Poster #41 - The Role of Sleep Variability in Adolescent Error Monitoring

• In Event: 1-115 - Poster Session 03
  In Poster Session: PS 03 Section - Cognitive Processes

Integrative Statement

Introduction: Engagement in risk behavior may have negative effects on adolescent health (CDC, 2015) in that risk behavior is tied to risk appraisal and error monitoring. Both of these traits are controlled in part by the anterior cingulate cortex (ACC), which is integral in communication between the prefrontal cortex and limbic system. Error monitoring has been characterized using electroencephalography (EEG) to record event-related potentials (ERPs) produced by the ACC, using electrodes placed at fronto-central regions. ERPs can be quantified by the amplitude of the error-related negativity (ERN) following an incorrect response in cognitive performance tasks (Gehring, 2018). The ERN is a construct drawing on three neural networks: cognitive control, reward perception, and threat perception (Hanna, 2016). ERN amplitude can be influenced by a number of factors, including sleep deprivation (Tsai 2005), monetary incentives (Hsieh, 2010), and mental disorders (Carrasco, 2014). Sleep deprivation is associated with impairments in error monitoring and response inhibition (Renn, 2012), but the effect of variability in habitual sleep on ERN in adolescents is underexplored. This study investigates the role of variability in habitual sleep as it relates to error monitoring in adolescents, measured by actigraphy (sleep data) and electroencephalography.

Methods: The analytic sample is derived from the first wave of an ongoing longitudinal study characterizing neurocognitive correlates associated with adolescents’ self-reported engagement in risk behavior (N = 2017, recruited from public high schools in Southeast Michigan). A total of 60 adolescents at varying levels of risk behavior completed daily sleep diaries accompanied by wrist actigraphy for approximately one week prior to completing a computer administered Flanker task (512 trials) and EEG recording. From the EEG, average ERN and CRN waveforms were quantified for each participant using the Cz electrode. ERN amplitudes were measured in a timeframe of approximately 80 milliseconds following error responses, relative to a pre-response peak 200 to 50 milliseconds prior to responses. ERN and CRN amplitudes were corrected using this pre-response baseline. Habitual sleep duration derived from actigraphy and variability in sleep duration were used as predictors of the ERN/CRN ratio in linear regression models.

Results: Participants completed 379 person-days of actigraphy recordings (Table 1). Sleep duration was not associated with error monitoring, however, variability in habitual sleep duration was negatively associated (r = -.368, p = 0.004) with a decrease in the normalized amplitude of their ERN (Figure 1).

Conclusion: Error-related negativity in adolescents decreases with increased variability in total sleep duration. Given the decreased error-related response to mistakes in their task performance, these results suggest that adolescents who have more variable sleep schedules have a decreased ability to weigh the potential risks accompanying a situation.

Authors

• Caroline E Scheuing, University of Michigan
  Presenting Author

• Edward D. Huntley, Institute for Social Research, University of Michigan
  Non-Presenting Author

• Christine Wagner, University of Michigan
  Non-Presenting Author

• Sarah Churchill, University of Michigan
  Non-Presenting Author

• Joshua R.B. Hatfield, Institute for Social Research, University of Michigan
  Non-Presenting Author

• Daniel Keating, University of Michigan
  Non-Presenting Author