N400 in Depression 1

Running head: N400 and Emotional Words in Depression and Schizophrenia

Semantic Processing of Emotional Words in Depression and Schizophrenia

Heide Klumpp1, Jennifer Keller2, Gregory A. Miller3, Brooks R. Casas4,

Jennifer L. Best5, and Patricia J. Deldin1

1University of Michigan 2Stanford University School of Medicine 3University of Illinois at Urbana-Champaign 4Baylor University 5Duke University Number of text pages: 11 Number of tables: 3 Heide Klumpp, Ph.D. University of Michigan 4250 Plymouth Rd. Ann Arbor, MI 48109-5766 Phone: (734) 764-5348 Fax: (734) 764-4031

E-mail: heidek@umich.edu

N400 in Depression 2

Abstract

Major depressive disorder is associated with dysfunction in brain regions involved in language

and emotion processing. Despite evidence of emotion processing biases in depression,

neurophysiological evidence of language dysfunction for emotional words in depression has

been inconsistent. This series of three studies evaluated whether depressed individuals

exhibited abnormal semantic processing of emotionally valenced words. During the passive

viewing of sentences with mood congruent and incongruent sentence endings, the N400

component of the event-related brain potential was measured in patients with depression,

dysthymia, or schizophrenia and in healthy controls. In each study, results revealed normal

semantic processing in depression. That is, N400 was similar for both mood-incongruent

(positive and neutral) endings and mood-congruent (negative) endings. In contrast, the small

sample of individuals with schizophrenia exhibited a significantly exaggerated N400 for negative

word endings compared to the depressed and healthy control groups. These data suggest

anomalies in semantic network interactions with emotion processing in schizophrenia.

Key Words: depression, schizophrenia, semantic processing, N400, emotional words, biases,

ERP

N400 in Depression 3

Semantic Processing of Emotional Words in Depression and Schizophrenia

Introduction

Major depressive disorder (MDD) is a prevalent and disabling psychiatric illness that has

been hypothesized to be associated with a general dysfunction in the left frontal lobe (e.g.,

Davidson et al., 2002; Heller and Nitschke, 1997). Bilateral dysfunction in the dorsolateral

prefrontal cortex (DLPFC) has also been proposed (Johnstone et al., 2007; Levin et al., 2007;

Mayberg, 2006). Although these brain regions are involved in language processing, individuals

with depression do not have consistent language deficits (for a review see Klumpp and Deldin,

this issue) and do not show semantic processing difficulties with neutral words manifested in the

N400 component of the event-related brain potential (ERP) (Deldin et al., 2006; Iakimova et al.,

2009), an index of semantic processing (Kutas and Hillyard, 1980a, b). The failure to find

consistent language dysfunction suggests that the purported left frontal or bilateral DLPFC

deficits may be task-specific and would necessitate a revision of models away from particular

cognitive deficits towards more complex neural network models that account for interactions

among brain regions (e.g., cortical-subcortical). Support for this hypothesis and the potential for

individual differences in neural function is supported by a review of brain functioning in

depression (Fitzgerald et al., 2008), which noted inconsistent results across studies. These

investigators propose that the pathophysiology of depression is likely complex and not

amenable to simple models. For example, it may be the case that language dysfunction in

depression is more likely to manifest when emotional stimuli are being presented, as the

potential for dysfunctional neural interactions that involve emotion processing is increased.

Behavioral evidence in support of this includes studies that show that individuals diagnosed with

MDD are more likely to have cognitive biases towards negative stimuli (Caseras et al., 2007;

Gotlib et al., 2004; Mathews et al., 1996) or away from positive stimuli (e.g., Deldin et al., 2001;

Joormann and Gotlib, 2006; Surguladze et al., 2004; Yoon et al., 2009) than general non-

emotional language processing deficits.

N400 in Depression 4

Regarding semantic processing, behavioral studies of depression show equivocal

evidence of dysfunction for emotional words. For example, some provide evidence of biased

responses to negative words (enhanced response latency for negative words; Weisbrod et al.,

1999), partial evidence of such bias (Atchley et al., 2003), and lack of evidence of semantic

dysfunction (Dannlowski et al., 2006). Other behavioral studies show normal semantic

processing for neutral words in MDD (Besche-Richard et al., 2002; Georgieff et al., 1998).

Thus, relatively few behavioral studies have investigated semantic processing of emotional

words, and the results of this limited literature are inconsistent.

