Asymmetry pattern of resting EEG for different IQ levels. A.H. Jahidin*, M.N. Taib, N.M. Tahir, M.S.A. Megat Ali and S. Lias

Transcription

1 Available online at ScienceDirect Procedia - Social and Behavioral Scien ce s 97 ( 2013 ) The 9 th International Conference on Cognitive Science Asymmetry pattern of resting EEG for different IQ levels A.H. Jahidin*, M.N. Taib, N.M. Tahir, M.S.A. Megat Ali and S. Lias Faculty of Electrical Engineering, Universiti Teknologi MARA, Shah Alam, Selangor, Malaysia Abstract The aim of this work is to analyze asymmetry pattern of resting brainwave for different intelligence levels. EEG signals were measured from fifty samples an alpha and beta waves were computed by subtracting ln-transformed ESD of the left from right hemisphere. The study elucidate that individuals in high IQ level exhibit a balanced brain with smaller asymmetry score for alpha and beta waves compared to medium and low IQ levels. Meanwhile, the medium and low IQ levels exhibit unbalanced brain alpha and beta activity with greater asymmetry values The Authors. Published by by Elsevier Ltd. Ltd. Open access under CC BY-NC-ND license. Selection and/or peer-review under under responsibility of the of Universiti the Universiti Malaysia Malaysia Sarawak. Sarawak Keywords: 1. Introduction Intelligence encompasses a human ability to comprehend complex ideas, to adapt effectively with the environment, to learn from experience, to engage forms of reasoning, and to overcome obstacles by taking thoughts. The ability differs between individuals [1] and can be gauged via conventional psychometric tests such as the Wechsler Intelligence Scale [2] [3] and mental rotation tasks [4]. A mainstream study suggests that the brain is less mentally active in brighter individuals compared to the lesser intelligent people [5]. This leads to the hypothesis of neural efficiency and is supported by higher alpha, but lower beta power in the prefrontal region of brighter individuals [6]. Apart from an efficient brain, other related studies have also related specific functioning of the brain hemispheres with intelligence [7-9]. These have been studied in conjunction with other behavioural and cognitive-related parameters such as attentiveness [10] and information processing strategies [11]. For that purpose, the with a defined cognitive task. The left hemisphere is involved in sequential processes [12], analogic reasoning [13], non-verbal semantic-categorical processing [14] and linguistic [15]. Meanwhile, the right hemisphere is involved in visuo-spatial processing [16]. Hence, lateralization of hemispheric functioning results in brain asymmetry with specific hemisphere dominating the other in a defined cognitive task. These have been observed in terms of alpha and beta band power, and have also extended into a relational asymmetry index between both hemispheres. A related study shows that brain activation is * Corresponding author. Tel.: ; fax: address: aisyah23@gmail.com The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license. Selection and/or peer-review under responsibility of the Universiti Malaysia Sarawak. doi: /j.sbspro

2 A.H. Jahidin et al. / Procedia - Social and Behavioral Sciences 97 ( 2013 ) symmetrical when solving simple arithmetic operations, but is left dominant when solving complex mathematical problems [17]. Recent finding also suggests that activation asymmetry to left be related to high motivation [18]. Frontal asymmetry to left has also been associated with verbal fluency [19]. These however, contradict with findings that relate asymmetry characteristics with giftedness, where gifted individuals are right dominant. The phenomenon may be contributed by the creative ability that is much contributed by the right hemisphere [20]. This paper elaborates on the brain asymmetry pattern in relation to different intelligence level. The uniqueness of the finding is defined through the analysis of resting EEG and non-verbal reasoning task to cluster subjects into different levels of intelligence quotient (IQ). 2. Methods As shown in Fig. 1, several procedures were performed to accomplish the research work. These involve EEG measurement, signal pre-processing and filtering, extraction of normalized energy spectral density (ESD), and calculating the brain asymmetry score (AS). Detailed explanation is described in the following sub-sections and the patterns of AS among the different IQ levels are elaborated in Section 3. EEG recording and discrimination of IQ levels (resting and stimulated states) Signal pre-processing and filtering Feature extraction (ESD of alpha and beta waves) Normalization of ESD using ln-transformation technique Computation of brain asymmetry scores 2.1. EEG recording and establishment of IQ levels Fig. 1. Flow chart for assessment of resting asymmetry 50 healthy postgraduate and undergraduate students; mean age = 23.9 and standard deviation (SD) = 3.5 from various disciplines have volunteered in this research work. Real-time EEG signal was recorded using G-MobiLAB. Fp1 (left hemisphere-lhs) and Fp2 (right hemisphere-rhs) are connected to the prefrontal area with Fpz as the reference. Both ear lobes were used for grounding and the electrode placements conform with the international system. The EEG is sampled at a frequency of 256 Hz. EEG were recorded in two sessions; relax closed eyes for 3 minutes during the first session and with the brain stimulated with IQ test for 10 min in the second session. A non-verbal reasoning-based IQ test, adopted from RPM has been employed in the study. The IQ test has been set up in a user-friendly graphical user interface (GUI) to reduce body motion. The focus of study will be on resting EEG, hence further analysis on the second recording session will not be discussed further. The scores obtained from the IQ test are used to group the samples into different IQ levels [8, 9, 21].

