Gamma activation in young people with autism spectrum disorders and typically-developing controls when viewing emotions on faces

Abstract

Background: Behavioural studies have highlighted irregularities in recognition of facial affect in children and young people with autism spectrum disorders (ASDs). Recent findings from studies utilising electroencephalography (EEG) and magnetoencephalography (MEG) have identified abnormal activation and irregular maintenance of gamma (>30 Hz) range oscillations when ASD individuals attempt basic visual and auditory tasks.

Methodology/principal findings: The pilot study reported here is the first study to use spatial filtering techniques in MEG to explore face processing in children with ASD. We set out to examine theoretical suggestions that gamma activation underlying face processing may be different in a group of children and young people with ASD (n = 13) compared to typically developing (TD) age, gender and IQ matched controls. Beamforming and virtual electrode techniques were used to assess spatially localised induced and evoked activity. While lower-band (3-30 Hz) responses to faces were similar between groups, the ASD gamma response in occipital areas was observed to be largely absent when viewing emotions on faces. Virtual electrode analysis indicated the presence of intact evoked responses but abnormal induced activity in ASD participants.

Conclusions/significance: These findings lend weight to previous suggestions that specific components of the early visual response to emotional faces is abnormal in ASD. Elucidation of the nature and specificity of these findings is worthy of further research.

Figure 1. Response times for identifying specific emotions in ASD (red) and control (green) participants.

Figure 2. Significant differences in gamma band (30–80

Hz) response, 250–450 ms, between ASD and TD participants. Red  =  TD > ASD; Blue  =  ASD > TD; p<.001 for all highlighted voxels. Slice shown MNI: −36, −86, 2 (voxel of maximum difference).

Figure 3. The gamma band response (30–80 Hz), 250–450

ms, in TD (left) and ASD (right) participants. Red  =  maxima, blue  =  minima. P<.05 for all highlighted voxels. Slice shown MNI: −16, −56, 8.

Figure 4. The lower band response (3–30 Hz), in TD (left) and ASD (right) participants.

Red  =  maxima, blue  =  minima p<.05 for all highlighted voxels. Slice shown MNI: 32, −72, −2.

Figure 5. Virtual electrode responses in the right lateral occipital cortex, superior division (R-LOP; 36, −86, 2) for ASD (a) and TD (b) participants.

Evoked responses are displayed in the upper row of each figure; induced responses are displayed in the lower row. All responses indicate within subjects changes from baseline; changes significant at p<.05 level are indicated within dotted lines.

Figure 6. Virtual electrode responses in the left occipital pole (L-OP; −20, −92, 16) for ASD (a) and TD (b) participants.

Evoked responses are displayed in the upper row of each figure; induced responses are displayed in the lower row. All responses indicate within subjects changes from baseline; changes significant at p<.05 level are indicated within dotted lines.

Figure 7. Virtual electrode responses in the right fusiform gyrus (R-FG; 32, −57, −3) for ASD (a) and TD (b) participants.

Evoked responses are displayed in the upper row of each figure; induced responses are displayed in the lower row. All responses indicate within subjects changes from baseline; changes significant at p<.05 level are indicated within dotted lines.