Attentional Modulation of the P50 Suppression Deficit in Recent-Onset and Chronic Schizophrenia

doi:10.1037/a0018265

J Abnorm Psychol. Author manuscript; available in PMC 2011 February 1.

Published in final edited form as:

J Abnorm Psychol. 2010 February; 119(1): 31–39.

doi: 10.1037/a0018265

PMCID: PMC2844773

NIHMSID: NIHMS186069

Attentional Modulation of the P50 Suppression Deficit in Recent-Onset and Chronic Schizophrenia

Cindy M. Yee, Ph.D.,^1,² Terrance J. Williams, Ph.D.,¹ Patricia M. White, Ph.D.,¹ Keith H. Nuechterlein, Ph.D.,^1,² Donna Ames, M.D.,^2,³ and Kenneth L. Subotnik, Ph.D.²

¹Department of Psychology, University of California, Los Angeles

²Department of Psychiatry and Biobehavioral Sciences, University of California, Los Angeles

³Veterans Administration Greater Los Angeles Health Care System

Corresponding Author: Cindy M. Yee-Bradbury, Ph.D., Department of Psychology, University of California, Los Angeles, 1285 Franz Hall, Box 951563, Los Angeles, CA 90095-1563

The publisher's final edited version of this article is available at J Abnorm Psychol

See other articles in PMC that cite the published article.

Publisher's Disclaimer

Abstract

Schizophrenia is associated with deficits in P50 suppression to the second stimulus in a pair, a process often conceptualized as a preattentive index of sensory gating. This study assessed malleability of the deficit by determining whether early attentional control can influence P50 gating across different phases of schizophrenia. Participants included 28 patients in the recent-onset (n=16) or chronic (n=12) phase of illness, and 28 healthy comparison subjects. During the standard paradigm, chronic schizophrenia patients exhibited impaired P50 suppression relative to healthy subjects, while recent-onset schizophrenia patients were intermediate. Directing voluntary attention toward the initial stimulus yielded substantial improvements in the P50 ratio; recent-onset schizophrenia patients achieved ratio scores comparable to those of healthy participants while chronic patients also improved and could no longer be distinguished clearly from the healthy comparison sample. Directing attention toward the second stimulus enhanced P50 amplitude to the second stimulus across groups, possibly because activation of the inhibitory mechanism was overridden or circumvented by task demands. Thus, P50 suppression may be primarily preattentive under standard conditions but manipulation of early attention can exert a modulatory influence on P50, indicating that the suppression deficit is malleable in schizophrenia without pharmacological agents.

Keywords: schizophrenia, P50, sensory gating, attention, stage of illness

Individuals with schizophrenia frequently experience difficulty discriminating relevant information from the unimportant (Freedman et al., 1996; McGhie & Chapman, 1961). Confronted with an ever-changing environment and the need to focus selectively while filtering out distracting information, ongoing challenges present as patients with schizophrenia attempt to work productively, live independently, and function socially. The capacity to maintain and control attention, frequently referred to as top-down processing, along with the inhibition of responses to distracters, typically achieved with bottom-up processes, are central to effective neurocognitive functioning but are impaired in schizophrenia (Gold, Fuller, Robinson, Braun, & Luck, 2007; Gur et al., 2007). Though often viewed as separate processes involving distinct brain networks (Posner, 2004), these two systems also can interact dynamically, such that attention to sensory stimuli may be modulated by bottom-up and top-down processes (Hillyard, Vogel, & Luck, 1998; Posner, 2004; Woldorff et al., 1993). Given such a relationship between the two systems, it is possible that deficits in one domain are amenable to improvement by intact aspects of the other network among individuals with schizophrenia. If it can be established, for example, that deficits in the inhibitory response to distracting information are responsive to improvement by attentional control systems, this information would provide a potential pathway for future pharmacological interventions and cognitive remediation techniques.

Although neurocognitive dysfunction is extensive in schizophrenia, not all attentional processes are compromised by the illness. Notably, top-down modulation of early auditory attention appears to be spared. Relying upon the high temporal resolution offered by event-related potentials (ERPs), Mathalon, Heinks and Ford (2004) demonstrated that frontally-mediated control of attention is intact in schizophrenia patients up until about 75 ms in the processing stream. It is possible, therefore, that early attention can be utilized to influence disrupted processes occurring within this temporal domain.

