Skip to main content
ARS Home » Southeast Area » Little Rock, Arkansas » Arkansas Children's Nutrition Center » Research » Publications at this Location » Publication #151187
Title: BLINK-FREE DATA: ARE WE THROWING A BABY OUT WITH THE BATHWATER?

Author
item PIVIK, R - ACNC
item DYKMAN, ROSCOE - ACNC

Submitted to: Society of Psychophysiological Research
Publication Type: Abstract Only
Publication Acceptance Date: 5/6/2003
Publication Date: 10/15/2003
Citation: Pivik, R.T., Dykman, R.A. 2003. Blink-free data: are we throwing a baby out with the bathwater [abstract]? Society Of Psychophysiological Research. p. S67.

Interpretive Summary: Spontaneous eyeblinks protect, clean, and moisten the eyes, but other than briefly interrupting vision, there is no known effect of these events on behavior. In this study blink activity was measured while healthy 8-11 year old children performed an attentional task that required pressing a button to target stimuli. These studies were done in the morning before these children ate breakfast. Reaction times to target stimuli were faster when subjects blinked, and over time response speed slowed on trials without blinks, but not on trials with blinks. Many studies measuring brain activity eliminate blink-associated data because of the interference of blink signals with brain waves. These findings suggest that this practice does not provide an accurate measure of performance.

Technical Abstract: Much effort is made to eliminate the field potential influences associated with eye-blinks on brain potentials with the implicit assumption that blinks are sources of artifact without behavioral effects other than a brief interruption of vision. To explore the possibility that there may be more to blink-associated data than meets the eye, covariations of blinks and reaction time measures were examined in 40 healthy children (8-11 yrs. old; IQ >80; 24 females) performing an attentional task. Subjects were tested in the morning while continuing overnight fasting. Measures of sleep (overnight actigraphy) and blood glucose (finger stick) were also obtained. Blinks were recorded (vertical EOG) while subjects performed a Continuous Performance Task (300 letters: .2 s duration; 1/ 2.2 s) pressing a button to target letters (X after A; 17% random occurrence). Recordings were digitized for off-line blink determination (>150µV, >150ms). Data were analyzed using ANOVA with post-hoc t-tests. Subjects were well-rested (~8.5 hrs of sleep) and morning blood glucose levels were normal. Reaction times were ~ 80 ms faster in the presence of blinks (p<.001) and blinks protected against a time on task slowing in reaction time present on blink-free trials [target trials during the 1st vs 2nd 150 trials: with blinks (ns); without blinks (p<.003)]. These findings suggest that blink-free data may provide an inaccurate representation of behavioral and physiological status.