- Email: [email protected]
Brain oscillations are highly influenced by gender differences
International Journal of Psychophysiology 65 (2007) 294 – 299 www.elsevier.com/locate/ijpsycho
Fast track report
Brain oscillations are highly influenced by gender differences Bahar Güntekin, Erol Başar ⁎ Istanbul Kültür University, Brain Dynamics, Cognition and Complex Systems Research Unit, Faculty of Science and Letters, Istanbul, Turkey Received 12 February 2007; received in revised form 23 March 2007; accepted 23 March 2007 Available online 30 March 2007
Abstract There are few studies describing gender differences in event related oscillations (EROs). In the present report we demonstrate that gender differences are apparent in EROs even during simple visual stimulation, possibly activating very basic sensory networks. The data of 32 (16 males) healthy subjects were recorded from thirteen different scalp locations (F3, F4, Cz, C3, C4, T3, T4, T5, T6, P3, P4, O1, O2). Analysis was performed in the delta (0.5–3.5 Hz), theta (5–8.5 Hz), alpha (9–13 Hz), beta (15–24 Hz), and gamma (28–48 Hz) frequency ranges. The results showed that the maximum peak-to-peak delta response amplitudes for women were significantly higher than for men over occipital, parietal, central and temporal electrode locations. There were also differentiations in the beta and gamma oscillatory responses. These gender differences were most pronounced over the electrode site O2, that is, over primary visual areas. It is suggested that this study might serve as a standard to investigate gender differences in electrophysiology. © 2007 Elsevier B.V. All rights reserved. Keywords: Gender differences; Brain oscillations; Occipital cortex; EEG; Visual evoked potentials
1. Introduction In the decade of the brain Mountcastle (1992) indicated a paradigm change in neuroscience by stating that slow brain oscillatory potentials now provide evidence of being real signals of the CNS and that they are correlated with cognitive functions of the brain. Accordingly, experimental designs examining oscillatory brain dynamics became a very attractive research field in the neurosciences, since it is now possible to perform measurements on scalp electrodes of human subjects in various states of behavior and learning. Possibly, the first systematic work on the analysis of functional correlates of event related oscillations (EROs) had started in the 1970s. Freeman (1975) emphasized gamma band oscillations as the major physiological network property for understanding vegetative and cognitive brain processes. Parallel to this trend, our group indicated the necessity of using the brain's multiple oscillatory activities for the analysis of all brain functions in both animals and humans (Başar, 1980; Başar et al., 1975). Currently, the neuroscience literature contains a large number of publications dealing with ⁎ Corresponding author. Tel.: +90 212 498 43 92; fax: +90 212 498 45 46. E-mail address: [email protected] (E. Başar). 0167-8760/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ijpsycho.2007.03.009
different psychophysiological paradigms or strategies to find correlations between brain oscillatory responses and most complicated cognitive functions (Güntekin and Başar, 2007). Although in the 1990s mainly the modulations of gamma oscillations by sensory-cognitive demands were investigated, the importance of oscillations in the lower frequency bands is also emphasized in the literature since the last decade. It is a now widely accepted view that the EEG reflects/derives from the activity of an ensemble of generators producing rhythmic activity in several frequency ranges. During rest, these oscillators are usually active in a random way. However, during sensory stimulation these generators are coupled and act coherently. This synchronization and enhancement of EEG activity gives rise to “evoked” or “induced rhythms”. Evoked potentials are considered as a result of the transition of neural population responses from a disordered to an ordered state. The compound of the event related potential (ERP) is manifested in a superposition of evoked oscillations of different frequency bands (“natural frequencies of the brain”: delta: 0.5–3.5 Hz, theta: 3.5–7 Hz, alpha: 8–13 Hz, beta: 15–24 Hz, and gamma: 30–70 Hz) (Başar, 1999; Yordanova and Kolev, 1998). Although the strategy of analyzing EROs has gained considerable interest in the last years, there are only a small
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299
number of studies investigating the influences of basic causal factors on EROs. One of the most important aspects is genetic causality; such studies were pioneered by Begleiter and Porjesz (2006). A recent publication by Rangaswamy et al. (2007) excellently reviews the competing theories about the genesis of ERPs which dominate the current literature. Another important causal factor influencing the event related EEG response is gender differences. However, there are only few conventional ERP and almost no ERO studies investigating the influence of gender differences. The present report aims to describe the influence of gender difference on EROs in response to simple light stimulation in all the EEG-frequency bands, namely delta, theta, beta, alpha, and gamma oscillations. It is important to use a simple sensory stimulation in order to show whether it is possible to find a difference in the basic sensory circuit. 2. Methods
295
used. All electrode impedances were less then 5 kΩ. The EEG was amplified by means of a Nihon Kohden EEG-4421 G apparatus with band limits 0.1–100 Hz 24 dB/octave. A BrainData-System device was used for signal analysis and evaluation of oscillatory dynamics. The EEG was digitized on-line with a sampling rate of 512 Hz and a total recording time of 2000 ms, 1000 ms of which served as the pre-stimulus baseline. 2.4. Computation of event related potentials (ERPs) and event related oscillations (EROs) Before the averaging procedure, epochs containing artifacts were rejected by an off-line technique, i.e. single sweep EOG recordings were visually studied and trials with eye-movement or blink artifacts rejected. Subject averages and grand averages were calculated for each electrode site and experimental condition. The data were digitally filtered according to the frequency bands of interest (see below for determination of frequency bands).