The compromised capacity to gate or inhibit responses to distracting information in schizophrenia has been studied extensively by presenting paired auditory stimuli to assess the degree of suppression of the P50 ERP component to the second stimulus (“S2”) relative to the response elicited by the first stimulus (“S1”), when separated by a 500 ms interstimulus interval. The relationship between the two stimuli is often quantified as the S2/S1 ratio, with healthy individuals typically scoring below .50 and schizophrenia patients exceeding this value (Adler et al., 2004; Freedman et al., 2000; Patterson et al., 2008; Waldo et al., 1991). Meta-analytic studies reveal a robust effect size for the P50 deficit in schizophrenia that exceeds most other abnormalities (Bramon, Rabe-Hesketh, Sham, Murray, & Frangou, 2004; Heinrichs, 2001; Patterson et al., 2008). Although deficits in P50 suppression can be restored by pharmacological agents, such as nicotine and a subset of second-generation antipsychotics (Adler, Hoffer, Wiser, & Freedman, 1993; Adler et al., 2004; Light, Geyer, Clementz, Cadenhead, & Braff, 2000; Olincy et al., 2006), it has yet to be established whether voluntary control of attention can be engaged to improve or even normalize P50 gating in schizophrenia. Thus, the objective of the present study was to determine the extent to which early attentional control might improve inhibitory processing of the P50 component of the ERP in schizophrenia within the context of a sensory gating paradigm.

In evaluating another neurophysiological deficit in schizophrenia, we found that enhanced attention to the target stimulus improved eye tracking performance to such a degree that schizophrenia patients with a recent onset of illness were able to track as accurately as healthy individuals (Yee, Nuechterlein, & Dawson, 1998). Similar findings have been obtained in chronic schizophrenia patients (Holzman, Levy, & Proctor, 1976; Iacono, Tuason, & Johnson, 1981) with the level of improvement generally less than that observed in patients closer to their first episode of illness, possibly as a result of degenerative processes, sustained neuroleptic treatment, adaptation to a recurring illness or other factors associated with the chronic phase of schizophrenia.

The relative absence of research investigating the malleability of the P50 in schizophrenia patients, aside from pharmacological interventions, can be attributed to two factors. From a theoretical perspective, P50 gating has been conceptualized within the framework of an automatic, inhibitory process model (Braff & Light, 2004; Freedman et al., 1996). The prevailing view is that P50 suppression provides an index of sensory gating whereby S1 automatically activates an inhibitory process which then suppresses responsiveness to S2. In schizophrenia patients, the suggestion is that the initial stimulus fails to activate an inhibitory response to the second stimulus and thereby allows S2 to interfere with the processing of S1. The initial stimulus, therefore, is viewed as failing to close off processing of a subsequent auditory event that ensues almost immediately. Under this model, P50 suppression does not require voluntary attention and occurs automatically (Freedman et al., 1987; Freedman et al., 1996).

Second, the preattentive nature of P50 suppression is supported by studies in which healthy individuals were instructed to allocate voluntary attention towards each of the two stimuli but without any detectable influence on P50 or its suppression (Jerger, Biggins, & Fein, 1992; White & Yee, 1997). Relying upon normal variations in attentiveness among healthy participants, Cardenas, Gill and Fein (1997) further determined that P50 and its suppression are not affected by differing levels of wakefulness. There is some suggestion, however, that maintaining a running count of the second stimulus may serve to disrupt suppression (Guterman, Josiassen, & Bashore, 1992).

To our knowledge, there have been no investigations examining whether the P50 deficit in schizophrenia is also impervious to attentional manipulations. P50 suppression may be unaffected by voluntary attention in healthy individuals because the inhibitory process is intact and fully functional whereas, in schizophrenia patients, recruitment of voluntary attention could improve the deficit by compensating for a faulty preattentive inhibitory process. Alternatively, allocation of attentional processes might serve to further disrupt a process that is already compromised, depending upon how or when voluntary attention is engaged. Because it remains to be demonstrated how voluntary attention might influence P50 suppression in schizophrenia patients, three possible outcomes were evaluated in the current study.

