www.elsevier.com/locate/ijpsycho

International Journal of Psychop

Beta oscillations in face recognition

Murat Ozgorena,b, Canan BaYar-Eroglub,c, Erol BaYara,b,*

aBrain Dynamics Multidisciplinary Research Center and Faculty of Medicine, Department of Biophysics, Dokuz Eylul University,

Izmir 35340, TurkeybTUBITAK Brain Dynamics Multidisciplinary Research Network, Ankara, Turkey

cBremen, Institute of Psychology and Cognition Research and Center for Cognitive Sciences, University of Bremen,

P.O. Box 330440, Bremen D-28334, Germany

Received 4 May 2004; received in revised form 3 June 2004; accepted 8 June 2004

Available online 5 August 2004

Abstract

This report presents an analysis of the brain’s beta oscillations in face recognition. We performed experiments on 26 subjects

with a strategy consisting of two types of stimulations: (1) the picture of an elder anonymous lady (unknown face) and (2) the

picture of the subject’s own grandmother (known face). The subjects were healthy, young people between the ages of 15–32

years. Data were analyzed by means of amplitude frequency characteristics and digital filtering. Our results show the significant

role of beta response in face recognition and the differentiation of known and unknown faces. Furthermore, this report supports

our former view that the presentation of grandmother face evokes selectively distributed multiple oscillations in the brain.

Together with the scope of other frequencies (e.g., delta, theta, and alpha), this method can serve as a tool for research studies or

clinical studies in memory and cognition.

D 2004 Elsevier B.V. All rights reserved.

Keywords: Beta; Delta; Alpha; Theta; Oscillations; ERP; Face recognition; Semantic; Episodic; Memory; Laterality; Prolongation

1. Introduction

The aim of the present study is to investigate the

role of beta oscillations in face recognition. In our

long-standing work, we have empirically shown that

the integrative brain functions are manifested with the

0167-8760/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.ijpsycho.2004.06.005

* Corresponding author. Brain Dynamics Multidisciplinary

Research Center and Faculty of Medicine, Department of Bio-

physics, Dokuz Eylul University, Izmir 35340, Turkey. Tel.: +90

232 4124480; fax: +90 232 27291888.

E-mail address: erol.basar@deu.edu.tr (E. BaYar).

superposition of multiple oscillations. In order to give

further experimental support to this working hypoth-

esis, we recently have started measurements with

facial stimuli and described that delta, theta, and alpha

responses of the brain have clear differentiation upon

presentation of known and unknown faces (BaYar et

al., 2001, 2004; BaYar, 2004).

According to some authors, face selective activity

takes place in fusiform gyrus, which has also been

known as face fusiform area (Klopp et al., 1999;

Kanwisher et al., 1998). However, a number of studies

suggest that not only the occipito-temporal areas but

hysiology 55 (2005) 51–59

M. Ozgoren et al. / International Journal of Psychophysiology 55 (2005) 51–5952

also other areas in the brain are activated during face

recognition (Begleiter et al., 1995; Schweinberger et

al., 2002; Eimer, 2000).

The brain’s beta oscillatory responses remained in

the shadow of the analyses in the gamma band. These

oscillations, however, merit important consideration.

As the present study clearly shows, the distributed

beta oscillatory responses play an essential role in face

recognition and the differentiation of known and

unknown faces.

2. Materials and methods

2.1. Experimental strategy

We used a strategy consisting of application of

three different types of stimulation:

1) A simple light stimulation as control signal: its

luminance was approximately at the same level of

stimulations 2 and 3 described in the following

(approximately 30 cd/m2).

2) The picture of an unknown face (an anonymous

elder lady)

3) The picture of the dknown faceT: the subject’s

own grandmother.

A total of 26 subjects in the age range of 15–36

years (17 females and 9 males) participated in the

study. They had normal or corrected-to-normal

binocular visual acuity and were right-handed. The

pictures of the faces of the subjects’ own grandmother

were prepared prior to measurements. The pictures

were black and white (17�17 cm) and were displayed

on a screen located at a distance of 120 cm from the

subjects. The stimulus duration was set to 1000 ms

with intervals varying between 3.5 and 7.5 s. The

subjects, who were instructed to minimize blinking

and eye movements, sat in a sound-proof and dimly

illuminated echo-free room.

