- Email: [email protected]
Right hemisphere pitch-mismatch negativity reduction in patients with major depression: An MEG study
Journal of Affective Disorders 215 (2017) 225–229
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research paper
Right hemisphere pitch-mismatch negativity reduction in patients with major depression: An MEG study
MARK
Noriaki Hirakawaa, Yoji Hiranoa,b, Itta Nakamuraa,c, Shogo Hiranoa, Jinya Satoa, Naoya Oribea,d, ⁎ Takefumi Uenoa,d, Shigenobu Kanbaa, Toshiaki Onitsukaa, a
Department of Neuropsychiatry, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Department of Psychiatry, Harvard Medical School and Neural Dynamics Laboratory, VA Boston Healthcare System, Boston, MA, USA c Division of Clinical Research, National Hospital Organization, Kokura Medical Center, Fukuoka, Japan d Division of Clinical Research, National Hospital Organization, Hizen Psychiatric Center, Saga, Japan b
A R T I C L E I N F O
A BS T RAC T
Keywords: Major depression Magnetoencephalography Mismatch negativity Preattentive dysfunction
Background: The mismatch negativity (MMN) component of the event-related potential and its magnetic counterpart, the MMNm, are generated by a mismatch between the physical features of a deviant stimulus and a neuronal sensory-memory trace produced by repetitive standard stimuli. Deficits in the MMN/MMNm have been reported in patients with major depression; however, the results are inconsistent. The present study investigated the pitch-MMNm in patients with major depression using whole-head 306-channel magnetoencephalography (MEG). Methods: Twenty patients with major depression and 36 healthy subjects participated in this study. Subjects were presented with two sequences of auditory stimuli. One consisted of 1000 Hz standard signals (probability=90%) and 1200 Hz deviant signals (probability=10%), while the other consisted of 1200 Hz standard (90%) and 1000 Hz deviant signals (10%). Event-related brain responses to standard tones were subtracted from responses to deviant tones. Results: Major depressive patients showed significantly reduced magnetic global field power (GFP) of MMNm in the right hemisphere (p=0.02), although no significant MMNm reduction was observed in the left hemisphere (p=0.81). Additionally, patients with major depression showed significantly earlier bilateral MMNm peak latencies (p=0.004). No significant associations were observed between MMNm variables and demographic data/clinical variables within the patients. Limitations: We could not exclude the effects of antidepressants, mood stabilizers, or neuroleptics on the MMNm abnormalities found in patients with major depression. Sample size was also insufficient to permit subgroup analyses. Conclusions: Patients with major depression exhibited reduced GFP of MMNm in the right hemisphere. The present study suggested that patients with major depression may have right hemispheric dominant preattentive dysfunction.
1. Introduction Mismatch negativity (MMN) is a preattentive auditory eventrelated potential elicited when a sequence of repetitive sounds is interrupted by a deviant sound (Näätänen et al., 1978). MMN waveforms can be obtained by subtracting the event-related response to a repeated regular event (standard stimulus) from that of an oddball or deviant event (deviant stimulus). Näätänen (1990) proposed that the MMN is generated by a mismatch between the physical features of a deviant stimulus and the neuronal sensory-memory trace produced by
⁎
repetitive standard stimuli. MMN can be elicited by changes in the frequency, duration, intensity, and location of a stimulus. MMN abnormalities are frequently reported in patients with schizophrenia (e.g., Suga et al., 2016). Deficits in the MMN component have been also reported in patients with major depression; however, the results are inconsistent and mixed. For example, Qiao et al. (2013) reported significant duration-MMN reductions in patients with major depression. Chen et al. (2015) also demonstrated significant decreased duration-MMN in first-episode and recurrent major depression. In contrast, Umbricht et al. (2003) showed no significant duration-MMN
Correspondence to: Department of Neuropsychiatry, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashiku, Fukuoka 812-8582, Japan. E-mail address: [email protected] (T. Onitsuka).
http://dx.doi.org/10.1016/j.jad.2017.03.046 Received 27 December 2016; Received in revised form 13 March 2017; Accepted 17 March 2017 Available online 19 March 2017 0165-0327/ © 2017 Elsevier B.V. All rights reserved.
Journal of Affective Disorders 215 (2017) 225–229
N. Hirakawa et al.
sessions was counterbalanced across subjects. The inter-stimulus interval was a constant 400 ms (offset-to-onset). The stimuli were 80-dB sound-pressure-level sinusoidal tone bursts with a duration of 100 ms (including a linear rise/fall time of 10 ms).
differences between patients with major depression and healthy subjects. Qiao et al. (2015) reported that significant duration-MMN reductions were observed only in female subjects with major depression. Pang et al. (2014) presented that MMN of sad syllable stimuli was absent in patients. Electroencephalography (EEG) is widely used to measure MMN; it has a high temporal resolution, and most of the reports on MMN deficits have been based on EEG studies. Magnetoencephalography (MEG) also has a high temporal resolution, and detects the magnetic counterpart of the EEG MMN (the MMNm) with a higher spatial resolution. The EEG MMN is usually distributed over fronto-central scalp locations, with maximum amplitudes found at the fronto-central electrode positions (Garrido et al., 2009). MEG is superior for the investigation of the laterality of the MMNm, as MEG can accurately detect neural activities tangential to the surface of the cortex for both hemispheres, with minimal influence from conductivity through tissues such as brain tissue, skull, and scalp (Hämäläinen et al., 1993). As far as we are aware, there are currently only two studies on MEG measurement of the MMNm in major depression (Takei et al., 2009; Kähkönen et al., 2007) with conflicting results. Takei et al. (2009) investigated the MMNm associated with deviant pitch, duration, and vowel stimuli in 14 patients with major depression. The authors found a significant bilateral overall MMNm reduction in the patients. However, Kähkönen et al. (2007) reported no significant difference in pitch-MMNm amplitudes between patients with major depression and healthy subjects. The present MEG study investigated the pitchMMNm, which may lead to a better understanding of the neurophysiological feature of major depression.
