Neuroscience Letters 448 (2008) 24–28
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/neulet
Mismatch negativity to the patient’s own name in chronic disorders
Pengmin Qina, Haibo Dib, Xiaodan Yana, Senming Yuc, Dan Yuc, Steven Laureysd, Xuchu Wenga,∗
aLaboratory for Higher Brain Functions, Institute of Psychology, Chinese Academy of Sciences, 4A Datun Road, Chaoyang District, Beijing 100101, China
bHangzhou Normal University, Hangzhou, China
cRehabilitation Center for Brain Damage, Wujing Hospital of Hangzhou City, Hangzhou, China
dComa Science Group, Cyclotron Research Centre and Neurology Department, University of Liège, Liège, Belgium
a r t i c l e i n f o
Received 26 June 2008
Received in revised form 3 October 2008
Accepted 7 October 2008
Disorders of consciousness
Minimally conscious state
a b s t r a c t
Previous studies implicated potential value of mismatch negativity (MMN) in predicting recovery of con-
sciousness in patients with disorders of consciousness (DOC). We have adopted a novel MMN evoked by
subject’s own name (SON), a self-referential stimulus thought to be powerful in evoking residual brain
activity, and examined the correlation between the MMN and recovery of consciousness in patients with
chronic (>1 month) DOC. Twelve patients and 12 age-matched healthy controls were investigated. The
patients were diagnosed as coma (n=4), vegetative state (VS, n=6), and minimally conscious state (MCS,
n=2), mainly based on the JFK Coma Recovery Scale-Revised. The SON-evoked MMN (SON-MMN) was
present in seven patients. Critically, the presence of SON-MMN was significantly correlated with recovery
of consciousness. While four of the five patients (three VS and two coma) showing SON-MMN changed to
MCS 3 months later, the rest of the patients (three VS and two coma) without SON-MMN failed to show
any clinical improvement. Our study thus illustrates that the subject’s own name is effective in evoking
MMN in patients with DOC, and that SON-MMN has potential prognostic values in predicting recovery of
© 2008 Elsevier Ireland Ltd. All rights reserved.
Self-referential stimuli such as the subject’s own name (SON) are
thought to be more effective to induce brain activities in patients
with disorders of consciousness (DOC) compared to other stim-
ulation types . The SON could evoke cognitive event-related
potential (ERP) waveforms [11,16,18,21] and induce activation in
widespread brain areas [4,15,22] in patients with DOC who remain
In our recent fMRI study, we have showed that the SON-induced
brain activation seems to be highly correlated with the recovery
of consciousness in VS patients . However, the availability and
feasibility of fMRI is relatively poor as compared to ERP technique,
Moreover, this technique can be used at the bedside, which is par-
ticularly significant for clinical use on patients with severe brain
injury. Among the ERP components, mismatch negativity (MMN)
has been given special attention in the field. For example, MMN
evoked by non-self-related stimuli is shown to be able to pre-
dict awakening in acute coma . This MMN also seems to have
∗Corresponding author. Tel.: +86 10 64854476; fax: +86 10 64872070.
E-mail address: email@example.com (X. Weng).
potential value in predicting clinical improvement in patients with
chronic DOC [13,25]. Nonetheless, the prognostic value of MMN
evoked by self-referential stimuli has not yet been properly evalu-
ated. Based on our previous study  and those of others , we
predict that the presence of SON-evoked MMN (SON-MMN) will
MMN is regarded as an indicator of pre-attentive sensory mem-
ory processes. The conventional MMN is obtained by subtracting
of MMN is prone to both physical properties and neural refractori-
ness, in the present study we have employed a protocol designed
to minimize the effects of these factors similar to Jacobsen and
Schroger . In the protocol, the brain response to a deviant stim-
ulus in the Oddball block (consisting standard and deviant stimuli)
is compared with the brain response to the same stimulus in the
Control blocks (i.e., the stimulus has the same probability as the
deviant in the Oddball block).