2.1.1. Data recording set

All three stimulations described above have been

applied in a random sequence. Seventy-five stimula-

tion signals have been applied with approximately the

same distribution of each type of stimulation. Follow-

ing the recordings, grandmother’s (known) face,

unknown face, and light responses have been divided

into subsets. In all experiments, subjects reported that

they clearly recognized and differentiated the face of

their own grandmother.

2.1.2. Electrophysiological recording

The electroencephalogram (EEG) was recorded

from F3, F4, Cz, C3, C4, P3, P4, O1, and O2

locations according to the 10–20 system (Jasper,

1958). For the recordings, an EEG-CAP was used.

For the ears, electromyogram (EMG) and electro-

olfactogram (EOG) recordings with Ag/AgCl electro-

des were used. Linked earlobe electrodes (A1+A2)

served as reference. The EOG medial upper and

lateral orbital rim of the right eye was also

registered. The EEG was amplified by means of a

Nihon Kohden EEG-4421 G apparatus with band

limits 0.1–100 Hz, 24 dB/octave. 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 prestimulus baseline and was stored on

the hard disc of the computer.

2.2. Data Analysis

Before the averaging procedure, the epochs con-

taining artifacts were rejected by an off-line technique.

The frequency responses of the system were deter-

mined using the amplitude frequency characteristics

(AFC) method (for a more detailed explanation of

AFC method, please see BaYar et al., 2001; KarakaY et

al., 2000).

According to the general systems theory, all

information about the frequency characteristics of a

linear system is contained in the transient response of

the system and vice versa. In other words, knowledge

of the transient response of the system allows one to

predict how the system will react to different

stimulation frequencies if the stimulating (input)

signal is sinusoidally modulated. If the step response

c(t) of the system—in our case, the ERP—is known,

the frequency characteristics G(jx) of this system can

be obtained with a Laplace (or one-sided Fourier)

transform of the following form:

G jxð Þ ¼Zl

0

exp � jxtð Þd c tð Þgf

Fig. 1. AFC as grand averages (N=26). Solid line: Grandmothe

(known) face responses at F4. Dashed line: Unknown face responses

at F4. Along the x-axis: frequency is in logarithmic scale. Along the

y-axis: relative amplitude in decibels. The amplitudes are normalized

in such a way that the amplitude at 1 Hz is equal to 0. Vertical dotted

lines indicate the beta range (15–30 Hz).

M. Ozgoren et al. / International Journal of Psychophysiology 55 (2005) 51–59 53

where G (jx)=frequency characteristics of the system;

c(t)=step response of the system; x=2Sf=angularfrequency, and f=frequency of the input signal.

The frequency characteristics, including the infor-

mation on amplitude changes of forced oscillations

and the phase angle between the output and the input,

are also called the frequency response function. It is a

special case of transfer function and is, in practice,

identical with the transfer function.

For numerical evaluation, a fast Fourier transform

(FFT) is used. Let Xn be a discrete time series

Xn=X(nDt), T=((N�1)Dt). Then the Fourier transform

Yk of Xn is:

Yk ¼ Y ðxkÞ ¼XN�1

n¼0

Xnexpð � i2pN�1nkÞ;

xk ¼ 2pkT�1

where Yk=ak+ibk are the complex Fourier coeffi-

cients, the geometric mean of which is the amplitude

spectrum.

The AFC is expressed in relative units and reflects

the amplification in the studied frequency channels.

The presence of a peak in the AFC thus reveals the

frequency selectivities of the system, and these are

interpreted as its most preferred oscillations when

responding to stimuli. The AFC method has also the

advantage of showing the status of all frequencies in a

combined manner.

