Resting-State Functional Thalamo-Cortical Overconnectivity Is Associated with Block Design Peak in Autism

Background: The thalamus relays sensory information to the cortex, but is also implicated in cognition trough cortico-thalamo-cortical connections (Sherman, 2012). Thus, this subcortical structure’s intrinsic activity might have a role in autistics’ atypical perceptual processing, which is related to their superior performance in visual tasks (Falter et al., 2013), like the Block Design (BD) task (Muth et al., 2014). There are a few reports of thalamic differences between autistic and non-autistic individuals with regards to its structure, activity and connectivity, but we do not have yet a clear and specific pattern of thalamic differences and their associations with cognitive performances.

Objectives: To investigate intrinsic functional connectivity involving the thalamus during rest in autistic versus non-autistic individuals and to verify the presence of associations with cognitive performance.

Methods: Thirty-five autistic (31 males) and 43 non-autistic (42 males) individuals underwent a resting state fRMI scan (Siemens Tim Trio 3T). Groups were matched on age (14-38 years; Autistics M=21.3, SD=6.1; Non-autistics M=22.6, SD=5.3), Wechsler Performance Intelligence Quotient (IQ) (Autistics M=106.0, SD=12.4; Non-autistics M=104.8, SD=13.4), Raven’s Progressive Matrices (RPM) percentiles (Autistics M=72.2, SD=23.7; Non-autistics M=72.4, SD=22.3) and manual preference. Seed-to voxel analyses were performed between the thalamic nuclei and the rest of the brain in both groups. Connectivity values of regions showing significant differences between autistics and non-autistics were then correlated with the participants’ Wechsler Full-scale, Verbal and Performance IQ, their RPM percentile and their BD scaled score and peak (the difference between the BD scaled score and the mean of all other Wechsler subtests).

Results: A pattern of greater thalamo-cortical connectivity was observed in autistic compared to non-autistic participants. Specifically, the autistic group presented stronger connections between the thalamus and the fusiform gyrus, the superior, middle and inferior temporal gyri, the temporal pole, the superior occipital gyrus and the inferior frontal regions than the non-autistic group. Interestingly, these over-connections were observed for a specific thalamic region, the medio-dorsal nucleus. The analysis of the correlations between the connectivity values of each of the cortical regions showing stronger connections with the medio-dorsal nucleus in the autistic relative to non-autistic participants with cognitive measures revealed significant positive correlation with BD peak values.

Conclusions: A pattern of stronger intrinsic connectivity between thalamus and posterior regions was observed in autistic participants relative to non-autistic participants. This stronger connectivity mainly involved a specific region from the thalamus, the medio-dorsal nucleus. The medio-dorsal thalamic nucleus typically has reciprocal connections with the prefrontal cortex and has a role in executive functions and learning (Ferguson & Gao, 2014). Its resources could be partially reallocated from the prefrontal cortex to the occipito-temporal regions in autistic individuals (as they exhibited stronger thalamo-occipitotemporal connectivity). As the strength of these connections correlates with visuospatial performance, this thalamo-occipitotemporal enhanced connectivity seen in autistic individuals may contribute to their visuospatial and cognitive strengths.