Available online at www.sciencedirect.com

01 (2008) 320–330www.elsevier.com/locate/schres

Schizophrenia Research 1

Electrophysiological basis for the ability of olanzapine to improveverbal memory and functional outcome in patients with

schizophrenia: A LORETA analysis of P300

Yuko Higuchi a, Tomiki Sumiyoshi a,b,⁎, Yasuhiro Kawasaki a,b, Mie Matsui b,c,Hirofumi Arai b,d, Masayoshi Kurachi b,d

a Department of Neuropsychiatry, University of Toyama Graduate School of Medicine and Pharmaceutical Sciences, Toyama, Japanb Core Research for Evolutional Science and Technology, Japan Science and Technology Corporation, Tokyo, Japan

c Department of Psychology, University of Toyama Graduate School of Medicine and Pharmaceutical Sciences, Toyama, Japand Department of Psychiatric Early Intervention, University of Toyama Graduate School of Medicine and Pharmaceutical Sciences, Toyama, Japan

Received 1 October 2007; received in revised form 31 December 2007; accepted 10 January 2008Available online 5 March 2008

Abstract

Abnormality of P300 waveforms of event-related potentials (ERPs) has been suggested to represent an aspect of the pathophysiologyof schizophrenia. Previous work points to the contribution of altered neural function in discrete brain regions in the left hemisphere topsychotic symptoms and cognitive deficits of schizophrenia. In this study, we sought to determine: 1) if patients with schizophrenia elicita decreased P300 current source density in brain areas, such as the superior temporal gyrus (STG); 2) if decreased P300 generator densityin the left STG is recovered by treatment with the most widely-used antipsychotic drug olanzapine; and 3) if the recovery of P300 sourcedensity is associatedwith improvements of cognitive and functional status. P300 in response to an auditory oddball task, aswell as verballearning memory, psychopathology, and quality of life were evaluated in 16 right-handed patients with schizophrenia before and aftertreatment with olanzapine for 6 months. ERP data were also obtained from 16 right-handed age and gender-matched normal volunteers.Low resolution electromagnetic tomography (LORETA) analysis was used to obtain current density images of P300. Patients withschizophrenia showed significantly smaller LORETA values in several brain regions in the left side, particularly STG, middle frontalgyrus, and precentral gyrus, compared with control subjects. Six-month treatment with olanzapine significantly increased P300 sourcedensity only in the left STG. Positive symptoms, negative symptoms, verbal learning memory, and quality of life were also improvedduring treatment. Significant correlations were found between the increase in LORETAvalues of left STG vs. improvements of negativesymptoms, as measured by Scale for the Assessment of the Negative Symptoms, and verbal learning memory, as measured by theJapanese Verbal Learning Test. Improvement of quality of life, as evaluated by the Quality of Life Scale, were significantly associatedwith an increase in LORETAvalues of middle frontal gyrus, and tended to correlate with that of precentral gyrus. The results of this studysuggest that changes in cortical activity, as measured by ERPs, are responsible for the ability of some antipsychotic drugs to improvecognition and functional outcome in patients with schizophrenia.© 2008 Elsevier B.V. All rights reserved.

Keywords: Event-related potentials; LORETA; Cognition; Negative symptoms; Quality of Life; Superior temporal gyrus; Atypical antipsychotic drugs;Schizophrenia

⁎ Corresponding author. Department of Neuropsychiatry, University of Toyama Graduate School of Medicine and Pharmaceutical Sciences, 2630Sugitani, Toyama 930-0194, Japan. Tel.: +81 76 434 7323; fax: +81 76 434 5030.

E-mail address: tomikisumiyoshi840@hotmail.com (T. Sumiyoshi).

0920-9964/$ - see front matter © 2008 Elsevier B.V. All rights reserved.doi:10.1016/j.schres.2008.01.020

321Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

1. Introduction

Cognitive function, such as verbal memory, attention,and executive function, is a major determinant of outcomein patients with schizophrenia (Meltzer and Sumiyoshi,2003; Sumiyoshi and Meltzer, 2003). Treatment with thesecond generation antipsychotics, or so-called “atypicalantipsychotic drugs (AAPDs)”, has been found topartially improve cognitive disturbances of schizophrenia(Keefe et al., 2006a,b). There is accumulated evidence forthe ability of AAPDs, e.g. clozapine, olanzapine,risperidone, quetiapine, melperone, and ziprasidone toameliorate cognitive impairments in patients with schizo-phrenia (Sumiyoshi et al., 2006a; reviewed by Sumiyoshiand Meltzer, 2003), although, their effects have recentlybeen found to be smaller than previously considered(Woodward, 2006; Woodward et al., 2005; Keefe et al.,2007). AAPDs have been shown to possess relativelyhigh serotonin-5-HT2A/dopamine-D2 affinity ratios com-pared with those of typical antipsychotic drugs, such ashaloperidol (Meltzer et al., 1989; Stockmeier et al., 1993;Sumiyoshi et al., 1995), which is assumed to be related tothe ability of these agents to enhance DA and acetylcho-line release in the prefrontal cortex and hippocampus(Chung et al., 2004; Huang et al., 2006). For example, 5-HT2A antagonist action of AAPDs inhibits the 5-HTsystem that regulates activity of DA neurons (Kuroki etal., 1999; Meltzer et al., 2003). These pharmacologicalproperties have been suggested to be critical to the abilityof AAPDs to improve negative symptoms and cognitivedisturbances of schizophrenia (Farde et al., 1998; Meltzeret al., 1999; Kuroki et al., 1999).

Neurophysiological measurements, such as event-related potentials (ERPs), have been suggested to providea biological substrate for some aspects of cognitivedisturbances of schizophrenia. Particularly, P300 has beenshown to provide an electrophysiological measure ofattention-dependent information processing (Kawasakiet al., 1997, 2007). Its amplitude, thought to reflect activa-tion of immediate memory, is reduced in subjects withschizophrenia (Kawasaki et al., 1997; Nieman et al., 2002;Umbricht et al., 1998; Sumiyoshi et al., 2006b).

