Proceedings of the 29th Annual International Conference of the IEEE EMBS Cit Internationale, Lyon, France August 23-26,

2007.

FrP1D5.12

A High Resolution EEG Study of Dynamic Brain Activity during Video Game Play

C. Sheikholeslami, H. Yuan, Student Member, IEEE, E.J. He, X. Bai, L. Yang, and B. He, Fellow, IEEE
(9-12 Hz) band over the parietal area during the game playing. But the dynamic signature of game-related modulation in response to a long-lasting playing was not clear. Spectral components of human EEG have been shown to be sensitive to the brain activity associated with mental tasks. The frontal midline theta rhythm is associated with performance of various mental tasks. The dipole localization method has localized this signal to the region of the anterior cingulated cortex (ACC) [5]. The increases in theta activity with task difficulty suggest that this signal might be related to the mental load allocated to task performance. In contrast, posterior alpha signals decreased with increased mental load [4]. However, oscillations in the alpha band were found to increase with mental loads in a short-term memory task [6]. In the present study, we investigate the dynamic brain activity during continuous video game play using the high resolution EEG. Subjects played a competitive video game, Mario Power Tennis, on a Nintendo Gamecube while their EEG signals were recorded at evenly distributed time segments. Spectral analysis was performed at each time segment of spontaneous signals to determine the dynamic affects of long-lasting video game play. II. METHODS Two healthy subjects participated in the present high-resolution EEG study. A competitive game - Mario Power Tennis (Nintendo Gamecube) was selected. Mario Power Tennis is a quick paced tennis video game with the classic Mario Brothers characters. The game requires real-time active responses from the player, but little high-level strategy. Both subjects had not played the game before the study. They were given the instructions to play the game before starting the game play. The EEG signals were recorded with 128 channels using Neuroscan Labs Synamps 2 (Compumedics, Inc., El Paso, TX). The EEG signals were digitized with a sampling rate of 1000 Hz and band-pass filtered from 0.15 - 70 Hz. Vertical and horizontal EOGs were also recorded for artifact rejection. The experimental recording protocol was broken up into 3-minute recording segments evenly distributed throughout the experiment. For each recording segment the subject was asked to minimize eye blinks and muscle movements, as well as having their eyes open as much as they can. Before the subjects started to play the game a 3-minute open eye resting condition was taken as baseline recording. Afterwards, the subject played the video game continuously for 65 minutes.

Abstract A high resolution EEG study was conducted on healthy human subjects during video game play. Throughout the game playing experiment short time segments of spontaneous activity were recorded. Spectral analysis was performed on these segments for the theta-wave (4-8 Hz) and alpha-wave (8-13 Hz) bands to investigate the modulatory effects of long-lasting game play and the dynamic changes of spectral contribution in range of alpha- and theta-wave. The present results revealed that a) the frontal midline theta-wave activity increased over time relative to the eye open resting condition and b) the parietal alpha-wave activity initially decreased relative to the resting condition, then followed by a slow increase. These experimental results indicate the high resolution EEG provides a useful quantitative analysis tool for studying dynamic brain activity. Keywords high resolution EEG, spontaneous EEG, theta wave, alpha wave, video game, brain mapping
I.

video games have increased in popularity, there is an emerging interest to study the cognitive and emotional process of brain activities during game playing. One of the most characteristic traits of human behavior is the competitive brain state and intensive mental loads. The goal of the present study is to investigate the brain responses to a competitive video game during a long period of continuous play. Recent advances in neuroimaging techniques, like the functional Magnetic Resonance Imaging (fMRI) and high resolution electroencephalogram (EEG), allow game-related processing and its neural correlates to be studied noninvasively in humans. Efforts have been made to use fMRI to examine brain regions involved in economic games and to investigate the relative contributions of cognitive and emotional aspects to human decision-making [1-2]). In an EEG study, mental effort-induced changes in oscillatory cortical activity were examined during video game playing that requires attention to orientation and color constitution in juvenile subjects with epilepsy and normal control subjects [3]. Their results showed increased frontal midline EEG activity in theta (6-7 Hz) band and attenuated activity in alpha
S Manuscript received April 16, 2007. This work was supported in part by NIH RO1EB00178 and NSF BES-0411898. H.Y. was supported in part by an NIH Neuro-Physical-Computational Sciences Fellowship. C.S. was supported in part by an UROP fellowship from the University of Minnesota. Cameron Sheikholeslami, Han Yuan, Xiaoxiao Bai, Lin Yang and Bin He are with the Department of Biomedical Engineering, University of Minnesota, Minneapolis, MN 55455, USA (e-mail: binhe@umn.edu). Eric He is with the Mounds View High School, Arden Hills, MN 55112, USA.

INTRODUCTION

1-4244-0788-5/07/$20.00 2007 IEEE

2489

After each 10-minutes segment of playing the game, EEG signals were recorded for 3 minutes. During these recordings the subject would still be playing the game. Fig. 1 shows a block diagram of the experimental protocol and Fig. 2 depicts the experimental setup.

Fig. 1. Block diagram of experimental protocol

to the baseline recording. The percent difference map was calculated by subtracting and normalized by the baseline recordings power map. The five recording segments are displayed in progression from the beginning of the experiment to the end. The percent difference is displayed using a blue-to-red color scheme, with bright blue representing a theta-wave power decrease of 50% from the baseline and bright red representing a 50% increase from the baseline. Fig. 4 shows the 3-dimensional parietal alpha-wave percent difference maps, calculated and displayed in the same manner as the theta-wave percent difference maps. The color scheme goes from bright blue representing a 100% decrease in alpha-wave power to bright red representing a 100% increase from baseline.