Consistent with the automatic, inhibitory process model described above and with available evidence from healthy subjects, voluntary attention directed at either S1 or S2 should have no effect on P50 gating in schizophrenia patients. As an alternative, we examined the possibility that schizophrenia patients fail to allocate sufficient attention towards the incoming, initial stimulus which, in turn, leads to a failure to activate the P50 inhibitory process. In this case, P50 suppression is not entirely automatic but instead may be enhanced by top-down processes directed towards the auditory channel. Instructing schizophrenia patients to direct attention to S1, therefore, should help to fully engage the P50 inhibitory mechanism and lead to normal levels of P50 suppression. This option is described as a compensatory attention model.

A third scenario is that because the automatic P50 inhibitory mechanism is only partially intact in schizophrenia patients, it is easily disrupted when attention is allocated to other stimuli that may be generated internally or externally. Accordingly, impairments in P50 suppression should be exacerbated when voluntary attention is directed towards S2. This possibility, an attention disruption model, is not incompatible with the compensatory attention model because both alternatives allow for attentional effects but attribute them to different manners of action.

In the present study, moderating effects of phase of illness on the attentional manipulations also were examined, as the eye tracking results suggest that greater gains with voluntary attention may be achieved with individuals closer to the onset of schizophrenia. Consistent with a compensatory attention model, we predicted that directing voluntary attention towards S1 would activate an inhibitory response to S2 and result in improved P50 suppression, with more significant gains observed among patients with a recent onset of schizophrenia than those in the chronic phase. As suggested by an attention disruption model, we also hypothesized that because of its tenuous nature in schizophrenia, P50 and its suppression will be further compromised when voluntary attention is directed at the second, interfering stimulus, with the most deleterious effect on chronic patients who also were expected to show greatest P50 impairment overall (Brockhaus-Dumke et al., 2008).

Methods

Participants

Written informed consent was obtained from 28 outpatients diagnosed with schizophrenia and 28 healthy comparison subjects. Efforts were made to match each patient group with healthy comparison subjects on age, sex, level of parental education and ethnicity. To examine any effects associated with duration of illness, patients were classified as recent-onset (n=16) if they were within three years of their first psychotic episode (M=1.4 years; SD= 1.0) and as chronic (n=12) if more than five years had elapsed since their initial psychotic episode (M=13.2 years; SD= 5.3). Given possible anticholinergic effects on P50, antiparkinsonian medications were discontinued at least 24 hours prior to testing. The expanded Brief Psychiatric Rating Scale (Ventura et al., 1993) was administered to assess symptom levels during the preceding 2-week period. All patients were evaluated in a clinically stable state.

All recent-onset and 6 chronic schizophrenia patients were initially recruited for the UCLA “Developmental Processes in Schizophrenic Disorders” project. The remaining 6 chronic patients were participants in the naturalistic follow-up phase of a study of atypical medications conducted at the VA Greater Los Angeles Healthcare System. Comparison subjects were recruited from the community and were excluded if they had a history of schizophrenia, schizoaffective disorder or other major psychopathology, any alcohol or substance abuse in the last three months, or history of a major psychiatric disorder in a first-degree relative. Diagnoses were obtained with the Structured Clinical Interview for DSM-IV (First, Spitzer, GIbbon, & Williams, 2001; Ventura, Liberman, Green, Shaner, & Mintz, 1998). Additional exclusion criteria for all participants included neurological disorders, history of a head trauma or loss of consciousness for greater than five minutes, and current significant or habitual alcohol or substance abuse. Although one patient from each group smoked 15-20 minutes prior to the number task, this time interval is sufficient for P50 suppression to return to near baseline values (Adler et al., 1993). All other smokers refrained from cigarettes for at least 45 minutes before any condition.