Neurophysiological studies examining emotional language processing are also

equivocal, and results depend upon on the specific process examined. For example, some

ERP components detect enhanced processing of emotional words in depression (e.g., P300,

P2, N2, PINV; Nikendei et al., 2005; Serfaty et al., 2002; Shimizu et al., 2006), whereas N100

and CNV (Nikendei et al., 2005; Serfaty et al., 2002) indicate normal language processing in

depression. However, semantic processing of emotional words as indexed by N400 has not

been evaluated in MDD.

In the present series of three studies, MDD and healthy control participants were

presented sentences with mood congruent and incongruent emotional endings in order to

assess whether individuals diagnosed with mood disorders have normal emotional semantic

processing. Participants passively viewed the sentences while their ERPs were recorded. If

emotional semantic processing errors occurs in depression, participants will display increased

N400 for emotionally incongruent (i.e., positive word) endings compared to healthy controls.

Alternatively, if language processing, regardless of emotional valence is normal in depression,

no differences between the depressed and healthy control groups will be found.

One of the three studies included dysthymic participants in order to determine if

semantic processing deficits extended to a second mood disorder group. Another of the studies

included individuals diagnosed with schizophrenia, as language disturbance (e.g., abnormal

N400 in Depression 5

associations) is a hallmark of this illness and should be associated with N400 abnormalities

(Adams et al., 1993; Hokama et al., 2003; Kiang et al., 2007; Kostova et al., 2003, 2005; Ohta et

al., 1999; Sitnikova et al., 2002; Strandburg et al., 1997). Moreover, there is evidence of

abnormal emotion processing in schizophrenia (An et al., 2003; Bell et al., 1997; Berenbaum et

al., 2008; Dougherty et al., 1974; Kring and Moran, 2008; Suslow et al., 2003). Hence,

schizophrenia is a relevant comparison group for MDD and may highlight the sensitivity of the

task in detecting N400 anomalies.

Methods Participants

Individuals diagnosed with major depressive disorder (MDD) were recruited from an

acute inpatient psychiatry unit in a moderate-size, mid-western city (Studies 1 and 2) and from a

large eastern city (Study 3). Individuals diagnosed with schizophrenia (SCZ) were recruited

from an inpatient unit (Study 2). Individuals with dysthymia and non-patient comparison

participants were recruited from both communities via newspaper advertisements and paid $5-

$10/per hour for participation. Inpatient MDD and SCZ participants were not paid for

participation. For all studies, left-handed participants were excluded. Participants were

diagnosed with current MDD (Studies 1-3), dysthymic disorder (Study 3), or schizophrenia

(Study 2) using the Structured Clinical Interview for the DSM-III-R (Study 1) or DSM-IV (Studies

2 and 3) (SCID; First et al., 1996; Spitzer et al., 1990). To ensure diagnostic consensus,

interviews for Studies 1 and 2 were completed by advanced doctoral students in clinical

psychology closely supervised by a clinical psychologist (GAM). In Study 3, a clinical

psychologist (PJD) and advanced graduate students conducted the clinical interviews.

Participants were included in the study only if a consensus diagnosis was reached by trained

supervisees of PJD in a review of taped interviews.

Physiological Recording

N400 in Depression 6

The electroencephalogram (EEG) was recorded from nine International 10-20 System

sites (Jasper, 1958): Fz, Cz, Pz, F3, C3, P3, F4, C4, and P4, referenced to the left mastoid

(Studies 1 and 2) or left earlobe (Study 3). EEG activity recorded from the right mastoid or right

earlobe was subsequently used to compute an average mastoid reference ([A1+A2]/2; Miller et

al., 1991). In Studies 1 and 2, Beckman miniature Ag-AgCl electrodes were used for the EEG

recording. In Study 3, EEG was collected using a tin-electrode cap (Electro-Cap International,

Inc., Eaton, OH). For all three studies, two orthogonal channels of the electro-oculogram (EOG)

recorded vertical and horizontal eye movements using electrodes placed near the outer canthi

and at left supra- and suborbital sites. Impedances were kept below 10 kΩ.

EEG was amplified using a Grass Model 12 (Studies 1 and 2) or an S. A. Instruments

Custom Bioamp polygraph (Study 3) with a .01-30 Hz analog bandpass filter. Data were

sampled at 125 Hz (Studies 1 and 2) or 512 Hz (Study 3) for 1400 ms beginning 200 ms prior to

the onset of the last word in each sentence.

