standard cortical reconstruction and parcellation pipeline in FreeSurfer. Cortical segmentations were automatically and manually inspected, then corrected for white- and pial-surface boundary errors. Patterns of HG duplication were identified based on virtual dissections revealing the 3D-reconstructed gray matter surface of the superior temporal plane and confirmed in the volume. Two independent labelers manually demarcated HG, HG2 (when present), and PT on the cortical surface following standard gyral and sulcal landmarks. Surface labels were transformed into the volume and masked by the cortical ribbon; boundary disagreements were resolved by consensus. Using these labels, we measured the gray matter volume, cortical surface area, and cortical thickness of each structure. We analyzed these for differences between groups and hemispheres, controlling for age, sex, nonverbal IQ (KBIT), and total cerebral volume. In the ASD group, we investigated relationships between these morphological measurements and communication skills based on CELF core language, digit span, SCQ communication, and ADOS calibrated severity scores. Results: The probability of HG duplication did not differ between groups, whether in both hemispheres or each separately (all p>0.46). There were no relationships between HG duplication and language ability or symptom severity in ASD. Both HG and PT were larger in ASD than NT, including gray matter volume (HG: 2.22cc/1.94cc, p<0.006; PT: 1.56cc/1.34cc, p<0.025) and cortical surface area (HG: 743mm²/646mm², p<0.0013; PT: 598mm²/499mm², p<0.008), but not cortical thickness (HG: 2.82mm/2.83mm, p=0.89; PT: 2.74mm/2.76mm, p<0.60). However, PT cortical thickness was significantly related to core language scores in ASD (p<0.002) (but not NT; p=0.69). PT cortical thickness in ASD was also related to digit span (p<0.006), SCQ communication (p<0.012), and ADOS calibrated severity (p<0.008). These relationships were not significant in HG. Discussion: HG duplication was unrelated to either ASD diagnosis or language ability. The sizes of both HG and PT were larger in ASD than NT (controlling for age, sex, IQ, and total cerebral volume), but the sizes of these structures were not related to language abilities in ASD. Instead, PT cortical thickness was related to several measures of language and communication abilities in ASD. These results suggest that, while exaggeration of primary and association auditory areas may occur in ASD, language abilities in this group are specifically related to differences in PT morphology that may reflect microstructural tissue differences in this area.

Topic Areas: Disorders: Developmental, Development