Abstract

The magnetic equivalent (MMNm) of mismatch negativity may reflect auditory discrimination and sensory memory. To study whether temporal lobe epilepsy (TLE) affects automatic central auditory-change processing, we recorded magnetoencephalographic (MEG) responses to standard and duration-deviant sounds in 12 TLE patients and 12 age-matched controls, and repeated MEG measurement in 8 patients 6–30 months following epilepsy surgery and in 6 controls 3–8 months after their first measurement. We compared the MMNm between patients and controls, and also evaluated intertrial phase coherences as indexed by phase-locking factors (PLF) using wavelet-based analyses. We observed longer MMNm latencies for patients than for controls. Dipole modeling and minimum-current estimates together showed bi-frontotemporal sources for MMNm. The phase locking across trials was dominant at the 4- to 14-Hz band, and the main difference in PLF between deviant- and standard-evoked responses occurred in the time frame of 150–250 ms after stimulus onset. Notably, in the 5 patients who became seizure free after removal of right temporal epileptic focus, the phase-locking phenomena resulting from deviant stimuli were enhanced, and even more distributed in the frontotemporal regions. We conclude that mesial TLE might affect auditory-change detection, and a successful surgery causes a possible plastic change in phase locking of deviant-evoked signals.

Introduction

Chronic temporal lobe epilepsy (TLE) is associated with memory impairment due to damage in the hippocampus and neighboring regions (Helmstaedter et al. 2003; Alessio et al. 2006). In patients with medically intractable TLE, successful surgical treatment offers not only satisfactory seizure control (Engel et al. 1993; Wiebe et al. 2001) but also has a beneficial effect on memory performance (Helmstaedter et al. 2003). Previous behavioral studies (Massaro 1972; Cowan 1984, 1995) have suggested that human short-term memory is an ensemble of sensory and working memory components. Recent studies have also shown that the hippocampus and adjacent mesial temporal structures play an important role in both long-term and short-term working memory functions (Squire and Zola-Morgan 1991; Squire 1992; Cabeza et al. 2002; Karlsgodt et al. 2005). However, it still remains uncertain whether TLE patients would exhibit any deficits in initial sensory memory encoding.

Mismatch negativity (MMN) (Naatanen et al. 1978) and its magnetic counterpart (MMNm) (Hari et al. 1984) are generated when the afferent input, conveyed by a deviant stimulus, is compared with a neural sensory memory trace that encodes the physical features of a repetitive standard stimulus (Naatanen 1992). This signal of cerebral reactivity as a response to infrequent auditory changes may be a reflection of sensory memory processing (Naatanen 1992; Alho 1995). MMN has therefore received considerable interest because of its potential application value in clinical research. For example, several studies have even reported the presence of attenuated MMN in some neurological diseases such as Parkinson's disease (Pekkonen et al. 1995) and Alzheimer's disease (Pekkonen et al. 1994). MMN may also be used to predict recovery from coma (Kane et al. 1996) and of stroke patients (Ilvonen et al. 2003). However, there is still a lack of research of MMN in epilepsy disorders. Therefore, we felt it of scientific interest to study whether the presence of an epileptic focus in the temporal lobe would affect MMN.

Furthermore, when based on the magnetoencephalographic (MEG) dipole analysis (Hari et al. 1984; Sams et al. 1985) and the results of intracranial recordings (Rosburg et al. 2005) of deviant-evoked responses in humans, previous studies have shown that the neuronal source for the MMN and MMNm is mainly located in the vicinity of the primary auditory cortex. Intracranial recordings in animal studies have suggested that the hippocampus may contribute to the generation of MMN (Vinogradova 1975; Csepe et al. 1987; Ruusuvirta et al. 1995). Other studies have also demonstrated that hippocampal lesions may cause impairment in the performance of auditory memory tasks (Luria and Karasseva 1968; Mukhamedrakhimov 1989; Chao and Knight 1995) or lead to a delay in neural reactivity to novel stimuli (Knight 1996). In contrast, however, Alain et al. (1998) reported no significant change in MMN amplitude in patients with hippocampal damage. Thus, the question remains open as to whether the mesial temporal structure is involved in sensory memory encoding in humans.

Clinically, patients with medically intractable TLE may obtain satisfactory seizure control after undergoing temporal lobe surgery (Engel et al. 1993; Wiebe et al. 2001). It is therefore worthwhile investigating the cerebral reactivity to infrequent auditory inputs in TLE patients, as they provide a good opportunity to clarify the role of the epileptic hippocampal structure in the formation of MMN in humans. In this study, we investigated the cerebral reactivity to deviant auditory inputs in patients with TLE before and after they underwent temporal lobe surgery. We used a whole-head neuromagnetometer to record the neuromagnetic responses of 12 adult TLE patients and 12 age-matched control subjects to standard sounds and duration-deviant sounds. We then repeated these measurements in 8 of the 12 TLE patients, 6–30 months after they had undergone temporal lobe surgery. We also repeated the same in 6 of the 12 control subjects 3–8 months following the first measurements.

The generation of event-related cerebral activations involves not only the evoked neuronal activity but also the phase synchronization of the ongoing cortical oscillations. Amplitude and phase modulation of these oscillations have been found to be able to operate together in the brain (Basar et al. 1984; David et al. 2005; Fuentemilla et al. 2006). Fuentemilla et al. (2006) have further demonstrated that it is perhaps only through the reorganization of the phase of the scalp electroencephalographic (EEG) activity, while it is ongoing, that certain event-related responses would arise. And considering the negative results in the study of Alain et al. (1998), which showed no remarkable change in MMN amplitude in patients with hippocampal lesions, we inferred that some neuronal plasticity information may be hidden in the averaged evoked signals. Thus, in the present study, we not only analyzed the averaged time-domain waveforms of mismatch responses by using equivalent current dipole (ECD) modeling and minimum-current estimation but also measured the intertrial phase-locking values of single epochs by employing wavelet-based analyses. Therefore, our study goals were 1) to see whether the MMNm in mesial TLE patients differ from those in healthy control subjects; and 2) to evaluate the plastic change of the cerebral reactivity to deviants in patients after they had undergone temporal lobe surgery, by analyzing both the MMNm signals and the phase-locking phenomenon across trials.

Materials and Methods

Participants

We studied, with their informed consent, 12 epileptic patients (5 women and 7 men; age 18–40 years, mean 30 years; all right-handed) and 12 healthy control subjects (5 women and 7 men; age 20–42 years, mean 29 years; all right-handed). None of the patients or the control subjects had hearing deficits.

Each patient had medically intractable TLE and was admitted for intensive presurgical workup, which included intensive video-scalp EEG monitoring, magnetic resonance (MR) imaging, positron emission tomography, interictal and postictal single-photon emission computed tomography, and neuropsychological assessment (Lin et al. 1997, 2003). Scalp EEG was recorded with gold-disc electrodes placed according to the International 10–20 System with the addition of bilateral cheek electrodes. Seizure documentation was based on the International League Against Epilepsy classification (Commission on Classification and Terminology of the International League Against Epilepsy 1989).

