Hot executive function following moderate-to-late preterm birth: altered delay discounting at 4 years of age

Abstract

Interest in monitoring long-term neurodevelopmental outcomes of children born moderate-to-late preterm (32-36 weeks gestation) is increasing. Moderate-to-late preterm birth has a negative impact on academic achievement, which may relate to differential development of executive function (EF). Prior studies reporting deficits in EF in preterm children have almost exclusively assessed EF in affectively neutral contexts in high-risk preterm children (< 32 weeks gestation). Disrupted function in motivational or emotionally charged contexts (hot EF) following preterm birth remains uninvestigated, despite evidence that preterm children show differential development of neural circuitry subserving hot EF, including reduced orbitofrontal cortex volume. The present study is the first to examine whether low-risk, healthy children born moderate-to-late preterm exhibit impairments in the development of hot EF. Preterm children at age 4.5 years were less likely to choose larger, delayed rewards across all levels of reward magnitude on a delay discounting task using tangible rewards, but performed more similarly to their full-term peers on a delay aversion task involving abstract rewards and on measures of cool EF. The relationship between gestational age at birth and selection of delayed rewards extended across the entire gestational age range of the sample (32-42 weeks), and remained significant after controlling for intelligence and processing speed. Results imply that there is not a finite cut-off point at which children are spared from potential long-term neurodevelopmental effects of PT birth. Further investigation of reward processing and hot EF in individuals with a history of PT birth is warranted given the susceptibility of prefrontal cortex development to early environmental variations.

Figure 1.

Group differences in delay discounting at age 4. PT children were less likely to choose the larger delayed reward across all levels of reward magnitude in comparison to FT children. Means ± standard errors are illustrated.

Figure 2.

Gestational age at birth and delay likelihood on the delay discounting task. Gestational age at birth was a predictor of likelihood to delay on the first task trial (as well as the overall percentage of delayed reward choices) across the entire sample of children. Figure illustrates the fitted logistic regression function with 95% confidence interval.

Figure 3.

Parent ratings of EF and delay likelihood on delay discounting task. Lower parent reported EF difficulties in everyday contexts predicted a higher likelihood to delay on the first task trial (as well as the overall percentage of delayed reward choices) across the entire sample of children. Figure illustrates the fitted logistic regression function with 95% confidence interval.