2.2.1. Algorithm for evaluation of prolongation of

oscillations

We performed an analysis to evaluate the magni-

tudes and prolongations of the beta responses: the

maximum peak-to-peak (pp) amplitude and the

latency (lat) of the beta responses were measured. In

order to assess the duration and the damping of these

responses, the time periods for reaching half (1/2 lat)

and third (1/3 lat) of the maximum pp amplitudes of

the responses have been measured. This measure

revealed the temporal characteristics of the oscilla-

tions, as the onset, duration, and damping of these

responses. The same procedures were repeated for all

sets and locations.

The Wilcoxon matched pair test was used as a

statistical procedure in order to constitute the dis-

tinction between responses to grandmother and

unknown faces.

3. Results

3.1. Beta responsiveness in the frequency and time

domains

Fig. 1 shows the grand average AFC obtained from

responses to grandmother and unknown faces at F4. It

is noteworthy that the relative amplitude of beta

response to the unknown faces is higher in compar-

ison to grandmother faces. This AFC is also useful to

choose limits of filters for determination of beta

responses in the time domain. As illustrated in Fig. 1,

the frequency response to facial stimuli had more

prominent and distinct maxima in the beta range,

peaking around 22 Hz, in comparison to the gamma

range. Accordingly, we recommend the use of

frequency characteristics in order to determine opti-

mum filter limits (15–30 Hz) instead of using given

templates.

3.2. Beta responses from single subjects and grand

average responses

Beta (15–30 Hz) responses of individual subjects

to grandmother and unknown faces were recorded at

all electrode sites. These responses were similar to

the results obtained from grand averages of 26

r

M. Ozgoren et al. / International Journal of Psychophysiology 55 (2005) 51–5954

subjects. In regard to this similarity, Fig. 2 illustrates

beta responses to unknown face (upper curve) and

grandmother face (lower curve) of a single subject

(subject 16) and (B) grand average responses to

grandmother and unknown faces are illustrated. In

this illustration, the global linear envelopes of

oscillatory responses are also indicated. The prolon-

gation time PT is the time difference between the

decay of responses to grandmother face and the

anonymous face. Latencies evaluated firstly with the

maximal pp response and also with the minimum of

the dampening amplitude, as vertical dashed lines

indicate. Note the smaller amplitude and prolonged

character of the lower curves (beta responses to

Fig. 2. (A) F4 beta responses (15–30 Hz) to grandmother (known)

and unknown faces from a single subject. The asterisks point to the

pp response. (B) F4 grand average beta responses (15–30 Hz) to

grandmother (known) and unknown faces from 26 subjects.

PT=prolongation time. The response to grandmother (known) faces

is represented by a solid line, and response to unknown faces is

represented by a dashed line.

grandmother face). The responses to the unknown

face have larger amplitudes and are less prolonged.

Furthermore, the responses to grandmother faces

showed prolonged oscillations in comparison to the

responses to unknown faces.

3.2.1. Topography of beta responses

In Fig. 3, the topographical distribution of grand

average beta responses to grandmother and unknown

faces is shown. Please note that the responses to

grandmother faces are more prolonged than the

responses to unknown faces at the frontal and central

areas. The latencies of beta responses to grandmother

faces varied between 150 and 220 ms in different

recording sites. Similarly, the latencies of beta

responses to unknown faces varied between 130 and

190 ms in different recording sites.

3.2.2. Amplitudes of beta responses

Statistical analysis showed that amplitudes of beta

responses to unknown face stimuli were significantly

larger than the amplitudes of beta responses to

grandmother stimuli at F3, F4, Cz, and P3 recording

sites (Pb0.05). Accordingly, at the F3 recording site,

the amplitude of beta response to grandmother face

was 0.98 AV pp and the amplitude of beta response to

unknown face was 1.32 AV pp, which was signifi-

cantly larger (P=0.023).

At the F4 recording site, the amplitude of beta

response to unknown face (1.40 AV pp) was

significantly larger (P=0.035) than the amplitude of

beta response to grandmother face (0.98 AV pp). At

the Cz recording site, the amplitude of beta response

to unknown face (0.93 AV pp) was significantly larger

(P=0.020) than the amplitude of beta response to

grandmother face (0.63 AV pp). Also, at the P3

recording site, the amplitude of beta response to

unknown face (0.74 AV pp) was significantly larger

(P=0.023) than the amplitude of beta response to

grandmother face (0.36 AV pp).