Procedure

After an initial interview that screened prospective participants for handedness, head

injury, seizures, normal or corrected-to-normal vision, major medical illness, and developmental

disabilities, eligible participants were given an individual lab tour and provided written consent

prior to the SCID diagnostic interview.

Participants returned within 30 days for the electrophysiological portion of the study. On

average, the length of time between the diagnostic interview and EEG study was 2 weeks.

Following electrode application, participants were instructed to relax and focus on the screen in

front of them. Consistent with other depression and schizophrenia studies (e.g., Adams et al.,

1993; Deldin et al., 2006; Kuperberg et al., 2006), sentences were presented one word at a

time. Thus, participants were told that the briefly presented words formed sentences. They

were asked to read the sentences silently and told that they would be asked questions about the

N400 in Depression 7

sentences at the end of the session. Subjects were then presented with a block of 30

sentences; each sentence stem was repeated three times, once each with the terminal word

being a positive, neutral, or negative word (e.g., Today I am feeling (happy, okay, or sad)). The

order of sentence terminal word ending (i.e., negative, positive, neutral) was random within the

constraints that no more than four of the same type occurred consecutively. Additionally,

sentences were counterbalanced such that the first presentation, used for N400 analyses,

varied in emotional ending. Sentences were 6-8 words in length. The inter-stimulus interval

was 400 ms: each word appeared on the screen for 200 ms, with 200 ms of blank screen

between each word. An inter-trial interval of 2600 ms occurred between the offset of the last

word of a sentence, indicated by a period, and the onset of the next sentence.

Stimuli. First, a list of emotional adjectives was compiled from several studies

(Anderson, 1968; Bellezza et al., 1986) and was supplemented by the experimenters. Students

then rated each word on five-point Likert scale for happy/sad, arousing/calm, and

dominant/passive. Valence ratings for the chosen types of words (positive, neutral, and

negative) were found to be significantly different, F(4,85)=346.58, p<.001. Words were also

classified as either high or low arousal. Emotional words (positive and negative) had reliably

high and similar arousal ratings relative to neutral words, F(4,85)=152.28, p<.001. In addition,

average length, F(4,85)=1.66, p<.17, and frequency, F(4,85)=.14, p<.97, of the terminal words

were comparable across stimulus conditions. A total of 90 emotion sentences were then

developed using a base sentence and various terminal words chosen using the ratings

discussed above. The sentences were the same across the three studies.

Data Reduction and Analyses

Eye movement correction procedures were performed (Gratton et al., 1983; Miller et al.,

1988, for Studies 1 and 2; James Long Co., unpublished, for Study 3). Sentence type was

categorized according to the valence of the word (positive, neutral, or negative) that ended the

N400 in Depression 8

sentence. EEG was manually inspected for the presence of residual eye movements or EMG

artifact, and contaminated epochs were removed from the analysis. Within-subject average

ERP waveforms were computed for each site and valence type. N400 scores were computed

as the average voltage between 300 and 600 ms deviated from a 200 ms pre-stimulus baseline.

______________________________

Insert Fig 1 about here

______________________________

Data analysis. For each study, N400 scores collapsed across the midline (Fz, Cz, Pz)

were submitted to a Group [MDD, nondepressed (Studies 1, 2 and 3); MDD, SCZ,

nondepressed controls (Study 2); or MDD, dysthymia, nondepressed (Study 3)] x Valence

(positive, neutral, negative) repeated-measures analysis of variance (ANOVA). ANOVAs were

supplemented with simple-effects ANOVAs and Bonferroni tests as needed to interpret effects

significant at the two-tailed, 0.05 level. Reported probability values reflect the Huynh and Feldt

(1976) degrees of freedom correction. Finally, consistency of the three individual studies were

then confirmed using the Stouffer meta-analytic procedure (Bush et al., 1954) to determine

whether each group exhibited statistically significant N400 scores (area measurement) for

positive, neutral, or negative words. In this procedure (Bush et al., 1954), p values for the

studies were converted to z scores, which were summed and divided by the square root of the

number of studies.

Results

Participant Characteristics

Demographics. Across studies demographic characteristics were similar among MDD,

dysthymia, and healthy control groups (see Table 1). The schizophrenia group had

proportionally more males and was less educated than the other groups. These characteristics

N400 in Depression 9

are consistent with evidence that males have a higher lifetime risk of developing schizophrenia

(Tandon et al., 2008) and are less likely to complete high school (e.g., Carr et al., 2004).

