0169 Distinct Dimensions in Psychomotor Vigilance Test (PVT) Metrics

Abstract Introduction The PVT measures sustained attention through speeded responses to rolling millisecond counter stimuli, which appear at random 2-10s intervals across the 10min task. Performance impairment on the PVT has been quantified using a variety of outcome metrics that describe aspects of...

Full description

Saved in:
Bibliographic Details
Published in:Sleep (New York, N.Y.) N.Y.), 2018-04, Vol.41 (suppl_1), p.A66-A67
Main Authors: Riedy, S M, Van Dongen, H
Format: Article
Language:English
Subjects:
Principal components analysis
Sleep deprivation
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Introduction The PVT measures sustained attention through speeded responses to rolling millisecond counter stimuli, which appear at random 2-10s intervals across the 10min task. Performance impairment on the PVT has been quantified using a variety of outcome metrics that describe aspects of the response time (RT) distribution. We investigated whether PVT metrics used in the literature reflect essentially the same or different aspects of performance impairment due to sleep loss. Methods N=99 healthy subjects (ages 22–37, 50 females) participated in 1 of 4 laboratory studies, each involving a baseline sleep opportunity (10h TIB) followed by at least 38h of extended wakefulness. The 10min PVT was administered every 2h of scheduled wakefulness. Thirty-six metrics were calculated for each PVT bout. Principal component analysis (PCA) and mixed-effects ANOVA were used to analyze the data. Results A total of 1,034 test bouts were completed. PCA indicated that there were four distinct dimensions among the 36 outcome metrics, with some metrics associating with more than one dimension. Dimension 1 captured response slowing; dimensions 2 and 3 captured response variability; and dimension 4 captured premature responses (false starts). Dimension 3 captured PVT metrics that showed no reliable sensitivity to sleep deprivation, and this dimension was therefore discarded. Dimensions 1, 2 and 4 exhibited stereotypical profiles of change over time (F10,980>4.5, p
ISSN:0161-8105
1550-9109
DOI:10.1093/sleep/zsy061.168