2.1. Participants 2.5. Amplitude frequency characteristics and digital filtering Sixteen women and sixteen men in the age range from 16 to 39 years volunteered for the measurements. The mean age of the female subjects was 23.06 ± 4.64 and the mean age of the male subjects was 23.37 ± 4.30 years. Fourteen of the female and fifteen of the male subjects were right handed. All subjects had completed at least 10 years of education. All subjects completed a questionnaire about their family history, demographics, medical profile and drinking habits. No subjects reported any known current or past neurological or psychiatric history. All participants had normal or corrected to normal vision. All subjects signed an approved consent form. The participating women reported the time of their last menses prior to the testing. It was found that 9 of the women were between 1 and 16 days after the first day of their menses and 7 of the women were between 20 and 27 days after the first day of their menses. 2.2. Stimuli and experimental procedure For stimulation a simple square visual stimulus with an intensity of approximately 30 cd/m2 was displayed on a computer screen. Viewing distance was 120 cm, resulting in a stimulus size of 8.1° of visual angle (absolute stimulus size was 17 × 17 cm). Seventy-five light stimuli were presented with a stimulus duration of 1000 ms and a blank interval between each stimulus presentation. The duration of the blank interval varied randomly between 3 and 7 s. 2.3. Electrophysiological recording The EEG was recorded from F3, F4, Cz, C3, C4, T3, T4, T5, T6, P3, P4, O1 and O2 locations according to the 10–20 system (Jasper, 1958). For the recordings an EEG-CAP (Ag/AgCl electrodes) was used. Linked earlobe electrodes (A1 + A2) served as reference. EOG from medial upper and lateral orbital rim of the right eye was also registered. For the reference electrodes, and EOG recordings, Ag/AgCl electrodes were
The numerical evaluation of the frequency characteristics was accomplished using a Fast Fourier transform (FFT). According to the amplitude frequency characteristics (AFC), the frequencies of interest were determined and the frequency ranges for the digital filtering defined. In most of the studies on EROs, conventional EEG-frequency bands are used. Based on our long-standing analyses we have concluded that this simple choice leads to errors (Başar, 1999). Hence, in the adaptive filtering we chose the limits of the frequency bands according to the major maxima, and defined the cut-off frequencies as the minima in the AFCs. For the frequency ranges, grand averages were computed based on single subjects' averages of the AFCs for each condition and location. As oscillatory responses we defined the peak-to-peak maximum amplitude, i.e. an oscillatory component of an evoked potential in the frequency range determined as a frequency of interest by the AFC analysis. That is, for further analysis we measured the peak-to-peak amplitude of each subject's averaged responses filtered in the range of delta (0.5–3.5 Hz) between 0 and 600 ms, theta (5–8.5 Hz) between 0 and 500 ms, alpha (9–13 Hz) between 0 and 500 ms, beta (15–24 Hz) between 0 and 300 ms, and gamma (28–48 Hz) between 0 and 300 ms for both gender groups. The time period, (or latencies) of EROs merits important consideration: As a rule, for higher frequencies shorter periods, and for the lower frequencies longer analysis periods, were used in order to cover only the immediate time period following the stimulation. The computational program SPSS was used for statistical analysis. Peak-to-peak maximum amplitude responses were separately analyzed for each frequency band by means of a repeated measures ANOVA including the between subjects factor gender (female, male) and the within subject factor location (F3, F4, Cz, C3, C4, T3, T4, T5, T6, P3, P4, O1 and O2). Greenhouse–Geisser corrected p values have been used. Post-hoc analysis was conducted using the Wilcoxon paired sample test.
296
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299
responses revealed a significant effect for location (F = 15.05, p = 0.000), indicating that the alpha response was largest over occipital locations.
3. Results 3.1. Delta frequency window Maximum peak-to-peak delta responses were analyzed in the 0–600 ms time window after stimulation. The largest gender differences were observed at the parietal and occipital locations. From the values of Table 1 one can recognize that in parietal and occipital locations the peak-to-peak amplitudes can be up to 50% larger for females than for males. The ANOVA on delta responses revealed a significant effect for location (F = 740, p = 0.000) indicating an increased delta response over posterior recording sites, and for gender (F = 6.76), p = 0.014), indicating a larger delta response for women than for men (see Fig. 1A). Post-hoc comparisons using the Wilcoxon paired sample test revealed that the peak-to-peak delta response was significantly larger for women than for men over the electrode sites Cz, T5, P3, P4, O1 and O2 (p b 0.05 for all comparisons, see Table 1 and Fig. 1A for further information). 3.2. Theta frequency window Maximum peak-to-peak theta responses were analyzed in the 0–500 ms time window after stimulation. In the theta frequency range no significant differences between men and women were found. The ANOVA on theta responses revealed a significant effect for location (F = 10.71, p = 0.000), indicating that the theta response was larger for frontal, central and parietal recording sites. 3.3. Alpha frequency window In the alpha frequency range, the maximum peak-to-peak response was analyzed in the 0–400 ms time window. No significant differences between men and women were found. We also have to mention that there is a tendency that male subjects to show higher amplitude in T6 and O2 locations (Fig. 1A), however this tendency is not significant. The ANOVA on alpha
Table 1 For female and male subjects the mean, standard deviations and p values of the maximum peak-to-peak amplitude values are shown for the delta, beta and gamma frequency ranges at different locations Location– frequency
