The test-retest reliability of the contingent negative variation (CNV) in children and adults before and after removing aberrant CNV segments
Date
2014
Authors
Cabral, Brittany Kristine, author
Gavin, William, advisor
Davies, Patricia, committee member
Fidler, Deborah, committee member
Seger, Carol, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
The contingent negative variation (CNV) is a slow, negative drift in electroencephalographic potential that occurs between two stimuli. Researchers have examined the CNV and its embedded components, the O-wave and E-wave, in the study of development and dysfunction in attentional processing. However, few studies have tested the reliability of these components, and never in a paradigm with two visual stimuli or in children. The present study investigated the test-retest reliability of the visually-evoked CNV components in 58 children and 32 adults. The efficacy of a newly-developed procedure for reducing trial-to-trial variability in ERPs was also tested. Participants performed a visual Go-NoGo task while EEG data were recorded during two sessions scheduled one-to-two weeks apart. Developmental data agreed with previous literature such that children had significantly less negative CNV amplitudes than adults, though each component presented with a significant Group by Session interaction. Adult amplitudes became less negative from one session to the next, and children's data shifted in the opposite direction. Correlational analyses also indicated that developmental trends were present among children; amplitudes became more negative with increasing age. Reliability analyses revealed significantly lower indices than previous findings using auditory paradigms. Although children seemed to have higher reliability (r = .34 - .53) than adults (r = .05 - .58), analyses revealed no significant differences between these groups. The newly developed procedure for reducing variability did not significantly improve reliability, but it did significantly change the amplitude of the total CNV. Future investigations should further examine the efficacy of this new procedure in producing averaged ERPs. The data from the present study suggest that researchers and clinicians should be careful in interpreting visually-evoked CNV components. Changes noted over time through the course of development or intervention may largely be the result of normal fluctuations in brain processing. Further research is required to better understand what underlying mechanisms may be affecting the reliability of the CNV components, and how to improve the reliability through adjustments to data collection procedures and measurement techniques.