3 248 A.H. Jahidin et al. / Procedia - Social and Behavioral Sciences 97 ( 2013 ) Signal pre-processing, filtering and feature extraction The EEG was pre-processed offline using MATLAB 2010b. The trials which were affected by noise; characterized by amplitudes exceeding ±100 μv, has been rejected. As no EOG (EMG) signal was recorded during data collection, automatic rejection approach [22] on the EEG has been employed. The EEG is then filtered into alpha (8 13 Hz) and beta (13 30 Hz) waves by means of bandpass filter. The spectral characteristics of the artifactfree EEG signals are computed via Fast Fourier Transform (FFT) algorithm. The power spectral density (PSD) has been signal was divided into 50% overlapping epochs with a window size of 256. The length of FFT is 1024 and extracted using a Hamming window. Then, the ESD which covers the entire energy distribution for each range of the frequency band is calculated Brain asymmetry score in resting alpha and beta waves Initially, the normalized ESD of alpha and beta waves for both hemispheres have been computed via lntransformation method and plotted to observe the pattern in RHS and LHS. As mathematically expressed by Eq. 1, the AS is then derived from the difference of normalized magnitude between both hemispheres. AS j ln( ESD j, Fp 2) ln( ESD j, Fp1) where j represent brainwave frequency bands. Positive scores of AS indicate right dominance while negative scores refer to left dominance. 3. Results and Discussion 3.1. Establishment of IQ levels The IQ scores based RPM range from 0 to 150 and were statistically analyzed to group the samples into different IQ levels. The population of scores (N=50) is in normal distribution with the mean (96.9) and SD (24.8). Therefore, three distinct IQ levels have been established with the corresponding number of samples as shown in Fig. 2. Due to the existing outliers, one sample each has been removed from the low and medium IQ levels; ending up with a total of 48 samples. (1) Fig. 2. Number of samples among IQ levels

4 A.H. Jahidin et al. / Procedia - Social and Behavioral Sciences 97 ( 2013 ) Assessment on alpha and beta asymmetry The ln-transformed ESD of LHS and RHS with respect to low, medium and high IQ groups are as shown in Fig. 3. It is observed that the normalized ESD for alpha wave in RHS is higher compared to LHS in medium and high IQ levels. In contrast, the normalized ESD for alpha wave in RHS is lesser than LHS for low IQ level. An inversed pattern can be observed in high IQ level for beta wave where the normalized ESD in RHS is lower than LHS. The medium and low IQ levels however attained similar pattern as the normalized ESD for alpha wave. Fig. 3. Normalized ESD for alpha (on the left) and beta waves (on the right) for LHS and RHS among different IQ levels Subsequently, Fig. 4 shows the mean of AS for alpha and beta waves. The high IQ level tends to be more balanced with smaller asymmetry value for alpha and beta waves. A high IQ level is characterized by a right dominant alpha activity. In contrast, beta activity is left dominant. Meanwhile, medium and low IQ levels indicate unbalanced brain activity with greater asymmetry values. It can be observed that alpha and beta activities are right dominant for medium IQ level. However, low IQ level exhibit brain activities that are left dominant. The findings support previous evidence, where the low asymmetry indicates efficient attention shifting between both hemispheres and is associated with higher intelligence [23]. Furthermore, deviation from symmetrical brain activities in less intelligent individuals is associated with less efficient network configuration in the brain [24]. Fig. 4. Resting alpha and beta asymmetry in distinct IQ levels

