The Effects of an Extended Period of Mental Arithmetic on Response-Locked Electroencephalographic Activity

Jennifer Wade

Department of Nursing

Faculty Sponsor: Dr. Karla Kubitz

(2004)

The purpose of this study was to examine the effects of an extended period of mental arithmetic on response-locked electroencephalographic (EEG) activity. This study was part of a larger data-collection effort, which was carried out by Dr. Karla Kubitz of Towson University and Dr. Leonard Trejo of the NASA Ames Research Center, focused on EEG monitoring of mental fatigue. The participants in the study were 17 volunteers (13 males and 4 females) between the ages of 18 and 38. All had normal vision and hearing and were asked to refrain from consuming caffeine or other stimulants prior to testing. Most (14 out of 17) were right-handed. Participants signed an informed consent and were compensated for their participation in this study. The materials and equipment used in this study included: (1) two self-report measures of mood states, the Activation Deactivation Adjective Checklist (AD-ACL, Thayer, 1967; 1978) and the Visual Analogue Mood Scale (VAMS, Stern, 1997; (2) a computerized mental arithmetic task; and (3) a Neuroscan ^TM data acquisition and analysis system. The computerized mental arithmetic task asked participants to repetitively solve simple math problems (i.e., to find the algebraic sum of 4 single-digit numbers) and to indicate their answers by pressing buttons on a keypad. The Neuroscan ^TM system included their: (a) Quick Cap ^TM electrode positioning system; (b) QuickGel ^TM electrode gel; (c) SynAmps ^TM amplifiers; (d) SCAN ^TM data acquisition software; and (e) Edit ^TM data analysis software. The research design was a repeated measures design. The independent variable was time-on-task. The dependent variables were EEG activity (i.e., the power in each of the four traditional frequency bands, including the delta [0-4 Hz], theta [4-7 Hz], alpha [8-12 Hz], and beta [16-20 Hz] bands), response times, and self-reported mood. The procedures were that the participants: (1) completed a 10-minute familiarization session on the computerized mental arithmetic task; (2) were prepared for data collection by having the electrode cap attached; (3) self-reported pre-task mood states using the AD-ACL and the VAMS); (4) completed an extended, three-hour, version of the computerized mental arithmetic task [i.e., repeatedly solving mental arithmetic problems and pressing buttons on a keypad indicating solutions]; (5) self-reported post-task mood states [again using the AD-ACL and the VAS]. The data were processed to remove artifact, including manual artifact rejection, eye blink artifact correction, low pass (<30 Hz) filtering, and baseline correction. The 'cleaned' data were epoched around the response (i.e., they were time-locked to the period between -5000 ms before and +500 ms after the button press) and processed via Fast Fourier Transformation and power density analysis. Power densities were analyzed using SPSS. It was hypothesized that, consistent with the known effects of physical fatigue, mental fatigue would shift the power from the slower to the faster EEG frequency bands. That is, it was hypothesized that the power in the beta and alpha bands would decrease and the power in the theta and delta bands would decrease concomitant with time-on-task. It was also hypothesized that self-reported fatigue and response time (during the mental arithmetic task) would likewise increase as time-on-task increased. The results will be discussed in terms of their implications for EEG monitoring of mental fatigue.

Links

Updated July 9, 2004