Jeon and Polich (2003) reported that P300 amplitudesare overall smaller in patients with schizophreniacompared to control subjects, and differs in its effectsize topography across themidline and temporal electrodesites, with the strongest effect sizes obtained for the Pz andTCP1 lateral electrodes. Bruder et al. (1999) found thatnormal controls showed greater P300 amplitudes,whereas patients did not. Specifically, Kawasaki et al.(1997) found negative correlations between auditoryP300 amplitudes and severity of psychotic symptoms of

schizophrenia. Renoult et al. (2007) report a positivecorrelation between differences in P300 amplitudes attemporal sites (T4–T3) and severity of positive symptomsand worse global functioning, consistent with an associa-tion between low P300 amplitudes and verbal memorydeficits in schizophrenia (Nagasawa et al., 1999; Niemanet al., 2002).

Efforts have been made to elucidate mechanisms forthe ability of AAPDs to ameliorate psychophysiologicalimpairments of schizophrenia. Honey et al. (1999)reported that treatment with risperidone increased bloodflow in right prefrontal cortex, as revealed by fMRI.Lieberman et al. (2005) measured the brain volumes ofsubjects with first episode schizophrenia, and found thatpatients treated with haloperidol showed a reduced graymatter volume, whereas olanzapine did not have such aneffect. Umbricht et al. (1998) found that treatment withclozapine but not haloperidol increased P300 amplitudesin patients with schizophrenia. Specifically, Niznikiewiczet al. (2005) observed an increase in P300 amplitudes inleft temporal electrodes during treatment with clozapine,indicating a region-specific response to pharmacologicaltreatment. However, localization of the electrical gen-erator of P300 has been difficult due to limited spatialresolution with traditional ERP methods.

Low Resolution Electromagnetic Tomography (LOR-ETA) provides three-dimensional images of brain elec-trical activity (Pascual-Marqui et al., 1999). To date, therehas been little study on the effect of AAPDs on LORETAimages of electrophysiological activity. In a previousstudy, Sumiyoshi et al. (2006b) reported preliminaryresults that 6-month treatment with olanzapine restoredthe left-dominant laterality of the P300 current density inthe superior temporal gyrus (STG) in a small sample ofsubjects with schizophrenia who showed improvedperformance on a test of verbal learning and memory.

In this study, we sought to test the hypotheses that 1)patients with schizophrenia elicit a decreased P300current source density in brain regions relevant to thepathophysiology of schizophrenia, such as the left STG;2) decreased P300 activity is restored by treatment withthe most widely-used antipsychotic drug olanzapine;and 3) the recovery of P300 source density is correlatedwith improvement of cognition and functional statuswhich are relevant to quality of life for patients.

2. Methods

2.1. Subjects

Datawere obtained from 16 patients (male/female=11/5) meeting DSM-IV-R criteria for schizophrenia (APA,

322 Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

1994). Demographic data of the patients are shown inTable 1. All subjects were right-handed, and were treatedat University of Toyama Hospital. Diagnosis was madebased on the Structured Clinical Interview for DSM-IV(SCID). Psychiatric and treatment history was obtainedfrom the patients, informants, and medical records.Subjects with current history of substance abuse ordependence, seizure or head injury were excluded fromthe study. Eligible patients had a complete physicalexamination. Standard laboratory testing (blood count,liver and renal function, blood sugar, total cholesterol,triglyceride) was normal. Clinical staff explained thenature of the study to the subjects, the risks and benefits,and the option not to participate in research. If the mentalstatus of a subject was impaired to the point where s/hecould not understand the nature of the study, its risks andbenefits, or the option not to participate, the subject wasnot approached to be in research. This protocol wasapproved by the Committee on Medical Ethics ofUniversity of Toyama. After complete description of thestudy to the subjects, written informed consent wasobtained. At baseline, the patients had been receiving thefollowing antipsychotic drugs: haloperidol (4), risperi-done (3), perospirone (4), chlorpromazine (1), orrisperidone+perospirone (1). Three subjects were neuro-leptic-free. All of the schizophrenia samples were out-patients, and were lightly to moderately ill who weretreated without hospitalization. No other supports, e.g.,cognitive behavioral therapy or cognitive rehabilitation,were provided. No patient was admitted to hospital duringthe study period.

Sixteen age and gender-matched right-handedhealthy volunteers also participated in the study ascontrol subjects. Demographic data of the controlsubjects are shown in Table 1.

Table 1Demographic data of participants

Schizophrenia(n=16)

Healthy controls(n=16)

Male/female 11/5 11/5Age (years) 31.0 (10.0)

(range, 17–56)31.2 (8.2)(range, 24–50)

Education (years) 13.5 (2.1) ⁎ 16.6 (1.1)Age of onset (years) 23.3 (6.2) –Duration of illness (years) 8.3 (8.8) –Drug (mg/day, risperidoneequivalence)

2.1 (1.9) –

JART 100.4 (15.7) –

Values represent mean (SD).JART, Japanese Adult Reading Test.*pb0.05, significantly different from control values.

2.2. Clinical assessments

The Scale for the Assessment of Positive Symptoms(SAPS; Andreasen, 1983b), Scale for the Assessment ofNegative Symptoms (SANS; Andreasen, 1983a), andQuality of Life Scale (QLS; Heinrichs et al., 1984; Bilkeret al., 2003) were assessed by an experienced psychiatristwho was not informed of medication status. Cognitiveassessment was performed with the Japanese VerbalLearning Test (JVLT; Matsui et al., 2007), and DigitSpan from the Wechsler Adult Intelligence Scale—Revised (WAIS-R; Wechsler, 1981). The JVLT iscomposed of 16-common Japanese words (Ogawa,1972). Subjects are required to recall words after a 16-word list is presented. The total number of words recalledfrom the list for three trials was calculated. We usedalternate forms of the JVLT at baseline and 6 months(Sumiyoshi et al., 2006b). To evaluate premorbid IQ, theJapanese Adult Reading Test (JART; Matsuoka et al.,2006) was administered. Neuropsychological evaluationwas conducted by Master's level psychologists, whowere not informed of other clinical data or medicationstatus. Immediately after the baseline assessment, anti-psychotic medications were switched stepwise to olan-zapine during the initial 6weeks. The treating psychiatristsadjusted the dose to optimize improvement in psycho-pathology, while attempting to keep the side effects of thedrug tolerable. Concomitant medications were restrictedto small dose benzodiazepines. The treatment regimenwas based on our previous report (Sumiyoshi et al.,2006b).