Female Mean value μv
Standard deviation
Male Mean value μv
Standard deviation
Cz Delta T5 Delta P3 Delta P4 Delta O1 Delta O2 Delta O1 Beta O2 Beta O2 Gamma
6.1936 3.8313 6.3035 5.5824 6.2502 6.6248 2.8032 2.7716 1.397
±2.6156 ±1.1303 ±3.2922 ±2.3954 ±2.2661 ±2.2345 ±1.5162 ±0.9632 ±0.8319
4.4631 2.9454 3.8094 4.2996 3.7551 4.5105 1.7943 2.1288 0.9484
±1.7111 ±1.4977 ±1.8960 ±2.2202 ±2.3748 ±3.1163 ±0.6089 ±0.5588 ±0.3585
p value
0.006 0.015 0.002 0.044 0.003 0.034 0.013 0.007 0.03
3.4. Beta frequency window Maximum peak-to-peak beta responses were analyzed in the 0–300 ms time window. The ANOVA on beta responses revealed a significant effect for location (F = 9.23, p = 0.000) indicating that the beta response over occipital sites was increased. ANOVA revealed no significant results for location × gender (F = 1.60, p = 0.14) or for gender (F = 2.09, p = 0.15). The Wilcoxon paired sample test showed that the peak-to-peak beta response amplitudes of the female subjects were significantly higher than the peak-to-peak beta response amplitudes of the male subjects at the O1 and O2 locations (O1, p = 0.013; O2, p = 0.007) (Table 1). 3.5. Gamma frequency window Maximum peak-to-peak gamma responses were analyzed in the 0–300 ms time window. The ANOVA on gamma responses revealed a significant effect for location (F = 7.14, p = 0.000), indicating that the gamma response over occipital sites was increased. ANOVA revealed no significant results for location × gender (F = 1.80), p = 0.12) or for gender (F = 0.85, p = 0.36). The Wilcoxon paired sample test showed that peakto-peak gamma response amplitudes of the female subjects were significantly higher than the peak-to-peak gamma response amplitudes of the male subjects at O2 (p = 0.03) (Table 1). Our use of the Wilcoxon test, despite non-significant results on ANOVA, could be criticized. There are different opinions about this issue from the viewpoint of statistical procedures. We performed our ANOVA analysis using a large number of electrode locations, reducing the chance of obtaining overall significant differences. In this context, we find it useful to add the information from a significant Wilcoxon test despite the non-significant ANOVA analysis. We think that the description of these results merits consideration. Fig. 1A presents all significant results on a topological map. The delta response was larger in women than in men over the electrode sites Cz,T5, P3, and P4. The increase of the delta response in women was 37% over Cz, 29% over P4, and 65% over P3. The increase of the delta response was even more pronounced over occipital recording sites (66% for O1 and 46% for O2). The occipital beta and gamma responses were also larger in women than in men. Globally, the dominant differentiations between women and men occur at the posterior sites in the delta frequency range. There are no significant differences at the frontal locations or in the theta frequency range (see Results). The right occipital site seems to play a special role, as we find gender differences in three frequency ranges; this means that the right occipital location depicts the most important differentiation between genders in visual processing of simple light stimulation. In addition to the histograms of Fig. 1A we add an extending illustration showing the grand averages in the delta,
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299
297
Fig. 1. A: Histograms of the mean of the averages in delta (0.5–3.5 Hz), beta (15–24 Hz), gamma (28–48) Hz and alpha (9–13 Hz) frequency range. “⁎” represents significant results. B: Grand averages of oscillatory responses of female and male subjects in delta, beta, gamma and alpha. Female subjects are represented by black lines and male subjects are represented by gray lines.
beta, gamma and alpha frequency windows (Fig. 1B). The differences in peak-to-peak amplitudes are clearly seen in this illustration, which supports the statistical information of Fig. 1A.
4. Discussion Only few studies compare ERPs and EROs between male and female subjects. The common finding of already published
298
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299
studies is the higher amplitude of female evoked potentials in different latency ranges. Orozco and Ehlers (1998) reported longer latency and higher amplitude in the P450 component to facial expressions. During a recognition memory task Guillem and Mograss (2005) recorded higher amplitudes of ERPs in female subjects. According to Hoffman and Polich (1999) male subjects produce smaller P200, P300 and N200 components during a classical oddball paradigm. Proverbio et al. (2006) published results on infant happy/distressed expression indicating much larger occipital P110 response in women compared to men. Karakaş et al. (2006) used auditory stimulation and found that the differences in gamma band are independent of gender. We have to note that none of these studies considered a simple physiological stimulation; they have been focused rather on facial expression, also reflecting emotional behavior, attention and working memory. According to Fuster (1995) the phyletic memory is constituted by the structure of primary sensory and motor cortex at birth, which is common to all organisms of the same species. Phyletic memory would be the most basic of all memories, the genetic memory that the organism has formed in the course of evolution by interactions with the surrounding world. The genetically defined structures of primary cortex dedicated to analyzing elementary sensory features and to integrating elementary primitives of movement would form the basic template on which the memory of the individual would develop. In the light of Fuster's definition Başar (2004) emphasized that even “simple light” triggers sensory memory (iconic memory) and that all experiments related to sensorycognitive processing are interrelated or interwoven with the sensory memory. This is a simple physiological consideration, which