Table 1 summarizes the patients' clinical information. The duration of epilepsy ranged from 2 to 31 years (mean 13.3 years). All of them were taking 2 or more variations of the antiepileptic drugs carbamazepine, valproate, topiramate, lamotrigine, and phenytoin for seizure control. Eight of the 12 TLE patients underwent surgical intervention, and of these patients 6 had anterior temporal lobectomy and the other 2 had lesionectomy and marginal resection for cavernous hemangioma in one patient and for astrocytoma in the other patient. Also, the antiepileptic medications for all 8 of these patients were kept unchanged after surgery. Seizure outcome after epilepsy surgery was assessed according to Engel's classification (Engel et al. 1993). This study was approved by the local Ethics Committee of the Taipei Veterans General Hospital, Taipei, Taiwan.

Table 1

Clinical information of the 12 patients with mesial TLE

Patient number Sex Age (years) Seizure onset (years) Spike focus MRI Surgery/pathology Postsurgical follow-up Surgical outcome 
22 10 F7-Ch1 HS, L ATL/gliosis 18 months only auras 
38 F8-Ch2 HS, R ATL/gliosis 12 months SF 
27 20 F8-Ch2 HS, R ATL/gliosis 30 months SF 
40 17 F8-Ch2 HS, R ATL/gliosis 12 months SF 
32 11 F8-Ch2 HS, R ATL/gliosis 6 months SF 
25 17 F8-T4 negative ATL/gliosis 30 months 50% SR 
34 32 F7-Ch1 mesial temporal cavernoma, L LN + MR/cavernous hemangioma 6 months SF 
18 16 F8-Ch2 mesial temporal tumor, R LN + MR/astrocytoma 6 months SF 
29 17 F7-Ch1 HS, L ND   
10 26 21 F8-Ch2 HS, R ND   
11 36 13 F8-Ch2 HS, R ND   
12 29 15 F8-Ch2 mesial temporal cavernoma, R ND   
Patient number Sex Age (years) Seizure onset (years) Spike focus MRI Surgery/pathology Postsurgical follow-up Surgical outcome 
22 10 F7-Ch1 HS, L ATL/gliosis 18 months only auras 
38 F8-Ch2 HS, R ATL/gliosis 12 months SF 
27 20 F8-Ch2 HS, R ATL/gliosis 30 months SF 
40 17 F8-Ch2 HS, R ATL/gliosis 12 months SF 
32 11 F8-Ch2 HS, R ATL/gliosis 6 months SF 
25 17 F8-T4 negative ATL/gliosis 30 months 50% SR 
34 32 F7-Ch1 mesial temporal cavernoma, L LN + MR/cavernous hemangioma 6 months SF 
18 16 F8-Ch2 mesial temporal tumor, R LN + MR/astrocytoma 6 months SF 
29 17 F7-Ch1 HS, L ND   
10 26 21 F8-Ch2 HS, R ND   
11 36 13 F8-Ch2 HS, R ND   
12 29 15 F8-Ch2 mesial temporal cavernoma, R ND   

Note: Determination of spike focus was based on interictal and ictal scalp EEG recordings. M, male; F, female; Ch, cheek electrode; L, left; R, right; HS, hippocampus sclerosis; ATL, anterior temporal lobectomy; LN, lesionectomy; MR, marginal resection; ND, not done; SF, seizure free; SR, seizure reduction.

Stimuli

Auditory stimuli (70-dB sound pressure level) were delivered binaurally through plastic tubes and earpieces. Our oddball paradigm consisted of the standard (probability of 85%) and deviant pure tones (probability of 15%), given randomly with an interstimulus interval (from onset to onset) of 0.5 s. The standard stimuli were 1000 Hz in frequency, with 100 ms in duration including 10-ms rise and fall times. Each subject received the stimuli under 2 kinds of experimental conditions: with either duration deviants (50 ms in duration including 10-ms rise and fall times, 1000 Hz in frequency) or frequency deviants (1070 Hz, 100 ms in duration).

In order to direct the subject's attention away from the auditory stimulation (Naatanen et al. 1993), we instructed the subject to ignore the tones during the MEG recordings and to concentrate on watching a self-chosen movie that was being silently presented on a screen at a distance of about 1.2 m.

MEG Recordings

Auditory-evoked fields were measured while each subject was sitting in a magnetically shielded room with the head well supported by the helmet-shaped bottom of a whole-scalp neuromagnetometer (Vectorview, Elekta Neuromag, Helsinki, Finland). Our MEG device comprises 102 identical triple sensor elements. Each sensor element consists of 2 orthogonal planar gradiometers and one magnetometer. The planar gradiometers detect the largest signal just above the activated brain area (Hämäläinen et al. 1993). Details of the principles of MEG recordings have been described in previous studies (Hämäläinen et al. 1993; Lin et al. 2003).

We determined the exact location of the head with respect to the MEG sensors by measuring magnetic signals, produced by currents that were led to 4 head indicator coils at known sites on the scalp. The coil locations with respect to the anatomical landmarks on the head were then found with a 3D digitizer. This allows for further alignment of the MEG and MR imaging coordinate systems. The positive x-, y-, and z-axes in the head-coordinate system go toward the right preauricular point, the nasion, and the head vertex, respectively. The brain MR images of all subjects were acquired with a 3-T Bruker Medspec300 scanner (Germany).

The MEG epochs were 530 ms in duration, which included a 50-ms prestimulus baseline. We estimated the noise as the standard deviation (SD) of the values within the baseline. The epochs were averaged separately online for standard and deviant tone stimuli. The recording passband was 0.1–130 Hz and the data were digitized at 400 Hz. We discarded the responses that coincided with prominent vertical electrooculogram activity (>150 μV). One hundred accepted epochs were averaged for deviant stimuli, and more than 800 epochs for standard stimuli. We repeated the MEG measurement for each experimental condition to ensure the reliability of evoked responses.

For those patients who underwent epileptic surgery, we conducted MEG recordings again at 6–30 months after surgery (mean 15 months, SD = 10). In 6 out of the 12 healthy volunteers, we repeated the MEG measurement 3–8 months (mean 5 months, SD = 2) after the first recording.

Magnetic Counterpart of MMN

The averaged data were filtered with a passband of 1–30 Hz. We subtracted the responses to standard tones from the responses to deviant tones, and accordingly identified the MMNm signals from the difference waves (Naatanen et al. 1978; Hari et al. 1984; Giard et al. 1990; Alho 1995). The MMNm latency and strength were obtained at the individually determined gradiometer which showed the maximum amplitude over each hemisphere.

To identify the cerebral sources of the MMNm, we analyzed the subtraction signals using the ECD modeling technique (Hämäläinen et al. 1993) of the Neuromag software system (Vectorview, Elekta Neuromag, Ltd, Helsinki, Finland). We identified the ECDs by carrying out a least-squares search in a spherical volume conductor model, using subsets of 20–30 channels around the maximum response area. The goodness-of-fit (g) of the ECD model was also calculated to find out what percentage of the measured signal variance was accounted for by the ECD. Only ECDs which showed a g value >80% at selected periods of time in the subset of channels were accepted for the subsequent analysis. We then further evaluated the ECDs using a time-varying multidipole model in which all the 204 gradiometer channels were taken into account. We evaluated the validity of individual ECDs within the multidipole model by assessing the resemblance between the measured responses in our tests and the signals predicted by the model. If signals of some brain region were left inadequately explained, the data were reevaluated for more accurate estimation of the sources (Hämäläinen et al. 1993).

