Presented as a poster at the Neonatal Society 2016 Summer Meeting.
Lee CW1,2, Lloyd-Fox S3, Blasi A3, Chalia M1,2, Hebden JC1,4, Cooper RJ1,4, Austin T1,2
1 neoLAB, The Evelyn Perinatal Imaging Centre, Cambridge University Hospitals NHS FT
2 Department of Neonatology, Cambridge University Hospitals NHS FT
3 Centre for Brain and Cognitive Development, Birkbeck College, London
4 Department of Medical Physics and Biomedical Engineering, University College London
Background: Early cerebral auditory processing facilitates the development of language and cognition. How the newborn brain perceives contrasting sound types can provide fundamental information on early brain development (1,2). In adults, vocal specialisation is located in the superior temporal sulcus (3) while the developmental axis in infancy is unclear. Using noninvasive monitoring at the cot-side, we present the first diffuse optical images (DOI) of selective auditory responses in a group of healthy term infants within the first week of life. We predict cortical activation of sound sensitive regions in the auditory cortex with distinct responses to voice and non-voice auditory stimuli.
Methods: Written consent was obtained from parents for infants to participate. DOI provides spatiotemporal information of the haemodynamic correlates of neural function. DOI scans were obtained from 9 healthy full-term infants (median gestation at birth: 40 weeks, female infants: 5). Two types of auditory stimuli: human non-speech voice (V) and environmental non-vocal sounds (N), lasting 8.5 s each, were presented with 10-15 s silence intervals. These were presented 10 times in a pseudorandom sequence. Using a block-average approach and image reconstruction techniques, DOI images of the cortical response to the V and N conditions were obtained.
Results: The DOI images demonstrate haemodynamic changes that are spatially broad in response to the N condition, while in contrast these changes are focal to the left temporal cortex in the V condition. The average haemo-dynamic signal over a peak 6 s window was calculated for each condition in 3 regions of interest (ROI): primary auditory (PA), inferior frontal (IF) and posterior superior temporal (PST) regions. An omnibus general linear model (GLM) approach for each condition was conducted with the following repeated measures factors: ROI (PA, IF, PST) and hemisphere (left, right), to assess the spatial response. The GLM (and post-hoc analysis) revealed a significant response in the left PA region in the V condition (p=0.02), and no interactions or main effects in the N condition. To identify N and V selective regions, paired t-tests were carried out between N and V responses at each ROI in each hemisphere which revealed the left PST region to be N selective.
Conclusion: This study demonstrates the cortical responses to human voice and non-vocal sounds are distinct in the newborn period suggestive of early cognitive competencies that may facilitate learning and vocal perception. Moving forward, this can provide valuable prognostic markers at an early stage in infants at risk of cognitive and developmental impairment.
Corresponding author: firstname.lastname@example.org
1. Peña, M. et al., 2003, Proc Natl Acad Sci U S A 100, 11702-11705; 2. Sato, H. et al., 2012, Hum Brain Mapp 33, 2092-2103; 3. Belin, P. et al., 2000, Nature, 403, 309-312.