______________________________

Insert Table 1 about here

______________________________

Within-Study Results for Positive, Neutral, and Negative Words

Figure 1 present ERP grand averages for each study, and Tables 2 and 3 provide

ANOVA results. Across all three studies, there was no difference in N400 amplitude between

the healthy controls and the depressed groups (both major depression and dysthymia). This

result was confirmed with the meta-analytical procedure (see Table 2). The only significant

result concerned the SCZ group (Study 2). Follow-up 3 (Group: MDD, SCZ, control) x 3

(Valence: negative, positive, neutral) simple-effects ANOVAs were conducted for each word

type. Differences were found for negative (F (2, 32)= 4.28, p<.02) but not positive (F (2, 32)=

1.51, p=.24) or neutral (F (2, 32)= 2.34, p=.11) words. Post-hoc Bonferroni comparisons for

negative word endings revealed larger N400 in schizophrenia patients compared to major

depression (p<.05) and healthy control (p<.03) groups. There was no difference between the

MDD and healthy controls (p=.99).

______________________________

Insert Tables 2 and 3 about here

______________________________

Discussion

N400 results suggest normal language function during emotion processing in

depression. Thus, N400 as a measure of semantic integration (Brown and Hagoort, 1993) is

not impaired for emotionally-valenced words. These results point to models of depression that

N400 in Depression 10

emphasize diverse functional disruptions in neural networks (e.g., Fitzgerald et al., 2008;

Mathews et al., 2008) rather than discrete cognitive deficits. For example, cortical-subcortical

responses to cognitive and emotional tasks have been shown to differ among depressed

individuals (Siegle et al., 2007). Additionally, brain imaging studies do not consistently show

lateralized effects in depression or reliable neural patterns (e.g., Fitzgerald et al., 2008). Taken

together, accumulating data indicate that abnormal function among brain regions, possibly to

varying degrees (e.g., strength of cortical-subcortical interactions, neural compensatory

processes) may be a factor in individual differences in depression rather than specific deficits

(e.g., hypoactive left frontal lobe). Thus, although particular neural networks are implicated in

depression, the varying ability of a given task to tap into abnormal function and individual

differences in neural network function may help explain mixed results in language processing for

emotional words.

Of note, schizophrenia patients exhibited a larger N400 (more negative response) to

negative words than the depressed or healthy controls, and there was no difference between

these groups in response to positive or neutral sentence endings. This provocative finding

suggests a violation of expectancy only for the negative words, indicating a greater surprise to

negative sentence endings. This finding is consistent with studies that show impaired

processing of emotional stimuli in schizophrenia (e.g., for review see Gold et al., 2009;

Rockstroh et al., 2006). Interestingly, the results in this study were specific to negative valence,

whereas other studies have found differential processing for both positive and negative

valenced stimuli compared to neutral stimuli. While the specific finding argues against a general

deficit, we cannot rule out the possibility that factors such as reading ability and/or working

memory deficits may have impacted results.

Given the small sample size of the schizophrenia group, results are in need of

replication. Nevertheless, results suggest that schizophrenia patients have a relatively large

violation of expectancy to negative sentence endings than do depressed patients and healthy

N400 in Depression 11

controls. This finding may help elucidate cognition-emotion interactions in schizophrenia. For

example, operation of the semantic network structure in schizophrenia is not entirely clear

though it has been recently proposed that problems in schizophrenia involve both an initial

overly activated semantic network and later inhibition difficulties (e.g., failure to rely on context)

(Niznikiewicz et al., in press). The present significant results were restricted to negative word

endings to ambiguous sentences. Therefore, it may be the case that these stimuli reflect problems

integrating negatively-valenced stimuli within a vague context though further study is needed to

better understand the extent to which results reflect early or later semantic network abnormalities.

Limitations of this study include the broad scoring window used to measure N400 (i.e.,

300 ms to 600 ms) to ensure N400 was captured in patient groups. It is possible that results for

some participants may include some later positivity. Additionally, the sample sizes in each

study, particularly schizophrenia patients, were small, which may have reduced our ability to find

within group differences based on the meta-analytic calculations. All the schizophrenia patients

were on medications, and we could not control for medication in that group. Also, although the

word endings were in the first person and intended to be self-referential, the same words were

used in each study. Therefore, it is possible that individual differences for self-referent

emotional words reduced sensitivity to detect N400 effects for mood-incongruent words.