According to the statistical analysis, the amplitudes

of beta responses to grandmother and unknown face

stimuli were not significantly different at the record-

ing sites of P4, O1, and O2.

3.2.3. Laterality at parietal electrodes

The beta response amplitudes were also compared

in between two hemispheres for each stimulation and

Fig. 3. Topography of the grand averages of filtered (15–30 Hz) EEG–ERPs to the known face (own grandmother) and unknown face

(anonymous) stimuli. Beta response to grandmother faces is represented by a solid line, and beta response to unknown faces is represented by a

dashed line. Stimulation applied at d0 msT time point.

M. Ozgoren et al. / International Journal of Psychophysiology 55 (2005) 51–59 55

location. This analysis was significant only for the

beta response to grandmother stimuli at parietal

locations: there was significant difference between

the amplitudes of beta responses to grandmother face

at P3 and P4 recording sites (P=0.017).

3.2.4. Prolongation of beta responses

An important feature of the results is the significant

differentiation of beta responses along time axis. In

Fig. 4, the prolongations of beta responses in different

recording sites are indicated. As this figure shows, the

Fig. 4. The time periods for reaching a third of the maximum

amplitude values of grand average filtered beta (15–30 Hz)

responses to different face stimuli. Beta responses to grandmother

(known) faces are represented by black bars, and beta responses to

unknown faces are represented by gray bars.

M. Ozgoren et al. / International Journal of Psychophysiology 55 (2005) 51–5956

responses to grandmother faces are more prolonged in

comparison to responses to unknown faces at F3, F4,

and Cz.

Accordingly, at the F3 recording site, the time

period for reaching a third of its pp amplitude (1/3

lat) of beta response to unknown face was 255 ms

and to grandmother face was 400 ms. Thus, the beta

response to grandmother face was significantly more

prolonged than the beta response to unknown face

(P=0.004). Similarly, at F4, the beta response to

unknown face was prolonged to 275 ms and the beta

response to grandmother face was prolonged to 400

ms. At this recording site, there was significant

prolongation in favor of beta response to grand-

mother face (P=0.006). The largest prolongation

(575 ms) was of the beta response to grandmother

face at Cz, which was also significantly more

prolonged than the beta response (260 ms) to the

unknown face (P=0.028).

The other recording sites did not show significant

changes of prolongations of beta responses to grand-

mother and unknown faces.

4. Discussion

4.1. Selectively distributed beta responses

Several research groups have reported differ-

entiated ERP components during face presentation

localized at temporal areas (Botzel et al., 1989;

Bentin and Golland, 2002; Eimer, 2000; Zhang et al.,

2001). Only a few papers are related to the frontal

regions during face recognition. Begleiter et al.

(1995) showed that for recognition of familiar faces,

both the temporal and frontal regions are involved.

Schweinberger et al. (2002) also reported that only

for the familiar faces, responses were recorded from

parietal, central, and prefrontal areas.

Our results showed that the beta responses to both

grandmother faces and unknown faces were widely

distributed at all recording sites. At F3, F4, Cz, and P4

areas, the maximal amplitudes of beta responses to

grandmother faces were smaller than the maximal

amplitudes of responses to unknown faces. However,

at P4 and O2, the beta responses to grandmother faces

were larger than unknown face responses. The

selective distribution of beta responses in all recording

sites is suggestive that not only temporo-parietal

regions but also the cortex are involved during the

processing of faces.

4.2. Beta responses in time domain

A major part of the papers indicated the

presence of face-specific responses in the range

of 120–210 ms (Botzel et al., 1989; Linkenkaer-

Hansen et al., 1998; Bentin and Golland, 2002). In

a recent study, potentials specific to familiar faces

extending to 400–600 ms were recorded as the

largest at midline electrodes (Eimer, 2000). It was

suggested that those results might reflect the

activation of stored representations of familiar

faces.