2.3. Data acquisition and processing MEG signals were acquired using a whole-head, 306-channel sensor array (Vectorview; ELEKTA Neuromag, Helsinki, Finland) that consisted of 102 identical triple-sensor elements. Each sensor element consisted of two orthogonal planar-type gradiometers and one magnetometer. Prior to the recording, four head position indicator (HPI) coils were attached to the scalp and a three-dimensional (3D) digitizer was used to measure the anatomical landmarks of the head with respect to the HPI coils. During the recording in a magnetically shielded room, the subjects were instructed to sit relaxed and read a book with their heads positioned inside the helmet-shaped sensor array. The precise location of the head with respect to the sensor array was determined using the HPI coils. A bandpass filter of 0.01–30 Hz and a sampling rate of 1 kHz were used for recording. During the stimulation, the acquired raw data was stored for off-line analysis. The data were averaged separately for standard and deviant stimuli. The entire sweep time was 450 ms (100 ms before stimulus onset to 350 ms after stimulus onset). A spatiotemporal signal space separation (tSSS) method was applied to the recorded raw data off-line (Taulu and Simola, 2006), with signal variations exceeding 1500 femtotesla being excluded. The Final MMNm responses were calculated using the epochs of both sessions.
2. Methods
2.4. Data analysis
2.1. Subjects
2.4.1. Magnetic global field power of the MMNm The data obtained from the 54 channels positioned over each temporal region were analyzed according to the methods of previous studies (Shimano et al., 2014, see Fig. 1). The magnetic global field power (mGFP) of responses was calculated separately for each hemisphere, using the following formula (Takei et al., 2010, 2009; Kreitschmann-Andermahr, 1999; Lehmann and Skrandies, 1980):
Twenty patients with major depression and 36 healthy subjects participated in this study. All subjects had normal hearing, were aged 21–60 years, and were right-handed as determined by the Edinburgh Inventory (Oldfield, 1971). All participants signed an informed consent form after being given a complete description of the study, which was approved by the Kyushu University Institutional Review Board for Clinical Trials. The exclusion criteria were: (i) neurological illness or major head trauma that could result in abnormal EEG findings; (ii) electroconvulsive therapy; (iii) alcohol or drug dependence during the subject's lifetime; (iv) alcohol or drug abuse within the past five years; and (v) a verbal intelligence quotient below 75 (Wechsler, 1997). Healthy subjects were screened using the Structured Clinical Interview (SCID) non-patient edition. None of the healthy subjects or their firstdegree relatives had an Axis I psychiatric disorder (Spitzer et al., 1990). All patients were recruited from the Kyushu University Hospital, and were diagnosed with major depression on the basis of the SCID for DSM-IV Axis I disorders and the patient's medical records (First et al., 1996). Two senior clinical psychiatrists confirmed that all subjects had the ability to consent to participate in the examination. The socioeconomic statuses (SES) of the subjects and their parents were measured using the Hollingshead Two-Factor Index (Hollingshead, 1965). The severity of mood symptoms was assessed using the Structured Interview Guide for the Hamilton Depression Rating Scale (SIGH-D; Williams, 1988).
mGFP(ft/cm) =
2 n n ⎛ ∑ j =1 Uj ⎞ 1 ⎜ ⎟ ⋅ ∑ Ui − ⎟ , n i =1 ⎜⎝ n ⎠
where n is the number of channels and U is the amplitude for each channel. The variable global field power (GFP) corresponds to the spatial standard deviation, and quantifies the amount of neural activity at each time-point. First, GFP waveforms to standard and deviant stimuli were
2.2. Stimuli and procedure Subjects were presented with two sessions of 1000 stimuli. These consisted of two types of auditory stimuli sequences, administered binaurally through earphones in a random order. One sequence consisted of 1000 Hz standard (probability =90%) and 1200 Hz deviant stimuli (probability =10%), while the other consisted of 1200 Hz standard (90%) and 1000 Hz deviant stimuli (10%). The order of
Fig. 1. Layout of the measurement channels. The MEG signals were acquired using a whole-head 306-channel sensor array that consisted of 102 identical triple-sensor elements. Each sensor consisted of two orthogonal planar-type gradiometers and one magnetometer. We used 27 sensors for each hemisphere (a 54-channel orthogonal gradiometer). The two circles indicate the sensors used for the analyses.
226
Journal of Affective Disorders 215 (2017) 225–229
N. Hirakawa et al.
calculated for each hemisphere with 27 channels. Second, MMNm responses were obtained by subtracting the averaged responses to standard stimuli from those of deviant stimuli for each channel, and then GFP waveforms of MMNm was finally calculated using 27 channels for each hemisphere. The responses were digitally filtered using a bandpass range of 1–20 Hz. The peak amplitude of the MMNm component was defined as the largest amplitude of an individual GFP between 90 and 250 ms, and the MMNm peak latency was defined as the latency with this largest amplitude.
Table 2 mGFP latencies and amplitudes of the MMNm.
HS (n=36) Mean (SD)
MDD (n=20) Mean (SD)
mGFP latency (ms)
left
left
138.19 (21.85) 137.17 (25.20) 18.48 (8.52)
right
23.03 (11.63)
right mGFP power (ft/ cm)
2.5. Statistical analyses
df
t
p-value
125.10 (18.65) 119.10 (18.40) 19.02 (6.49)
54
2.26
0.03
54
2.81
0.01
54
−0.25
0.81
17.45 (5.75)
54
2.40
0.02
HS = healthy subjects; MDD = major depressive disorders; mGFP = magnetic global field power; MMNm = mismatch negativity magnetic counterpart.
Independent samples t-tests were used to assess group differences in age, handedness score, SES, parental SES, education years, and sleeping-scale score. The GFP peak latencies and powers of the MMNm were analyzed using repeated-measures analysis of covariates (ANCOVA) with group (depression or control) as a between-subject factor, hemisphere (left or right) as a within-subject factor, and age as a covariate, as the patients with major depression were significantly older than the healthy subjects (t[54] =−2.67, p=0.01). Degrees of freedom were adjusted using the Huynh–Feldt epsilon for factors with more than two levels. Exploratory analyses of the relationships between MMNm variables and the demographic/clinical data were evaluated using Pearson correlations with a Bonferroni correction for multiple correlations (16 correlations for healthy subjects and 100 correlations for patients with major depression).