Despite limitations, this study provides further evidence for normal N400 in MDD and

dysthymic participants. This study expands on N400 studies of neutral stimuli by suggesting

N400 to emotional stimuli is also normal in MDD and dysthymia. Finally, this study

demonstrates that schizophrenia patients exhibit abnormal information processing as indexed

by enhanced N400 to negative stimuli thus providing further evidence of the specificity of

cognitive biases in mood disordered participants and disruption of emotional and language

processing in schizophrenia.

N400 in Depression 12

References

Adams, J., Faux, S.F., Nestor, P.G., Shenton, M.E., Marcy, B., Smith, S., 1993. ERP

abnormalities during semantic processing in schizophrenia. Schizophr. Res. 10, 247-

257.

An, S.K., Lee, S.J., Lee, C.H., Cho, H.S., Lee, P.G., Lee, C.I., Lee, E., Roh, K.S., Namkoong,

K., 2003. Reduced P3 amplitudes by negative facial emotional photographs in

schizophrenia. Schizophr. Res. 64, 125-135.

Anderson, N.H., 1968. Likableness ratings of 555 personality-trait words. J. Pers. Soc. Psychol.

9, 272-279.

Atchley, R.A., Ilardi, S.S., Enole, A., 2003. Hemispheric asymmetry in the processing of

emotional content in word meanings: the effect of current and past depression. Brain

Lang. 84, 105-119.

Bell, M., Bryson, G., Lysaker, P., 1997. Positive and negative affect recognition in

schizophrenia: a comparison with substance abuse and normal control subjects.

Psychiatry Res. 73, 73-82.

Bellezza, F.S., Greenwald, A.G., Banaji, M.R., 1986. Words high and low in pleasantness as

rated by male and female college students. Behav. Res. Meth. Instrum. Comput. 18,

299-303.

Berenbaum, H., Kerns, J.G., Vernon, L.L., Gomez, J.J., 2008. Cognitive correlates of

schizophrenia signs and symptoms: II. Emotional disorders. Psychiatry Res. 159, 157-

162.

Besche-Richard, C., Passerieux, C., Hardy-Baylé, M.-C., 2002. Lexical decision tasks in

depressive patients: semantic priming before and after clinical improvement. Eur.

Psychiatry 17, 69-74.

Brown, C., Hagoort, P., 1993. The processing nature of the N400: evidence from masked

priming. J. Cogn. Neurosci. 5, 34-44.

N400 in Depression 13

Bush, R.R., Mosteller, F., Thompson, G.L., 1954. A formal structure for multiple-choice

situations. In: Thrall, R.M., Coombs, C.H., Davis, R.L. (Eds), Decision Processes, viii.

332, Oxford, England: John Wiley and Sons, Inc., pp. 99-126.

Carr, V.J., Lewin, T.J., Neil, A.L., Halpin, S.A., Holmes, S., 2004. Premorbid, psychosocial and

clinical predictors of the costs of schizophrenia and other psychoses. Br. J. Psychiatry

184, 517-525.

Caseras, X., Garner, M., Bradley, B.P., Mogg, K., 2007. Biases in visual orienting to

negative and positive scenes in dysphoria: an eye movement study. J. Abnorm.

Psychol. 116, 491-497.

Dannlowski, U., Kersting, A., Arolt, V., Lalee-Mentzel, J., Donges, U.S., Suslow, T., 2006.

Unimpaired automatic processing of verbal information in the course of clinical

depression. Depress. Anxiety 23, 325-330.

Davidson, R.J., Pizzagalli, D., Nitschke, J.B., Putnam, K., 2002. Depression: perspectives from

affective neuroscience. Annu. Rev. Psychol. 53, 545-574.

Deldin, P.J., Deveney, C.M., Kim, A.S., Casas, B.R., Best, J.L., 2001. A slow wave

investigation of working memory biases in mood disorders. J. Abnorm. Psychol. 110,

267-281.

Deldin, P., Keller, J., Casas, B.R., Best, J., Gergen, J., Miller, G.A., 2006. Normal N400 in mood

disorders. Biol. Psychol. 71, 74-79.

Dougherty, F.E., Bartlett, E.S., Izard, C.E., 1974. Responses of schizophrenics to expressions

of the fundamental emotions. J. Clin. Psychol. 30, 243-246.

First, M.B., Spitzer, R.L., Gibbon, M., Williams, J.B.W., 1996. Structured Clinical Interview for

DSM-IV Axis I Disorders, Clinician Version (SCID-CV). Washington, DC: American

Psychiatric Press, Inc.

