Using fNIRS to assess cognition in urban Bangladeshi infants and toddlers

• In Event: 2-043 - Objective measures of early development in low- and middle-income countries: EEG, Eye-tracking, fNIRS

Integrative Statement

Introduction:
Prior work using functional near-infrared spectroscopy (fNIRS) has shown differential brain responses in infants to social vs. nonsocial stimuli in urban European (Lloyd-Fox, 2013) and rural African cohorts (Lloyd-Fox, 2014). This experimental paradigm has been proposed as an objective measure of social cognition that can be used in many different cohorts, but is particularly well-suited for use in low- and middle-income countries. In this study, our participants were 6-month-old and 36-month-old Bangladeshi children. Study participants lived in a poor urban neighborhood and were exposed to a broad range of adversity early in life including poverty, malnourishment, recurrent infections, lack of maternal education, and high levels of maternal psychosocial stress.

Methods:
Infants (n=130) and children (n = 130) were tested with a 38-channel NIRS system that recorded brain activity over the frontal and temporal cortices. The experimental stimuli presented were videos of women moving their eyes or hands, accompanied by vocal sounds, non-vocal sounds, or silence. Social videos were alternated with control blocks of pictures of local modes of transportation presented with no sounds. The oxyhemoglobin responses to each condition were calculated using HOMER2 (Huppert, 2009) by finding the mean response from 12-16 seconds after the start of the stimulus presentation. Statistical testing compared the silent visual social condition with the silent control condition, and the auditory social condition (vocal stimuli) with the auditory nonsocial stimuli (non-vocal stimuli) to find brain regions that were selective for auditory or visual social stimuli. Socially selective brain responses were correlated with socioeconomic status risk factors (degree of poverty, maternal education) and social environmental risk factors (maternal psychosocial stress and family care indicators).

Results:
Brain regions that had higher responses to social vs. nonsocial stimuli (ps<.05) for each age group are shown in Figure 1. All tested brain regions were socially selective for auditory and visual stimuli in the 6-month-old cohort. In the 36-month-old cohort, all tested brain regions were selective for visual social stimuli, but only the bilateral middle temporal regions were socially selective for auditory stimuli. Correlations were found between adversity-related risk factors and the auditory social contrast magnitude in both the 6-month-old cohort and the 36-month-old cohort (ps<.05, see Figure 2).

Conclusions:
We find similar visual social-selective regions in 6-month-old infants and 36-month-old children. However, we found social selectivity to auditory social stimuli in more brain regions at 6-months than at 36-months, indicating that there may be some cortical specialization in social processing in this age range. We also found relationships between social selectivity of brain regions and risk factors for poor neurodevelopment including poverty, low maternal education, maternal psychosocial stress, and family care. The social environmental risk factors were correlated in the 36-month cohort but not the 6-month cohort, potentially indicating that the social environment has a larger impact on social brain responses at later developmental timepoints.

Authors

• Katherine Perdue, Harvard Medical School
  Presenting Author

• Sarah K.G. Jensen, Boston Children's Hospital
  Non-Presenting Author

• Swapna Kumar, Boston Children's Hospital
  Non-Presenting Author

• John Richards, University of South Carolina
  Non-Presenting Author

• Shahria Hafiz Kakon, icddrb
  Non-Presenting Author

• Rashidul Haque, ICDDR
  Non-Presenting Author

• William A. Petri, University of Virginia
  Non-Presenting Author

• Sarah Lloyd-Fox, Birkbeck, University of London
  Non-Presenting Author

• Clare E Elwell, University College London
  Non-Presenting Author

• Charles A. Nelson, Harvard Medical School
  Non-Presenting Author