includes the necessity to use EPs as control signal. Accordingly, we have to filter out effects of simple light causation in analyzing responses to facial expressions, to attentive processes and also to various types of cognitive loads. In consideration of these facts, the most basic question in our discussion is: “Does simple light evoke differentiations in EROs in comparing female and male subjects? If responses to simple light are anchored with some basic physiological mechanisms, we have to use these results in order to be able to discuss perceptual processes such as face recognition or recognition of facial expressions and/or emotions in a more refined manner. The results of this study also indicate that there are very important differences in the delta band and smaller differences in the beta and gamma frequency ranges. The histogram results presented in Fig. 1A merit important attention. Firstly, we have to note that no differentiations are observed at frontal locations or in the theta frequency range. According to Klimesch et al. (2000) and Başar et al. (1999) a large number of cognitive functions occur in the theta frequency range and at frontal locations. Since we used here a simple light as stimulation, and the subjects did not perform any cognitive tasks, the absence of differentiation in the frontal theta is not surprising. Two other aspects should be mentioned also here. In the vast literature on brain oscillations, the role of gamma oscillations has been mentioned as one of the most important functional key oscillations. In the present study, the gamma response depicts
only a differentiating role at the right occipital site. According to our results, the right occipital location is the area which is the most active indicating the differentiation between males and female subjects. 5. Conclusion Based also on the results of our previous publications related to oscillatory brain dynamics, we come to the following conclusions: (1) The brain response is a construct in a “multidimensional state” incorporating amplitudes of oscillatory responses, topological coordinates, and changes in the time axis following presentation of the percepts, including delays and prolongations, and coherence between locations. Only a new Cartesian System embracing all these parameters can represent the dynamics of functionality in the brain. (2) It is also important to note that in order to analyze and interpret EROs it seems to be extremely important to consider several causal factors (Başar and Güntekin, 2007). Among these, gender differences should be considered as a fundamental causal factor in the search for functional correlates of EROs. Acknowledgements The authors are thankful to Dr. Birgit Mathes for helpful comments on the manuscript. References Başar, E., 1980. EEG-Brain Dynamics. Relation Between EEG and Brain Evoked Potentials. Elsevier, Amsterdam. Başar, E., 1999. Brain Function and Oscillations: II. Integrative Brain Function. Neurophysiology and Cognitive Processes. Springer-Verlag, Heidelberg. Başar, E., 2004. Memory and Brain Dynamics: Oscillations Integrating Attention, Perception, Learning and Memory. CRC Press, Florida. Başar, E., Gönder, A., Özesmi, C., Ungan, P., 1975. Dynamics of brain rhythmic and evoked potentials. II. Studies in the auditory pathway, reticular formation, and hippocampus during the waking stage. Biol. Cybern. 5, 145–160. Başar, E., Başar-Eroglu, C., Karakaş, S., Schurmann, M., 1999. Are cognitive processes manifested in event-related gamma, alpha, theta and delta oscillations in the EEG? Neurosci. Lett. 15 (259(3)), 165–168. Başar, E., Güntekin, B., 2007. A breakthrough in neuroscience needs a “Nebulous Cartesian System" Oscillations, quantum dynamics and chaos in the brain and vegetative system. Int. J. Psychophysiol. 64 (1) (April), 108–122. Begleiter, H., Porjesz, B., 2006. Genetics of human brain oscillations. Int. J. Psychophysiol. 60 (2) (May), 162–171. Freeman, W.J., 1975. Mass Action in the Nervous System. Academic Press, New York. Fuster, J.M., 1995. Memory in the Cerebral Cortex. A Bradford Book. The MIT Press, Cambridge. Guillem, F., Mograss, M., 2005. Gender differences in memory processing: evidence from event-related potentials to faces. Brain Cogn. 57, 84–92. Güntekin, B., Başar, E., 2007. Emotional face expressions are differentiated with brain oscillations. Int. J. Psychophysiol. 64 (1) (Apr), 91–100. Hoffman, L.D., Polich, J., 1999. P300, handedness, and corpus callosal size: gender, modality, and task. Int. J. Psychophysiol. 31, 163–174. Jasper, H.H., 1958. The ten-twenty electrode system of the International Federation. Electroencephalogr Clin. Neurophysiol. 10, 371–375. Karakaş, S., Tüfekçi, I., Bekçi, B., Çakmak, E.D., Doğutepe, E., Erzengin, O.U., Özkan, A., Arıkan, O., 2006. Early time-locked gamma response and gender specificity. Int. J. Psychophysiol. 60, 225–239.
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299 Klimesch, W., Doppelmayr, M., Schwaiger, J., Winkler, T., Gruber, W., 2000. Theta oscillations and the ERP old/new effect: independent phenomena? Clin. Neurophysiol. 111 (5), 781–793. Mountcastle, V.B., 1992. Preface. In: Başar, E., Bullock, T.H. (Eds.), Induced Rhythms in the Brain. Birkhäuser, Boston, MA, pp. XVII–XIX. Orozco, S., Ehlers, C.L., 1998. Gender differences in electrophysiological responses to facial stimuli. Biol. Psychiatry 15 (4), 281–289. Rangaswamy, M., Jones, K.A., Porjesz, B., Chorlian, D.B., Padmanabhapillai, A., Kamarajan, C., Kuperman, S., Rohrbaugh, J., O'connor, S.J., Bauer, L.O.,
299
Schuckit, M.A., Begleiter, H., 2007. Delta and theta oscillations as risk markers in adolescent offspring of alcoholics. Int. J. Psychophysiol. 63 (1) , 3–15 Jan. Proverbio, A.M., Brignone, V., Matarazzo, S., Zotto, M.D., Zani, A., 2006. Gender and parental status affect the visual cortical response to infant facial expression. Neuropsychologia 44, 2987–2999. Yordanova, J., Kolev, V., 1998. Single-sweep analysis of the theta frequency band during an auditory oddball task. Psychophysiology 35 (1), 116–126.