We studied 12 patients with chronic (>1 month) DOC (Table 1)
and 12 age-matched healthy controls (mean±S.D., 42±10 years;
rological examinations and the Glasgow-Liège Scale (GLS)  (i.e.,
0304-3940/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved.
P. Qin et al. / Neuroscience Letters 448 (2008) 24–28
Patients’ demographic and clinical data.Patient number/gender/age
Time of ERP recording
(months after insult)
Bilateral occipital and cerebellar lesions, ex-vacuo hydrocephalus
Right frontal, temporal and parietal lesions, ex-vacuo hydrocephalus
Right parietal and temporal hemorrhage
Bilateral frontal and temporal subdural collections
Bilateral frontal and temporal contusions, ex-vacue hydrocephalus
atrophy with ex-vacuo hydrocephalus
Right temporoparietal edema and hemorrhage
Bilateral temporal and right frontal cortex lesions
Right frontal, temporal and thalamic lesions
Right frontal and left temporal, occipital, parietal cortex lesions
Bilateral temporal and frontal cortex lesions
GLS (Glasgow-Liège Scale): E (eye response); V (verbal); M (motor) are scored as in the Glasgow Coma Scale (GCS); R (brainstem reflexes: score of 5 indicates preserved fronto-orbicular, 4 indicates vertical oculovestibular). T:
the Glasgow Coma Scale  and standardized evaluation of brain-
stem reflexes). Clinical diagnoses of VS and MCS were based on
the JFK Coma Recovery Scale-Revised (CRS-R), which comprises
six subscales addressing auditory, visual, motor, oromotor, com-
munication, and arousal functions, respectively . Accordingly,
the patients were diagnosed as coma (n=4), VS (n=6), and MCS
(n=2), at the time of ERP recording. Informed written consents
were obtained from the healthy participants and the families of the
patients, and the study was approved by the local Ethics Commit-
tees of the Institute of Psychology at Chinese Academy of Sciences
and Zhejiang University School of Medicine.
There were two block conditions in the ERP recording: the Odd-
ball block and the Control block. In the Oddball block, SONs, called
by the same female voice, served as deviant stimuli (12.5% of all
stimuli), and tones of 800Hz as standard stimuli. The mean dura-
tion of SONs was 637±71ms, while tones of 800Hz had a duration
of 30ms. The tones were presented binaurally at an 85dB and the
SONs at an 80dB sound pressure level maximal intensity. The Con-
trol block included SONs and seven control stimuli (three names of
other persons and four sine tones in different frequencies, 12.5% for
each type of stimuli). There were 720 events in each block. Stim-
ulus onset asynchrony (SOA) was randomized between 1100 and
1300ms. In the Oddball block, deviants occurred in a pseudoran-
in immediate successions, and stimuli in the Control block were
counter-balanced across the subjects.
Healthy controls sat in a comfortable chair and were instructed
to watch a self-selected silent movie on a computer during the
experiment. For patients, data was acquired at the bedside. ERPs
were recorded at Fz, Cz, and Pz (according to the International
Ten-Twenty System) (a reference was put on the nose) using
a Neuroscan NuAmps Amplifier with Ag/AgCl electrodes. Data
was sampled at 1000Hz with an analog bandpass of 0.1–100Hz.
Impedance of each electrode was kept below 5k? and 50Hz was
notched. The electro-oculogram was recorded from two pairs of
electrodes, one above and below the right eye and the other on the
outer canthi of the two eyes. A ground electrode was placed on the
middle of the forehead.
EOG artifacts were corrected using the method proposed by
Semlitsch et al. . All continuous EEG data was digitally filtered
with a low pass filter at 30Hz for N100. All continuous EEG data
was digitally filtered with a bandpass of 3–30Hz for MMN . The
uli onset, and included a pre-stimulus period of 100ms. All epochs
including voltage changes exceeding ±50?V were automatically
rejected and were averaged according to same stimuli type.