2.3. ERP recording

ERPs were recorded at the time of the clinicalassessments using an auditory odd-ball paradigm, basedon an established method (Kawasaki et al., 1997, 2007;Sumiyoshi et al., 2006b). Electroencephalograms (EEGs)were recorded with a 32-channel DC-amplifier (EEG-2100 version 2.22J, Nihon Kohden Corp., Tokyo, Japan).Recordings were performed using an electrocap (Electro-cap Inc., Eaton, OH) in a sound-attenuated room.Auditory stimuli were delivered binaurally throughheadphones with variable inter-stimulus intervals rangingfrom 1.5 to 2.5 s. Target tones of 2000 Hz were randomlypresented in a series of standard tones of 1000Hz,with thepresentation probability of 0.2 for target tones. All toneswere 100 ms in duration with a rise–fall time of 10 ms.Subjects were requested to press a button promptly andaccurately in response to infrequent target tones. EEGwasrecorded with 19 electrodes located at FP1, FP2, F3, F4,F7, F8, C3, C4, P3, P4, O1, O2, T3, T4, T5, T6, Fz, Cz,

Fig. 1. Grandaverage of ERPwaveforms. Black lines represent responsesto target stimuli of healthy controls. Blue and red lines represent res-ponses to target stimuli before and after treatment with olanzapine,respectively.

323Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

and Pz, according to the international 10–20 system. Allelectrodes were referred to the average amplitude of earelectrodes (bandwidth=0.16–120 Hz, 60 Hz notch filter).Electrode impedance was less than 10 kΩ. Data werecollected with a sampling rate of 500 Hz. Averaging ofERP waves and related procedures were performed usingEPLYZER II software (Kissei Comtec, Co. Ltd. Nagano,Japan). The epoch was 700 ms, including a 100-ms pre-stimulus baseline. We excluded a response if patientmissed performance. We did not set a cut-off time. EEGresponses with the mean of 37.1±8.7 times to target toneswere averaged off-line. Eye movement artifacts (blinksand eye movements) were rejected off-line. For artifactrejection, trials were excluded if their voltage exceeded±50 μV.

2.4. LORETA analysis

LORETA images were obtained by estimating thecurrent source density distribution for epochs of brainelectric activity on a dense grid of 2394 voxels at 7-mmspatial resolution applied to the digitized Talairach andTournoux (1988), based on the established method(Pascual-Marqui et al., 1999). LORETA made use of thethree-shell spherical head model registered to theTalairach atlas available as a digitized MRI from theBrain Imaging Centre, Montreal Neurologic Institute.Registration between spherical and realistic headgeometry used EEG electrode coordinates reported byTowle et al. (1993). The solution space was restricted tocortical gray matter and hippocampus, as determined bythe corresponding digitized Probability Atlas alsoavailable from the Brain Imaging Centre. A voxel waslabeled as gray matter if it met the following threeconditions: its probability of being gray matter washigher than that of being white matter, its probability ofbeing gray matter was higher than that of beingcerebrospinal fluid, and its probability of being graymatter was higher than 33% (Pascual-Marqui et al.,1999). We calculated LORETA images for each subjectin the time frame 250–500 ms post-stimulus based onprevious reports (Kawasaki et al., 2007; Sumiyoshiet al., 2006b). LORETA values from each voxel weredivided by the average whole brain LORETAvalue, andwere used to determine current source density.

2.5. Data analysis

Effects of treatment with olanzapine on clinicalvariables were analyzed by paired t-test. The effect ofbaseline medications (typical antipsychotic drugs, atypi-cal antipsychotic drugs, antipsychotic-free) on ERP and

clinical measures were assessed by one-way ANOVA.Correlation analyses were performed by Pearson'scorrelation test using SPSS 11.0 software. Comparisonsbetween healthy controls and patients on LORETA sourceimaging were conducted using voxel-by-voxel unpairedt-statistics, after logarithmic transformation of the data.Holmes' non-parametric correction for multiple compar-isons was applied (Holmes et al., 1996). Significancelevel for all statistical tests was set at pb0.05 (two-tailed).The corresponding corrected alpha was 0.01 (t=2.66).

3. Results

3.1. LORETA images of normal controls andschizophrenia

Grandaverage waveforms from patients and healthycontrols are shown in Fig. 1. P300 current source density

324 Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

was lower for patients compared to normal controls inseveral brain regions, especially the left STG (t=−6.81,pb10−4), left precentral gyrus (t=−6.92, pb10−4), leftmiddle frontal gyrus (t=−6.86, pb10−4), and leftprecuneus (t=−6.16, pb10−4) (Fig. 2). Therefore, weassessed the effect of olanzapine treatment on LORETAvalues in these four brain regions.

3.2. LORETA images before and after treatment witholanzapine

All patients completed the study. The mean (SD)dose of olanzapine at 6 months was 9.1 (5.5) mg/day. Nopatient experienced noticeable side effects.

Fig. 3 shows LORETA images of the grandaverageof P300 from healthy controls and all patients before andafter treatment with olanzapine. Overall, schizophreniapatients showed lower P300 current density in the left

Fig. 2. Comparison of P300 source imaging in patients with schizophrenia at bmapping (SnPM) voxel-wise LORETA comparisons for independent samplesblue, respectively. T-thresholds for pb0.01, pb0.05, and pb0.1, were set −

hemisphere compared with normal controls. However,after 6-month treatment with olanzapine, a left-domi-nant pattern of P300 current source density was evident,which was similar to the pattern of healthy controls(Fig.3; upper and lower panels). In fact, as shown inTable 2, treatment with olanzapine significantlyincreased LORETA values only for the left STG butnot other three regions.