Fast track report
Brain oscillations are highly influenced by gender differences Bahar Güntekin, Erol Başar ⁎ Istanbul Kültür University, Brain Dynamics, Cognition and Complex Systems Research Unit, Faculty of Science and Letters, Istanbul, Turkey Received 12 February 2007; received in revised form 23 March 2007; accepted 23 March 2007 Available online 30 March 2007
Abstract There are few studies describing gender differences in event related oscillations (EROs). In the present report we demonstrate that gender differences are apparent in EROs even during simple visual stimulation, possibly activating very basic sensory networks. The data of 32 (16 males) healthy subjects were recorded from thirteen different scalp locations (F3, F4, Cz, C3, C4, T3, T4, T5, T6, P3, P4, O1, O2). Analysis was performed in the delta (0.5–3.5 Hz), theta (5–8.5 Hz), alpha (9–13 Hz), beta (15–24 Hz), and gamma (28–48 Hz) frequency ranges. The results showed that the maximum peak-to-peak delta response amplitudes for women were significantly higher than for men over occipital, parietal, central and temporal electrode locations. There were also differentiations in the beta and gamma oscillatory responses. These gender differences were most pronounced over the electrode site O2, that is, over primary visual areas. It is suggested that this study might serve as a standard to investigate gender differences in electrophysiology. © 2007 Elsevier B.V. All rights reserved. Keywords: Gender differences; Brain oscillations; Occipital cortex; EEG; Visual evoked potentials
1. Introduction In the decade of the brain Mountcastle (1992) indicated a paradigm change in neuroscience by stating that slow brain oscillatory potentials now provide evidence of being real signals of the CNS and that they are correlated with cognitive functions of the brain. Accordingly, experimental designs examining oscillatory brain dynamics became a very attractive research field in the neurosciences, since it is now possible to perform measurements on scalp electrodes of human subjects in various states of behavior and learning. Possibly, the first systematic work on the analysis of functional correlates of event related oscillations (EROs) had started in the 1970s. Freeman (1975) emphasized gamma band oscillations as the major physiological network property for understanding vegetative and cognitive brain processes. Parallel to this trend, our group indicated the necessity of using the brain's multiple oscillatory activities for the analysis of all brain functions in both animals and humans (Başar, 1980; Başar et al., 1975). Currently, the neuroscience literature contains a large number of publications dealing with ⁎ Corresponding author. Tel.: +90 212 498 43 92; fax: +90 212 498 45 46. E-mail address: [email protected] (E. Başar). 0167-8760/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ijpsycho.2007.03.009
different psychophysiological paradigms or strategies to find correlations between brain oscillatory responses and most complicated cognitive functions (Güntekin and Başar, 2007). Although in the 1990s mainly the modulations of gamma oscillations by sensory-cognitive demands were investigated, the importance of oscillations in the lower frequency bands is also emphasized in the literature since the last decade. It is a now widely accepted view that the EEG reflects/derives from the activity of an ensemble of generators producing rhythmic activity in several frequency ranges. During rest, these oscillators are usually active in a random way. However, during sensory stimulation these generators are coupled and act coherently. This synchronization and enhancement of EEG activity gives rise to “evoked” or “induced rhythms”. Evoked potentials are considered as a result of the transition of neural population responses from a disordered to an ordered state. The compound of the event related potential (ERP) is manifested in a superposition of evoked oscillations of different frequency bands (“natural frequencies of the brain”: delta: 0.5–3.5 Hz, theta: 3.5–7 Hz, alpha: 8–13 Hz, beta: 15–24 Hz, and gamma: 30–70 Hz) (Başar, 1999; Yordanova and Kolev, 1998). Although the strategy of analyzing EROs has gained considerable interest in the last years, there are only a small
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299
number of studies investigating the influences of basic causal factors on EROs. One of the most important aspects is genetic causality; such studies were pioneered by Begleiter and Porjesz (2006). A recent publication by Rangaswamy et al. (2007) excellently reviews the competing theories about the genesis of ERPs which dominate the current literature. Another important causal factor influencing the event related EEG response is gender differences. However, there are only few conventional ERP and almost no ERO studies investigating the influence of gender differences. The present report aims to describe the influence of gender difference on EROs in response to simple light stimulation in all the EEG-frequency bands, namely delta, theta, beta, alpha, and gamma oscillations. It is important to use a simple sensory stimulation in order to show whether it is possible to find a difference in the basic sensory circuit. 2. Methods
295
used. All electrode impedances were less then 5 kΩ. The EEG was amplified by means of a Nihon Kohden EEG-4421 G apparatus with band limits 0.1–100 Hz 24 dB/octave. A BrainData-System device was used for signal analysis and evaluation of oscillatory dynamics. The EEG was digitized on-line with a sampling rate of 512 Hz and a total recording time of 2000 ms, 1000 ms of which served as the pre-stimulus baseline. 2.4. Computation of event related potentials (ERPs) and event related oscillations (EROs) Before the averaging procedure, epochs containing artifacts were rejected by an off-line technique, i.e. single sweep EOG recordings were visually studied and trials with eye-movement or blink artifacts rejected. Subject averages and grand averages were calculated for each electrode site and experimental condition. The data were digitally filtered according to the frequency bands of interest (see below for determination of frequency bands).