Fitzgerald, P.B., Laird, A.R., Maller, J., Daskalakis, Z.J., 2008. A meta-analytic study of

changes in brain activation in depression. Hum. Brain Mapp. 29, 683-695.

N400 in Depression 14

Georgieff, N., Dominey, P.F., Michel, F., Marie-Cardine, M., Dalery, J., 1998. Semantic priming

in major depressive state. Psychiatry Res. 78, 29-44.

Gold, J.M., Hahn, B., Strauss, G.P., Waltz, J.A., 2009. Turning it upside down: areas of

preserved cognitive function in schizophrenia. Neuropsychol. Rev. 19, 294-311.

Gotlib, I.H., Krasnoperova, E., Yue, D.N., Joormann, J., 2004. Attentional biases for

negative interpersonal stimuli in clinical depression. J. Abnorm. Psychol. 113, 127-135.

Gratton, G., Coles, M.G.H., Donchin, E., 1983. A new method for off-line removal of ocular

artifact. Electroenceph. Clin. Neurophysiol. 55, 468–484.

Heller, W., Nitschke, J., 1997. Regional brain activity in emotion: a framework for understanding

cognition in depression. Cogn. Emot. 11, 637-661.

Hokama, H., Hiramatsu, K.I., Wang, J., O’Donnell, B.F., Ogura, C., 2003. N400 abnormalities

in unmedicated patients with schizophrenia during a lexical decision task. Int.

J. Psychophysiol. 48, 1-10.

Huynh, H., Feldt, L.S., 1976. Estimation of the box correction for degrees of freedom from

sample data in randomized block and split-plot designs. J. Educ. Behav. Stat.

1, 69–82.

Iakimova, G., Passerieux, C., Foynard, M., Fiori, N., Besche, C., Laurent, J.P., Hardy-Baylé,

M.C., 2009. Behavioral measures and event-related potentials reveal different

aspects of sentence processing and comprehension in patients with major depression.

J. Affect. Disord. 113, 188-194.

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

Electroencephalogr. Clin. Neurophysiol. 10, 371-375.

Johnstone, T., van Reekum, C.M., Urry, H.L., Kalin, N.H., Davidson, R.J., 2007. Failure to

regulate: counterproductive recruitment of top-down prefrontal-subcortical circuitry in

major depression. J. Neurosci. 27, 8877-8884.

Joormann, J., Gotlib, I.H., 2006. Is this happiness I see? Biases in the identification of

N400 in Depression 15

emotional facial expressions in depression and social phobia. J. Abnorm. Psychol. 115,

705-714.

Kiang, M., Kutas, M., Light, G.A., Braff, D.L., 2007. Electrophysiological insights into

conceptual disorganization in schizophrenia. Schizophrenia Res. 92, 225-236.

Klumpp, H., Deldin, P., in press. Review of brain functioning in depression for semantic

processing and verbal fluency. Int. J. Psychophysiol.

Kostova, M., Passerieux, C., Laurent, J.P., Hardy-Baylé, M.C., 2003. An electrophysiologic

study: can semantic context processes be mobilized in patients with thought-disordered

schizophrenia? Can. J. Psychiatry 48, 615-619.

Kostova, M., Passerieux, C., Laurent, J.P., Hardy-Baylé, M.C., 2005. N400 anomalies in

schizophrenia are correlated with the severity of formal thought disorder. Schizophrenia

Res. 78, 285-291.

Kring, A.M., Moran, E.K., 2008. Emotional response deficits in schizophrenia: insights from

affective science. Schizophr. Bull. 34, 819–834.

Kuperberg, G.R., Sitnikova, T., Goff, D., Holcomb, P.J. 2006. Making sense of sentences in

schizophrenia: electrophysiological evidence for abnormal interactions between

semantic and syntactic processing. J. Abnorm. Psychol. 115, 251-265.

Kutas, M., Hillyard, SA., 1980a. Reading senseless sentences: brain potential reflects semantic

incongruity. Science 207, 204-206.

Kutas, M., Hillyard, S.A., 1980b. Reading between the lines: event-related brain potentials

during natural sentence processing. Brain Lang. 11, 354-373.

Levin, R., Heller, W., Mohanty, A., Herrington, J., Miller, G., 2007. Cognitive deficits in

depression and functional specificity of regional brain activity. Cognit. Ther. Res., 31, 211-

233.