Minimum-Current Estimate

The sources of MMNm were also evaluated by using the minimum-current estimate method (Uutela et al. 1999; Stenbacka et al. 2002), which measured the neural currents in points of a 3D cubical lattice within the brain. For this analysis (Neuromag MCE version 1.3), we employed a standard boundary element model as the source space, and used thirty singular values for regularization.

Measurement of Phase-Locking Factors

To unveil the phase consistency across trials of signals elicited by standard and deviant sounds, we used Morlet wavelet analysis (Kronland-Martinet et al. 1987) to calculate the continuous phase at the target frequency. This is a function of time t and frequency f0, defined as 

graphic

The width of the wavelet (m = f0f) was 7 (Tallon-Baudry et al. 1996; Lachaux et al. 1999; Rodriguez et al. 1999; Jensen et al. 2002), and the representation of the phase for trial i at time t and frequency f0 is given from the result of the convolution of the complex wavelet w(t, f0) with the signal si(t), normalized by the amplitude 

graphic

The phase-locking factor (PLF) over N trials is then defined at time t and frequency f0 as 

graphic

The PLF ranges from 0 to 1, with the value 1 indicating a δ-function distribution and the value approaching 0 indicating a uniform distribution. To get a time–frequency representation of the PLF in the neuromagnetic responses, we used a set of wavelets with f0 ranging from 0.5 to 40 Hz in steps of 1 Hz. We referred to individual frequency bands as delta (0.5–4 Hz), theta (4–8 Hz), alpha (8–14 Hz), beta (14–25 Hz), and gamma bands (25–40 Hz).

Data Presentations and Statistical Analysis

The average values of measured data were shown as mean ± standard error of the mean (SEM). The statistical analysis for MMNm in varying conditions was conducted with a SPSS statistical package, version 14.0, and P < 0.05 was taken as the significance threshold.

To compare the overall difference of the mismatch responses between the patient and the control groups, the data of all participants (n = 12 for each group) were included. We used a generalized linear model with univariate option to evaluate the difference in MMNm latency and amplitude between the 2 groups.

To demonstrate the effect of successful epileptic surgery on the evoked activations, we used the data from only the 5 patients (patients 2–5 and 8), who had right temporal lobe seizure focus and became seizure free after surgery. Two patients with left seizure focus were excluded from this analysis because the case number was too little to do a feasible comparison. Therefore, we used a generalized linear model with repeated measurement to compare the difference between pre- and postsurgical MMNm of the 5 right-TLE patients in terms of the peak latencies, amplitudes, and ECD locations. The difference of MMNm between first and second measurement for the 6 normal controls was also assessed by a generalized linear model.

The statistical significance of the PLFs was assessed by using a Rayleigh test (Fisher 1993). For 100 trials (n = 100), a PLF above 0.18 and 0.22 is statistically significant at P < 0.05 and P < 0.01, respectively.

Results

In this paper, we have only presented the results of the MMNm of the duration-deviant condition because those for the frequency-deviant condition showed relatively poor signal-to-noise ratios and therefore were inadequate to be retained for the present investigation. The average noise level was 3.5 ± 0.1 fT/cm for planar gradiometers.

Figure 1 shows the spatial distributions of the auditory-evoked fields of patient 2 before surgery (left upper panel) and control 2 (right upper panel)—as representative of the whole sample population of the 12 TLE patients and their 12 age-matched healthy controls—to 50-ms deviant and 100-ms standard stimuli. The difference waves between the responses to the deviant and standard stimuli, the so called MMNm, are displayed in red. As shown in the inserts, the clear upward deflections of the MMNm waveforms from the TLE patient peaked at 184 ms in the left and 174 ms in the right hemisphere. For control 2, the MMNm peaked at 155 ms in the left hemisphere, and 141 ms in the right hemisphere. The bottom panels in Figure 1 demonstrate the superior temporal localizations for the ECDs of the bilateral MMNm in both patient 2 and control 2.

Figure 1.

(Top) Spatial distribution of auditory-evoked fields of patient 2 (left panel) and control subject 2 (right panel) to standard (1000 Hz, 100 ms) and deviant tone stimuli (1000 Hz, 50 ms). The signals from the MEG channels are projected onto a plane; the head is viewed from the top and the nose pointed upwards. Each response pair illustrates signals recorded by the 2 orthogonal gradiometers of a signal sensor unit. The MMNm was obtained by subtracting the standard-elicited waveform from the deviants-elicited waveform. (Bottom) The ECD location (circled dots) of the MMNm in each hemisphere was superimposed on the surface rendition of the brain MR images. L, left; R, right; LH, left hemisphere; RH, right hemisphere.

Figure 1.

(Top) Spatial distribution of auditory-evoked fields of patient 2 (left panel) and control subject 2 (right panel) to standard (1000 Hz, 100 ms) and deviant tone stimuli (1000 Hz, 50 ms). The signals from the MEG channels are projected onto a plane; the head is viewed from the top and the nose pointed upwards. Each response pair illustrates signals recorded by the 2 orthogonal gradiometers of a signal sensor unit. The MMNm was obtained by subtracting the standard-elicited waveform from the deviants-elicited waveform. (Bottom) The ECD location (circled dots) of the MMNm in each hemisphere was superimposed on the surface rendition of the brain MR images. L, left; R, right; LH, left hemisphere; RH, right hemisphere.

Based on a 2-dipole modeling approach—one ECD in each hemisphere—the generators of the MMNm were localized in the bilateral supratemporal cortices of all the patients and control participants.

Table 2 shows the mean peak latencies and amplitudes of MMNm across the 12 medically intractable TLE patients as well as the 12 healthy age-matched controls. The mean latency value was longer in the patients than the control subjects (P = 0.04) but we did not observe any statistical difference in amplitudes between the 2 groups (P = 0.4 for the data in the left hemisphere; P = 0.3 for the right hemisphere). The MMNm amplitude was clearly larger in the right than the left hemisphere in both the patient (P < 0.05) and the control group (P < 0.005).

Table 2

Mean latencies and amplitudes (SEM) of MMNm for duration-deviant condition in 12 normal controls and 12 temporal lobe epilepsy patients

Subjects Peak latencies (ms) Amplitudes (fT/cm) 
 Left H Right H Left H Right H 
Controls 167 (6) 161 (5) 28.5 (4.8) 57.0 (6.8)** 
Patients 189 (8)# 181 (7)# 35.2 (5.5) 47.1 (7.4)* 
Subjects Peak latencies (ms) Amplitudes (fT/cm) 
 Left H Right H Left H Right H 
Controls 167 (6) 161 (5) 28.5 (4.8) 57.0 (6.8)** 
Patients 189 (8)# 181 (7)# 35.2 (5.5) 47.1 (7.4)* 

Note: H indicates hemisphere. *P < 0.05, **P < 0.005 compared with the amplitude in the left hemisphere of the same subject group. #P < 0.05 compared with the latency in the corresponding hemisphere of the control group.

Table 3 showed the mean dipole parameters of MMNm for the 12 TLE and 12 control subjects. We did not find a significant difference in dipole locations between patient and control groups.