2.1. Participants 2.5. Amplitude frequency characteristics and digital filtering Sixteen women and sixteen men in the age range from 16 to 39 years volunteered for the measurements. The mean age of the female subjects was 23.06 ± 4.64 and the mean age of the male subjects was 23.37 ± 4.30 years. Fourteen of the female and fifteen of the male subjects were right handed. All subjects had completed at least 10 years of education. All subjects completed a questionnaire about their family history, demographics, medical profile and drinking habits. No subjects reported any known current or past neurological or psychiatric history. All participants had normal or corrected to normal vision. All subjects signed an approved consent form. The participating women reported the time of their last menses prior to the testing. It was found that 9 of the women were between 1 and 16 days after the first day of their menses and 7 of the women were between 20 and 27 days after the first day of their menses. 2.2. Stimuli and experimental procedure For stimulation a simple square visual stimulus with an intensity of approximately 30 cd/m2 was displayed on a computer screen. Viewing distance was 120 cm, resulting in a stimulus size of 8.1° of visual angle (absolute stimulus size was 17 × 17 cm). Seventy-five light stimuli were presented with a stimulus duration of 1000 ms and a blank interval between each stimulus presentation. The duration of the blank interval varied randomly between 3 and 7 s. 2.3. Electrophysiological recording The EEG was recorded from F3, F4, Cz, C3, C4, T3, T4, T5, T6, P3, P4, O1 and O2 locations according to the 10–20 system (Jasper, 1958). For the recordings an EEG-CAP (Ag/AgCl electrodes) was used. Linked earlobe electrodes (A1 + A2) served as reference. EOG from medial upper and lateral orbital rim of the right eye was also registered. For the reference electrodes, and EOG recordings, Ag/AgCl electrodes were
The numerical evaluation of the frequency characteristics was accomplished using a Fast Fourier transform (FFT). According to the amplitude frequency characteristics (AFC), the frequencies of interest were determined and the frequency ranges for the digital filtering defined. In most of the studies on EROs, conventional EEG-frequency bands are used. Based on our long-standing analyses we have concluded that this simple choice leads to errors (Başar, 1999). Hence, in the adaptive filtering we chose the limits of the frequency bands according to the major maxima, and defined the cut-off frequencies as the minima in the AFCs. For the frequency ranges, grand averages were computed based on single subjects' averages of the AFCs for each condition and location. As oscillatory responses we defined the peak-to-peak maximum amplitude, i.e. an oscillatory component of an evoked potential in the frequency range determined as a frequency of interest by the AFC analysis. That is, for further analysis we measured the peak-to-peak amplitude of each subject's averaged responses filtered in the range of delta (0.5–3.5 Hz) between 0 and 600 ms, theta (5–8.5 Hz) between 0 and 500 ms, alpha (9–13 Hz) between 0 and 500 ms, beta (15–24 Hz) between 0 and 300 ms, and gamma (28–48 Hz) between 0 and 300 ms for both gender groups. The time period, (or latencies) of EROs merits important consideration: As a rule, for higher frequencies shorter periods, and for the lower frequencies longer analysis periods, were used in order to cover only the immediate time period following the stimulation. The computational program SPSS was used for statistical analysis. Peak-to-peak maximum amplitude responses were separately analyzed for each frequency band by means of a repeated measures ANOVA including the between subjects factor gender (female, male) and the within subject factor location (F3, F4, Cz, C3, C4, T3, T4, T5, T6, P3, P4, O1 and O2). Greenhouse–Geisser corrected p values have been used. Post-hoc analysis was conducted using the Wilcoxon paired sample test.
296
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299
responses revealed a significant effect for location (F = 15.05, p = 0.000), indicating that the alpha response was largest over occipital locations.
3. Results 3.1. Delta frequency window Maximum peak-to-peak delta responses were analyzed in the 0–600 ms time window after stimulation. The largest gender differences were observed at the parietal and occipital locations. From the values of Table 1 one can recognize that in parietal and occipital locations the peak-to-peak amplitudes can be up to 50% larger for females than for males. The ANOVA on delta responses revealed a significant effect for location (F = 740, p = 0.000) indicating an increased delta response over posterior recording sites, and for gender (F = 6.76), p = 0.014), indicating a larger delta response for women than for men (see Fig. 1A). Post-hoc comparisons using the Wilcoxon paired sample test revealed that the peak-to-peak delta response was significantly larger for women than for men over the electrode sites Cz, T5, P3, P4, O1 and O2 (p b 0.05 for all comparisons, see Table 1 and Fig. 1A for further information). 3.2. Theta frequency window Maximum peak-to-peak theta responses were analyzed in the 0–500 ms time window after stimulation. In the theta frequency range no significant differences between men and women were found. The ANOVA on theta responses revealed a significant effect for location (F = 10.71, p = 0.000), indicating that the theta response was larger for frontal, central and parietal recording sites. 3.3. Alpha frequency window In the alpha frequency range, the maximum peak-to-peak response was analyzed in the 0–400 ms time window. No significant differences between men and women were found. We also have to mention that there is a tendency that male subjects to show higher amplitude in T6 and O2 locations (Fig. 1A), however this tendency is not significant. The ANOVA on alpha
Table 1 For female and male subjects the mean, standard deviations and p values of the maximum peak-to-peak amplitude values are shown for the delta, beta and gamma frequency ranges at different locations Location– frequency