Fig. 2. Experimental setup

Post-recording analysis was performed on the EEG recordings using EEGLAB [7]. The EEG signals were band-pass filtered again from 0.5 35 Hz using an FIR filter. Eye blinks were identified using the vertical and horizontal EOG channels. The data were then visually inspected and data segments corrupted by blinks, residual eye movements and body movement artifacts were rejected from subsequent analysis. Following the artifact rejection the signals were re-referenced using a common average of the potentials at all electrodes. Spectral analysis was performed with Welchs power spectral density method. The EEG signals for each channel were split into short sequences (2048 points, 2.05s) and windowed with a Hamming window. The discrete Fourier transform was then computed on the short sequences, followed by computing the squared magnitude of the result. A 50% overlap was applied and the results were averaged over each 3-minute recording segment. A frequency resolution of 0.195 Hz was achieved. The theta-wave band power was determined by averaging the spectra power from 4 8 Hz and the alpha-wave band power was determined by averaging the spectra power from 8 13 Hz. The scalp distributions of spectral power were then visualized in a 3-dimensional standard head model using the SCAN software (Compumedics, Inc., El Paso, TX). III. RESULTS Fig. 3 shows the frontal theta-wave difference maps using a standard head model. The averaged power over the theta band during each 3-minute recording segment was compared

Fig. 3 Theta-wave percent difference maps in a subject. Red and blue colors refer to increase and decrease in theta activity with reference to the open eye resting condition before starting the game play. Note that theta activity increased gradually on the frontal lobe during the game play.

Fig. 4 Alpha-wave percent difference maps in a subject. Blue color refers to decrease in alpha activity with reference to the open eye resting condition before starting the game play. Note that the alpha activity decreased on the parietal lobe during the game play.

Theta-wave powers in the region of interest over the frontal cortex were averaged and computed for each recording segment. The time-domain analysis of the recording segments theta-wave power is shown in Fig. 5 a) and c) for subjects #1 and #2, respectively. Alpha-wave powers in the parietal electrodes were averaged and computed for each recording segment and shown in Fig. 5 b) and d) for subjects #1 and #2, respectively. The 3rd recording segment in subject #2 was very noisy due to outside interference, therefore was rejected. IV. DISCUSSION It has been shown that frontal midline theta-wave power increases as a result of video game play [3]. The present study revealed the changes of spontaneous brain activity in response to long-lasting video game play. A time-domain

2490

the parietal alpha-wave power during the period of playing video while there is general attenuation relative to the baseline activity. The pattern is observable in both Fig. 4 and Fig. 5 b) and d). It is also worthy to mention that Fig. 5 d) shows an initial further decrease in the parietal alpha-wave power followed by a rebound as the subject played the game longer. The parietal alpha-wave band rebound is an interesting phenomenon that requires further investigation into its significance. These results indicate that long, continuous periods of video game play increases the mental load of the players, which has mental health implications. Further investigations are needed to delineate the cortical sources that account for the dynamic modulatory effects of long-lasting video game play.
Fig. 5 Average frontal electrode theta-wave percent difference parietal alpha-wave percent difference. a) Subject 1 frontal theta-wave percent difference, b) Subject 1 parietal alpha-wave percent difference; c) Subject 2 frontal theta-wave percent difference; d) Subject 2 parietal alpha-wave percent difference.

REFERENCES
[1] [2] [3] Sanfey AG, Rilling JK, Aronson JA, Nystrom LE, Cohen JD. The neural basis of economic decision-making in the Ultimatum Game. Science. 2003 Jun 13;300(5626):1755-8 King-Casas B, Tomlin D, Anen C, Camerer CF, Quartz SR, Montague PR. Getting to Know You: Reputation and Trust in a Two-Person Economic Exchange. Science. 2005 Apr 1;308(5718):78-83. Pellouchoud E, Smith ME, McEvoy L, Gevins A. Mental effort-related EEG modulation during video-game play: comparison between juvenile subjects with epilepsy and normal control subjects. Epilepsia. 1999;40 Suppl 4:38-43. Gundel A, Wilson GF. Topographical changes in the ongoing EEG related to the difficulty of mental tasks. Brain Topogr. 1992 Fall;5(1):17-25 Gevins A, Smith ME, McEvoy L, Yu D. High-resolution EEG mapping of cortical activation related to working memory: effects of task difficulty, type of processing, and practice. Cereb Cortex. 1997 Jun;7(4):374-85. Jensen O, Gelfand J, Kounios J, Lisman JE. Oscillations in the alpha band (9-12 Hz) increase with memory load during retention in a short-term memory task. Cereb Cortex. 2002 Aug;12(8):877-82. Delorme A, Makeig S. EEGLAB: an open source toolbox for analysis of single-trial EEG dynamics including independent component analysis. J Neurosci Methods. 2004 Mar 15;134(1):9-21.

analysis of these increases shows that the longer a subject plays the video game, the larger the frontal theta-wave increase becomes. This trend is shown in Fig. 3 and Fig. 5 a) and c). The topographical location (frontal region) confirms previous studies indications [5] that the ACC is involved in tasks of concentration. The larger increase in theta activity in the frontal lobe suggests that long term competitive video game play may represent a significant mental load to the subjects. Parietal alpha-wave power has been shown to attenuate as a result of video game play [3]. It has also been shown to increase with mental load [6]. However, the temporal aspects of these phenomena have not been explored for continuous video game play. Our results revealed the trend of rebound in

[4] [5]

[6] [7]

2491