GFP peak latency of the MMNm, the repeated-measures ANCOVA showed a significant main effect of group [F(1,54) =9.2, p=0.004], but no significant main effect of hemisphere [F(1,54) =0.35, p=0.56], and no significant hemisphere-by-group interaction [F (1,54) =0.20, p=0.66]. This indicates that patients with major depression showed significantly earlier MMNm peak latencies. For the GFP of the MMNm, a repeated-measures ANCOVA showed no significant main effects of group [F(1,54) =0.19, p=0.67] or hemisphere [F(1,54) =0.54, p=0.47]. However, there was a significant hemisphere-by-group interaction [F(1,54) =4.4, p=0.041]. Thus, post hoc t-tests were performed for each hemisphere. In the left hemisphere, there was no significant group difference (t[54] =−0.25, p=0.81), but in the right hemisphere, the patients with major depression showed significantly reduced GFP of MMNm (t[54] =2.4, p=0.02).
3. Results 3.2.1. Correlations between the MMNm variables and demographic/ clinical data In healthy subjects, no significant correlations were observed between the MMNm variables and the demographic data (−0.41≤r < 0.51; 0.002≤p≤0.73). In the patients with depression, there were no significant correlations between the MMNm variables and the demographic data or SIGH-D scores (−0.54 ≤ r ≤0.44; 0.02 ≤ p ≤0.98). Additionally, antidepressant dose (imipramine equivalent) did not significantly correlate with any of the MMNm variables in the patients with major depression (−0.14≤r ≤0.10; 0.55≤p ≤0.97).
3.1. Demographic characteristics Demographic and clinical data for all subjects are presented in Table 1. There were no significant intergroup differences in handedness, SES, parental SES, and years of education. According to the criteria used to rate depression severity (Furukawa et al., 2007), at the time of the study, two patients showed severe depression, three showed moderate depression, and eight showed mild depression. Seven further patients demonstrated a euthymic mood. Seventeen patients were receiving antidepressants, five were receiving mood stabilizers such as lithium carbonate, valproate, and lamotrigine, and eleven patients were receiving neuroleptic medication.
4. Discussion The present study examined the pitch-MMNm in patients with major depression. In comparison with non-depressed subjects, these patients showed significantly reduced MMNm amplitudes in the right hemisphere, although no significant MMNm reduction was observed in the left hemisphere. Additionally, patients with major depression showed significantly earlier bilateral MMNm peak latencies. With respect to previous findings on pitch-MMN and MMNm in general, there are various reports and no consensus has been formed. As noted in the introduction section, one MEG study reported significant bilateral MMNm reductions (Takei et al., 2009), whereas another study failed to demonstrate MMNm reductions in patients with major depression (Kähkönen et al., 2007). The previous results from EEG studies are also inconsistent. Ogura et al. (1995) found significant MMN reductions in patients with major depression. Umbricht et al. (2003) and Mu et al. (2016) reported no significant MMN differences compared to HS. However, He et al. (2010) detected significantly larger MMN in patients with major depression. Restuccia et al. (2016) reported that significantly increased MMN can be observed in patients by using hiintensity (90 dB) stimulation. Our MEG study came to a certain conclusion that patients with major depression may have right hemispheric dominant preattentive dysfunction, The MMNm may be useful for differentiating between major depression and bipolar disorder. Our group has previously reported
3.2. Magnetic global field of MMNm Table 2 shows the GFP and latencies of the MMNm components. Fig. 2 illustrates the group mean waveforms of the MMNm. For the Table 1 Demographic and clinical characteristics of the study participants.
Sex, male/female, n Age, years Handedness SES Parental SES Education, years Duration of illness, years Imipramine equivalent, mg SIGH-D
HS Mean (SD)
MDD Mean (SD)
df
t
p-value
17/19 34.06 (13.39) 97.78 (5.79)
9/11 43.05 (9.14)
54
−2.67
0.01
54
1.13
0.27
54 54 54
−1.88 −0.36 1.43
0.066 0.72 0.16
2.19 (0.79) 2.67 (0.93) 15.00 (2.35)
93.50 (21.59) 2.60 (0.75) 2.75 (0.64) 14.10 (2.05) 9.30 (6.20) 172.5 (114.6) 9.60 (5.83)
HS = healthy subjects; MDD = major depressive disorders; SES = socio-economic status; SIGH-D = Structured Interview Guide for the Hamilton Depression Rating Scale.
227
Journal of Affective Disorders 215 (2017) 225–229
N. Hirakawa et al.
Fig. 2. Grand mean magnetic global field power (GFP) waveforms of responses to standard stimuli, deviant stimuli, and mismatch fields (MMNm) in patients with major depressive disorder (MDD) [solid line] and healthy subjects (HS) [dotted line]. Left: left hemisphere, Right: right hemisphere.
References
that patients with bipolar disorder showed a significant bilateral MMNm reduction (Shimano et al., 2014), while this study found such a reduction only in the right hemisphere of patients with major depression. Takei et al. (2010) reported significantly delayed MMNm peak latencies in the right hemispheres of patients with bipolar disorder, although our group reported no significant latency changes. In the present study, patients with major depression exhibited significantly earlier bilateral MMNm latencies. The results of our study suggest that the MMNm has potential as a biomarker for the differentiation between major depression and bipolar disorder.
