pubmed.ncbi.nlm.nih.gov

Neural processing of facial identity and emotion in infants at high-risk for autism spectrum disorders - PubMed

  • ️Tue Jan 01 2013

Neural processing of facial identity and emotion in infants at high-risk for autism spectrum disorders

Sharon E Fox et al. Front Hum Neurosci. 2013.

Abstract

Deficits in face processing and social impairment are core characteristics of autism spectrum disorder. The present work examined 7-month-old infants at high-risk for developing autism and typically developing controls at low-risk, using a face perception task designed to differentiate between the effects of face identity and facial emotions on neural response using functional Near-Infrared Spectroscopy. In addition, we employed independent component analysis, as well as a novel method of condition-related component selection and classification to identify group differences in hemodynamic waveforms and response distributions associated with face and emotion processing. The results indicate similarities of waveforms, but differences in the magnitude, spatial distribution, and timing of responses between groups. These early differences in local cortical regions and the hemodynamic response may, in turn, contribute to differences in patterns of functional connectivity.

Keywords: autism; cognitive development; fNIRS; face processing; near-infrared spectroscopy.

PubMed Disclaimer

Figures

Figure 1
Figure 1

Video stimuli and probes. (A) Mothers of infant participants were instructed to answer a series of questions with both a neutral and smiling expression. Sound was removed from the videos, and 16 video clips of each emotion were combined to achieve a 32 s stimulus with transition from neutral to smiling expression. Stranger videos were selected for similarity in appearance to the mother. A visual baseline of moving objects was used to direct infant attention to the screen at the beginning of the experimental session, and a fixation cross was present between stimulus presentation. (B) Two chevron arrays of probes placed into a custom designed neoprene cap. Probes were placed in the frontal and right lateral panels, and adjusted to proper fit and location with the colored Velcro tabs.

Figure 2
Figure 2

Results of Face Identity × Emotion × Group ANOVA by Channel. (A) Difference in Oxy-hemoglobin Response; (B) Difference in Deoxy-hemoglobin Response. (M = Mother, S, Stranger; E, Emotion; N, Neutral; HRA, High-Risk Autism; LRC, Low-risk Control). Channels with significant main effects of Face Identity (magenta) and Emotion (green) occurred in frontal regions, while the main effect of Group was seen in right lateral channels (yellow). The difference in population marginal means with corresponding 95% confidence intervals, Tukey’s HSD used for correction for multiple comparisons. Bonferroni correction was employed for correction of comparisons across channels. Marginally significant effects are indicated by “♦” (full statistics in Table A1 in Appendix). Significant interactions were seen across frontal and lateral channels (gray circles). Further information regarding interactions is available in Table 1.

Figure 3
Figure 3

(A) Channels significant for main effects and interactions on oxy-hemoglobin response. Channels with main effects of Face Identity (magenta) and Emotion (green) occurred in frontal regions, while the main effect of Group was seen in right lateral channels (yellow). Significant interactions were seen across frontal and lateral channels (dashed circles). (B) Channels significant for main effects and interactions on deoxy-hemoglobin response. Channels with main effects of Face Identity (magenta), Emotion (green), and Group occurred in frontal regions, while significant interactions were seen across frontal and lateral channels neighboring those significant for oxy-hemoglobin responses (dashed circles).

Figure 4
Figure 4

ICA Data Analysis Diagram (adapted from Katura et al. (2008)). IC, independent component; CR-IC, condition-related independent component; MITC, mean inter-trial cross-correlation.

Figure 5
Figure 5

Centroid Waveforms of Component Clusters. Each waveform (centroid) is the mean of the components of the cluster, after centering and normalizing points to zero mean and unit standard deviation. Three waveforms were obtained for each of the oxy-hemoglobin (HbO: red = HRA, pink = LRC) and deoxy-hemoglobin (HbD: blue = HRA, cyan = LRC) components, for both the smiling and neutral conditions. Waveforms from the HRA and LRC groups are shown together based upon correlation.

