Preterm infants in the NICU show higher engagement with mothers, but contingent communication with fathers

Introduction. Little work assesses infant capacity for bi-directional, contingent communication at birth. No work to our knowledge assesses this capacity with very preterm infants in the neonatal period. Presence vs. absence of such coordination makes a difference for our theories of development. For fullterms, Lavelli and Fogel (2013) showed mother-infant bi-directional effects in the first 2 weeks of life, with fine-grained coding/sequential analysis. For preterms, Feldman and Eidelman (2007) showed conditional probability of maternal affiliative behavior during infant alert state, at 37 weeks, using 10s coding-unit; direction of effects was not addressed. Aim: To examine bi-directional second-by-second contingent communication in mother-infant vs. father-infant dyads with very-preterm infants in the NICU. Hypotheses. Comparing mother- vs. father-infant dyads, (a) engagement levels will be higher in mother-infant dyads; (b) interactive contingency will be higher in mother-infant dyads. Study population. Very-preterm infants (N=20), born 27-33 weeks, weighing 665-2100g (mean 1450g), were studied at 35 weeks when medically stable, lying in heated-cot. Methods. Spontaneous face-to-face communication was videotaped, first with mother, then father, and coded on 1s time-base with engagement scales (Table 1). Multi-level time-series models evaluated self-contingency (auto-correlation) and interactive contingency (lagged cross-correlation). Covariates included twin-status, infant sex, infant medical status, maternal age, maternal depression, paternal depression (CES-D). Posthoc descriptive probability explication identified engagement levels involved in significant effects. Results. Percent time in engagement levels. Mothers (vs. fathers) spent more time in affiliative behavior (highest engagement levels 8+7+6), t(19)=2.43, p=.025, interpreted as more arousing. Fathers (vs. mothers) spent more time in midrange engagement level 5 (Table 1), t(19)=-2.56, p=.019, interpreted as more relaxed, less “demanding” of infant engagement. Infants were more gaze-on-parent’s-face and gaze-on-environment with mothers, t(19)=2.38, p=.028, and more eyes-closed with fathers, t(19)=-2.70, p=.014, interpreted as more alert with mothers. Self-and interactive contingency. In separate mother-infant and father-infant time-series models, all partners showed significant self-contingency, but only infant coordination with father (F→I) was significant (Figure 1). Combining mother-infant and father-infant data, and testing differences for interactive contingency, only fathers and infants showed bi-directional contingency (Figure 1). Posthoc explication of infants→parents showed that, given infants in lower engagement levels 1s prior, fathers (vs. mothers) stayed in lower engagement levels in the current second, closer to infants. Posthoc explication of parents→infants showed that, given both parents in lower (or higher) engagement levels 1s prior, infants stayed in lower (or higher) engagement in the current second closer to fathers, thus coordinating more with fathers than mothers. Conclusions: We conjecture that fathers and infants were more coordinated because fathers were less “demanding” infants to be in high engagement, so that fathers could stay closer to infants’ engagement levels, facilitating infant-father coordination. In conclusion, bi-directional contingent communication is a robust interpersonal capacity form birth, even in very preterm infants.