Procedure

Baseline data were obtained while participants were seated comfortably and presented with 80 trials of paired high-intensity stimuli. During the attention tasks, four types of stimuli were delivered in random order: a) moderate-intensity single stimuli, b) moderate-intensity paired stimuli, c) high-intensity single stimuli, and d) high-intensity paired stimuli. A total of 160 trials was presented during each of the attention tasks with 52% of trials (n=82) comprised of high-intensity paired stimuli that were identical to those delivered during baseline. Each of the three remaining types of stimuli accounted for 16% of the total trials per task; the resulting data were not included in any of the subsequent analyses.

During the intensity task, participants were instructed to respond as quickly as possible with a button press whenever they detected a high-intensity stimulus, irrespective of whether the stimulus involved single or paired stimuli; this task served to direct attention to S1. During the number task, participants were instructed to respond as quickly as possible whenever they detected a pair of stimuli, irrespective of stimulus intensity; this task served to direct attention to S2. Order of presentation of the attention tasks was counterbalanced.

Auditory Stimuli

Auditory stimuli were 3 ms in duration, with a 500 ms interstimulus interval and a variable intertrial interval of 9-11 s between pairs. For 6 chronic patients, 2 recent-onset patients, and 14 healthy subjects, high-intensity stimuli were delivered at 90-dB SPL and moderate-intensity stimuli at 80-dB SPL. Stimuli were presented against 40-dB SPL background white noise over headphones. The auditory stimuli and background noise were created by amplifying white noise generated by a San Diego Instruments Sound Generator board. Stimuli for the remaining 6 chronic patients, 14 recent-onset patients, and 14 healthy subjects were presented through foam-insert earphones. Sound threshold levels were determined for each ear with high-intensity stimuli at 55 dB above threshold and moderate-intensity stimuli at 20 dB below the sound level of high-intensity stimuli. Auditory stimuli consisted of amplified white noise that was created using the Neuroscan STIM presentation system. Background noise was unnecessary because recordings were obtained in a soundproof chamber rather than the sound-attenuated room used previously which had a background noise level measuring at 46 dB SPL, emanating primarily from the amplifier system in the adjoining room and the ventilation system. As recommended by Veneklasen Associates, consultants in acoustics who provided sound measurements and recommendations on maintaining comparable conditions between rooms, ambient noise in the soundproof chamber was introduced by a small ventilation fan that did not exceed the noise criteria (NC) curve of 30. The change in labs and equipment was necessitated by local seismographic retrofitting. To ensure comparability of data across labs, extensive testing was undertaken with a separate sample of 16 healthy subjects. Auditory stimuli, for example, were generated using the two methods while electrophysiological data were acquired with the same recording equipment; mean P50 amplitudes to S1 (M=3.36, SD=1.64 versus M=3.34, SD=1.43) and S2 (M=1.24, SD=.77 versus M=1.27, SD=.43) were effectively identical.

Data Acquisition and Analysis

For the first set of participants described above, EEG was recorded from electrodes placed at Fz, Cz and Pz midline sites, and referenced to linked earlobes. EOG activity was recorded from electrodes placed above and below the left eye. Signals were acquired with a Model 12 Neurodata system (Grass, West Warwick, RI) with half-amplitude analog filters set at 0.1 and 1000 Hz. For remaining participants, EEG recordings were obtained with a high-density cap containing Ag/AgCl sintered electrodes (Falk Minow Services, Herrsching, Germany). Signals were acquired with a SynAmps system (Neuroscan, Charlotte, NC) with filters set at 0.5 and 200 Hz. As with stimulus presentation, the recording systems were assessed for comparability with the separate healthy sample by obtaining data from each unit while using identical auditory stimuli; again, no notable or significant differences were detected. Chronic patients and healthy subjects also were equally distributed between the two labs. Although data from a greater proportion of recent-onset patients were acquired with a SynAmps system, their P50 data and that of the healthy comparison group reported below (i.e., suppression ratios of M=.50, SD=.29 and M=.36, SD=.21, respectively) are largely comparable to data obtained with Grass instrumentation from previous samples of recent-onset patients and healthy comparison subjects (mean ratios of M=.55, SD=.32 and M=.38, SD=.18, respectively; see (Yee, Nuechterlein, Morris, & White, 1998). Across all participants, the sampling rate was 1000 Hz and only data obtained from Cz are included in this study.