Chen, J., Zhang, Y., Wei, D., Wu, X., Fu, Q., Xu, F., Wang, H., Ye, M., Ma, W., Yang, L., Zhang, Z., 2015. Neurophysiological handover from MMN to P3a in first-episode and recurrent major depression. J. Affect. Disord. 174, 173–179. First, M.B., Spitzer, R.L.S., Gibbon, M., Williams, J.B.W., 1996. Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV). American Psychiatric Press Inc, Washington, DC. Furukawa, T.A., Akechi, T., Azuma, H., Okuyama, T., Higuchi, T., 2007. Evidence-based guidelines for interpretation of the Hamilton Rating Scale for Depression. J. Clin. Psychopharmacol. 27, 531–534. Garrido, M.I., Kilner, J.M., Stephan, K.E., Friston, K.J., 2009. The mismatch negativity: a review of underlying mechanisms. Clin. Neurophysiol. 120, 453–463. Hämäläinen, M., Hari, R., Ilmoniemi, R.J., Knuutila, J., Lounasmaa, O.V., 1993. Magnetoencephalography - theory, instrumentation, and applications to noninvasive studies of the working human brain. Rev. Mod. Phys. 65, 413–497. He, W., Chai, H., Zheng, L., Yu, W., Chen, W., Li, J., Chen, W., Wang, W., 2010. Mismatch negativity in treatment-resistant depression and borderline personality disorder. Prog. Neuropsychopharmacol. Biol. Psychiatry 34, 366–371. Hollingshead, A., 1965. Two Factor Index of Social Position. Yale University Press, New Haven, CT. Kähkönen, S., Yamashita, H., Rytsälä, H., Suominen, K., Ahveninen, J., Isometsä, E., 2007. Dysfunction in early auditory processing in major depressive disorder revealed by combined MEG and EEG. J. Psychiatry Neurosci. 32, 316–322. Kiang, M., Braff, D.L., Sprock, J., Light, G.A., 2009. The relationship between preattentive sensory processing deficits and age in schizophrenia patients. Clin. Neurophysiol. 120, 1949–1957. Kreitschmann-Andermahr, I., Rosburg, T., Meier, T., Volz, H.P., Nowak, H., Sauer, H., 1999. Impaired sensory processing in male patients with schizophrenia: a magnetoencephalographic study of auditory mismatch detection. Schizophr. Res. 35, 121–129. Lehmann, D., Skrandies, W., 1980. Reference-free identification of components of checkerboard-evoked multichannel potential fields. Electroencephalogr. Clin. Neurophysiol. 48, 609–621. Mu, Z., Chang, Y., Xu, J., Pang, X., Zhang, H., Liu, X., Zheng, Y., Liu, X., Liu, X., Wan, Y., 2016. Pre-attentive dysfunction of musical processing in major depressive disorder: a mismatch negativity study. J. Affect. Disord. 194, 50–56. Näätänen, R., 1990. The role of attention in auditory information processing as revealed by event-related potentials and other brain measures of cognitive function. Behav. Brain Sci. 13, 201–288. Näätänen, R., Gaillard, A.W., Mäntysalo, S., 1978. Early selective-attention effect on evoked potential reinterpreted. Acta Psychol. (Amst. ) 42, 313–329. Ogura, C., Nageishi, Y., Omura, F., 1995. Reduced mismatch negativity and
4.1. Limitations Limitations of the current study include the fact that we could not exclude any effects of the antidepressants, mood stabilizers, or neuroleptics on the MMNm abnormalities found in the patients with major depression. Cross-sectional studies with a more homogenous patient group (drug-free vs medicated), as well as studies assessing participants before and after treatment with specific medications, are required in the future. Second, although no MMNm variables were significantly correlated with age, in the current study we did not exclude any aging effects of MMNm in the findings of the current study, since it was reported that MMN amplitudes declined with age (e.g., Kiang et al., 2009). Finally, the sample size was also insufficient to permit subgroup analyses; it will be important to investigate the MMNm with a larger and more homogeneous sample of patients. In summary, the current study showed that in comparison with non-depressed normal volunteers, patients with major depression exhibited reduced GFP of MMNm in the right hemisphere and significantly earlier bilateral MMNm peak latencies. The present study suggested that patients with major depression may have right hemispheric dominant preattentive dysfunction. 228
Journal of Affective Disorders 215 (2017) 225–229
N. Hirakawa et al.
Interview for DSM-III-R (SCID-NP)–Non-Patient Edition. American Psychiatric Association, Washington, DC. Suga, M., Nishimura, Y., Kawakubo, Y., Yumoto, M., Kasai, K., 2016. Magnetoencephalographic recording of auditory mismatch negativity in response to duration and frequency deviants in a single session in patients with schizophrenia. Psychiatry Clin. Neurosci. 70, 295–302. Takei, Y., Kumano, S., Hattori, S., Uehara, T., Kawakubo, Y., Kasai, K., Fukuda, M., Mikuni, M., 2009. Preattentive dysfunction in major depression: a magnetoencephalography study using auditory mismatch negativity. Psychophysiology 46, 52–61. Takei, Y., Kumano, S., Maki, Y., Hattori, S., Kawakubo, Y., Kasai, K., Fukuda, M., Mikuni, M., 2010. Preattentive dysfunction in bipolar disorder: a MEG study using auditory mismatch negativity. Prog. Neuropsychopharmacol. Biol. Psychiatry 34, 903–912. Taulu, S., Simola, J., 2006. Spatiotemporal signal space separation method for rejecting nearby interference in MEG measurements. Phys. Med. Biol. 51, 1759–1768. Umbricht, D., Koller, R., Schmid, L., Skrabo, A., Grübel, C., Huber, T., Stassen, H., 2003. How specific are deficits in mismatch negativity generation to schizophrenia*. Biol. Psychiatry 53, 1120–1131. Wechsler, D., 1997. Wechsler Adult Intelligence Scale Third edition. The Psychological Corporation, San Antonio. Williams, J.B., 1988. A structured interview guide for the Hamilton Depression Rating Scale. Arch. Gen. Psychiatry 45, 742–747.