5 250 A.H. Jahidin et al. / Procedia - Social and Behavioral Sciences 97 ( 2013 ) Conclusion Significant pattern of resting alpha and beta waves can be observed from the asymmetry score with varying IQ levels. EEG activity in relaxed state may reflect the relationship between balanced brain and intelligence. As a conclusion, findings show that the highly intelligent individuals are able to maintain relatively balanced attention between the two hemispheres, as compared to the medium and low IQ groups. Acknowledgements This work was supported by the Ministry of Higher Education (Grant No.: 600-RMI/FRGS 5/3 (72/2012)), as well as Research Management Institute (Grant No: 600-RMI/DANA 5/3/RIF (121/2012)) and Faculty of Electrical Engineering, Universiti Teknologi MARA, Malaysia. The corresponding author is sponsored through the Malaysian MyPhD scholarship programme. References [1] Neubauer AC, Fink A. Intelligence and neural efficiency. Neuroscience & Biobehavioral Reviews 2009; 33: [2] Benson N, Hulac DM, Kranzler JH. Independent Examination of the Wechsler Adult Intelligence Scale-(WAIS-IV): What Does the WAIS-IV Measure? Psychological Assessment 2010; 22: [3] Kaufman SB, DeYoung CG, Gray JR et al. Associative learning predicts intelligence above and beyond working memory and processing speed. Intelligence 2009; 37: [4] ncy and serial processing. Neuroscience Letters 2010; 482: [5] Fink A, Grabner RH, Benedek M et al. The creative brain: Investigation of brain activity during creative problem solving by means of EEG and fmri. Human Brain Mapping 2009; 30: [6] Neubauer AC, Fink A. Intelligence and neural efficiency: Measures of brain activation versus measures of functional connectivity in the brain. Intelligence 2009; 37: [7] van der Knaap LJ, van der Ham IJM. How does the corpus callosum mediate interhemispheric transfer? A review. Behavioural brain research 2011; 223: [8] Jahidin AH, Taib MN, Md Tahir N et al. Brainwave sub-band power ratio characteristics in intelligence assessment. Control and System Graduate Research Colloquium (ICSGRC), 2012 IEEE; July 2012; p [9] Jahidin AH, Taib MN, Ali MSAM et al. Evaluation of brainwave sub-band spectral centroid in human intelligence. Signal Processing and its Applications (CSPA), 2013 IEEE 9th International Colloquium on; March 2013; p [10] Posner MI, Petersen SE. The Attention System of the Human Brain. Annual Review of Neuroscience 1990; 13: [11] Riding RJ, Glass A, Butler SR, Pleydell Pearce CW. Cognitive Style and Individual Differences in EEG Alpha During Information Processing. Educational Psychology 1997; 17: [12] Monfort M, Martin SA, Frederickson W. Information-processing differences and laterality of students from different colleges and disciplines. Perceptual and Motor Skills 1990; 70: [13] Boroojerdi B, Phipps M, Kopylev L et al. Enhancing analogic reasoning with rtms over the left prefrontal cortex. Neurology 2001; 56: [14] -categorical information in a non-verbal semantic format. Brain and Cognition 2013; 81: [15] Corballis PM. Visuospatial processing and the right-hemisphere interpreter. Brain and Cognition 2003; 53: [16] Deary IJ, Penke L, Johnson W. The neuroscience of human intelligence differences. Nature Reviews Neuroscience 2010; 11: [17] Kwon H, Cho J, Lee E. EEG Asymmetry Analysis of the Left and Right Brain Activities During Simple versus Complex Arithmetic Learning. Journal of Neurotherapy 2009; 13: [18] Harmon-Jones E, Gable PA, Peterson CK. The role of asymmetric frontal cortical activity in emotion-related phenomena: A review and update. Biological Psychology 2010; 84: [19] Papousek I, Schulter Gn. Manipulation of frontal brain asymmetry by cognitive tasks. Brain and Cognition 2004; 54: [20] Mrazik M, Dombrowski SC. The neurobiological foundations of giftedness. Roeper Review 2010; 32:

6 A.H. Jahidin et al. / Procedia - Social and Behavioral Sciences 97 ( 2013 ) [21] Lias S, Sulaiman N, Murat ZH, Taib MN. IQ Index using Alpha-Beta correlation of EEG power spectrum density (PSD). Industrial Electronics & Applications (ISIEA), 2010 IEEE Symposium on; 2010: IEEE; p [22] Fatourechi M, Bashashati A, Ward RK, Birch GE. EMG and EOG artifacts in brain computer interface systems: A survey. Clinical Neurophysiology 2007; 118: [23] Nakagawa A. Attentional balance and intelligence. Intelligence 1996; 22: [24] Langer N, Pedroni A, Gianotti LRR et al. Functional brain network efficiency predicts intelligence. Human Brain Mapping 2012; 33: