Alesha Majors

As children age, they can learn increasingly complex features of environmental structure—a key prerequisite for adaptive decision-making. Yet when we tested children ( N = 304, 4–13 years old) in the Children’s Gambling Task, an age-appropriate variant of the Iowa Gambling Task, we found that age was negatively associated with performance. However, this paradoxical effect of age was found only in children who exhibited a maladaptive deplete-replenish bias, a tendency to shift choices after positive outcomes and repeat choices after negative outcomes. We found that this bias results from sensitivity to incidental nonrandom structure in the canonical, deterministic forms of these tasks—and that it would actually lead to optimal outcomes if the tasks were not deterministic. Our results illustrate that changes in decision-making across early childhood reflect, in part, increasing sensitivity to environmental structure.

Background: Traditional methods for recruiting and maintaining contact with participants in cohort studies include print-based correspondence, which can be unidirectional, labor intensive, and slow. Leveraging technology can substantially enhance communication, maintain engagement of study participants in cohort studies, and facilitate data collection on a range of outcomes. Objective: This paper provides an overview of the development process and design of a cohort management platform (CMP) used in the Newborn Epigenetic STudy (NEST), a large longitudinal birth cohort study. Methods: The platform uses short message service (SMS) text messaging to facilitate interactive communication with participants; it also semiautomatically performs many recruitment and retention procedures typically completed by research assistants over the course of multiple study follow-up visits. Results: Since February 2016, 302 participants have consented to enrollment in the platform and 162 have enrolled with active engagement in the system. Daily reminders are being used to help improve adherence to the study’s accelerometer wear protocol. At the time of this report, 213 participants in our follow-up study who were also registered to use the CMP were eligible for the accelerometer protocol. Preliminary data show that texters (138/213, 64.8%), when compared to nontexters (75/213, 35.2%), had significantly longer average accelerometer-wearing hours (165.6 hours, SD 56.5, vs 145.3 hours, SD 58.5, P=.01) when instructed to wear the devices for 1 full week. Conclusions: This platform can serve as a model for enhancing communication and engagement with longitudinal study cohorts, especially those involved in studies assessing environmental exposures.

BACKGROUND Traditional methods for recruiting and maintaining contact with participants in cohort studies include print-based correspondence, which can be unidirectional, labor intensive, and slow. Leveraging technology can substantially enhance communication, maintain engagement of study participants in cohort studies, and facilitate data collection on a range of outcomes. OBJECTIVE This paper provides an overview of the development process and design of a cohort management platform (CMP) used in the Newborn Epigenetic STudy (NEST), a large longitudinal birth cohort study. METHODS The platform uses short message service (SMS) text messaging to facilitate interactive communication with participants; it also semiautomatically performs many recruitment and retention procedures typically completed by research assistants over the course of multiple study follow-up visits. RESULTS Since February 2016, 302 participants have consented to enrollment in the platform and 162 have enrolled with active engagement in the system. Daily reminders are being used to help improve adherence to the study’s accelerometer wear protocol. At the time of this report, 213 participants in our follow-up study who were also registered to use the CMP were eligible for the accelerometer protocol. Preliminary data show that texters (138/213, 64.8%), when compared to nontexters (75/213, 35.2%), had significantly longer average accelerometer-wearing hours (165.6 hours, SD 56.5, vs 145.3 hours, SD 58.5, P=.01) when instructed to wear the devices for 1 full week. CONCLUSIONS This platform can serve as a model for enhancing communication and engagement with longitudinal study cohorts, especially those involved in studies assessing environmental exposures.