3.3. Correlations between changes in clinical variablesand those in LORETA values

Scores of the SAPS, SANS, QLS and JVLT, but notDigit Span were significantly improved during treat-ment with olanzapine (Table 3).

Table 4 demonstrates correlations between changesof LORETA values and those of clinical variables.Significant correlations were noted for JVLT and SANS

aseline against normal controls as revealed by statistical non-parametric. Positive, zero, and negative t-values are represented in red, white and3.31, −2.66, and −2.29, respectively.

Fig. 3. LORETA images of P300 grandaverage in normal controls (upper panel) and patients at baseline (middle panel) and after 6-month treatmentwith olanzapine (lower panel).

325Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

scores vs. LORETA values of the left STG (Table 4,Fig. 4A,B). Significant correlations were also noted forthe change of QLS scores vs. LORETAvalues of the left

Table 2Change in LORETA values in the left hemisphere by olanzapine treatment i

Baseline (n=16)

Superior temporal gyrus 1.12 (0.17)Precentral gyrus 0.65 (0.12)Middle frontal gyrus 0.82 (0.19)Precuneus 0.98 (0.24)

Values represent mean (SD).

middle frontal gyrus (Table 4, Fig. 5A), while thechange in QLS scores tended to be correlated with thechange of LORETA values of precentral gyrus (Table 4,

n patients with schizophrenia

6 months (n=16) Difference (t-test) p-value

1.28 (0.21) 0.0220.67 (0.11) n.s.0.87 (0.14) n.s.0.97 (0.19) n.s.

Table 3Change in clinical variables by olanzapine treatment

Baseline (n=16) 6 months (n=16) Difference (t-test)p-value

SAPS 25.2 (22.7) 12.5 (17.4) 0.001SANS 51.6 (18.9) 41.0 (21.4) 0.004QLS 44.3 (14.7) 62.0 (21.9) b0.001JVLT 25.0 (9.5) 27.8 (8.1) 0.007Digit Span 14.4 (5.3) 16.1 (4.1) n.s.

Values represent mean (SD).SAPS, Scale for the Assessment of Positive Symptoms.SANS, Scale for the Assessment of Negative Symptoms.QLS, Quality of Life Scale.JVLT, Japanese Verbal Learning Test.

326 Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

Fig. 5B). There were no significant correlations betweenchanges in LORETA values and those in SAPS andDigit Span scores (Table 4).

There were significant correlations between changesof SANS vs. JVLT scores (r=−0.696, p=0.003) andSANS vs. Digit Span Scores (r=−0.503, p=0.047), butnot for other combinations (e.g. SAPS score vs.cognitive measures) (data not shown).

3.4. Effect of baseline medications

We divided the patients according to type of baselinemedications (3 for antipsychotic-free, 5 for typicalantipsychotics, 8 for atypical antipsychotics), andevaluated its effect on changes of ERP and clinicalmeasures. There was no significant influence of baselinemedication status on changes of LORETA values in theleft STG (F=0.431, p=0.657) or scores of SAPS(F=1.114, p=0.358), SANS (F=0.530, p=0.601),QLS (F = 0.001, p = 0.999), JVLT (F = 0.561,p=0.584), and Digit Span (F=0.174, p=0.842).

Table 4Pearson correlation coefficients between changes of LORETA values for P3variables

STG Precentral gyrus

r p r p

ΔSAPS 0.243 n.s. −0.112 nΔSANS −0.631 0.009 −0.162 nΔQLS 0.189 n.s. 0.464 0ΔJVLT 0.613 0.011 0.143 nΔDigit Span 0.227 n.s. 0.278 n

SAPS, Scale for the Assessment of Positive Symptoms.SANS, Scale for the Assessment of Negative Symptoms.QLS, Quality of Life Scale.JVLT, Japanese Verbal Learning Test.STG, superior temporal gyrus.

4. Discussion

Treatment with olanzapine for 6 months wasassociated with improvements in positive and negativesymptoms, performance on verbal learning memory, andquality of life. P300 current source density in the leftSTG, as represented by LORETA values, was increasedby olanzapine treatment. Importantly, the degree ofrecovery of P300 LORETA values in the left STG waspositively correlated with improvements in negativesymptoms and verbal learning memory. Moreover, theincrease in P300 current source density in the left middlefrontal gyrus and precentral gyrus were associated withimprovement in QLS scores.

Enhancement of verbal learning and memory inpatients treated with olanzapine is similar to Mori et al.(2004) and Suzuki et al. (2006), who found improve-ment of immediate memory after 4-week treatment witholanzapine, while Keefe et al. (2004) reported improvedperformance on the California Verbal Learning Testafter 3-month treatment in subjects with schizophrenia.These findings suggest the beneficial influence ofolanzapine on learning and memory in subjects withschizophrenia. Also, Revicki et al. (1999) found that 6-week olanzapine treatment improved negative symp-toms and QOL. Further, Phillips et al. (2006) reportedthat 28-week treatment with olanzapine or ziprasidonesignificantly improved both measures. These previousobservations are consistent with the results of the currentstudy, confirming the ability of olanzapine to improvenegative symptoms and quality of life in schizophreniapatients.

Several studies report the effect of olanzapine on P300activity in schizophrenia. Gonul et al. (2003) observed anincrease in P300 amplitudes in the Fz electrode after 6-week treatment with olanzapine in 26 schizophrenia

00 current source density in the left hemisphere and those of clinical

Middle frontal gyrus Precuneus

r p r p

.s. −0.136 n.s. 0.009 n.s.

.s. −0.028 n.s. −0.051 n.s.

.07 0.565 0.023 −0.271 n.s.

.s. −0.212 n.s. −0.054 n.s.

.s. 0.222 n.s. −0.318 n.s.

Fig. 4. Scatter plots and least squares regression lines depicting the relationship between changes of JVLT (A) and SANS (B) vs. changes of LORETAvalues for the P300 current source density in the left superior temporal gyrus (STG).

327Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

patients. On the other hand, Molina et al. (2004) found nosignificant change either in amplitudes or latency, asmeasured at Pz and Fz electrodes after 6-month treatmentwith olanzapine in 11 patients. Korostenskaja et al. (2005)observed that 5–10 mg olanzapine for 4 weeks had nosignificant effect on P300 latency and amplitude in 11patients with schizophrenia. The conflicting results fromthese previous studies indicate observation period, as wellas sample size, may determine whether or not P300changes occur as a result of olanzapine treatment. In the

Fig. 5. Scatter plots and least squares regression lines depicting the relationshicurrent source density in the left middle frontal gyrus (A) and left precentra

current study, we evaluated P300 activity during treatmentwith olanzapine for 6 months, which was sufficient toimprove some of the neurophysiological measures inpatients with schizophrenia.

There is accumulated evidence for the advantage ofanalyzing EEG activity with LORETA. Gallinat et al.(2002) reported that the strength of LORETA is a betterrepresentation of a widely distributed or multiple-oriented activity, which is difficult to model with dipolemethods such as BESA. Generally, the EEG recording

p between changes of QLS vs. changes of LORETAvalues for the P300l gyrus (B).

328 Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

procedure is simple and produces less demand onpatients than does fMRI imaging.

In control subjects, we found left-dominant later-alized activations in response to an auditory odd-balltask, similar to Mulert et al. (2004). On the other hand,Alexander et al. (1996) found that greater P300amplitudes over right hemisphere were found in 80healthy subjects. Bruder et al. (1998) also reportedgreater current flow over right than left in 22 normalcontrol patients. The discrepancy between the previousresults and our finding may be due to the difference inthe mode of P300 activity measurement, i.e. topogra-phical P300 amplitudes in the former studies vs. currentsource analysis in our study. LORETA is a method thatexpresses electrical current source of brain activity;therefore, there would be an instance where traditionalwaveforms are not sensitive enough to reflect electro-physiological changes as detected by LORETA analysis.

In patients with schizophrenia asymmetrical topo-graphy of auditory P300 has been reported, withattenuated amplitudes in the left hemisphere and aright-dominant P300 electrical power (Strik et al.,1994). Localization of P300 generators with LORETAhas indicated a reduction in the P300 current density inthe left medial temporal area and left inferior parietalarea in schizophrenia patients compared with healthycontrols (Pae et al., 2003). Winterer et al. (2001)compared LORETA images of P300 between schizo-phrenia patients and normal subjects, and found that thepatients showed decreased activation in the left corticalregions than normal subjects. These previous studiesreporting altered anatomical configuration of P300 inschizophrenia are in line with our observations pre-sented here.

The STG in the dominant hemisphere is responsiblefor language comprehension, and injury of this regioncauses sensory aphasia (Shapleske et al., 1999). Schizo-phrenia patients show lesser anatomical and functionalasymmetries than healthy subjects, and the failure todevelop left dominancy may reflect disturbances oflanguage (Crow, 1997). In a pilot study, Sumiyoshiet al. (2006b) reported 6-month treatment with olanzapinerestored the left-dominant pattern of P300 current densityin the STG, and improved performance on the JVLT infive subjects with schizophrenia. In the current study, weevaluated these measures with more than a tripled numberof subjects. Further, we have reported, for the first time,that the current source density itself (not laterality ratios)in the left STG was increased by olanzapine, which wassignificantly correlated with the improvement of verballearning memory and negative symptoms. This finding ispartly consistent with a study reporting that after 6-month

treatment with olanzapine, a specific association emergedbetween P300 amplitude and negative symptoms (Gonulet al., 2003). The current study also found that the increasein P300 current source density in the left middle frontalgyrus and left precentral gyrus was associated withimprovement in quality of life in patients with schizo-phrenia. Frontal lobe function has been suggested togovern social functions and higher cognitive functions,e.g. memory organization (Nohara et al., 2000) and scripttasks (Sumiyoshi et al., 2008). In sum, the results of thepresent study indicate the ability of brain imaging ofelectrophysiological measures, such as ERPs, to predicttreatment response in terms of functional outcomes inmajor psychiatric disorders.

There were significant correlations between changesof SANS vs. JVLT scores and SANS vs. Digit Spanscores. These findings indicate robust associations be-tween negative symptoms and cognitive measures inpatients treated with olanzapine, and suggest commonbiological substrates which underlie negative symptomsand deficits in some types of cognitive domains seen inschizophrenia (Sumiyoshi et al., 2003).

Themajor limitation of the present study is the lack of alongitudinal assessment of control subjects. However,P300 activities have been reported not to be affected byprevious procedure (Fabiani et al., 1987). We usedalternate forms of JVLT to rule out the possibility of apractice effect. Scores of the Digit Span were not changedafter 6-month treatment. In fact, the lengthy intervalbetween the two assessment points may make it unlikelythat these variables were affected by practice effect.

In conclusion, the results of the present study suggestthe usefulness of LORETA imaging of P300 as anindicator of treatment response in some aspects of thepsychopathology and cognitive disturbances of schizo-phrenia. Further study is warranted to determine if ourobservations will be applicable to other antipsychoticdrugs and cognition-enhancing agents.

Role of the funding sourceThis work was supported by a Grant-in-Aid for Science Research

from Japan Society for the Promotion of Science (Nos. 16591126 and19591345), a Pharmacopsychiatry Research Grant from the MitsubishiPharma Research Foundation. These funding bodies had no furtherrole in the study design; in the collection, analysis, and interpretationof data; in the writing of the report; and in the decision to submit thepaper for publication.

ContributorsAuthor Yuko Higuchi undertook clinical assessments, ERP imaging,

and statistical analysis, andwrote the first draft of themanuscript. AuthorTomiki Sumiyoshi designed, the study and wrote the protocol, and edited

329Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

the manuscript. Author Yasuhiro Kawasaki provided technical supportfor LORETA analysis. Author Mie Matsui designed the neuropsycho-logical assessments. Author Hirofumi Arai administered neuropsycho-logical assessments. Author Masayoshi Kurachi managed literaturesearches. All authors contributed to and have approved the finalmanuscript.

