Note: Most Internet Explorer 8 users encounter issues playing the presentation videos. Please update your browser or use a different one if available.

Neural Correlates of Implicit Contextual Learning in Adults with ASD

Thursday, 2 May 2013: 09:00-13:00
Banquet Hall (Kursaal Centre)
10:00
P. S. Powell1, M. R. Klinger2, L. G. Klinger3 and R. K. Kana4, (1)Psychology, University of North Carolina - Chapel Hill, Chapel Hill, NC, (2)Allied Health Sciences, University of North Carolina - Chapel Hill, Chapel Hill, NC, (3)TEACCH Autism Program, Department of Psychiatry, University of North Carolina, Chapel Hill, NC, (4)University of Alabama at Birmingham, Birmingham, AL
Background:  

The ability to use environment cues (i.e., context learning) to make sense of the world is an important cognitive skill.  Previous studies examining implicit contextual learning in individuals with autism spectrum disorder (ASD) have reported mixed results (Barnes et al., 2008; Brown et al., 2010; Kourkoulou, et al., 2012; Powell et al., 2011).  We hypothesize that, regardless of whether behavioral performance is intact, individuals with ASD may be using a more explicit rather than implicit learning approach.  To date, there are no studies that have examined brain responses associated with a contextual cueing task in individuals with ASD. 

Objectives:  

The main goal of this fMRI study was to examine the neural circuitry underlying implicit contextual learning in ASD using a novel visual search task. We predicted that individuals with ASD would show less neural activation in areas typically associated with implicit learning.

Methods:  

Fifteen high-functioning adolescents and young adults with ASD and 17 age and IQ-matched individuals with typical development completed a visual search task.  Participants located a target Disney character (i.e., Jiminy Cricket) as quickly and accurately as possible while hidden amongst an array of 19 other distractor characters.  Contextual information (the arrangement of the other 19 characters) predicted the location of the target.  Participants had limited awareness that the arrangement of the characters predicted the location of the target.  The data were collected on a 3 Tesla fMRI scanner. Before the MRI session, participants were given eight blocks of predictable trials to learn the contextual relations between characters.  In the MRI scanner, the participants received a predictable block, an unpredictable block and a final predictable block of trials.

Results:  

Current neuroimaging results are based on only six participants per group as analyses are ongoing.  Individuals with ASD and individuals with TD demonstrated similar behavioral performance on the contextual cueing task.  However, differences in neural activation emerged between the two groups.  Individuals with ASD demonstrated lower levels of activation than individuals with typical development in the right precental gyrus, right insula, left cerebellum, and left superior temporal gyrus (p <.005, uncorrected) during predictable trials.  However, individuals with ASD demonstrated greater activation than individuals with typical development in the left caudate nucleus.  Finally, across both predictable and unpredictable trials, individuals with ASD showed reduced activation in bilateral inferior parietal lobule (IPL) compared in individuals with typical development. 

Conclusions:  

Preliminary findings suggest that individuals with ASD demonstrate unique neural activation during a contextual cueing task compared to individuals with typical development.  Greater activation of the left caudate during the predictable trials for individuals with ASD along with reduced activation in left and right IPL during both predictable and unpredictable trials suggests that individuals with ASD have different activation of both implicit learning and strategic attentional areas of the brain compared to individuals with typical development.

| More