Female Mean value μv
Standard deviation
Male Mean value μv
Standard deviation
Cz Delta T5 Delta P3 Delta P4 Delta O1 Delta O2 Delta O1 Beta O2 Beta O2 Gamma
6.1936 3.8313 6.3035 5.5824 6.2502 6.6248 2.8032 2.7716 1.397
±2.6156 ±1.1303 ±3.2922 ±2.3954 ±2.2661 ±2.2345 ±1.5162 ±0.9632 ±0.8319
4.4631 2.9454 3.8094 4.2996 3.7551 4.5105 1.7943 2.1288 0.9484
±1.7111 ±1.4977 ±1.8960 ±2.2202 ±2.3748 ±3.1163 ±0.6089 ±0.5588 ±0.3585
p value
0.006 0.015 0.002 0.044 0.003 0.034 0.013 0.007 0.03
3.4. Beta frequency window Maximum peak-to-peak beta responses were analyzed in the 0–300 ms time window. The ANOVA on beta responses revealed a significant effect for location (F = 9.23, p = 0.000) indicating that the beta response over occipital sites was increased. ANOVA revealed no significant results for location × gender (F = 1.60, p = 0.14) or for gender (F = 2.09, p = 0.15). The Wilcoxon paired sample test showed that the peak-to-peak beta response amplitudes of the female subjects were significantly higher than the peak-to-peak beta response amplitudes of the male subjects at the O1 and O2 locations (O1, p = 0.013; O2, p = 0.007) (Table 1). 3.5. Gamma frequency window Maximum peak-to-peak gamma responses were analyzed in the 0–300 ms time window. The ANOVA on gamma responses revealed a significant effect for location (F = 7.14, p = 0.000), indicating that the gamma response over occipital sites was increased. ANOVA revealed no significant results for location × gender (F = 1.80), p = 0.12) or for gender (F = 0.85, p = 0.36). The Wilcoxon paired sample test showed that peakto-peak gamma response amplitudes of the female subjects were significantly higher than the peak-to-peak gamma response amplitudes of the male subjects at O2 (p = 0.03) (Table 1). Our use of the Wilcoxon test, despite non-significant results on ANOVA, could be criticized. There are different opinions about this issue from the viewpoint of statistical procedures. We performed our ANOVA analysis using a large number of electrode locations, reducing the chance of obtaining overall significant differences. In this context, we find it useful to add the information from a significant Wilcoxon test despite the non-significant ANOVA analysis. We think that the description of these results merits consideration. Fig. 1A presents all significant results on a topological map. The delta response was larger in women than in men over the electrode sites Cz,T5, P3, and P4. The increase of the delta response in women was 37% over Cz, 29% over P4, and 65% over P3. The increase of the delta response was even more pronounced over occipital recording sites (66% for O1 and 46% for O2). The occipital beta and gamma responses were also larger in women than in men. Globally, the dominant differentiations between women and men occur at the posterior sites in the delta frequency range. There are no significant differences at the frontal locations or in the theta frequency range (see Results). The right occipital site seems to play a special role, as we find gender differences in three frequency ranges; this means that the right occipital location depicts the most important differentiation between genders in visual processing of simple light stimulation. In addition to the histograms of Fig. 1A we add an extending illustration showing the grand averages in the delta,
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299
297
Fig. 1. A: Histograms of the mean of the averages in delta (0.5–3.5 Hz), beta (15–24 Hz), gamma (28–48) Hz and alpha (9–13 Hz) frequency range. “⁎” represents significant results. B: Grand averages of oscillatory responses of female and male subjects in delta, beta, gamma and alpha. Female subjects are represented by black lines and male subjects are represented by gray lines.
beta, gamma and alpha frequency windows (Fig. 1B). The differences in peak-to-peak amplitudes are clearly seen in this illustration, which supports the statistical information of Fig. 1A.
4. Discussion Only few studies compare ERPs and EROs between male and female subjects. The common finding of already published
298
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299
studies is the higher amplitude of female evoked potentials in different latency ranges. Orozco and Ehlers (1998) reported longer latency and higher amplitude in the P450 component to facial expressions. During a recognition memory task Guillem and Mograss (2005) recorded higher amplitudes of ERPs in female subjects. According to Hoffman and Polich (1999) male subjects produce smaller P200, P300 and N200 components during a classical oddball paradigm. Proverbio et al. (2006) published results on infant happy/distressed expression indicating much larger occipital P110 response in women compared to men. Karakaş et al. (2006) used auditory stimulation and found that the differences in gamma band are independent of gender. We have to note that none of these studies considered a simple physiological stimulation; they have been focused rather on facial expression, also reflecting emotional behavior, attention and working memory. According to Fuster (1995) the phyletic memory is constituted by the structure of primary sensory and motor cortex at birth, which is common to all organisms of the same species. Phyletic memory would be the most basic of all memories, the genetic memory that the organism has formed in the course of evolution by interactions with the surrounding world. The genetically defined structures of primary cortex dedicated to analyzing elementary sensory features and to integrating elementary primitives of movement would form the basic template on which the memory of the individual would develop. In the light of Fuster's definition Başar (2004) emphasized that even “simple light” triggers sensory memory (iconic memory) and that all experiments related to sensorycognitive processing are interrelated or interwoven with the sensory memory. This is a simple physiological consideration, which includes the necessity