Table 3

Mean dipole parameters (SEM) of mismatch fields for duration-deviant condition in 12 normal controls and 12 temporal lobe epilepsy patients

Subjects Left hemispheric dipole Confidence volume (V/mm3) Right hemispheric dipole Confidence volume(V/mm3) 
 Location coordinates (mm) Dipole moment vectors (nAm)  Location coordinates (mm) Dipole moment vectors (nAm)  
 x y z Qx Qy Qz  x y z Qx Qy Qz  
Controls −49 (2) 3.1 (1.3) 50.8 (3.3) −2.1 (0.9) −11.7 (1.7) −9.5 (1.4) 2592.8 (967.8) 57.7 (1.7) 5.5 (3.1) 52.3 (1.9) 3.1 (1.1) −11.7 (2.5) −11.3 (2.4) 467 (156.6) 
Patients −46.2 (2.5) 11.5 (4.3) 48.9 (3.8) −3.1 (2.6) −7.1 (3.1) −11.4 (3.4) 2101.6 (706.2) 54.1 (2.8) 11.6 (3.6) 52.1 (2.2) 5.5 (1.1) −12.6 (1.9) −10.7 (1.4) 1367.4 (639.1) 
Subjects Left hemispheric dipole Confidence volume (V/mm3) Right hemispheric dipole Confidence volume(V/mm3) 
 Location coordinates (mm) Dipole moment vectors (nAm)  Location coordinates (mm) Dipole moment vectors (nAm)  
 x y z Qx Qy Qz  x y z Qx Qy Qz  
Controls −49 (2) 3.1 (1.3) 50.8 (3.3) −2.1 (0.9) −11.7 (1.7) −9.5 (1.4) 2592.8 (967.8) 57.7 (1.7) 5.5 (3.1) 52.3 (1.9) 3.1 (1.1) −11.7 (2.5) −11.3 (2.4) 467 (156.6) 
Patients −46.2 (2.5) 11.5 (4.3) 48.9 (3.8) −3.1 (2.6) −7.1 (3.1) −11.4 (3.4) 2101.6 (706.2) 54.1 (2.8) 11.6 (3.6) 52.1 (2.2) 5.5 (1.1) −12.6 (1.9) −10.7 (1.4) 1367.4 (639.1) 

Figure 2 shows that in the 5 TLE patients (patients 2–5 and 8) who became seizure free after surgical removal of right temporal seizure focus, we did not see a significant change in the latency or amplitude of MMNm.

Figure 2.

Mean latencies and amplitudes (+SEM) of the MMNm in 5 patients and 6 healthy controls. Pre- (pre) and postoperative (post) data were presented separately for the patient group, whereas data obtained in the first (1st) and second measurement (2nd; about 3–8 months after the first one) were shown for the control group. **P < 0.005.

Figure 2.

Mean latencies and amplitudes (+SEM) of the MMNm in 5 patients and 6 healthy controls. Pre- (pre) and postoperative (post) data were presented separately for the patient group, whereas data obtained in the first (1st) and second measurement (2nd; about 3–8 months after the first one) were shown for the control group. **P < 0.005.

For the 6 control subjects who were studied twice, we did not see a significant difference in MMNm data between the 2 separate recordings either. Moreover, the larger MMNm amplitude in the right than the left hemisphere was identified in both the first (64.7 ± 8.7 vs. 27.3 ± 7.9 fT/cm, P < 0.005) and the second recording (60.7 ± 7.7 vs. 28.9 ± 5.8 fT/cm, P < 0.005).

Figure 3 shows the locations of the MMNm sources evaluated by minimum-current estimates in Patient 2 and in Control subject 2 as representative of the whole sample population of the 12 TLE patients and their 12 age-matched healthy controls. In addition to noting superior temporal localization bilaterally, we also observed activation in the frontal and parietal areas in both the patient and the control. On the whole, the superior temporal sources were identified in all 12 patients and 12 controls, whereas the frontal lobe activations were found in 9 patients and 10 controls. We were able to identify parietal sources in 7 patients and 8 controls.

Figure 3.

Minimum-current estimates of the cerebral sources of mismatch magnetic fields in one healthy subject (control 2) and in one TLE patient (patient 2). LH, left hemisphere; RH, right hemisphere.

Figure 3.

Minimum-current estimates of the cerebral sources of mismatch magnetic fields in one healthy subject (control 2) and in one TLE patient (patient 2). LH, left hemisphere; RH, right hemisphere.

Figure 4 depicts the time–frequency representations of the grand-averaged PLFs for the deviant-evoked magnetic fields in the 5 right-TLE patients (patients 2–5 and 8) prior to surgery (left upper panel) and in the 6 controls during the first recording (right upper panel). In both the patients and their age-matched healthy controls, the phase-locking phenomena were mainly observed for theta (4–8 Hz) as well as alpha (8–14 Hz) oscillations over the bilateral temporal regions. The latency of peak PLF occurred at about 200 ms after the stimulus onset.

Figure 4.

PLF across trials of cerebral responses to deviant sound stimuli in 5 right-TLE patients and 6 controls. Left upper: The grand-averaged time–frequency representations of PLFs for preoperative 0.5- to 40-Hz responses to deviant stimuli in TLE patients. The head is flattened and viewed from above. Each insert shows the average plot of 18 gradiometer channels in the individual hemispheres. The PLFs are color coded: Small values are indicated with blue, and large, with red. PLF > 0.18 indicates P < 0.05. Left bottom: The grand-averaged PLFs for individual oscillatory bands over time with respect to the stimulus onset (time 0). Right upper: The grand-averaged time–frequency representations of PLFs for 0.5–40 Hz responses to deviant stimuli in the control subjects in the first measurement. Right bottom: The grand-averaged PLFs for individual oscillatory bands over time with respect to the stimulus onset (time 0).

Figure 4.

PLF across trials of cerebral responses to deviant sound stimuli in 5 right-TLE patients and 6 controls. Left upper: The grand-averaged time–frequency representations of PLFs for preoperative 0.5- to 40-Hz responses to deviant stimuli in TLE patients. The head is flattened and viewed from above. Each insert shows the average plot of 18 gradiometer channels in the individual hemispheres. The PLFs are color coded: Small values are indicated with blue, and large, with red. PLF > 0.18 indicates P < 0.05. Left bottom: The grand-averaged PLFs for individual oscillatory bands over time with respect to the stimulus onset (time 0). Right upper: The grand-averaged time–frequency representations of PLFs for 0.5–40 Hz responses to deviant stimuli in the control subjects in the first measurement. Right bottom: The grand-averaged PLFs for individual oscillatory bands over time with respect to the stimulus onset (time 0).

As shown in Figure 5, the peak PLF in the TLE patients occurred at 206 ± 32 ms in the left and at 213 ± 14 ms in the right hemisphere. For the control subjects, the PLF peaked at 183 ± 8 ms in the left hemisphere, and at 173 ± 7 ms in the right hemisphere. For the right-hemispheric responses, the patient group showed a longer peak latency (213 vs. 173 ms, P < 0.05) and a smaller PLF (0.18 vs. 0.26, P < 0.005) than the control subjects.

Figure 5.

The mean peak PLFs and latencies after deviant sound stimulation in 5 patients and 6 healthy controls. *P < 0.05; **P < 0.005.

Figure 5.

The mean peak PLFs and latencies after deviant sound stimulation in 5 patients and 6 healthy controls. *P < 0.05; **P < 0.005.