Our data show that the beta responses to known

faces vary between 150 and 220 ms and beta

responses to unknown faces are between 130 and

190 ms. These results are in agreement with earlier

findings reporting face specific activity between

120 and 210 ms (Botzel et al., 1989; Linkenkaer-

Hansen et al., 1998). Beta responses to grand-

mother faces had prolonged oscillations in compar-

ison to beta responses to unknown faces at F3, F4,

and Cz. Furthermore, beta response prolongations

(extending to 575 ms) do occur with the same

latency of the late evoked potential components

(400–600 ms) that are specific to familiar faces

(Eimer, 2000).

M. Ozgoren et al. / International Journal of Psychophysiology 55 (2005) 51–59 57

4.3. Laterality of responses

In a study of interhemispheric cooperation, Mohr et

al. (2002) obtained a highly significant familiarity by

visual field interaction showing that, only for familiar

faces, a bilateral advantage was obtained. Schutter et

al. (2001) reported parietal asymmetry in beta

frequency.

Our results indicate that the amplitudes of

responses to grandmother faces (but not to unknown

faces) show significant differences only between P3

and P4 recordings. Our results are in accordance with

studies of Schutter et al. (2001) who showed

asymmetries in the beta range.

4.4. Multiple functions of beta oscillations

Various functions have been assigned to beta such

as visual attention (Marrufo et al., 2001; Wrobel,

2000), movement-related changes (Muller et al.,

2003), excitation–inhibition (Whittington et al.,

2000), sensory memory (Haenschel et al., 2000),

cognitive tasks (Ray and Cole, 1985), and recollection

and familiarity (Burgess and Ali, 2002). Recently,

linkage and linkage disequilibrium between beta

frequency and a GABAA receptor gene have been

reported (Porjesz et al., 2002). Rangaswamy et al.

(2002) reported enhanced beta power in the resting

EEG of alcoholics.

4.5. Multiple oscillations in face recognition

Polich and Kok (1995) emphasized the factors

that contribute to neuroelectric activity related to

cognitive processes besides attention allocation and

immediate memory. In a study of cognitive process-

ing, Klopp et al. (1999) analyzed face-selective

spectral changes of ERPs. They concluded that

phase locking of event-related spectral power took

place not within a narrow frequency band but across

a broadband of frequencies from 5 to 45 Hz, thus

including the beta range and also the other major

frequency bands.

In most of the face-related reports, only the

amplitudes and latencies of conventional evoked

potentials were mentioned. The sensory and cogni-

tive processes related to facial stimuli of the brain

would need to be more complicated. The whole

brain and all oscillations are activated during

recognition or remembering of one’s own grand-

mother and an anonymous face. The ensemble of

responses behaves like a three-dimensional con-

struct, consisting of temporal, spatial, and frequency

spaces (BaYar, 2004). The responses to an anony-

mous face or to a grandmother face are not

represented solely by one location and a unique

frequency or at the same position along the

temporal axis, as differentiated delay and prolonga-

tion of multiple oscillations that are selectively

distributed on the whole cortex show (BaYar et al.,

2004).

Although beta responses have larger amplitudes

upon presentation of unknown faces, there is more

prolonged activity in the beta range upon presentation

of grandmother faces. On the contrary, theta responses

are larger and less prolonged to grandmother faces in

comparison to theta responses to the unknown faces

(BaYar, 2004).

Klimesch et al. (1994) and Sauseng et al. (2002)

showed the possibility of differentiating the role of

alpha and theta oscillatory activities during memory

tasks. Alpha activity has been also correlated with

phyletic memory, as well as episodic and semantic

memories, and selectively distributed over several

cortical areas (BaYar, 1999, 2004).

In the delta frequency range, significant amplitude

increases were observed in posterior areas to

responses upon grandmother and unknown face

stimuli, but not in frontal areas (BaYar et al., 2004).

The differentiation of the two types of facial stimuli is

more distinct in beta responses than alpha and delta

responses especially at the frontal and central areas,

thus showing the important role of beta oscillations in

face recognition.