Figure 6
Figure 6

Components within Clusters. Individual components assigned to each centroid (black) are depicted separately for each group (HRA and LRC) and condition (neutral and smiling). Oxy-hemoglobin (HbO) components are shown in red (HRA) and pink (LRC), while deoxy-hemoglobin components (HbD) are shown in blue (HRA) and cyan (LRC). The spatial assignment by laterality index is indicated as “frontal,” “lateral,” or “mixed” for each cluster, and numeric differences in mean normalized weight across regions are given (Frontal-Lateral).

Figure A1
Figure A1

(A) Comparison of response to neutral mother and neutral stranger faces within the LRC group. Channel placement and numbers are shown on the right lateral portions of the infant head. Colors within the channels indicate a T-value map across 10 LRC subjects comparing oxy-hemoglobin response to the neutral mother’s face as compared to the neutral stranger’s face at each channel location (color bar, p = 0.05, Bonferroni corrected). The channel circled in red indicates significant differences in oxy-hemoglobin response, and the channel circled in blue indicates significant differences in deoxy-hemoglobin response. (B) Comparison of Oxy-hemoglobin Response to Smiling Mother and Smiling Stranger Faces within the LRC group. Significant activation occurred in channels placed on the front of the head (numbered as shown). Colors within the channels indicate a T-value map across 10 LRC subjects comparing oxy-hemoglobin response to the smiling mother’s face as compared to the smiling stranger’s face. Channels circled in red indicate significant differences in oxy-hemoglobin response.

Figure A2
Figure A2

Spatial distribution of oxy-hemoglobin waveform 2. Colors within channels indicate normalized weights for this waveform according to the color bar. (A) LRC Group. Compared to the neutral condition, there is an increase in right lateral distribution of Cluster 2 during the smiling condition, and a focally increased frontal weight. (B) HRA group. There is increased right lateral distribution of Cluster 2, but without a focal increase of weight. The spatial distribution of frontal weights is reduced between the neutral and smiling conditions.

Figure A3
Figure A3

Spatial distribution of deoxy-hemoglobin waveform 1. Colors within channels indicate normalized weights for this waveform according to the color bar. (A) LRC Group. The waveform has greater frontal weight in the neutral condition, and is distributed between focal frontal and right lateral regions in the smiling condition (B) HRA Group. Frontal weight increases during the smiling condition, resulting in a greater mean weight across frontal channels for the smiling as compared to neutral conditions (p = 0.0002).

Similar articles

Cited by

References

    1. Akgul C. B., Akin A., Sankur B. (2006). Extraction of cognitive activity-related waveforms from functional near-infrared spectroscopy signals. Med. Biol. Eng. Comput. 44, 945–95810.1007/s11517-006-0116-3 - DOI - PubMed
    1. Amaral D. G., Schumann C. M., Nordahl C. W. (2008). Neuroanatomy of autism. Trends Neurosci. 31, 137–14510.1016/j.tins.2007.12.005 - DOI - PubMed
    1. Aslin R. N., Mehler J. (2005). Near-infrared spectroscopy for functional studies of brain activity in human infants: promise, prospects, and challenges. J. Biomed. Opt. 10, 110009.10.1117/1.1854672 - DOI - PubMed
    1. Barnea-Goraly N., Lotspeich L. J., Reiss A. L. (2010). Similar white matter aberrations in children with autism and their unaffected siblings: a diffusion tensor imaging study using tract-based spatial statistics. Arch. Gen. Psychiatry 67, 1052–106010.1001/archgenpsychiatry.2010.123 - DOI - PubMed
    1. Bassett R. C., Rogers J. S., Cherry G. R., Cruzhit C. (1953). The effect of constant media on the blood-brain-barrier. J. Neurosurg. 10, 38–4710.3171/jns.1953.10.1.0038 - DOI - PubMed

LinkOut - more resources