Figure 6 shows the presence of clear phase locking of evoked responses to deviant or standard sounds from 50 ms following the stimulus onset. For both the patient and control groups, the main difference between the deviant- and standard-evoked responses in terms of phase locking occurred around 150–250 ms after stimulus onset. In this time frame, the PLF values over the temporal regions were clearly larger for the deviant-evoked than for the standard-evoked responses.

Figure 6.

Left panel: The topographic plots of PLFs for 4- to 14-Hz signals with respect to the onset of standard or deviant stimulation in TLE patients before and after epileptic surgery. The MEG helmet covering the entire scalp is projected onto a plane, with the nose pointing upward. Right panel: The topographic plots of phase locking for 4- to 14-Hz cerebral signals with respect to the onset of standard or deviant stimulation in 2 separate measurements of control subjects. Strong phase locking is indicated with red, and weak, with blue. L, left; R, right.

Figure 6.

Left panel: The topographic plots of PLFs for 4- to 14-Hz signals with respect to the onset of standard or deviant stimulation in TLE patients before and after epileptic surgery. The MEG helmet covering the entire scalp is projected onto a plane, with the nose pointing upward. Right panel: The topographic plots of phase locking for 4- to 14-Hz cerebral signals with respect to the onset of standard or deviant stimulation in 2 separate measurements of control subjects. Strong phase locking is indicated with red, and weak, with blue. L, left; R, right.

For control groups, the topographic mapping of PLF values for the evoked responses was similar between 2 separate recordings. In contrast, the PLF for the TLE patients was larger for the postsurgical than for the presurgical recordings in terms of statistical power and spatial distribution. Here we saw a more apparent increase in PLF for the deviant- than for the standard-evoked oscillations, especially in the time frame of 150–250 ms.

Discussion

Our finding that the peak latency of the MMNm was relatively longer for the TLE patients than for their age-matched controls may reflect the relative difficulty involved in detecting the deviation of sound features in patients. This observation seems in agreement with previous studies where the MMN latency becomes longer as the magnitude of deviation decreases (Naatanen 1992; Yabe et al. 2001).

In this study, the MMNm was also seen to dominate in the right hemisphere of both the patient and control groups, in line with the observation of an earlier study (Paavilainen et al. 1991). However, this hemispheric asymmetry was not consistently observed in the patients with a right-sided temporal epileptic focus. This finding suggests that the epileptic abnormality in the right mesial temporal area may affect the interhemispheric balance of auditory memory processing. Taken together, the longer latency in the whole of the patient group and the reduced hemispheric asymmetry in the right-TLE subgroup indicate that the mesial temporal structure is possibly involved in the neuronal network of cerebral reactivity to deviant sounds, even though this area may be not the principal area for the generation of scalp-recorded MMNm.

Using ECD modeling, we found that the MMNm arises from the supratemporal auditory cortex, in line with previous animal (Csepe et al. 1987; Javitt et al. 1992) and human (Hari et al. 1984; Sams and Hari 1991; Sams et al. 1991; Tiitinen et al. 1993) studies. In this study, in addition to identifying supratemporal localization of the MMNm, we further demonstrated that the minimum-current estimates of the MMNm reveal clear activations over the bilateral prefrontal regions. Also, minimum-current estimation, which does not need a priori information about the number of active brain sources (Hämäläinen et al. 1993; Uutela et al. 1999; Stenbacka et al. 2002), has been considered to be useful when the distribution of activity is poorly known (Hämäläinen et al. 1993; Hari et al. 2000).

The involvement of the frontal cortex in the formation of the MMNm is consistent with previous current density maps (Giard et al. 1990) and chaos analysis (Molnar et al. 1995) of the MMN. Some earlier studies in humans with prefrontal cortical lesions (Alho et al. 1994; Alain et al. 1998) have also shown an attenuation of MMN activity. Using subdural recordings of the MMN in patients with focal epilepsy, Rosburg et al. (2005) also demonstrated evidence of the participation of the frontal gyrus in the generation of MMN. The prefrontal MMN subcomponent may be associated with involuntary attention switching to changes in auditory sensory inputs (Giard et al. 1990; Naatanen 1992). Moreover, previous lesion studies also suggest a role of the parietal lobe in shifting attention (Posner et al. 1984; Alain et al. 1998). Our present study showed both frontal and parietal activations in some subjects, in line with one earlier functional MRI study on mismatch responses that have proposed the involvement of frontal and parietal cortices in reflecting recruitment of attention-switching mechanisms (Molholm et al. 2005). Future studies of patients with specified neocortical seizure foci will help us to clarify the functional roles of the frontal and parietal activations in sensory memory processing.

To further exemplify the auditory stimulus-induced changes in brain dynamics, we evaluated the PLF using a precise time course, and uncovered the spectral features of the activities which were induced by deviant and standard stimuli. In this study, we found that neural activities were strongly phase-locked in the bilateral temporal regions following deviant sound presentation (see Fig. 4). Notably, the main difference between the deviant- and standard-evoked responses in terms of PLF values was seen at around 150–250 ms after the stimulus onset, which is temporally consistent with the time frame for the mismatch signals obtained using the conventional subtraction procedure (Naatanen et al. 1978; Hari et al. 1984; Giard et al. 1990; Alho 1995). Our present result suggests that the phase-locking characteristics of cortical signals in the time frame of 150–250 ms following stimulus onset can also reflect the differentiation between cerebral reactivity to standard and to deviant sound stimuli.

In the patients with right TLE who became free from seizures after surgery, we observed an increase in phase locking for the event-related fields as a result of deviant auditory stimuli. The phase-locking enhancement for theta and alpha oscillations occurred over the bilateral temporal and frontal regions, which suggests a plastic change in TLE patients following removal of a mesial temporal lobe seizure focus. Previous studies have shown monosynaptic projections from the hippocampus onto prefrontal cortical neurons (Degenetais et al. 2003; Tierney et al. 2004). More recent studies further demonstrated prefrontal phase locking to hippocampus theta oscillations (Jensen 2005; Jones and Wilson 2005; Siapas et al. 2005). Our present results may suggest that, before surgical treatment, the epileptic nature of the abnormality in the mesial temporal region may affect the involvement of frontotemporal cortices in processing deviant inputs. Following the removal of epileptic lesions, we thus found a possible plastic increase in the phase locking of the 4- to 14-Hz oscillations in not only the temporal but also the frontal areas.

It has been debated among neuroscientists as to whether event-related responses arise as a fixed-polarity and fixed-latency superimposed neuronal contribution to background signals (Vaughan and Arezzo 1988; Schroeder et al. 1995; Makinen et al. 2004), or as an event-locked increase in the phase concentration of ongoing oscillations (Sayers et al. 1974; Basar 1980; Brandt et al. 1991; Makeig et al. 2002; Kruglikov and Schiff 2003). In this study, the analysis of the MMNm waveforms showed no remarkable difference before and after removal of the mesial temporal epileptogenic focus but our measurement of the phase-locking phenomena demonstrated enhanced phase locking after epileptic surgery, which agrees with an earlier observation that the evoked and oscillatory mechanisms behave differentially for the genesis of individual event-related responses (Fuentemilla et al. 2006). Our present data suggest, basically in line with recent studies (Basar et al. 1984; Sannita et al. 2001; David et al. 2005; Fuentemilla et al. 2006), that both evoked and oscillatory activities could contribute together to the cerebral basis for detecting deviant auditory events in the daily environment. Thus, separate analysis of event-related waveforms and phase-locking changes helps to reveal specific insights into the mechanisms of various cognitive processes (Fell et al. 2004).