We identified two ERP components: N100 evoked by each stim-
ulus type and the mismatch negativity evoked by subject’s own
name (denoted as SON-MMN), which was computed by subtract-
SON in the Oddball block. The time window appropriate for N100
was defined on the basis of inspection of the average waveform
(with peak within 100–200ms after stimulus onset), and the mean
amplitude in this window was measured in each single trial, sep-
arately at Fz and Cz . One sample t-test then was used. If the
Bonferroni correction), N100 was judged as being present. For the
SON-MMN, the mean amplitude in each trial of SON in both blocks
was measured in an appropriate time window, which was defined
on the basis of inspection of the difference waveform (with peak
within 100–400ms after stimulus onset), respectively, at Fz and Cz.
The amplitude difference of the Oddball block and Control block
at Fz or Cz was tested with an independent sample t-test. If the
mean amplitude of the Oddball block was significantly different
P. Qin et al. / Neuroscience Letters 448 (2008) 24–28
Fig. 1. Grand average event-related potentials at three midline electrodes. ERPs to SON in the Oddball block and in the Control block, and MMN differences in the control
ence corresponded to the conventional MMN, then the SON-MMN
was judged as being present. The significance level was set at 0.05
(without Bonferroni correction). As the latency of MMN in patients
was more delayed than in controls , an extended time window
ness of the two ERP components, the patients’ conscious state was
assessed again by CRS-R 3 months after ERP recording (6 months
after acute brain injury), and correlation between the transition of
mined using Fisher’s tests. Specifically, one variable was whether
the patients emerged from VS (or coma) to MCS 3 months after ERP
recording, and the other was whether the presence or absence of
the ERP components. The data were analyzed using SPSS version
13.0 (SPSS Inc., Chicago, IL).
We found that N100 and SON-MMN were present in each of the
was 186±30ms (mean±S.D.) and −3.06±0.88?V (mean±S.D.)
In contrast, SON-MMN was present only in seven (two coma,
three VS and two MCS) out of the 12 patients (Fig. 2). The ampli-
tude and latency of SON-MMN in the patients were 216±67ms
(mean±S.D.) and −2.80±1.23?V (mean±S.D.) (Cz), and they
was present in nine patients (three coma, four VS and two MCS)
(Table 2). All of the seven patients with SON-MMN also showed
N100, so did other two patients (patient 4 and patient 10) without
Three months after the ERP recording, we re-assessed patients’
conscious state. The data showed that six out of the seven patients
with the presence of SON-MMN were diagnosed as MCS (four
changed from coma or VS). None of the remaining five patients
without SON-MMN (three VS, two coma) showed any clinical
improvement in the 3 month follow-up (Table 2). The presence
of SON-MMN was significantly correlated with the transition from
N100 and such transition (p=0.20).
In this study, we obtained a reasonably high presence ratio
of both SON-MMN and N100 in the patients with DOC (SON-
Fig. 2. Individual event-related potentials: ERPs to SON in the Oddball block, ERPs to SON in the Control block and MMN differences are shown at Cz. SON-MMN is indicated
if significant (p<0.05).
P. Qin et al. / Neuroscience Letters 448 (2008) 24–28
Patient diagnosis and JFK CRS-R scores at the time of ERP recording and 3 months later, presence of ERPs.
Patients Time of ERP recording N100SON-MMN Three month after ERP recording
Clinical diagnosis JFK CRS-R scoreClinical diagnosis JFK CRS-R score
JFK Coma Recovery Scale-Revised (CRS-R) subscales: auditory function; visual function; motor function; oromotor/verbal function; communication; arousal. “+”, significant
at 0.05 level, “−”, not significant at 0.05 level.
MMN in 7/12, N100 in 9/12). Only SON-MMN showed a significant
prognostic value. Four out of five unconscious patients showing
a SON-MMN response subsequently recovered to be in MCS 3
months after ERP recording, while none of the five patients with-
out SON-MMN regained consciousness. The SON-MMN thus was
significantly correlated with transition from VS to MCS. In contrast,
N100 was a poor indicator of the recovery of consciousness. Four
out of the seven unconscious patients with N100 changed to MCS,
while the other three did not show any clinical improvement. One
dictor than MMN . Our current ERP results were in line with our
previous fMRI study showing that brain activation evoked by SON
could predict the outcome of patients with DOC. In our previous
study, subsequent recovery of consciousness was only correlated
with the SON-induced activation in the secondary auditory cortex
but not that in the primary auditory cortex.
