25541
An fMRI Investigation of Working Memory in Older Adults with Autism Spectrum Disorder: Fronto-Hippo-Striatal-Thalamic Network Differences

Thursday, May 11, 2017: 5:30 PM-7:00 PM
Golden Gate Ballroom (Marriott Marquis Hotel)
B. B. Braden1, C. J. Smith2, T. K. Glaspy3, E. Wood4, D. Vatsa5 and L. C. Baxter6, (1)Speech and Hearing Science, Arizona State University, Tempe, AZ, (2)Southwest Autism Research and Resource Center, Phoenix, AZ, (3)Tufts University, Boston, MA, (4)Xavier Preparatory Academy, Phoenix, AZ, (5)BASIS Charter School, Scottsdale, AZ, (6)Radiology, Barrow Neurological Institute, Phoenix, AZ
Background: The effects of aging in adults with autism spectrum disorder (ASD) are understudied, but of increasing importance in order to anticipate unique needs of this growing group of individuals. Executive functioning is particularly vulnerable in both high-functioning ASD and in normal aging. Deficits are thought to arise from structural and functional connectivity disturbances between the frontal lobe and posterior brain regions. Recently, a cross-sectional study revealed sharper age-related declines of structural connectivity in adults with ASD, relative to typical adults (Koolschijn et al., 2016). However, differences in functional connectivity disturbances are still unknown; yet, emerging evidence suggests functional connectivity is the best neurobiological predictor of age-related cognitive decline.

Objectives:  The current study investigated functional brain network recruitment during an fMRI executive function task in middle-aged men with ASD, compared to age- and IQ-matched typically developing (TD) men.

Methods:  We evaluated 16 ASD and 17 matched TD men from ages 40 to 64 and of average to high intellectual functioning (IQ: 83-131), For participants with ASD, diagnosis was confirmed via the Autism Diagnostic Observation Schedule-2 and developmental history assessment. TD participants were screened for presence of ASD symptoms via the Social Responsiveness Scale-2. All participants were given the Kaufman Brief Intelligence Test-2. Participants performed the n-back fMRI task, a working memory task that is good indicator of executive function. Participants monitored a series of letters and identified targets that range from simply matching a target (“0-back”) to identifying matches that are two letters apart (“2-back”). Functionally connected network activity was assessed via group independent component analysis. Working memory load comparisons were made by the 2 vs. 0 contrasts. Reaction time and accuracy were recorded. Participants performed the task twice.

Results:  Both groups performed well on all conditions of the n-back task (over 85% accuracy). The ASD participants’ reaction time was slower on the 2-back condition, but were similar for accuracy in all conditions. For the working memory load comparison (2 back vs. 0 back), both groups showed similar activation a classic cortical working memory network including the bilateral dorsolateral prefrontal cortex (dlPFC), parietal cortex, insula, and the anterior cingulate cortex, and deactivation the default mode network (DMN). However, only the TD group activated an additional network including the left inferior frontal lobe, and bilateral hippocampi, striatum, and thalamus (Fig. 1a). This working memory activation was significantly greater in TD older adults than older adults with ASD (Fig. 1b).

Conclusions:  Results showed reduced engagement a fronto-hippo-striatal-thalamic neural network during working memory performance in older adults with ASD, compared to matched TD adults. Dysfunction within this network may underlie working memory and other executive function struggles in high functioning older adults with ASD, and could impact independence as aging ensues. Longitudinal evaluation of differences in age-related cognitive and brain trajectories between older adults with ASD and TD adults are in progress.