Compared with the traditional Fast Fourier transform, Morlet wavelet transform with phase-locking analysis actually provides good temporal resolution (Samar et al. 1999), with a sufficient wavelet duration (e.g., 223 ms at 10 Hz; 446 ms at 5 Hz) (Tallon-Baudry et al. 1996) that seems appropriate to detect the transitional signal of MMN. Moreover, wavelet-based analysis has been applied to analyze various evoked-components, such as limbic P3 responses in hippocampal sclerosis (Fell et al. 2005), visual attention-related alpha activities (Makeig et al. 2002), and visual gamma responses (Tallon-Baudry et al. 1996).

The PLFs denote whether the cortical signals in the brain, at any given point in time, show a nonrandom phase distribution in response to repetitive stimuli (Sinkkonen et al. 1995). And because the PLF is amplitude independent, it is able to avoid the large-amplitude artifacts. It may therefore be able to reveal small-amplitude but well-locked signals better than conventional coherence analysis (Lachaux et al. 1999). In this study, we observed an enhancement of phase-locking phenomena, in contrast with the similarity in the MMNm strengths between the TLE patients and age-matched controls. Accordingly, we proposed that the measurement of PLF may be used to evaluate more effectively the plasticity change of sensory memory processing, which would otherwise remain undetected by analysis of the fixed-polarity evoked waveforms.

Conclusions

The present study yielded 2 main findings. First, the longer MMNm latencies for TLE patients and the reduced interhemispheric asymmetry of the MMNm amplitudes in those patients with right-sided epileptic focus, in contrast with non-TLE subjects, suggest that the mesial temporal structures are involved in cerebral reactivity to deviant stimuli. Second, the measurement of intertrial coherence showed a differentiation of phase locking in the time frame of 150–250 ms for deviant versus standard-evoked cortical responses. This finding suggests that phase-locking characteristics are capable of reflecting a neural mismatch between an incoming stimulus and the transient representation of sounds. Furthermore, we found that in the frontotemporal neuronal network, which subserves sensory memory traces, there was a possible neuroplastic change in the phase locking of evoked signals following the removal of the temporal lobe seizure focus of TLE patients.

We thank Dr Ming-Wei Lin for her invaluable assistance in our statistical analysis. We appreciated Mr Chih-Che Chou and Mr Chou-Ming Cheng for technical assistance in the acquisition of MR images. This study was supported in part by research grants from Taipei Veterans General Hospital (V95ER3-006 and V95C1-043) and from the National Science Council (NSC-95-2314-B-010-030-MY3), Taipei, Taiwan. We highly appreciated the suggestions and comments from anonymous reviewers that strengthened our present work. Conflict of Interest: None declared.

