Functional Connectivity MRI Lateralization in Autism Spectrum Disorder

Background: A left- and right-lateralized functional network is characteristic of typical development (Liu et al. 2009; Nielsen et al.2012). Classical language regions and regions of the default mode network make up the left-lateralized functional network, whereas regions involved in attention make up the right-lateralized functional network. Abnormal lateralization of functional activity in autism has been reported, however, it is unclear whether abnormal functional lateralization is restricted to classical language regions or if it diffusely affects one or both lateralized functional networks.

Objectives: To determine whether abnormal functional lateralization in autism occurs focally in specific regions of interest or diffusely across lateralized functional networks.

Methods: The preprocessing of 964 (447 autism scans and 517 typically developing control scans) resting-state functional MRI scans from the Autism Brain Imaging Data Exchange (http://fcon_1000.projects.nitrc.org/indi/abide/index.html) followed standard image processing procedures. Using an independent analysis done on 1,019 typically developing controls, 9 regions involved in the left-lateralized functional network and 10 regions involved in the right-lateralized functional network were identified as the regions involved in the most lateralized connections (Nielsen et al.2012). A functional lateralization index (fLI) was calculated for each of the 81 pairwise connections (36 pairwise connections between the 9 left-lateralized regions and 45 pairwise connections between the 10 right-lateralized regions) for each subject by subtracting the connection strength (determined by correlating the BOLD time series between two regions and taking hyperbolic arctangent of the correlation coefficient) in the right hemisphere from the connection strength in the left hemisphere. Group comparisons were made on the fLIs for the 81 connections using two-sample t-tests with a Bonferroni correction. The correlation between fLI and the following behavioral measures was also investigated: autism severity (as measured by the ADOS social + communication scores), handedness, and age. 

Results: Two connections in the left-lateralized network—Wernicke’s area to the posterior cingulate cortex (PCC) and Wernicke’s area to the temporoparietal junction (TPJ)—demonstrate a lack of left lateralization in the autism group compared to the typically developing group (Wern-PCC: t = 3.36, p = 0.0008; Wern-TPJ: t = 3.30, p = 0.001). The connection between Wernicke’s area and the PCC negatively correlates with autism severity (r = -0.13, p= 0.02).

Conclusions: Abnormal lateralization of functional connectivity during rest in autism is restricted to specific left-lateralized connections rather than diffusely affecting either the left- or right-lateralized functional networks. Also, as left lateralization increases between Wernicke’s area and the posterior cingulate cortex, autism severity decreases.

References: Liu, H, et al. Evidence from intrinsic activity that asymmetry of the human brain is controlled by multiple factors. (2009) PNAS, vol. 106, no. 48, pp. 20499–20503.

Nielsen, JA, et al. Left and Right Dominant Functional Connectivity Networks. (2012, June). Poster presented at the Organization for Human Brain Mapping’s 18th Annual Meeting, Beijing, China.