Conflict of interestNone.

AcknowledgementThe authors thank Drs. Toru Ito, Atsushi Furuichi, and Michio

Suzuki for assistance.

References

Alexander, J.E., Bauer, L.O., Kuperman, S., Morzorati, S., O'Connor,S.J., Rohrbaugh, J., Porjesz, B., Begleiter, H., Polich, J., 1996.Hemispheric differences for P300 amplitude from an auditoryoddball task. Int. J. Psychophysiol. 21, 189–196.

American Psychiatric Association, 1994. Diagnostic and StatisticalManual of Mental Disorders, 4th ed. American Psychiatric Press,Washington, DC.

Andreasen, N.C., 1983a. Scale for the Assessment of NegativeSymptoms (SANS). University of Iowa, Iowa City.

Andreasen, N.C., 1983b. Scale for the Assessment of PositiveSymptoms (SAPS). University of Iowa, Iowa City.

Bilker, W.B., Brensinger, C., Kurtz, M.M., Kohler, C., Gur, R.C.,Siegel, S.J., Gur, R.E., 2003. Development of an abbreviatedschizophrenia quality of life scale using a new method.Neuropsychopharmacology 28, 773–777.

Bruder, G.E., Tenke, C.E., Towey, J.P., Leite, P., Fong, R., Stewart, J.E., McGrath, P.J., Quitkin, F.M., 1998. Brain ERPs of depressedpatients to complex tones in an oddball task: relation of reduced P3asymmetry to physical anhedonia. Psychophysiology 35, 54–63.

Bruder, G., Kayser, J., Tenke, C., Amador, X., Friedman, M., Sharif, Z.,Gorman, J., 1999. Left temporal lobe dysfunction in schizophrenia:event-related potential and behavioral evidence from phonetic andtonal dichotic listening tasks. Arch. Gen. Psychiatry 56, 267–276.

Chung, Y.C., Li, Z., Dai, J., Meltzer, H.Y., Ichikawa, J., 2004.Clozapine increases both acetylcholine and dopamine release in ratventral hippocampus: role of 5-HT1A receptor agonism. Brain Res.1023, 54–63.

Crow, T.J., 1997. Schizophrenia as failure of hemispheric dominancefor language. Trends Neurosci. 20, 339–343.

Fabiani, M., Gratton, G., Karis, D., 1987. Definition, identification, andreliability of measurement of P300 component of the event relatedpotentials. Advances in Psychophysiology, vol. II, pp. 1–78.

Farde, L., Ito, H., Swahn, C.G., Pike, V.W., Halldin, C., 1998.Quantitative analyses of carbonyl-carbon-11-WAY-100635 bind-ing to central 5-hydroxytryptamine-1A receptors in man. J. Nucl.Med. 39, 1965–1971.

Gallinat, J., Mulert, C., Bajbouj, M., Herrmann, W.M., Schunter, J.,Senkowski, D., Moukhtieva, R., Kronfeldt, D., Winterer, G., 2002.Frontal and temporal dysfunction of auditory stimulus processingin schizophrenia. Neuroimage 17, 110–127.

Gonul, A.S., Suer, C., Coburn, K., Ozesmi, C., Oguz, A., Yilmaz, A.,2003. Effects of olanzapine on auditory P300 in schizophrenia.Prog. Neuro-Psychopharmacol. Biol. Psychiatry 27, 173–177.

Heinrichs, D.W., Hanlon, T.E., Carpenter, W.T.J., 1984. The quality oflife scale: an instrument for rating the schizophrenic deficitsyndrome. Schizophr. Bull. 10, 388–398.

Holmes, A.P., Blair, R.C., Watson, J.D., Ford, I., 1996. Nonparametricanalysis of statistic images from functional mapping experiments.J. Cereb. Blood. Flow Metab. 16, 7–22.

Honey, G.D., Bullmore, E.T., Soni, W., Varatheesan, M., Williams, S.C.,Sharma, T., 1999. Differences in frontal cortical activation by aworking memory task after substitution of risperidone for typicalantipsychotic drugs in patients with schizophrenia. Proc. Natl. Acad.Sci. U. S. A. 96, 13432–13437.

Huang, M., Li, Z., Ichikawa, J., Dai, J., Meltzer, H.Y., 2006. Effects ofdivalproex and atypical antipsychotic drugs on dopamine andacetylcholine efflux in rat hippocampus and prefrontal cortex.Brain Res. 1099, 44–55.

Jeon,Y.W.,Polich, J., 2003.Meta-analysisofP300and schizophrenia: patients,paradigms, and practical implications. Psychophysiology 40, 684–701.

Kawasaki, Y., Maeda, Y., Higashima, M., Nagasawa, T., Koshino, Y.,Suzuki, M., Ide, Y., 1997. Reduced auditory P300 amplitude,medial temporal volume reduction and psychopathology inschizophrenia. Schizophr. Res. 26, 107–115.

Kawasaki, Y., Sumiyoshi, T., Higuchi, Y., Ito, T., Takeuchi, M., Kurachi,M., 2007. Voxel-based analysis of P300 electrophysiologicaltopography associated with positive and negative symptoms ofschizophrenia. Schizophr. Res. 94, 164–171.

Keefe, R.S., Seidman, L.J., Christensen, B.K., Hamer, R.M., Sharma, T.,Sitskoorn, M.M., Lewine, R.R., Yurgelun-Todd, D.A., Gur, R.C.,Tohen, M., Tollefson, G.D., Sanger, T.M., Lieberman, J.A., 2004.Comparative effect of atypical and conventional antipsychotic drugson neurocognition in first-episode psychosis: a randomized, double-blind trial of olanzapine versus low doses of haloperidol. Am.J. Psychiatry 161, 985–995.