hyperactivation of N2b in depression. Electroencephalogr. Clin. Neurophysiol. Suppl. 44, 218–223. Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9, 97–113. Pang, X., Xu, J., Chang, Y., Tang, D., Zheng, Y., Liu, Y., Sun, Y., 2014. Mismatch negativity of sad syllables is absent in patients with major depressive disorder. PLoS One 21 (3), e91995. Qiao, Z., Yu, Y., Wang, L., Yang, X., Qiu, X., Zhang, C., Ning, N., Shi, J., Chen, L., Li, Z., Liu, J., Xu, J., Zhao, L., Yang, Y., 2013. Impaired pre-attentive change detection in major depressive disorder patients revealed by auditory mismatch negativity. Psychiatry Res. 211, 78–84. Qiao, Z., Yang, A., Qiu, X., Yang, X., Zhang, C., Zhu, X., He, J., Wang, L., Bai, B., Sun, H., Zhao, L., Yang, Y., 2015. Gender effect on pre-attentive change detection in major depressive disorder patients revealed by auditory MMN. Psychiatry Res. 234, 7–14. Restuccia, D., Vollono, C., Scaloni, L., Buccelletti, F., Camardese, G., 2016. Abnormality of auditory mismatch negativity in depression and its dependence on stimulus intensity. Clin. EEG Neurosci. 47, 105–112. Shimano, S., Onitsuka, T., Oribe, N., Maekawa, T., Tsuchimoto, R., Hirano, S., Ueno, T., Hirano, Y., Miura, T., Kanba, S., 2014. Preattentive dysfunction in patients with bipolar disorder as revealed by the pitch-mismatch negativity: a magnetoencephalography (MEG) study. Bipolar Disord. 16, 592–599. Spitzer, R.L., Williams, J.B.W., Gibbson, M., First, M., 1990. The Structured Clinical
229
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research paper
Right hemisphere pitch-mismatch negativity reduction in patients with major depression: An MEG study
MARK
Noriaki Hirakawaa, Yoji Hiranoa,b, Itta Nakamuraa,c, Shogo Hiranoa, Jinya Satoa, Naoya Oribea,d, ⁎ Takefumi Uenoa,d, Shigenobu Kanbaa, Toshiaki Onitsukaa, a
Department of Neuropsychiatry, Graduate School of Medical Sciences, Kyushu University, Fukuoka, Japan Department of Psychiatry, Harvard Medical School and Neural Dynamics Laboratory, VA Boston Healthcare System, Boston, MA, USA c Division of Clinical Research, National Hospital Organization, Kokura Medical Center, Fukuoka, Japan d Division of Clinical Research, National Hospital Organization, Hizen Psychiatric Center, Saga, Japan b
A R T I C L E I N F O
A BS T RAC T
Keywords: Major depression Magnetoencephalography Mismatch negativity Preattentive dysfunction
Background: The mismatch negativity (MMN) component of the event-related potential and its magnetic counterpart, the MMNm, are generated by a mismatch between the physical features of a deviant stimulus and a neuronal sensory-memory trace produced by repetitive standard stimuli. Deficits in the MMN/MMNm have been reported in patients with major depression; however, the results are inconsistent. The present study investigated the pitch-MMNm in patients with major depression using whole-head 306-channel magnetoencephalography (MEG). Methods: Twenty patients with major depression and 36 healthy subjects participated in this study. Subjects were presented with two sequences of auditory stimuli. One consisted of 1000 Hz standard signals (probability=90%) and 1200 Hz deviant signals (probability=10%), while the other consisted of 1200 Hz standard (90%) and 1000 Hz deviant signals (10%). Event-related brain responses to standard tones were subtracted from responses to deviant tones. Results: Major depressive patients showed significantly reduced magnetic global field power (GFP) of MMNm in the right hemisphere (p=0.02), although no significant MMNm reduction was observed in the left hemisphere (p=0.81). Additionally, patients with major depression showed significantly earlier bilateral MMNm peak latencies (p=0.004). No significant associations were observed between MMNm variables and demographic data/clinical variables within the patients. Limitations: We could not exclude the effects of antidepressants, mood stabilizers, or neuroleptics on the MMNm abnormalities found in patients with major depression. Sample size was also insufficient to permit subgroup analyses. Conclusions: Patients with major depression exhibited reduced GFP of MMNm in the right hemisphere. The present study suggested that patients with major depression may have right hemispheric dominant preattentive dysfunction.
1. Introduction Mismatch negativity (MMN) is a preattentive auditory eventrelated potential elicited when a sequence of repetitive sounds is interrupted by a deviant sound (Näätänen et al., 1978). MMN waveforms can be obtained by subtracting the event-related response to a repeated regular event (standard stimulus) from that of an oddball or deviant event (deviant stimulus). Näätänen (1990) proposed that the MMN is generated by a mismatch between the physical features of a deviant stimulus and the neuronal sensory-memory trace produced by
⁎
repetitive standard stimuli. MMN can be elicited by changes in the frequency, duration, intensity, and location of a stimulus. MMN abnormalities are frequently reported in patients with schizophrenia (e.g., Suga et al., 2016). Deficits in the MMN component have been also reported in patients with major depression; however, the results are inconsistent and mixed. For example, Qiao et al. (2013) reported significant duration-MMN reductions in patients with major depression. Chen et al. (2015) also demonstrated significant decreased duration-MMN in first-episode and recurrent major depression. In contrast, Umbricht et al. (2003) showed no significant duration-MMN
Correspondence to: Department of Neuropsychiatry, Graduate School of Medical Sciences, Kyushu University, 3-1-1 Maidashi, Higashiku, Fukuoka 812-8582, Japan. E-mail address: [email protected] (T. Onitsuka).
http://dx.doi.org/10.1016/j.jad.2017.03.046 Received 27 December 2016; Received in revised form 13 March 2017; Accepted 17 March 2017 Available online 19 March 2017 0165-0327/ © 2017 Elsevier B.V. All rights reserved.
Journal of Affective Disorders 215 (2017) 225–229
N. Hirakawa et al.
sessions was counterbalanced across subjects. The inter-stimulus interval was a constant 400 ms (offset-to-onset). The stimuli were 80-dB sound-pressure-level sinusoidal tone bursts with a duration of 100 ms (including a linear rise/fall time of 10 ms).
differences between patients with major depression and healthy subjects. Qiao et al. (2015) reported that significant duration-MMN reductions were observed only in female subjects with major depression. Pang et al. (2014) presented that MMN of sad syllable stimuli was absent in patients. Electroencephalography (EEG) is widely used to measure MMN; it has a high temporal resolution, and most of the reports on MMN deficits have been based on EEG studies. Magnetoencephalography (MEG) also has a high temporal resolution, and detects the magnetic counterpart of the EEG MMN (the MMNm) with a higher spatial resolution. The EEG MMN is usually distributed over fronto-central scalp locations, with maximum amplitudes found at the fronto-central electrode positions (Garrido et al., 2009). MEG is superior for the investigation of the laterality of the MMNm, as MEG can accurately detect neural activities tangential to the surface of the cortex for both hemispheres, with minimal influence from conductivity through tissues such as brain tissue, skull, and scalp (Hämäläinen et al., 1993). As far as we are aware, there are currently only two studies on MEG measurement of the MMNm in major depression (Takei et al., 2009; Kähkönen et al., 2007) with conflicting results. Takei et al. (2009) investigated the MMNm associated with deviant pitch, duration, and vowel stimuli in 14 patients with major depression. The authors found a significant bilateral overall MMNm reduction in the patients. However, Kähkönen et al. (2007) reported no significant difference in pitch-MMNm amplitudes between patients with major depression and healthy subjects. The present MEG study investigated the pitchMMNm, which may lead to a better understanding of the neurophysiological feature of major depression.
