Abnormal functional connectivity in autism spectrum disorders during face processing.

Abnormalities in the interactions between functionally linked brain regions have been suggested to be associated with the clinical impairments observed in autism spectrum disorders (ASD). We investigated functional connectivity within the limbic system during face identification; a primary component of social cognition, in 19 high-functioning adults with ASD and 21 age-and IQ-matched control adults. Activation during identification of previously viewed faces and houses using a one-back paradigm was compared. The fusiform face area (FFA) was individually localized in each participant and used as the seed point for functional connectivity analyses. The degree of correlation between FFA and the extended neural circuitry involved in face identification was tested. A whole brain analysis was also conducted in order to determine whether connectivity from the FFA to aberrant brain locations was present in the ASD group. Measures of clinical severity (ADOS social score and ADI-R social score) were included as independent variables into the functional connectivity analyses. Significant FFA-amygdala and FFA-superior temporal sulcus functional connectivity was found in both the ASD and control participants. However, the control group had significantly increased connectivity to the left amygdala and the posterior cingulate compared to ASD. Post hoc analyses additionally found increased connectivity to the thalamus in the controls. A significant relationship between abnormal functional connectivity and clinical severity in the ASD group was observed. Specifically, greater social impairment was associated with reduced FFA-amygdala connectivity and increased FFA-right inferior frontal connectivity. These results suggest that abnormal neural connections within the limbic system may contribute to the social impairments observed in ASD.

Increased amygdala activation to neutral faces is associated with better face memory performance.

Recent evidence suggests that the role of the amygdala may extend beyond threat detection to include processing socially relevant stimuli in general. Thus, we investigated perception and memory for neutral faces; a stimulus-type that lacks emotional valence yet contains relevant social information. Participants viewed neutral faces or houses when undergoing functional MRI. Neutral face memory testing was conducted outside the scanner. In the functional MRI of faces vs. houses contrast, significant bilateral activation in the amygdala and lateral fusiform gyrus was observed. Increased bilateral amygdala activation was associated with better delayed-memory performance. These findings indicate that the role of the amygdala may include processing neutral yet socially relevant stimuli. Further, amygdala activation, independent of emotional valence, appears to be associated with memory enhancement.

Motor cortex activation is preserved in patients with chronic hemiplegic stroke.

Many central nervous system conditions that cause weakness, including many strokes, injure corticospinal tract but leave motor cortex intact. Little is known about the functional properties of surviving cortical regions in this setting, in part because many studies have used probes reliant on the corticospinal tract. We hypothesized that many features of motor cortex function would be preserved when assessed independent of the stroke-affected corticospinal tract. Functional MRI was used to study 11 patients with chronic hemiplegia after unilateral stroke that spared regions of motor cortex. Activation in stroke-affected hemisphere was evaluated using 3 probes independent of affected corticospinal tract: passive finger movement, a hand-related visuomotor stimulus, and tapping by the nonstroke index finger. The site and magnitude of cortical activation were similar when comparing the stroke hemisphere to findings in 19 control subjects. Patients activated each of 8 cortical regions with similar frequency as compared to controls, generally with a smaller activation volume. In some cases, clinical measures correlated with the size or the site of stroke hemisphere activation. The results suggest that, despite stroke producing contralateral hemiplegia, surviving regions of motor cortex actively participate in the same proprioceptive, visuomotor, and bilateral movement control processes seen in control subjects.