Trials contaminated by subject movement were excluded and data were subtracted from a 200 ms prestimulus baseline. After correcting for eye movement artifact (Miller, Gratton, & Yee, 1988), single EEG trials were digitally filtered with a bandpass of 10-50 Hz to measure P50. For the intensity and number tasks, only trials with a correct behavioral response to high-intensity paired stimuli contributed to the ERP average. Trials with moderate-intensity stimuli were not examined further. P50 was identified as the most positive point between 40 and 70 ms after stimulus presentation. P30 amplitude and latency were scored as the maximum positivity occurring 20 to 40 ms after the stimulus. The maximum negativity between the P30 and P50 latencies was then used to measure P50 amplitude. P50 to S1 exceeded 0.5 microvolts for all participants and could be scored reliably. Reaction time (RT) was obtained from a button press made with the thumb of the dominant hand, and recorded from S1 for the intensity task and S2 for the number task.

Statistical Analysis

Preliminary analyses were conducted with lab as a factor to further verify comparability of the data. After determining that lab did not emerge as a significant main effect for any dependent variable and never interacted significantly with other factors, the data were merged. Additionally, no significant group differences were detected between the older and younger healthy comparison groups. Yoking of patient groups to separate comparison groups was eliminated in favor of a single, combined healthy sample to maximize the sample size for group contrasts. Although none of the P50 measures were associated with normal aging, P50 amplitude to S1 during the baseline correlated negatively with age in schizophrenia patients. We, therefore, undertook a hierarchical multiple regression analysis with group, age and their interaction. The absence of significant interaction effects confirmed that age was not a confounding variable. Analyses of variance (ANOVAs) were conducted with the Greenhouse-Geisser method to adjust degrees of freedom. Post hoc analyses relied upon Newman-Keuls and t-tests for between-group and within-group contrasts, respectively.

N100 was analyzed to examine comparability with the pattern of effects obtained previously (Jerger et al., 1992; White & Yee, 1997) and to verify along with the behavioral performance data that attention had been engaged. Because group differences were not detected, with the exception of healthy participants exhibiting a larger response to S1 during the intensity condition, these data are not included in the present report. Similarly, P50 latency was examined but no significant findings involving group or experimental conditions were obtained.

Results

Demographic and Clinical Characteristics

As indicated in Table 1, the patient groups were matched to their healthy comparison group on age, sex, level of parental education, and ethnicity with a single exception. Relative to the chronic schizophrenia patients, older healthy comparison subjects were significantly younger, F(1,22)=12.90, p=.01. Education levels were significantly higher in the young healthy comparison sample than in the recent-onset schizophrenia patient group, F(1,30)= 6.43, p=.05, and in the older comparison group relative to chronic schizophrenia patients, F(1,22)=8.90, p=.01. As compared with chronic schizophrenia patients, recent-onset patients had a significantly shorter duration of illness, F(1,23)=75.64, p=.001, were on a significantly lower dosage of antipsychotics, F(1,23)=11.13, p=.01, and had significantly lower BPRS scores, F(1,26)=7.29, p=.05

Table 1

Demographic and Clinical Characteristics of Participants

Behavioral Performance

Behavioral data appear in Table 2. The groups did not differ in mean percentage of correct responses during the intensity task, F(2,50)=.65, p=.53, or number task, F(2,50)=2.17, p=.12. Healthy comparison subjects were significantly faster in mean RT than the patient groups during the intensity task, with recent-onset schizophrenia patients providing significantly slower performance than either of the other two groups, F(2,53)=30.57, p=.001. During the number task, recent-onset patients were significantly slower again in performance than chronic schizophrenia patients and healthy comparison subjects, who did not differ from one another, F(2,53)=18.60, p=.001.

Table 2

Behavioral Performance

P50 Suppression Ratio

Grand-average ERP waveforms are presented in Figure 1. To assess for differences in P50 suppression prior to manipulating voluntary attention, the three groups were compared during the typical, passive P50 baseline condition. A significant main effect of group, F(2,53)=3.85, p=.028, revealed impaired P50 suppression in chronic schizophrenia patients (M=.61, SD=.36) relative to healthy subjects (M=.36, SD=.21), while recent-onset patients were intermediate (M=.50, SD=.29) and did not differ statistically from the other two groups.¹

Figure 1

Grand average event-related potential waveforms at the Cz recording site.