2.3. Data acquisition and processing MEG signals were acquired using a whole-head, 306-channel sensor array (Vectorview; ELEKTA Neuromag, Helsinki, Finland) that consisted of 102 identical triple-sensor elements. Each sensor element consisted of two orthogonal planar-type gradiometers and one magnetometer. Prior to the recording, four head position indicator (HPI) coils were attached to the scalp and a three-dimensional (3D) digitizer was used to measure the anatomical landmarks of the head with respect to the HPI coils. During the recording in a magnetically shielded room, the subjects were instructed to sit relaxed and read a book with their heads positioned inside the helmet-shaped sensor array. The precise location of the head with respect to the sensor array was determined using the HPI coils. A bandpass filter of 0.01–30 Hz and a sampling rate of 1 kHz were used for recording. During the stimulation, the acquired raw data was stored for off-line analysis. The data were averaged separately for standard and deviant stimuli. The entire sweep time was 450 ms (100 ms before stimulus onset to 350 ms after stimulus onset). A spatiotemporal signal space separation (tSSS) method was applied to the recorded raw data off-line (Taulu and Simola, 2006), with signal variations exceeding 1500 femtotesla being excluded. The Final MMNm responses were calculated using the epochs of both sessions.
2. Methods
2.4. Data analysis
2.1. Subjects
2.4.1. Magnetic global field power of the MMNm The data obtained from the 54 channels positioned over each temporal region were analyzed according to the methods of previous studies (Shimano et al., 2014, see Fig. 1). The magnetic global field power (mGFP) of responses was calculated separately for each hemisphere, using the following formula (Takei et al., 2010, 2009; Kreitschmann-Andermahr, 1999; Lehmann and Skrandies, 1980):
Twenty patients with major depression and 36 healthy subjects participated in this study. All subjects had normal hearing, were aged 21–60 years, and were right-handed as determined by the Edinburgh Inventory (Oldfield, 1971). All participants signed an informed consent form after being given a complete description of the study, which was approved by the Kyushu University Institutional Review Board for Clinical Trials. The exclusion criteria were: (i) neurological illness or major head trauma that could result in abnormal EEG findings; (ii) electroconvulsive therapy; (iii) alcohol or drug dependence during the subject's lifetime; (iv) alcohol or drug abuse within the past five years; and (v) a verbal intelligence quotient below 75 (Wechsler, 1997). Healthy subjects were screened using the Structured Clinical Interview (SCID) non-patient edition. None of the healthy subjects or their firstdegree relatives had an Axis I psychiatric disorder (Spitzer et al., 1990). All patients were recruited from the Kyushu University Hospital, and were diagnosed with major depression on the basis of the SCID for DSM-IV Axis I disorders and the patient's medical records (First et al., 1996). Two senior clinical psychiatrists confirmed that all subjects had the ability to consent to participate in the examination. The socioeconomic statuses (SES) of the subjects and their parents were measured using the Hollingshead Two-Factor Index (Hollingshead, 1965). The severity of mood symptoms was assessed using the Structured Interview Guide for the Hamilton Depression Rating Scale (SIGH-D; Williams, 1988).
mGFP(ft/cm) =
2 n n ⎛ ∑ j =1 Uj ⎞ 1 ⎜ ⎟ ⋅ ∑ Ui − ⎟ , n i =1 ⎜⎝ n ⎠
where n is the number of channels and U is the amplitude for each channel. The variable global field power (GFP) corresponds to the spatial standard deviation, and quantifies the amount of neural activity at each time-point. First, GFP waveforms to standard and deviant stimuli were
2.2. Stimuli and procedure Subjects were presented with two sessions of 1000 stimuli. These consisted of two types of auditory stimuli sequences, administered binaurally through earphones in a random order. One sequence consisted of 1000 Hz standard (probability =90%) and 1200 Hz deviant stimuli (probability =10%), while the other consisted of 1200 Hz standard (90%) and 1000 Hz deviant stimuli (10%). The order of
Fig. 1. Layout of the measurement channels. The MEG signals were acquired using a whole-head 306-channel sensor array that consisted of 102 identical triple-sensor elements. Each sensor consisted of two orthogonal planar-type gradiometers and one magnetometer. We used 27 sensors for each hemisphere (a 54-channel orthogonal gradiometer). The two circles indicate the sensors used for the analyses.
226
Journal of Affective Disorders 215 (2017) 225–229
N. Hirakawa et al.
calculated for each hemisphere with 27 channels. Second, MMNm responses were obtained by subtracting the averaged responses to standard stimuli from those of deviant stimuli for each channel, and then GFP waveforms of MMNm was finally calculated using 27 channels for each hemisphere. The responses were digitally filtered using a bandpass range of 1–20 Hz. The peak amplitude of the MMNm component was defined as the largest amplitude of an individual GFP between 90 and 250 ms, and the MMNm peak latency was defined as the latency with this largest amplitude.
Table 2 mGFP latencies and amplitudes of the MMNm.
HS (n=36) Mean (SD)
MDD (n=20) Mean (SD)
mGFP latency (ms)
left
left
138.19 (21.85) 137.17 (25.20) 18.48 (8.52)
right
23.03 (11.63)
right mGFP power (ft/ cm)
2.5. Statistical analyses
df
t
p-value
125.10 (18.65) 119.10 (18.40) 19.02 (6.49)
54
2.26
0.03
54
2.81
0.01
54
−0.25
0.81
17.45 (5.75)
54
2.40
0.02
HS = healthy subjects; MDD = major depressive disorders; mGFP = magnetic global field power; MMNm = mismatch negativity magnetic counterpart.
Independent samples t-tests were used to assess group differences in age, handedness score, SES, parental SES, education years, and sleeping-scale score. The GFP peak latencies and powers of the MMNm were analyzed using repeated-measures analysis of covariates (ANCOVA) with group (depression or control) as a between-subject factor, hemisphere (left or right) as a within-subject factor, and age as a covariate, as the patients with major depression were significantly older than the healthy subjects (t[54] =−2.67, p=0.01). Degrees of freedom were adjusted using the Huynh–Feldt epsilon for factors with more than two levels. Exploratory analyses of the relationships between MMNm variables and the demographic/clinical data were evaluated using Pearson correlations with a Bonferroni correction for multiple correlations (16 correlations for healthy subjects and 100 correlations for patients with major depression).
GFP peak latency of the MMNm, the repeated-measures ANCOVA showed a significant main effect of group [F(1,54) =9.2, p=0.004], but no significant main effect of hemisphere [F(1,54) =0.35, p=0.56], and no significant hemisphere-by-group interaction [F (1,54) =0.20, p=0.66]. This indicates that patients with major depression showed significantly earlier MMNm peak latencies. For the GFP of the MMNm, a repeated-measures ANCOVA showed no significant main effects of group [F(1,54) =0.19, p=0.67] or hemisphere [F(1,54) =0.54, p=0.47]. However, there was a significant hemisphere-by-group interaction [F(1,54) =4.4, p=0.041]. Thus, post hoc t-tests were performed for each hemisphere. In the left hemisphere, there was no significant group difference (t[54] =−0.25, p=0.81), but in the right hemisphere, the patients with major depression showed significantly reduced GFP of MMNm (t[54] =2.4, p=0.02).