References

Alain
C
Woods
DL
Knight
RT
A distributed cortical network for auditory sensory memory in humans
Brain Res
 , 
1998
, vol. 
812
 (pg. 
23
-
37
)
Alessio
A
Bonilha
L
Rorden
C
Kobayashi
E
Min
LL
Damasceno
BP
Cendes
F
Memory and language impairments and their relationships to hippocampal and perirhinal cortex damage in patients with medial temporal lobe epilepsy
Epilepsy Behav
 , 
2006
, vol. 
8
 (pg. 
593
-
600
)
Alho
K
Cerebral generators of mismatch negativity (MMN) and its magnetic counterpart (MMNm) elicited by sound changes
Ear Hear
 , 
1995
, vol. 
6
 (pg. 
38
-
51
)
Alho
K
Woods
DL
Algazi
A
Knight
RT
Naatanen
R
Lesions of frontal cortex diminish the auditory mismatch negativity
Electroencephalogr Clin Neurophysiol
 , 
1994
, vol. 
91
 (pg. 
353
-
362
)
Basar
E
EEG-brain dynamics: relation between EEG and brain evoked potentials
 , 
1980
New York
Elsevier
Basar
E
Basar-Eroglu
C
Rosen
B
Schutt
A
A new approach to endogenous event-related potentials in man: relation between EEG and P300-wave
Int J Neurosci
 , 
1984
, vol. 
24
 (pg. 
1
-
21
)
Brandt
ME
Jansen
BH
Carbonari
JP
Pre-stimulus spectral EEG patterns and the visual evoked response
Electroencephalogr Clin Neurophysiol
 , 
1991
, vol. 
80
 (pg. 
16
-
20
)
Cabeza
R
Dolcos
F
Graham
R
Nyberg
L
Similarities and differences in the neural correlates of episodic memory retrieval and working memory
Neuroimage
 , 
2002
, vol. 
16
 (pg. 
317
-
330
)
Chao
LL
Knight
RT
Human prefrontal lesions increase distractibility to irrelevant sensory inputs
Neuroreport
 , 
1995
, vol. 
6
 (pg. 
1605
-
1610
)
Commission on Classification and Terminology of the ILAE
Proposal for revised classification of epilepsies and epileptic syndromes
Epilepsia
 , 
1989
, vol. 
30
 (pg. 
389
-
399
)
Cowan
N
On short and long auditory stores
Psychol Bull
 , 
1984
, vol. 
96
 (pg. 
341
-
370
)
Cowan
N
Attention and memory: an integrated framework
 , 
1995
New York
Oxford University Press
Csepe
V
Karmos
G
Molnar
M
Evoked potential correlates of stimulus deviance during wakefulness and sleep in cat—animal model of mismatch negativity
Electroencephalogr Clin Neurophysiol
 , 
1987
, vol. 
66
 (pg. 
571
-
578
)
David
O
Harrison
L
Friston
KJ
Modelling event-related responses in the brain
Neuroimage
 , 
2005
, vol. 
25
 (pg. 
756
-
770
)
Degenetais
E
Thierry
AM
Glowinski
J
Gioanni
Y
Synaptic influence of hippocampus on pyramidal cells of the rat prefrontal cortex: an in vivo intracellular recording study
Cereb Cortex
 , 
2003
, vol. 
13
 (pg. 
782
-
792
)
Engel
J
Jr
Van Ness
PC
Rasmussen
TB
Ojemann
LM
Engel
JJR
Outcome with respect to epileptic seizures
Surgical treatment of the epilepsies
 , 
1993
2nd ed
New York
Raven Press
(pg. 
609
-
621
)
Fell
J
Dietl
T
Grunwald
T
Kurthen
M
Klaver
P
Trautner
P
Schaller
C
Elger
CE
Fernandez
G
Neural bases of cognitive ERPs: more than phase reset
J Cogn Neurosci
 , 
2004
, vol. 
16
 (pg. 
1595
-
1604
)
Fell
J
Kohling
R
Grunwald
T
Klaver
P
Dietl
T
Schaller
C
Becker
A
Elger
CE
Fernandez
G
Phase-locking characteristics of limbic P3 responses in hippocampal sclerosis
Neuroimage
 , 
2005
, vol. 
24
 (pg. 
980
-
989
)
Fisher
N
Statistical analysis of circular data
 , 
1993
Cambridge, UK
Cambridge University Press
Fuentemilla
L
Marco-Pallares
J
Grau
C
Modulation of spectral power and of phase resetting of EEG contributes differentially to the generation of auditory event-related potentials
Neuroimage
 , 
2006
, vol. 
30
 (pg. 
909
-
916
)
Giard
MH
Perrin
F
Pernier
J
Bouchet
P
Brain generators implicated in the processing of auditory stimulus deviance: a topographic event-related potential study
Psychophysiology
 , 
1990
, vol. 
27
 (pg. 
627
-
640
)
Hämäläinen
M
Hari
R
Ilmoniemi
RJ
Knuutila
J
Lounasmaa
OV
Magnetoencephalography—theory, instrumentation, and applications to noninvasive studies of the working human brain
Rev Mod Phys
 , 
1993
, vol. 
65
 (pg. 
413
-
497
)
Hari
R
Hamalainen
M
Ilmoniemi
R
Kaukoranta
E
Reinikainen
K
Salminen
J
Alho
K
Naatanen
R
Sams
M
Responses of the primary auditory cortex to pitch changes in a sequence of tone pips: neuromagnetic recordings in man
Neurosci Lett
 , 
1984
, vol. 
50
 (pg. 
127
-
132
)
Hari
R
Levanen
S
Raij
T
Timing of human cortical functions during cognition: role of MEG
Trends Cogn Sci
 , 
2000
, vol. 
4
 (pg. 
455
-
462
)
Helmstaedter
C
Kurthen
M
Lux
S
Reuber
M
Elger
CE
Chronic epilepsy and cognition: a longitudinal study in temporal lobe epilepsy
Ann Neurol
 , 
2003
, vol. 
54
 (pg. 
425
-
432
)
Ilvonen
TM
Kujala
T
Kiesilainen
A
Salonen
O
Kozou
H
Pekkonen
E
Roine
RO
Kaste
M
Naatanen
R
Auditory discrimination after left-hemisphere stroke: a mismatch negativity follow-up study
Stroke
 , 
2003
, vol. 
34
 (pg. 
1746
-
1751
)
Javitt
DC
Schroeder
CE
Steinschneider
M
Arezzo
JC
Jr
Vaughan HG
Demonstration of mismatch negativity in the monkey
Electroencephalogr Clin Neurophysiol
 , 
1992
, vol. 
83
 (pg. 
87
-
90
)
Jensen
O
Reading the hippocampal code by theta phase-locking
Trends Cogn Sci
 , 
2005
, vol. 
9
 (pg. 
551
-
553
)
Jensen
O
Gelfand
J
Kounios
J
Lisman
JE
Oscillations in the alpha band (9–12 Hz) increase with memory load during retention in a short-term memory task
Cereb Cortex
 , 
2002
, vol. 
12
 (pg. 
877
-
882
)
Jones
MW
Wilson
MA
Phase precession of medial prefrontal cortical activity relative to the hippocampal theta rhythm
Hippocampus
 , 
2005
, vol. 
15
 (pg. 
867
-
873
)
Kane
NM
Curry
SH
Rowlands
CA
Manara
AR
Lewis
T
Moss
T
Cummins
BH
Butler
SR
Event-related potentials—neurophysiological tools for predicting emergence and early outcome from traumatic coma
Intensive Care Med
 , 
1996
, vol. 
22
 (pg. 
39
-
46
)
Karlsgodt
KH
Shirinyan
D
Van Erp
TG
Cohen
MS
Cannon
TD
Hippocampal activations during encoding and retrieval in a verbal working memory paradigm
Neuroimage
 , 
2005
, vol. 
25
 (pg. 
1224
-
1231
)
Knight
R
Contribution of human hippocampal region to novelty detection
Nature
 , 
1996
, vol. 
383
 (pg. 
256
-
259
)
Kronland-Martinet
R
Morlet
J
Grossmann
A
Analysis of sound patterns through wavelet transforms
Int J Pattern Recogn Artif Intell
 , 
1987
, vol. 
1
 (pg. 
273
-
302
)
Kruglikov
SY
Schiff
SJ
Interplay of electroencephalogram phase and auditory-evoked neural activity
J Neurosci
 , 
2003
, vol. 
23
 (pg. 
10122
-
10127
)
Lachaux
JP
Rodriguez
E
Martinerie
J
Varela
FJ
Measuring phase synchrony in brain signals
Hum Brain Mapp
 , 
1999
, vol. 
8
 (pg. 
194
-
208
)
Lin
YY
Shih
YH
Hsieh
JC
Yu
HY
Yiu
CH
Wong
TT
Yeh
TC
Kwan
SY
Ho
LT
Yen
DJ
, et al.  . 
Magnetoencephalographic yield of interictal spikes in temporal lobe epilepsy. Comparison with scalp EEG recordings
Neuroimage
 , 
2003
, vol. 
19
 (pg. 
1115
-
1126
)
Lin
YY
Su
MS
Yiu
CH
Shih
YH
Yen
DJ
Kwan
SY
Wu
ZA
Chan
SH
Relationship between mesial temporal seizure focus and elevated serum prolactin in temporal lobe epilepsy