Mathews, A., Ridgeway, V., Williamson, D.A., 1996. Evidence for attention to threatening

stimuli in depression. Behav. Res. Ther. 34, 695-705.

N400 in Depression 16

Mathews, S.C., Strigo, I.A., Simmons, A.N., Yang, T.T., Paulus, M.P., 2008. Decreased

functional coupling of the amygdala and supragenual cingulate is related to increased

depression in unmedicated individuals with current major depressive disorder. J Affect

Disord. 111, 13-20.

Mayberg, H.S., 2006. Defining neurocircuits in depression: strategies toward treatment

selection based on neuroimaging phenotypes. Psychiatr Ann. 36, 259-268.

Miller, G.A., Gratton, G., Yee, C.M., 1988. Generalized implementation of an eye movement

correction procedure. Psychophysiology 25, 241–243.

Miller, G.A., Lutzenberger, W., Elbert, T., 1991. The linked-reference issue in EEG and ERP

recording. J. Psychophysiol. 5, 273–276.

Nikendei, C., Dengler, W., Wiedemann, G., Pauli, P., 2005. Selective processing of pain-related

word stimuli in subclinical depression as indicated by event-related brain potentials. Biol.

Psychol. 70, 52-60.

Niznikiewicz, M., Mittal, M.S., Nestor, P.G., McCarley, R.W. in press. Abnormal inhibitory

processes in semantic networks in schizophrenia. Int. J. Psychophysiol.

Ohta, K., Uchiyama, M., Matsushima, E., Toru, M., 1999. An event-related potential study in

schizophrenia using Japanese sentences. Schizophr. Res. 40, 159-170.

Rockstroh, B., Junghöfer, M., Elbert, T., Buodo, G., Miller, G.A., 2006. Electromagnetic brain

activity evoked by affective stimuli in schizophrenia. Psychophysiology 43, 431-439.

Serfaty, M.A., Bothwell, R., Marsh, R., Ashton, H., Blizard, R., Scott, J., 2002. Event-related

potentials and cognitive processing of affectively toned words in depression. J.

Psychophysiol. 16, 56-66.

Shimizu, H., Saito, H., Hoshiyama, M., 2006. Cognitive mechanism for meaning of emotive

words in depressed personality: an event-related potential study. Nagoya J. Med. Sci.

68, 35-44.

Siegle, G.J., Thompson, W., Carter, C.S., Steinhauer, S.R., Thase, M.E., 2007. Increased

N400 in Depression 17

amygdala and decreased dorsolateral prefrontal BOLD responses in unipolar

depression: related and independent features. Biol. Psychiatry 61, 198-209.

Sitnikova, T., Salisbury, D.F., Kuperberg, G., Holcomb, P.J., 2002. Electrophysiological

insights into language processing in schizophrenia. Psychophysiology 39, 851-860.

Spitzer, R.L., Williams, J.B.W., Gibbon, M., First. M.B.,1990. Structured clinical interview for the

DSM-III-R patient edn (SCID-P, Version 1.0). Washington, DC: American Psychiatric

Press, Inc.

Strandburg, R.J., Marsh, J.T., Brown, W.S., Asarnow, R.F., Guthrie, D., Harper, R., 1997.

Event-related potential correlates of linguistic information processing in schizophrenics.

Biol. Psychiatry 42, 596-608.

Surguladze, S.A., Young, A.W., Senior, C., Brebion, G., Travis, M.J., Phillips, M.L., 2004.

Recognition accuracy and response bias to happy and sad facial expressions in patients

with major depression. Neuropsychology 18, 212-218.

Suslow, T., Roestel, C., Arolt, V., 2003. Affective priming in schizophrenia with and without

affective negative symptoms. Eur. Arch. Psychiatry Clin. Neurosci. 253, 292-300.

Tandon, R., Keshavan, M.S., Nasrallah, H.A. 2008. Schizophrenia, “just the facts” what we

know in 2008. 2. Epidemiology and etiology. Schizophr. Res. 102, 1-18.

Weisbrod, M., Trage, J., Hill, H., Sattler, H.D., Maier, S., Kiefer, M., Grothe, J., Spitzer, M.,

1999. Emotional priming in depressed patients. Ger. J. Psychiatr. 2, 19-47.

Yoon, K.L., Joormann, J., Gotlib, I.H., 2009. Judging the intensity of facial expressions of

emotion: depression-related biases in the processing of positive affect. J. Abnorm.