3. Results 3.2.1. Correlations between the MMNm variables and demographic/ clinical data In healthy subjects, no significant correlations were observed between the MMNm variables and the demographic data (−0.41≤r < 0.51; 0.002≤p≤0.73). In the patients with depression, there were no significant correlations between the MMNm variables and the demographic data or SIGH-D scores (−0.54 ≤ r ≤0.44; 0.02 ≤ p ≤0.98). Additionally, antidepressant dose (imipramine equivalent) did not significantly correlate with any of the MMNm variables in the patients with major depression (−0.14≤r ≤0.10; 0.55≤p ≤0.97).
3.1. Demographic characteristics Demographic and clinical data for all subjects are presented in Table 1. There were no significant intergroup differences in handedness, SES, parental SES, and years of education. According to the criteria used to rate depression severity (Furukawa et al., 2007), at the time of the study, two patients showed severe depression, three showed moderate depression, and eight showed mild depression. Seven further patients demonstrated a euthymic mood. Seventeen patients were receiving antidepressants, five were receiving mood stabilizers such as lithium carbonate, valproate, and lamotrigine, and eleven patients were receiving neuroleptic medication.
4. Discussion The present study examined the pitch-MMNm in patients with major depression. In comparison with non-depressed subjects, these patients showed significantly reduced MMNm amplitudes in the right hemisphere, although no significant MMNm reduction was observed in the left hemisphere. Additionally, patients with major depression showed significantly earlier bilateral MMNm peak latencies. With respect to previous findings on pitch-MMN and MMNm in general, there are various reports and no consensus has been formed. As noted in the introduction section, one MEG study reported significant bilateral MMNm reductions (Takei et al., 2009), whereas another study failed to demonstrate MMNm reductions in patients with major depression (Kähkönen et al., 2007). The previous results from EEG studies are also inconsistent. Ogura et al. (1995) found significant MMN reductions in patients with major depression. Umbricht et al. (2003) and Mu et al. (2016) reported no significant MMN differences compared to HS. However, He et al. (2010) detected significantly larger MMN in patients with major depression. Restuccia et al. (2016) reported that significantly increased MMN can be observed in patients by using hiintensity (90 dB) stimulation. Our MEG study came to a certain conclusion that patients with major depression may have right hemispheric dominant preattentive dysfunction, The MMNm may be useful for differentiating between major depression and bipolar disorder. Our group has previously reported
3.2. Magnetic global field of MMNm Table 2 shows the GFP and latencies of the MMNm components. Fig. 2 illustrates the group mean waveforms of the MMNm. For the Table 1 Demographic and clinical characteristics of the study participants.
Sex, male/female, n Age, years Handedness SES Parental SES Education, years Duration of illness, years Imipramine equivalent, mg SIGH-D
HS Mean (SD)
MDD Mean (SD)
df
t
p-value
17/19 34.06 (13.39) 97.78 (5.79)
9/11 43.05 (9.14)
54
−2.67
0.01
54
1.13
0.27
54 54 54
−1.88 −0.36 1.43
0.066 0.72 0.16
2.19 (0.79) 2.67 (0.93) 15.00 (2.35)
93.50 (21.59) 2.60 (0.75) 2.75 (0.64) 14.10 (2.05) 9.30 (6.20) 172.5 (114.6) 9.60 (5.83)
HS = healthy subjects; MDD = major depressive disorders; SES = socio-economic status; SIGH-D = Structured Interview Guide for the Hamilton Depression Rating Scale.
227
Journal of Affective Disorders 215 (2017) 225–229
N. Hirakawa et al.
Fig. 2. Grand mean magnetic global field power (GFP) waveforms of responses to standard stimuli, deviant stimuli, and mismatch fields (MMNm) in patients with major depressive disorder (MDD) [solid line] and healthy subjects (HS) [dotted line]. Left: left hemisphere, Right: right hemisphere.
References
that patients with bipolar disorder showed a significant bilateral MMNm reduction (Shimano et al., 2014), while this study found such a reduction only in the right hemisphere of patients with major depression. Takei et al. (2010) reported significantly delayed MMNm peak latencies in the right hemispheres of patients with bipolar disorder, although our group reported no significant latency changes. In the present study, patients with major depression exhibited significantly earlier bilateral MMNm latencies. The results of our study suggest that the MMNm has potential as a biomarker for the differentiation between major depression and bipolar disorder.