Keefe, R.S., Seidman, L.J., Christensen, B.K., Hamer, R.M., Sharma, T.,Sitskoorn,M.M., Rock, S.L.,Woolson, S., Tohen,M., Tollefson,G.D.,Sanger, T.M., Lieberman, J.A., 2006a. Long-term neurocognitiveeffects of olanzapine or low-dose haloperidol in first-episodepsychosis. Biol. Psychiatry. 59, 97–105.

Keefe, R.S., Young, C.A., Rock, S.L., Purdon, S.E., Gold, J.M., Breier,A., 2006b. One-year double-blind study of the neurocognitiveefficacy of olanzapine, risperidone, and haloperidol in schizo-phrenia. Schizophr. Res. 81, 1–15.

Keefe, R.S., Bilder, R.M., Davis, S.M., Harvey, P.D., Palmer, B.W.,Gold, J.M., Meltzer, H.Y., Green, M.F., Capuano, G., Stroup, T.S.,McEvoy, J.P., Swartz, M.S., Rosenheck, R.A., Perkins, D.O., Davis,C.E., Hsiao, J.K., Lieberman, J.A., 2007. Neurocognitive effects ofantipsychotic medications in patients with chronic schizophrenia inthe CATIE Trial. Arch. Gen. Psychiatry 64, 633–647.

Korostenskaja, M., Dapsys, K., Siurkute, A., Maciulis, V., Ruksenas, O.,Kahkonen, S., 2005. Effects of olanzapine on auditory P300 andmismatch negativity (MMN) in schizophrenia spectrum disorders.Prog. Neuro-Psychopharmacol. Biol. Psychiatry 29, 543–548.

Kuroki, T., Meltzer, H.Y., Ichikawa, J., 1999. Effects of antipsychoticdrugs on extracellular dopamine levels in rat medial prefrontal cortexand nucleus accumbens. J. Pharmacol. Exp. Ther. 288, 774–781.

Lieberman, J.A., Tollefson, G.D., Charles, C., Zipursky, R., Sharma, T.,Kahn, R.S., Keefe, R.S., Green, A.I., Gur, R.E., McEvoy, J., Perkins,D., Hamer, R.M., Gu, H., Tohen, M., 2005. Antipsychotic drugeffects on brain morphology in first-episode psychosis. Arch. Gen.Psychiatry 62, 361–370.

Matsui, M., Sumiyoshi, T., Kato, K., Masayoshi, K., 2007. Develop-ment of alternate forms of Japanese Verbal Learning Test. SeishinIgaku (Clinical Psychiatry) 49 (1), 31–34.

330 Y. Higuchi et al. / Schizophrenia Research 101 (2008) 320–330

Matsuoka, K., Uno, M., Kasai, K., Koyama, K., Kim, Y., 2006.Estimation of premorbid IQ in individuals with Alzheimer'sdisease using Japanese ideographic script (Kanji) compoundwords: Japanese version of National Adult Reading Test.Psychiatry Clin. Neurosci. 60, 332–339.

Meltzer, H.Y., Sumiyoshi, T., 2003. Atypical antipsychotic drugsimprove cognition in schizophrenia. Biol. Psychiatry. 53, 265–267author reply 267–8.

Meltzer, H.Y., Matsubara, S., Lee, J.C., 1989. The ratios of serotonin2and dopamine2 affinities differentiate atypical and typicalantipsychotic drugs. Psychopharmacol. Bull. 25, 390–392.

Meltzer, H.Y., Park, S., Kessler, R., 1999. Cognition, schizophrenia, and theatypical antipsychotic drugs. Proc. Nat. Acad. Sci. 96, 13591–13593.

Meltzer, H.Y., Li, Z., Kaneda, Y., Ichikawa, J., 2003. Serotoninreceptors: their key role in drugs to treat schizophrenia. Prog.Neuro-Psychopharmacol. Biol. Psychiatry 27, 1159–1172.

Molina, V., Munoz, F., Martin-Loeches, M., Casado, P., Hinojosa, J.A.,Iglesias, A., 2004. Long-term olanzapine treatment and p300parameters in schizophrenia. Neuropsychobiology 50, 182–188.

Mori, K., Nagao, M., Yamashita, H., Morinobu, S., Yamawaki, S., 2004.Effect of switching to atypical antipsychotics on memory in patientswith chronic schizophrenia. Prog. Neuro-Psychopharmacol. Biol.Psychiatry 28, 659–665.

Mulert, C., Pogarell, O., Juckel, G., Rujescu, D., Giegling, I., Rupp, D.,Mavrogiorgou, P., Bussfeld, P., Gallinat, J., Moller, H.J., Hegerl, U.,2004. The neural basis of the P300 potential. Focus on the time-course of the underlying cortical generators. Eur. Arch. PsychiatryClin. Neurosci. 254, 190–198.

Nagasawa, T., Kamiya, T., Kawasaki, Y., Higashima, M., Urata, K.,Sakai, N., Koshino, Y., 1999. The relationship between auditoryERP and neuropsychological assessments in schizophrenia. Int.J. Psychophysiol. 34, 267–274.

Nieman,D.H.,Koelman, J.H., Linszen,D.H.,Bour, L.J.,Dingemans, P.M.,Ongerboer de Visser, B.W., 2002. Clinical and neuropsychologicalcorrelates of the P300 in schizophrenia. Schizophr. Res. 55, 105–113.

Niznikiewicz, M.A., Patel, J.K., McCarley, R., Sutton, J., Chau, D.T.,Wojcik, J., Green, A.I., 2005. Clozapine action on auditory P3response in schizophrenia. Schizophr. Res. 76, 119–121.

Nohara, S., Suzuki, M., Kurachi, M., Yamashita, I., Matsui, M., Seto,H., Saitoh, O., 2000. Neural correlates of memory organizationdeficits in schizophrenia: a single photon emission computedtomography study with 99m-Tc-ethyl-cysteinate dimmer during averbal learning task. Schizophr. Res. 42, 209–222.

Ogawa, T., 1972. Table of frequency in the recall of words belongingto 52 categories. J. Lit. Assoc. Kwansei Gakuin Univ. 22, 1–60.

