19479
Self-Related Processing and Its Reflections in Memory: An fMRI Study of Youth with and without ASD
Previous studies of memory in Autism Spectrum Disorder (ASD) have pointed to specific difficulties in autobiographical memory in ASD (Lind, 2010). Furthermore, several fMRI studies have shown abnormalities in brain activation during self-processing in persons with ASD (Lombardo et al., 2009; Kennedy and Courschesne; 2008; Pfeifer et al., 2013). Yet little is known about the neurobiological correlates of autobiographical memory in ASD. To our knowledge, this is the first study to examine the functional neurobiological correlates of self-related memory, in both ASD and Typically Developing (TD) youth.
Objectives:
To determine the brain activation associated with self-related words, and their subsequent memory, in ASD and TD youth.
Methods:
Participants: Twelve youths with ASD and 15 TD youths, matched for age (range: 8-18; mean=13.3±2.9), gender (70% male) and IQ (FSIQ range: 81-141; mean=115±15) participated in the study. ASD diagnosis was confirmed using ADOS and ADI-R.
Experimental Task: While in the scanner, subjects were presented with 60 trait-words, which participants evaluated as self-descriptive or not (Self Condition); and the same words with jumbled letters, which participants evaluated for presence of the letter ‘e’. Thirty minutes later, outside the scanner, subjects were presented with a recognition task including the 60 trait-words they had seen and 60 distracters.
Data analysis: BOLD activation was compared using a 2 x 2 random effects ANCOVAs, with age and VIQ entered as covariates. Two levels of thresholding were used: strict (False Discovery Rate correction with a q=0.005 and Cluster Size Threshold Estimation (CSTE) at p<0.05); and lenient (p<0.005, CSTE threshold of p < 0.05).
Results:
Behavioral: There were no significant group differences in self-endorsement nor in memory recognition rates.
Brain activation:
Self-processing: During Self (>Orthographic) processing, both ASD and TD participants significantly activated the left Superior Frontal Gyrus (SFG), the left Inferior Frontal Gyrus (IFG), Middle Frontal Gyrus (MFG) and the Middle Temporal Gyrus (MTG). During Orthographic (>Self) processing, participants significantly activated their precuneus bilaterally. No group differences were found at the stricter threshold, but at the more lenient threshold the left Precuneus was significantly more active in the TD than ASD participants.
Memory: The left Anterior Cingulate and Lingual Gyrus were significantly more active for Forgotten than Remembered words at the more lenient threshold; and no group differences nor interaction effects were found at either threshold.
Conclusions:
The regions activated by both ASD and TD participants during self-related orthographic and memory processes are in line with previous fMRI studies of self-related (e.g., Lombardo et al., 2009; Kennedy and Courschesne; 2008; Pfeifer et al., 2013), orthographic (e.g., Bolger et al., 2008); and memory processes (Kim, 2010) suggesting that our results are robust. Further, while it is possible that ASD individuals may be impaired on other, more complex, personal memory tasks, the fact that no group differences were found at the strict threshold for self-processing, and at either threshold for memory-related processing (and that no behavioral differences were found for this task), suggests that ASD and TD participants engage similar brain regions when processing, and remembering, self-related trait-words.