Authors
Bethan Dean, Lorna Ginnell, Victoria Ledsham, Emma Telford, Sarah Sparrow, Magda Rudnicka, Hilary Cruickshank, Sue Fletcher-Watson, James P. Boardman

Institution(s)
1. MRC Centre for Reproductive Health, University of Edinburgh, UK
2. Simpson Neonatal Unit, NHS Lothian, UK
3. Salvesen Mindroom Research Centre for Learning Difficulties, University of Edinburgh, UK

Introduction
Preterm birth is associated with atypical social cognition in infancy1, and with social difficulties and cognitive impairment in childhood2. Socioeconomic deprivation is also known to impact child social development3. The stability of an individual’s social cognition through childhood and its relationship with neurodevelopment are unknown. We used eye-tracking to investigate social attention over time in preterm and term-born infants, its relationship with standardised tests of neurodevelopment, and the modulating role of socioeconomic deprivation.

Methods
81 preterm and 66 term infants, with mean (range) gestational age at birth 28+5 (23+2-33+0) and 40+0 (37+0-42+1) respectively, completed eye-tracking in infancy, with a subset reassessed at 5 years. 3 free-viewing tasks presented social stimuli of increasing complexity. We measured fixation duration to areas of social interest and calculated a social preference score. We collected socioeconomic data and performed neurodevelopmental assessments at 2 and 5 years. We had NRES and NHS R&D approval and received funding from Theirworld. We used t-tests for group comparisons and assessed relationships using correlations and hierachical modelling.

Results
Preterm infants had lower social attentional preference in infancy compared to infants born at term (0.32 vs 0.39, p=0.02). Term infants’ social preference scores were stable between the two time points (0.43 vs 0.45, p=0.73), whereas preterm infants showed a significant increase over time (0.29 vs 0.43, p<0.005), reaching equivalent social preference scores to term controls by 5 years. In whole sample multivariate analyses, low gestational age and socioeconomic deprivation were associated with reduced social attention in infancy (rs(135)=0.20, p=0.02 and rs(135)=0.21, p=0.01), explaining a modest but significant proportion of the variance in hierarchical models.
The preterm infants scored lower in neurodevelopmental assessments at 5 years of age (Mullen Early Learning Composite 93 vs 104, p=0.003), particularly in language domains. Social attentional preference in infancy did not correlate with performance on standardised tests of neurodevelopment at 2 or 5 years.

Conclusions
Preterm infants have reduced social attentional preference in infancy compared to term-born controls, but catch up by 5 years. This is influenced by socioeconomic deprivation and gestational age. Further work is required to determine the role of infant social cognition in the ontogeny of cognitive impairment seen in preterm infants.

References: 1. EJ Telford et al., J Child Psychol Psychiatry 57 (7), 861 (2016). 2. S Johnson and N Marlow, Arch Dis Child 102 (1), 97 (2017). 3. SH Landry et al, J Clin Exp Neuropsychol 19 (2), 261 (1997).