The prognostic value of SON-MMN shown in the current study
stimulus-evoked MMN. However, the previous studies showed a
low presence ratio of MMN among patients. For example, Fischer
et al. recorded MMN evoked by sine tone in 25% of the patients
with DOC , whereas Kotchoubey et al. detected MMN in 16%
and 31% of their patients (evoked by sine tone and complex tone,
respectively) . The detection power was high in the present
study—SON-MMN was detected in 7 of 12 patients. Several rea-
in patients with DOC. First, SON is known to be a more potent
and attention-grabbing stimulus as illustrated by the own-name-
effect (for recent review see ). Second, the physical properties
of SON (deviant stimuli) and sine tones (standard stimuli) differ
in many aspects (such as duration and frequency) probably result-
ing in a comparatively larger amplitude of SON-MMN . Third,
familiarity of the SON could enhance the SON-MMN . Maybe so
could emotion . Future studies based on a larger sample size
are needed in order to directly compare SON-MMN with MMN
evoked by non-self-related stimuli for verifying such superiority
of SON-MMN statistically.
with DOC. The presence of SON-MMN in the present study is likely
to indicate that the patients retain the brain function that is impor-
tant for cognitive processing. The protocol in this study employed
one Control block to minimize the effects of physical properties
have a contextual effect  as compared to the targeted Oddball
condition. Compared with the vast number of different stimulation
features in the Oddball condition, the contextual effect during the
Control condition is negligible for SON-MMN since just a minimal
influence may be induced on the brain response to SON. This could
be verified by the results from our healthy subjects who showed
In conclusion, our study illustrates that the subject’s own name,
an emotion relevant self-referential stimulus, is effective to evoke
ERP MMN. The small sample size used in the present study pre-
vented us from making general claims about SON-MMN being a
reliable marker of self-consciousness, but our empirical data did
show potential prognostic values of this novel ERP waveform. Fur-
ther studies on a larger cohort of patients are warranted to confirm
the predictive value of SON-MMN in chronic DOC. Although the
direction would be valuable for the study of patients with DOC.
This work was partly supported by Chinese Ministry of Science
and Technology (No. 2007CB512300) and National Natural Science
and Nelson Klahr for their helpful comments on the manuscript.
 M. Beauchemin, L. De Beaumont, P. Vannasing, A. Turcotte, C. Arcand, P. Belin,
M. Lassonde, Electrophysiological markers of voice familiarity, Eur. J. Neurosci.
23 (2006) 3081–3086.
 J.D. Born, A. Albert, P. Hans, J. Bonnal, Relative prognostic value of best motor
response and brain stem reflexes in patients with severe head injury, Neuro-
surgery 16 (1985) 595–601.
low responsive patients outcome using event-related brain potentials: a meta-
analysis, Clin. Neurophysiol. 118 (2007) 606–614.
 H.B. Di, S.M. Yu, X.C. Weng, S. Laureys, D. Yu, J.Q. Li, P.M. Qin, Y.H. Zhu, S.Z.
Zhang, Y.Z. Chen, Cerebral response to patient’s own name in the vegetative
and minimally conscious states, Neurology 68 (2007) 895–899.
 C. Fischer, D. Morlet, P. Bouchet, J. Luaute, C. Jourdan, F. Salord, Mismatch
negativity and late auditory evoked potentials in comatose patients, Clin. Neu-
rophysiol. 110 (1999) 1601–1610.
ment characteristics and diagnostic utility, Arch. Phys. Med. Rehabil. 85 (2004)
 K.N. Goydke, E. Altenmuller, J. Moller, T.F. Munte, Changes in emotional tone
and instrumental timbre are reflected by the mismatch negativity, Brain Res.
Cogn. Brain Res. 21 (2004) 351–359.