5. Conclusion

(1) During activation of episodic memory (grand-

mother picture) and semantic memory (anony-

mous face), the topologic distribution of beta

responses is significantly differentiated by show-

ing varied degrees of amplitude enhancements

and prolongations.

(2) These results support and extend the concept that

integrative brain functions are based on multiple

M. Ozgoren et al. / International Journal of Psychophysiology 55 (2005) 51–5958

oscillations selectively distributed in the whole

brain.

(3) During face recognition, which is also involved

in memory differentiation, a parallel processing

in the beta frequency range takes place. Later-

ality, prolongation of oscillations, and topolog-

ical selectivity belong to the manifestations of

this parallel processing.

(4) Besides the parallel processing in beta frequency

range, the superposition of oscillations in delta,

theta, and alpha frequency windows indicates

that the recognition of faces is manifested with a

multifold of oscillatory activities including

prolongation of oscillations.

(5) This high degree of differentiation in recognition

of faces, which is reached by the present

analysis, cannot be realized by conventional

ERP analysis.

Acknowledgements

The experiments of this collaboration study were

supported by TUBITAK-DFG grant 446 TUR 112/14/

01, DEU 02 KB.SAG.012/ 027-03.KB.SAG.064.

References

BaYar, E., 1999. Brain Function and Oscillations: II. Integrative

Brain Function. Neurophysiology and Cognitive Processes.

Springer, Berlin, pp. 1–474.

BaYar, E., 2004. Memory and Brain Dynamics: Oscillations

Integrating Attention, Perception, Learning and Memory. CRC

Press LLC, pp. 153–168.

BaYar, E., Schurmann, M., Demiralp, T., BaYar-Eroglu, C.,

Ademoglu, A., 2001. Event-related oscillations are dreal brainresponsesT—wavelet analysis and new strategies. Int. J. Psy-

chophysiol. 39, 91–127.

BaYar, E., Ozgoren, M., KarakaY, S., BaYar-Eroglu, C., 2004. Super-

synergy in the brain: grandmother percept is manifested by

multiple oscillations. Int. J. Bifurcation and Chaos 14, 1–38.

Begleiter, H., Porjesz, B., Wang, W., 1995. Event-related brain

potentials differentiate priming and recognition to familiar and

unfamiliar faces. Electroencephalogr. Clin. Neurophysiol. 94,

41–49.

Bentin, S., Golland, Y., 2002. Meaningful processing of mean-

ingless stimuli: the influence of perceptual experience on early

visual processing of faces. Cognition 86, B1–B14.

Botzel, K., Grusser, O.J., Haussler, B., Nauman, A., 1989. The

search for face-specific evoked potentials. In: BaYar, E.,

Bullock, T.H. (Eds.), Brain Dynamics. Springer-Verlag, Berlin,

pp. 449–468.

Burgess, A.P., Ali, L., 2002. Functional connectivity of gamma

EEG activity is modulated at low frequency during conscious

recollection. Int. J. Psychophysiol. 46, 91–100.

Eimer, M., 2000. Event-related brain potentials distinguish process-

ing stages involved in face perception and recognition. Clin.

Neurophysiol. 111, 694–705.

Haenschel, C., Baldeweg, T., Croft, R.J., Whittington, M.,

Gruzelier, J., 2000. Gamma and beta frequency oscillations in

response to novel auditory stimuli: a comparison of human

electroencephalogram (EEG) data with in vitro models. Proc.

Natl. Acad. Sci. U. S. A. 20, 97 (13), 7645–7650.

Jasper, H.H., 1958. The ten-twenty electrode system of the

International Federation. Electroencephalogr. Clin. Neurophy-

siol. 10, 371–375.

Kanwisher, N., Tong, F., Nakayama, K., 1998. The effect of face

inversion on the human fusiform face area. Cognition 68,

B1–B11.

KarakaY, S., Erzengin, O.U., BaYar, E., 2000. A new strategy

involving multiple cognitive paradigms demonstrates that ERP

components are determined by the superposition of oscillatory

responses. Clin. Neurophysiol. 111, 1719–1732.

