Psychophysiological and Performance Indices from ERP Tasks Pose Unique Measurement Challenges

• In Event: 1-210 - Improving the Quality and Efficiency of Measurement in Early Life

Integrative Statement

Psychophysiological and neurobehavioral measures represent a class of individual difference marker that have promise for helping to bridge scientific work on cognitive, neurological, and emotional development. There are particular challenges when these strategies are used to model individual difference constructs and test questions about broader psychological processes and trait psychology, however. Shorthand borrowed from the language and techniques of individual difference work in other domains (e.g., temperament, questionnaire assessment) often do not translate well into these kinds of assessment contexts due to the constraints involved in the tasks, and the fundamental differences in underlying causal (and therefore measurement) models of how indices reflect the processes of interest. Behavioral task performance and neurophysiological markers are also subject to particular developmental forces that are related to quantifying stable individual difference factors unique to other measurement strategies. In this talk, we will describe an approach to analyzing, conceptualizing, and integrating child psychophysiological and behavioral performance indices to understand the development of neurobehavioral processes.
We collected data on multiple event-related potentials (ERPs) across a battery of psychophysiological tasks, using different stimuli and task requirements, in a sample of 166 children between the ages of 3 and 13 years of age. The sample consisted of participants from the community and from an ongoing, longitudinal, high-risk study of alcoholism. Children completed tasks involving error-monitoring (flankers), inhibition (Go-No Go), fear (startle potentiation), and reward processing.
We find evidence consistent with data from other samples that individual differences in the same ERP are only modestly correlated across different cognitive tasks taken during the same testing session. In some cases, associations across psychophysiological tasks was stronger than associations for the same ERP across different tasks. This suggests that conceptualizations that view a single ERP in a single task as a marker of a trait-like process are a misfit to the data. Associations between behavioral performance measures across tasks are larger, but performance indices are only modestly to moderately correlated with ERPs during the same tasks. We also found evidence that the association between behavioral performance indices and ERPs was stronger with increasing age.

Authors

• Emily Durbin, Michigan State University
  Presenting Author

• Jason M Moser, Michigan State University
  Non-Presenting Author

• Rachel Stacey, Michigan State University
  Non-Presenting Author

• Angus Clark, University of Michigan
  Non-Presenting Author