 I. Holeckova, C. Fischer, M.H. Giard, C. Delpuech, D. Morlet, Brain responses
to a subject’s own name uttered by a familiar voice, Brain. Res. 1082 (2006)
 T. Jacobsen, J. Horvath, E. Schroger, S. Lattner, A. Widmann, I. Winkler, Pre-
attentive auditory processing of lexicality, Brain Language 88 (2004) 54–67.
 T. Jacobsen, E. Schroger, Measuring duration mismatch negativity, Clin. Neuro-
physiol. 114 (2003) 1133–1143.
 B. Kotchoubey, S. Lang, E. Herb, P. Maurer, N. Birbaumer, Reliability of brain
28 Download full-text
P. Qin et al. / Neuroscience Letters 448 (2008) 24–28
in: N.C. Moore, M.K. Arikan (Eds.), Brainwaves and Mind: Recent Advances,
Kjellberg, Inc., New York, 2004, pp. 75–80.
 B. Kotchoubey, S. Lang, E. Herb, P. Maurer, D. Schmalohr, V. Bostanov, N. Bir-
baumer, Stimulus complexity enhances auditory discrimination in patients
with extremely severe brain injuries, Neurosci. Lett. 352 (2003) 129–132.
 B. Kotchoubey, S. Lang, G. Mezger, D. Schmalohr, M. Schneck, A. Semmler, V.
Bostanov, N. Birbaumer, Information processing in severe disorders of con-
sciousness: vegetative state and minimally conscious state, Clin. Neurophysiol.
116 (2005) 2441–2453.
 S. Laureys, F. Perrin, S. Bredart, Self-consciousness in non-communicative
patients, Conscious. Cogn. 16 (2007) 722–741.
 S. Laureys, F. Perrin, M.E. Faymonville, C. Schnakers, M. Boly, V. Bartsch, S.
state, Neurology 63 (2004) 916–918.
evoked potentials in severe traumatic brain injury, Arch. Phys. Med. Rehabil. 82
 R. Naatanen, P. Paavilainen, T. Rinne, K. Alho, The mismatch negativity (MMN)
in basic research of central auditory processing: a review, Clin. Neurophysiol.
118 (2007) 2544–2590.
 F. Perrin, C. Schnakers, M. Schabus, C. Degueldre, S. Goldman, S. Bredart,
M.E. Faymonville, M. Lamy, G. Moonen, A. Luxen, P. Maquet, S. Laureys, Brain
response to one’s own name in vegetative state, minimally conscious state, and
locked-in syndrome, Arch. Neurol. 63 (2006) 562–569.
 A. Roye, T. Jacobsen, E. Schroger, Personal significance is encoded automatically
by the human brain: an event-related potential study with ringtones, Eur. J.
Neurosci. 26 (2007) 784–790.
 H.V. Semlitsch, P. Anderer, P. Schuster, O. Presslich, A solution for reliable and
valid reduction of ocular artifacts, applied to the P300 ERP, Psychophysiology
23 (1986) 695–703.
 M. Signorino, S. D’Acunto, F. Angeleri, P. Pietropaoli, Eliciting P300 in comatose
patients, Lancet 345 (1995) 255–256.
 W. Staffen, M. Kronbichler, M. Aichhorn, A. Mair, G. Ladurner, Selective brain
activity in response to one’s own name in the persistent vegetative state, J.
Neurol. Neurosurg. Psychiatry 77 (2006) 1383–1384.
 R. Takegata, E. Brattico, M. Tervaniemi, O. Varyagina, R. Naatanen, I. Winkler,
Preattentive representation of feature conjunctions for concurrent spatially
distributed auditory objects, Brain Res. Cogn. Brain Res. 25 (2005) 169–
 G. Teasdale, B. Jennett, Assessment of coma and impaired consciousness. A
practical scale, Lancet 2 (1974) 81–84.
 V.J. Wijnen, G.J. van Boxtel, H.J. Eilander, B. de Gelder, Mismatch negativity
predicts recovery from the vegetative state, Clin. Neurophysiol. 118 (2007)