Klimesch, W., Schimke, H., Schwaiger, J., 1994. Episodic and

semantic memory: an analysis in the EEG theta and alpha band.

Electroencephalogr. Clin. Neurophysiol. 91, 428–441.

Klopp, J., Halgren, E., Marinkovic, K., Nenov, V., 1999. Face-

selective spectral changes in the human fusiform gyrus. Clin.

Neurophysiol. 110, 676–682.

Linkenkaer-Hansen, K., Palva, J.M., Sams, M., Hietanen, J.K.,

Aronen, H.J., Ilmoniemi, R.J., 1998. Face-selective processing

in human extrastriate cortex around 120 ms after stimulus onset

revealed by magneto and electroencephalography. Neurosci.

Lett. 253, 147–150.

Marrufo, M.V., Vaquero, E., Cardoso, M.J., Gomez, C.M., 2001.

Temporal evolution of a and h bands during visual spatial

attention. Cogn. Brain Res. 12, 315–320.

Mohr, B., Landgrebe, A., Schweinberger, S.R., 2002. Interhemi-

spheric cooperation for familiar but not unfamiliar face

processing. Neuropsychologia 40, 1841–1848.

Muller, G.R., Neuper, C., Rupp, R., Keinrath, C., Gerner, H.J.,

Pfurtscheller, G., 2003. Event-related beta EEG changes during

wrist movements induced by functional electrical stimulation of

forearm muscles in man. Neurosci. Lett. 340, 143–147.

Polich, J., Kok, A., 1995. Cognitive and biological determinants of

P300: an integrative review. Biol. Psychol. 41, 103–146.

Porjesz, B., Almasy, L., Edenberg, H.J., Wang, K., Chorlian, D.B.,

Foroud, T., et al., 2002. Linkage disequilibrium between the

beta frequency of the human EEG and a GABAA receptor gene

locus. Proc. Natl. Acad. Sci. 99, 3729–3733.

Rangaswamy, M., Porjesz, B., Chorlian, D.B., Wang, K., Jones,

K.A., Bauer, L.O., Rohrbaugh, J., O’Connor, S.J., Kuperman,

S., Reich, T., Begleiter, H., 2002. Beta power in the EEG of

alcoholics. Biol. Psychiatry 51, 831–842.

Ray, W.J., Cole, H.W., 1985. EEG alpha activity reflects attentional

demands, and beta activity reflects emotional and cognitive

processes. Science 228, 750–752.

M. Ozgoren et al. / International Journal of Psychophysiology 55 (2005) 51–59 59

Sauseng, P., Klimesch, W., Gruber, W., Doppelmayr, M., Stadler,

W., Schabus, M., 2002. The interplay between theta and alpha

oscillations in the human electroencephalogram reflects the

transfer of information between memory systems. Neurosci.

Lett. 324, 121–124.

Schutter, D.J.L.G., Putman, P., Hermans, E., van Honk, J., 2001.

Parietal electroencephalogram beta asymmetry and selective

attention to angry facial expressions in healthy human subjects.

Neurosci. Lett. 314, 13–16.

Schweinberger, S.R., Esther, C., Jentzsch, I., Burton, A.M.,

Kaufmann, J.M., 2002. Event-related brain potential evidence

for a response of inferior temporal cortex to familiar face

repetitions. Cogn. Brain Res. 14, 398–409.

Whittington, M.A., Traub, R.D., Kopell, N., Ermentrout, B., Buhl,

E.H., 2000. Inhibition-based rhythms: experimental and math-

ematical observations on network dynamics. Int. J. Psychophy-

siol. 38 (3), 315–336.

Wrobel, A., 2000. Beta activity: a carrier for visual attention. Acta

Neurobiol. Exp. (Wars) 60 (2), 247–260.

Zhang, Y., Wang, Y., Wang, H., Cui, L., Tian, S., Wang, D., 2001.

Different processes are involved in human brain for shape and

face comparisons. Neurosci. Lett. 303, 157–160.