Chen, J., Zhang, Y., Wei, D., Wu, X., Fu, Q., Xu, F., Wang, H., Ye, M., Ma, W., Yang, L., Zhang, Z., 2015. Neurophysiological handover from MMN to P3a in first-episode and recurrent major depression. J. Affect. Disord. 174, 173–179. First, M.B., Spitzer, R.L.S., Gibbon, M., Williams, J.B.W., 1996. Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version (SCID-CV). American Psychiatric Press Inc, Washington, DC. Furukawa, T.A., Akechi, T., Azuma, H., Okuyama, T., Higuchi, T., 2007. Evidence-based guidelines for interpretation of the Hamilton Rating Scale for Depression. J. Clin. Psychopharmacol. 27, 531–534. Garrido, M.I., Kilner, J.M., Stephan, K.E., Friston, K.J., 2009. The mismatch negativity: a review of underlying mechanisms. Clin. Neurophysiol. 120, 453–463. Hämäläinen, M., Hari, R., Ilmoniemi, R.J., Knuutila, J., Lounasmaa, O.V., 1993. Magnetoencephalography - theory, instrumentation, and applications to noninvasive studies of the working human brain. Rev. Mod. Phys. 65, 413–497. He, W., Chai, H., Zheng, L., Yu, W., Chen, W., Li, J., Chen, W., Wang, W., 2010. Mismatch negativity in treatment-resistant depression and borderline personality disorder. Prog. Neuropsychopharmacol. Biol. Psychiatry 34, 366–371. Hollingshead, A., 1965. Two Factor Index of Social Position. Yale University Press, New Haven, CT. Kähkönen, S., Yamashita, H., Rytsälä, H., Suominen, K., Ahveninen, J., Isometsä, E., 2007. Dysfunction in early auditory processing in major depressive disorder revealed by combined MEG and EEG. J. Psychiatry Neurosci. 32, 316–322. Kiang, M., Braff, D.L., Sprock, J., Light, G.A., 2009. The relationship between preattentive sensory processing deficits and age in schizophrenia patients. Clin. Neurophysiol. 120, 1949–1957. Kreitschmann-Andermahr, I., Rosburg, T., Meier, T., Volz, H.P., Nowak, H., Sauer, H., 1999. Impaired sensory processing in male patients with schizophrenia: a magnetoencephalographic study of auditory mismatch detection. Schizophr. Res. 35, 121–129. Lehmann, D., Skrandies, W., 1980. Reference-free identification of components of checkerboard-evoked multichannel potential fields. Electroencephalogr. Clin. Neurophysiol. 48, 609–621. Mu, Z., Chang, Y., Xu, J., Pang, X., Zhang, H., Liu, X., Zheng, Y., Liu, X., Liu, X., Wan, Y., 2016. Pre-attentive dysfunction of musical processing in major depressive disorder: a mismatch negativity study. J. Affect. Disord. 194, 50–56. Näätänen, R., 1990. The role of attention in auditory information processing as revealed by event-related potentials and other brain measures of cognitive function. Behav. Brain Sci. 13, 201–288. Näätänen, R., Gaillard, A.W., Mäntysalo, S., 1978. Early selective-attention effect on evoked potential reinterpreted. Acta Psychol. (Amst. ) 42, 313–329. Ogura, C., Nageishi, Y., Omura, F., 1995. Reduced mismatch negativity and
4.1. Limitations Limitations of the current study include the fact that we could not exclude any effects of the antidepressants, mood stabilizers, or neuroleptics on the MMNm abnormalities found in the patients with major depression. Cross-sectional studies with a more homogenous patient group (drug-free vs medicated), as well as studies assessing participants before and after treatment with specific medications, are required in the future. Second, although no MMNm variables were significantly correlated with age, in the current study we did not exclude any aging effects of MMNm in the findings of the current study, since it was reported that MMN amplitudes declined with age (e.g., Kiang et al., 2009). Finally, the sample size was also insufficient to permit subgroup analyses; it will be important to investigate the MMNm with a larger and more homogeneous sample of patients. In summary, the current study showed that in comparison with non-depressed normal volunteers, patients with major depression exhibited reduced GFP of MMNm in the right hemisphere and significantly earlier bilateral MMNm peak latencies. The present study suggested that patients with major depression may have right hemispheric dominant preattentive dysfunction. 228
Journal of Affective Disorders 215 (2017) 225–229
N. Hirakawa et al.
Interview for DSM-III-R (SCID-NP)–Non-Patient Edition. American Psychiatric Association, Washington, DC. Suga, M., Nishimura, Y., Kawakubo, Y., Yumoto, M., Kasai, K., 2016. Magnetoencephalographic recording of auditory mismatch negativity in response to duration and frequency deviants in a single session in patients with schizophrenia. Psychiatry Clin. Neurosci. 70, 295–302. Takei, Y., Kumano, S., Hattori, S., Uehara, T., Kawakubo, Y., Kasai, K., Fukuda, M., Mikuni, M., 2009. Preattentive dysfunction in major depression: a magnetoencephalography study using auditory mismatch negativity. Psychophysiology 46, 52–61. Takei, Y., Kumano, S., Maki, Y., Hattori, S., Kawakubo, Y., Kasai, K., Fukuda, M., Mikuni, M., 2010. Preattentive dysfunction in bipolar disorder: a MEG study using auditory mismatch negativity. Prog. Neuropsychopharmacol. Biol. Psychiatry 34, 903–912. Taulu, S., Simola, J., 2006. Spatiotemporal signal space separation method for rejecting nearby interference in MEG measurements. Phys. Med. Biol. 51, 1759–1768. Umbricht, D., Koller, R., Schmid, L., Skrabo, A., Grübel, C., Huber, T., Stassen, H., 2003. How specific are deficits in mismatch negativity generation to schizophrenia*. Biol. Psychiatry 53, 1120–1131. Wechsler, D., 1997. Wechsler Adult Intelligence Scale Third edition. The Psychological Corporation, San Antonio. Williams, J.B., 1988. A structured interview guide for the Hamilton Depression Rating Scale. Arch. Gen. Psychiatry 45, 742–747.
hyperactivation of N2b in depression. Electroencephalogr. Clin. Neurophysiol. Suppl. 44, 218–223. Oldfield, R.C., 1971. The assessment and analysis of handedness: the Edinburgh inventory. Neuropsychologia 9, 97–113. Pang, X., Xu, J., Chang, Y., Tang, D., Zheng, Y., Liu, Y., Sun, Y., 2014. Mismatch negativity of sad syllables is absent in patients with major depressive disorder. PLoS One 21 (3), e91995. Qiao, Z., Yu, Y., Wang, L., Yang, X., Qiu, X., Zhang, C., Ning, N., Shi, J., Chen, L., Li, Z., Liu, J., Xu, J., Zhao, L., Yang, Y., 2013. Impaired pre-attentive change detection in major depressive disorder patients revealed by auditory mismatch negativity. Psychiatry Res. 211, 78–84. Qiao, Z., Yang, A., Qiu, X., Yang, X., Zhang, C., Zhu, X., He, J., Wang, L., Bai, B., Sun, H., Zhao, L., Yang, Y., 2015. Gender effect on pre-attentive change detection in major depressive disorder patients revealed by auditory MMN. Psychiatry Res. 234, 7–14. Restuccia, D., Vollono, C., Scaloni, L., Buccelletti, F., Camardese, G., 2016. Abnormality of auditory mismatch negativity in depression and its dependence on stimulus intensity. Clin. EEG Neurosci. 47, 105–112. Shimano, S., Onitsuka, T., Oribe, N., Maekawa, T., Tsuchimoto, R., Hirano, S., Ueno, T., Hirano, Y., Miura, T., Kanba, S., 2014. Preattentive dysfunction in patients with bipolar disorder as revealed by the pitch-mismatch negativity: a magnetoencephalography (MEG) study. Bipolar Disord. 16, 592–599. Spitzer, R.L., Williams, J.B.W., Gibbson, M., First, M., 1990. The Structured Clinical
229