Psychol. 118, 223-228.

N400 in Depression 18

Table 1

Means and Standard Deviations in Parentheses for Participant Characteristics

______________________________________________________________ _______ Study 1 N Medication Age Education Female MDD 17 70.6% 35.1 (9.7) 14.12 (2.0) 57.1% Controls 12 0% 31.7 (12.9) 14.75(1.5) 36.4% Study 2 Schizophrenia 7 100% 33.2 (9.0) 11.00 (.71)* 14.3%* MDD 12 100% 31.0 (9.2) 14.29 (2.0) 75.0% Controls 14 0% 44.7 (15.5) 15.75 (1.5) 57.1% Study 3 MDD 18 27.8% 40.3 (13.1) 14.50 (2.7) 50.0% Dysthymia 14 14.3% 42.7 (13.4) 15.38 (2.1) 71.4% Controls 19 0 % 36.8 (14.6) 16.07 (2.0) 52.6%

______________________________________________________________ _______

*Significantly different from major depressive disorder, dysthymia, and healthy control groups

(p<.05).

N400 in Depression 19

Table 2. ANOVA Results Within Study and Meta-Analysis Results ____________________________________________________________________________ a. Group main effects ____________________________________________________________________________ Study Groups d.f. F p ____________________________________________________________________________ 1 Control, MDD 1, 27 .23 .64 2 Control, MDD 1, 24 1.97 .17 2 Control, MDD, Schizophrenia 2, 30 1.90 .17 3 Control, MDD 1, 35 .22 .64 3 Control, MDD, Dysthymia 2, 48 .33 .72 b. Valence main effects ____________________________________________________________________________ Study Groups d.f. F p ____________________________________________________________________________ 1 Control, MDD 2, 54 .10 .91 2 Control, MDD 2, 48 1.05 .36 2 Control, MDD, Schizophrenia 2, 60 4.74 .03 3 Control, MDD 2, 70 1.37 .26 3 Control, MDD, Dysthymia 2, 96 .81 .45 c. Group x valence interaction ____________________________________________________________________________ Study Groups d.f. F p ____________________________________________________________________________ 1 Control, MDD 2, 54 1.41 .25 2 Control, MDD 2, 48 2.50 .09 2 Control, MDD, Schizophrenia 4, 60 5.03 .01 3 Control, MDD 2, 70 .13 .87 3 Control, MDD, Dysthymia 4, 96 .28 .89 ____________________________________________________________________________

Stouffer Meta-Analytic Values and Standard Deviations in Parentheses

___________________________________________________________________________ Major Depression Dysthymia Schizophrenia Controls Negative words 0.08 (1.34) 0.36 (1.53) -1.72 (3.52) 0.04 (1.44) Positive words -0.01 (1.32) 0.23 (1.13) 0.92 (3.61) -0.15 (1.54) Neutral words 0.14 (1.40) 0.02 (1.35) -1.04 (3.22) 0.09 (1.39) Note: Value = 1.95 is statistically significant at 90% level.

N400 in Depression 20

Table 3 N400 means (and standard deviations) by group and study ____________________________________________________________________________

N400 Means and Standard Deviations in Parentheses

____________________________________________________________________________ Study 1 Major Depression Dysthymia Schizophrenia Controls Negative Words 2.58 (1.84) n/a n/a 2.69 (3.55) Positive Words 3.12 (2.22) n/a n/a 1.91 (4.53) Neutral Words 2.63 (2.92) n/a n/a 2.30 (2.48) Study 2 Negative Words 2.34 (1.71) n/a -6.80 (1.68) 2.98 (1.60) Positive Words 1.28 (2.05) n/a 5.34 (1.03) 3.09 (1.90) Neutral Words 2.41 (2.71) n/a -2.30 (1.10) 2.73 (1.64) Study 3 Negative Words 2.07 (3.23) 2.55 (2.71) n/a 1.58 (2.30) Positive Words 1.98 (2.70) 2.28 (1.84) n/a 1.52 (2.31) Neutral Words 2.42 (2.42) 2.38 (2.36) n/a 2.31 (2.97) ____________________________________________________________________________

N400 in Depression 21 Figure 1. Grand-average waveforms from Pz for Study 1, 2, and 3. Y axis is in microvolts. X axis is in milliseconds. Waveforms include a 200 pre-stimulus baseline subtraction. Zero is onset of the terminal, emotional word.