to use EPs as control signal. Accordingly, we have to filter out effects of simple light causation in analyzing responses to facial expressions, to attentive processes and also to various types of cognitive loads. In consideration of these facts, the most basic question in our discussion is: “Does simple light evoke differentiations in EROs in comparing female and male subjects? If responses to simple light are anchored with some basic physiological mechanisms, we have to use these results in order to be able to discuss perceptual processes such as face recognition or recognition of facial expressions and/or emotions in a more refined manner. The results of this study also indicate that there are very important differences in the delta band and smaller differences in the beta and gamma frequency ranges. The histogram results presented in Fig. 1A merit important attention. Firstly, we have to note that no differentiations are observed at frontal locations or in the theta frequency range. According to Klimesch et al. (2000) and Başar et al. (1999) a large number of cognitive functions occur in the theta frequency range and at frontal locations. Since we used here a simple light as stimulation, and the subjects did not perform any cognitive tasks, the absence of differentiation in the frontal theta is not surprising. Two other aspects should be mentioned also here. In the vast literature on brain oscillations, the role of gamma oscillations has been mentioned as one of the most important functional key oscillations. In the present study, the gamma response depicts
only a differentiating role at the right occipital site. According to our results, the right occipital location is the area which is the most active indicating the differentiation between males and female subjects. 5. Conclusion Based also on the results of our previous publications related to oscillatory brain dynamics, we come to the following conclusions: (1) The brain response is a construct in a “multidimensional state” incorporating amplitudes of oscillatory responses, topological coordinates, and changes in the time axis following presentation of the percepts, including delays and prolongations, and coherence between locations. Only a new Cartesian System embracing all these parameters can represent the dynamics of functionality in the brain. (2) It is also important to note that in order to analyze and interpret EROs it seems to be extremely important to consider several causal factors (Başar and Güntekin, 2007). Among these, gender differences should be considered as a fundamental causal factor in the search for functional correlates of EROs. Acknowledgements The authors are thankful to Dr. Birgit Mathes for helpful comments on the manuscript. References Başar, E., 1980. EEG-Brain Dynamics. Relation Between EEG and Brain Evoked Potentials. Elsevier, Amsterdam. Başar, E., 1999. Brain Function and Oscillations: II. Integrative Brain Function. Neurophysiology and Cognitive Processes. Springer-Verlag, Heidelberg. Başar, E., 2004. Memory and Brain Dynamics: Oscillations Integrating Attention, Perception, Learning and Memory. CRC Press, Florida. Başar, E., Gönder, A., Özesmi, C., Ungan, P., 1975. Dynamics of brain rhythmic and evoked potentials. II. Studies in the auditory pathway, reticular formation, and hippocampus during the waking stage. Biol. Cybern. 5, 145–160. Başar, E., Başar-Eroglu, C., Karakaş, S., Schurmann, M., 1999. Are cognitive processes manifested in event-related gamma, alpha, theta and delta oscillations in the EEG? Neurosci. Lett. 15 (259(3)), 165–168. Başar, E., Güntekin, B., 2007. A breakthrough in neuroscience needs a “Nebulous Cartesian System" Oscillations, quantum dynamics and chaos in the brain and vegetative system. Int. J. Psychophysiol. 64 (1) (April), 108–122. Begleiter, H., Porjesz, B., 2006. Genetics of human brain oscillations. Int. J. Psychophysiol. 60 (2) (May), 162–171. Freeman, W.J., 1975. Mass Action in the Nervous System. Academic Press, New York. Fuster, J.M., 1995. Memory in the Cerebral Cortex. A Bradford Book. The MIT Press, Cambridge. Guillem, F., Mograss, M., 2005. Gender differences in memory processing: evidence from event-related potentials to faces. Brain Cogn. 57, 84–92. Güntekin, B., Başar, E., 2007. Emotional face expressions are differentiated with brain oscillations. Int. J. Psychophysiol. 64 (1) (Apr), 91–100. Hoffman, L.D., Polich, J., 1999. P300, handedness, and corpus callosal size: gender, modality, and task. Int. J. Psychophysiol. 31, 163–174. Jasper, H.H., 1958. The ten-twenty electrode system of the International Federation. Electroencephalogr Clin. Neurophysiol. 10, 371–375. Karakaş, S., Tüfekçi, I., Bekçi, B., Çakmak, E.D., Doğutepe, E., Erzengin, O.U., Özkan, A., Arıkan, O., 2006. Early time-locked gamma response and gender specificity. Int. J. Psychophysiol. 60, 225–239.
B. Güntekin, E. Başar / International Journal of Psychophysiology 65 (2007) 294–299 Klimesch, W., Doppelmayr, M., Schwaiger, J., Winkler, T., Gruber, W., 2000. Theta oscillations and the ERP old/new effect: independent phenomena? Clin. Neurophysiol. 111 (5), 781–793. Mountcastle, V.B., 1992. Preface. In: Başar, E., Bullock, T.H. (Eds.), Induced Rhythms in the Brain. Birkhäuser, Boston, MA, pp. XVII–XIX. Orozco, S., Ehlers, C.L., 1998. Gender differences in electrophysiological responses to facial stimuli. Biol. Psychiatry 15 (4), 281–289. Rangaswamy, M., Jones, K.A., Porjesz, B., Chorlian, D.B., Padmanabhapillai, A., Kamarajan, C., Kuperman, S., Rohrbaugh, J., O'connor, S.J., Bauer, L.O.,
299
Schuckit, M.A., Begleiter, H., 2007. Delta and theta oscillations as risk markers in adolescent offspring of alcoholics. Int. J. Psychophysiol. 63 (1) , 3–15 Jan. Proverbio, A.M., Brignone, V., Matarazzo, S., Zotto, M.D., Zani, A., 2006. Gender and parental status affect the visual cortical response to infant facial expression. Neuropsychologia 44, 2987–2999. Yordanova, J., Kolev, V., 1998. Single-sweep analysis of the theta frequency band during an auditory oddball task. Psychophysiology 35 (1), 116–126.