Pae, J.S., Kwon, J.S., Youn, T., Park, H.J., Kim,M.S., Lee, B., Park,K.S.,2003. LORETA imaging of P300 in schizophrenia with individualMRI and 128-channel EEG. Neuroimage 20, 1552–1560.

Pascual-Marqui, R.D., Lehmann, D., Koenig, T., Kochi, K., Merlo, M.C.,Hell, D., Koukkou, M., 1999. Low resolution brain electromagnetictomography (LORETA) functional imaging in acute, neuroleptic-naive,first-episode, productive schizophrenia. Psychiatry Res. 90, 169–179.

Phillips, G.A., Van Brunt, D.L., Roychowdhury, S.M., Xu, W., Naber,D., 2006. The relationship between quality of life and clinicalefficacy from a randomized trial comparing olanzapine andziprasidone. J. Clin. Psychiatry 67, 1397–1403.

Renoult, L., Prevost, M., Brodeur, M., Lionnet, C., Joober, R., Malla,A., Debruille, J.B., 2007. P300 asymmetry and positive symptomseverity: a study in the early stage of a first episode of psychosis.Schizophr. Res. 93, 366–373.

Revicki, D., Genduso, L., Hamilton, S., Ganoczy, D., Beasley Jr., C.,1999. Olanzapine versus haloperidol in the treatment of schizo-

phrenia and other psychotic disorders: quality of life and clinicaloutcomes of a randomized clinical trial. Qual. Life Res. 8, 417–426.

Shapleske, J., Rossell, S.L., Woodruff, P.W., David, A.S., 1999. Theplanum temporale: a systematic, quantitative review of itsstructural, functional and clinical significance. Brain Res. BrainRes. Rev. 29, 26–49.

Stockmeier, C.A., DiCarlo, J.J., Zhang, Y., Thompson, P., Meltzer, H.Y.,1993. Characterization of typical and atypical antipsychotic drugsbased on in vivo occupancy of serotonin2 and dopamine2 receptors.J. Pharmacol. Exp. Ther. 266, 1374–1384.

Strik, W.K., Dierks, T., Franzek, E., Stober, G., Maurer, K., 1994. P300asymmetries in schizophrenia revisited with reference-independentmethods. Psychiatry Res. 55, 153–166.

Sumiyoshi, T., Meltzer, H., 2003. Pharmacological strategy forenhancement of social function and quality of life in patientswith schizophrenia: considerations of the effect of melperone, andatypical antipsychotic drug on cognitive function. Seishin Igaku(Clinical Psychiatry) 4 (12), 1279–1284.

Sumiyoshi, T., Stockmeier,C.A.,Overholser, J.C., Thompson, P.A.,Meltzer,H.Y., 1995. Dopamine D4 receptors and effects of guanine nucleotideson [3H]raclopride binding in postmortem caudate of subjects withschizophrenia or major depression. Brain Res. 681, 109–116.

Sumiyoshi, T., Jayathilake, K., Meltzer, H.Y., 2003. The effect ofmelperone, an atypical antipsychotic drug, on cognitive function inschizophrenia. Schizophr. Res. 59, 7–16.

Sumiyoshi, C., Sumiyoshi, T., Roy, A., Jayathilake, K., Meltzer, H.Y.,2006a. Atypical antipsychotic drugs and organization of long-termsemantic memory: multidimensional scaling and cluster analysis ofcategory fluency performance in schizophrenia. Int. J. Neuropsy-chopharmacol. 6, 677–683.

Sumiyoshi, T., Higuchi, Y., Kawasaki, Y.,Matsui,M., Kato, K., Yuuki, H.,Arai, H., Kurachi, M., 2006b. Electrical brain activity and response toolanzapine in schizophrenia: a study with LORETA images of P300.Prog. Neuro-Psychopharmacol. Biol. Psychiatry 30, 1299–1303.

Sumiyoshi, T., Matsui, M., Itoh, H., Higuchi, Y., Arai, H., Takamiya,C., Kurachi, M., 2008. Essential polyunsaturated fatty acids andsocial cognition in schizophrenia. Psychiatry Res. 157, 87–93.

Suzuki, H., Omoto, M., Inoue, Y., Sekiguchi, G., Akimoto, T., Gen, K.,Morokawa, Y., Aoba, A., 2006. Comparison of clinical symptomsand cognitive function before and after olanzapine switching inschizophrenia patients. Jpn. J. Clin. Psychopharmacol. 9, 881–891.

Talairach, J., Tournoux, P., 1988. Co-Planar Stereotaxic Atlas ofHuman Brain. Thieme, Stuttgart.

Towle, V.L., Bolanos, J., Suarez, D., Tan, K., Grzeszczuk, R., Levin, D.N., Cakmur, R., Frank, S.A., Spire, J.P., 1993. The spatial location ofEEG electrodes: locating the best-fitting sphere relative to corticalanatomy. Electroencephalogr. Clin. Neurophysiol. 86, 1–6.

Umbricht, D., Javitt, D., Novak, G., Bates, J., Pollack, S., Lieberman,J., Kane, J., 1998. Effects of clozapine on auditory event-relatedpotentials in schizophrenia. Biol. Psychiatry 44, 716–725.

Wechsler, D., 1981. WAIS-R Manual. Wechsler Adult IntelligenceScale—Revised. Psychological Corp., New York, NY.

Winterer,G.,Mulert, C.,Mientus, S.,Gallinat, J., Schlattmann, P.,Dorn,H.,Herrmann, W.M., 2001. P300 and LORETA: comparison of normalsubjects and schizophrenic patients. Brain Topogr. 13, 299–313.

Woodward, N.D., 2006. A meta-analysis of neuropsychologicalchange with second generation antipsychotics in schizophrenia.24th CINP Congress. Chicago, USA.

Woodward, N.D., Purdon, S.E.,Meltzer, H.Y., Zald, D.H., 2005. Ameta-analysis of neuropsychological change to clozapine, olanzapine,quetiapine, and risperidone in schizophrenia. Int. J. Neuropsycho-pharmacol. 8, 457–472.