Neurology
 , 
1997
, vol. 
49
 (pg. 
528
-
532
)
Luria
AR
Karasseva
TA
Disturbances of auditory-speech memory in focal lesions of the deep regions of the left temporal lobe
Neuropsychologia
 , 
1968
, vol. 
6
 (pg. 
97
-
104
)
Makeig
S
Westerfield
M
Jung
TP
Enghoff
S
Townsend
J
Courchesne
E
Sejnowski
TJ
Dynamic brain sources of visual evoked responses
Science
 , 
2002
, vol. 
295
 (pg. 
690
-
694
)
Makinen
V
May
P
Tiitinen
H
Transient brain responses predict the temporal dynamics of sound detection in humans
Neuroimage
 , 
2004
, vol. 
21
 (pg. 
701
-
706
)
Massaro
DW
Preperceptual images, processing time, and perceptual units in auditory perception
Psychol Rev
 , 
1972
, vol. 
79
 (pg. 
124
-
145
)
Molholm
S
Martinez
A
Ritter
W
Javitt
DC
Foxe
JJ
The neural circuitry of pre-attentive auditory change-detection: an fMRI study of pitch and duration mismatch negativity generators
Cereb Cortex
 , 
2005
, vol. 
15
 (pg. 
545
-
551
)
Molnar
M
Skinner
JE
Csepe
V
Winkler
I
Karmos
G
Correlation dimension changes accompanying the occurrence of the mismatch negativity and the P3 event-related potential component
Electroencephalogr Clin Neurophysiol
 , 
1995
, vol. 
95
 (pg. 
118
-
126
)
Mukhamedrakhimov
RZ
Role of the hippocampus in the short-term retention of moving and unmoving sound image localization in human beings
Sens Syst
 , 
1989
, vol. 
3
 (pg. 
150
-
158
)
Naatanen
R
Attention and brain function
 , 
1992
Hillsdale, NJ
Lawrence Erlbaum Associates
Naatanen
R
Gaillard
AW
Mantysalo
S
Early selective-attention effect on evoked potential reinterpreted
Acta Psychol
 , 
1978
, vol. 
42
 (pg. 
313
-
329
)
Naatanen
R
Paavilainen
P
Tiitinen
H
Jiang
D
Alho
K
Attention and mismatch negativity
Psychophysiology
 , 
1993
, vol. 
30
 (pg. 
436
-
450
)
Paavilainen
P
Alho
K
Reinikainen
K
Sams
M
Naatanen
R
Right hemisphere dominance of different mismatch negativities
Electroencephalogr Clin Neurophysiol
 , 
1991
, vol. 
78
 (pg. 
466
-
479
)
Pekkonen
E
Jousmaki
V
Kononen
M
Reinikainen
K
Partanen
J
Auditory sensory memory impairment in Alzheimer's disease: an event-related potential study
Neuroreport
 , 
1994
, vol. 
5
 (pg. 
2537
-
2540
)
Pekkonen
E
Jousmaki
V
Reinikainen
K
Partanen
J
Automatic auditory discrimination is impaired in Parkinson's disease
Electroencephalogr Clin Neurophysiol
 , 
1995
, vol. 
95
 (pg. 
47
-
52
)
Posner
MI
Walker
JA
Friedrich
FJ
Rafal
RD
Effects of parietal injury on covert orienting of attention
J Neurosci
 , 
1984
, vol. 
4
 (pg. 
1863
-
1874
)
Rodriguez
E
George
N
Lachaux
JP
Martinerie
J
Renault
B
Varela
FJ
Perception's shadow: long-distance synchronization of human brain activity
Nature
 , 
1999
, vol. 
397
 (pg. 
430
-
433
)
Rosburg
T
Trautner
P
Dietl
T
Korzyukov
OA
Boutros
NN
Schaller
C
Elger
CE
Kurthen
M
Subdural recordings of the mismatch negativity (MMN) in patients with focal epilepsy
Brain
 , 
2005
, vol. 
128
 (pg. 
819
-
828
)
Ruusuvirta
T
Korhonen
T
Penttonen
M
Arikoski
J
Kivirikko
K
Behavioral and hippocampal evoked responses in an auditory oddball situation when an unconditioned stimulus is paired with deviant tones in the cat: experiment II
Int J Psychophysiol
 , 
1995
, vol. 
20
 (pg. 
41
-
47
)
Samar
VJ
Bopardikar
A
Rao
R
Swartz
K
Wavelet analysis of neuroelectric waveforms: a conceptual tutorial
Brain Lang
 , 
1999
, vol. 
66
 (pg. 
7
-
60
)
Sams
M
Hamalainen
M
Antervo
A
Kaukoranta
E
Reinikainen
K
Hari
R
Cerebral neuromagnetic responses evoked by short auditory stimuli
Electroencephalogr Clin Neurophysiol
 , 
1985
, vol. 
61
 (pg. 
254
-
266
)
Sams
M
Hari
R
Magnetoencephalography in the study of human auditory information processing
Ann N Y Acad Sci
 , 
1991
, vol. 
620
 (pg. 
102
-
117
)
Sams
M
Kaukoranta
E
Hamalainen
M
Naatanen
R
Cortical activity elicited by changes in auditory stimuli: different sources for the magnetic N100m and mismatch responses
Psychophysiology
 , 
1991
, vol. 
28
 (pg. 
21
-
29
)
Sannita
WG
Bandini
F
Beelke
M
De Carli
F
Carozzo
S
Gesino
D
Mazzella
L
Ogliastro
C
Narici
L
Time dynamics of stimulus- and event-related gamma band activity: contrast-VEPs and the visual P300 in man
Clin Neurophysiol
 , 
2001
, vol. 
112
 (pg. 
2241
-
2249
)
Sayers
BM
Beagley
HA
Hensall
WR
The mechanism of auditory evoked EEG responses
Nature
 , 
1974
, vol. 
247
 (pg. 
481
-
483
)
Schroeder
CE
Steinschneider
M
Javitt
DC
Tenke
CE
Givre
SJ
Mehta
AD
Simpson
GV
Arezzo
JC
Vaughan
HG
Jr
Localization of ERP generators and identification of underlying neural processes
Electroencephalogr Clin Neurophysiol Suppl
 , 
1995
, vol. 
44
 (pg. 
55
-
75
)
Siapas
AG
Lubenov
EV
Wilson
MA
Prefrontal phase locking to hippocampal theta oscillations
Neuron
 , 
2005
, vol. 
46
 (pg. 
141
-
151
)
Sinkkonen
J
Tiitinen
H
Naatanen
R
Gabor filters: an informative way for analysing event-related brain activity
J Neurosci Methods
 , 
1995
, vol. 
56
 (pg. 
99
-
104
)
Squire
LR
Memory and the hippocampus: a synthesis from findings with rats, monkeys, and humans
Psychol Rev
 , 
1992
, vol. 
99
 (pg. 
195
-
231
)
Squire
LR
Zola-Morgan
S
The medial temporal lobe memory system
Science
 , 
1991
, vol. 
253
 (pg. 
1380
-
1386
)
Stenbacka
L
Vanni
S
Uutela
K
Hari
R
Comparison of minimum current estimate and dipole modeling in the analysis of simulated activity in the human visual cortices
Neuroimage
 , 
2002
, vol. 
16
 (pg. 
936
-
943
)
Tallon-Baudry
C
Bertrand
O
Delpuech
C
Pernier
J
Stimulus specificity of phase-locked and non-phase-locked 40 Hz visual responses in human
J Neurosci
 , 
1996
, vol. 
16
 (pg. 
4240
-
4249
)
Tierney
PL
Degenetais
E
Thierry
AM
Glowinski
J
Gioanni
Y
Influence of the hippocampus on interneurons of the rat prefrontal cortex
Eur J Neurosci
 , 
2004
, vol. 
20
 (pg. 
514
-
524
)
Tiitinen
H
Alho
K
Huotilainen
M
Ilmoniemi
RJ
Simola
J
Naatanen
R
Tonotopic auditory cortex and the magnetoencephalographic (MEG) equivalent of the mismatch negativity
Psychophysiology
 , 
1993
, vol. 
30
 (pg. 
537
-
540
)
Uutela
K
Hamalainen
M
Somersalo
E
Visualization of magnetoencephalographic data using minimum current estimates
Neuroimage
 , 
1999
, vol. 
10
 (pg. 
173
-
180
)
Vaughan Ju
HG
Arezzo
JC
Picton
TW
The neural basis of event-related potentials
Human event-related potentials
 , 
1988
, vol. 
Vol. 3
 
New York
Elsevier Science Publisher
(pg. 
45
-
94
)
Vinogradova
OS
Sokolov
EN
Vinogradova
OS
The hippocampus and the orienting reflex
Neuronal mechanisms of the orienting reflex
 , 
1975
Hillsdale, NJ
Erlbaum
(pg. 
128
-
154
)
Wiebe
S
Blume
WT
Girvin
JP
Eliasziw
M
A randomized, controlled trial of surgery for temporal-lobe epilepsy
N Engl J Med
 , 
2001
, vol. 
345
 (pg. 
311
-
318
)
Yabe
H
Koyama
S
Kakigi
R
Gunji
A
Tervaniemi
M
Sato
Y
Automatic discriminative sensitivity inside temporal window of sensory memory as a function of time
Brain Res Cogn Brain Res
 , 
2001
, vol. 
12
 (pg. 
39
-
48
)