Mean P50 suppression ratio scores for each task condition are displayed in Figure 2. As predicted, directing voluntary attention to S1 eliminated any group differences; recent-onset (M=.35, SD=.25) and chronic schizophrenia patients (M=.52, SD=.38) exhibited levels of P50 suppression that were statistically indistinguishable from that of healthy comparison subjects (M=.34, SD=.23), as indicated by the absence of a main effect for group during the intensity task, F(2,53)=2.12, p=.13. The number task, in contrast, interfered with P50 suppression to a similar degree across recent-onset patients (M=.66, SD=.78), chronic patients (M=.72, SD=.40) and healthy subjects (M=.54, SD=.34), given significantly elevated levels from baseline as reflected by a main effect for condition, F(1,53)=4.67, p=.035, and the absence of a main effect for group, F(2,53)=0.64, p=.53.

Figure 2

Mean P50 suppression ratios.

In terms of the magnitude of P50 suppression differences between patients and healthy participants during the passive baseline condition, chronic schizophrenia patients exhibited the larger effect size of Cohen’s d=.95 as compared with a medium effect size of d=.58 for recent-onset patients. Directing attention to the initial stimulus served to eliminate any differences between recent-onset schizophrenia patients and healthy individuals (d=.04) while reducing the effect size in chronic patients (d=.64). Likewise, the amount of interference introduced by directing attention to the second stimulus exerted less of an impact on recent-onset patients (d=.22) than chronic patients (d=.50) relative to healthy comparison subjects.

P50 Amplitude

Mean P50 amplitudes are shown in Figure 3. To evaluate the extent to which group differences in passive baseline ratio scores reflect the response elicited by each of the paired stimuli and to provide a difference score measure of P50 suppression, a repeated-measures ANOVA for the effects of group and stimulus (S1 versus S2) was undertaken. The main effect of stimulus, F(1,53)=78.89, p=.001, was modified by a significant group by stimulus interaction, F(2,53)=3.54, p=.036. Post hoc comparisons determined that P50 amplitude to S1 was comparable between recent-onset schizophrenia patients (M=4.67, SD=2.91) and healthy comparison subjects (M=4.26, SD=2.08) but reduced in chronic patients (M=2.82, SD=1.03). Although the relative magnitude of P50 amplitude to S2 was in the expected direction for recent-onset (M=1.99, SD=1.06) and chronic schizophrenia patients (M=1.63, SD=0.90) relative to healthy individuals (M=1.53, SD=1.10), the group difference did not reach statistical significance. Thus, there appears to have been greater mediation of the P50 ratio score by the smaller S1 response in the chronic group.

Figure 3

Mean P50 amplitude to paired stimuli.

To examine a priori hypotheses concerning the specific impact of voluntary attention on P50 within each group while again providing a difference score measure of P50 suppression, condition (baseline versus intensity task versus number task) by stimulus (S1 versus S2) ANOVAs were conducted for each sample. These within-subjects analyses further serve to mitigate any differences that may have resulted from obtaining ERP recordings in two different settings, as the data for any given participant were always obtained in the same location.

P50 amplitude in healthy comparison subjects exhibited condition, F(2,54)=5.80, p=.006, stimulus, F(1,27)=100.59, p=.001, and condition by stimulus, F(2,54)=3.75, p=.034, effects. In keeping with the pattern of directed attention, P50 to S1 was significantly larger during the intensity task than during baseline. Similarly, P50 to S2 was enhanced when attention was directed to the second stimulus during the number task relative to levels observed during baseline and the intensity task.

Among recent-onset schizophrenia patients, P50 amplitude showed significant effects for stimulus, F(1,15)=25.92, p=.001, and for the condition by stimulus interaction, F(2,30)=4.26, p=.027. Consistent with hypothesized improvements in P50 suppression as a result of directing attention to S1, P50 to S2 was significantly inhibited or reduced during the intensity task relative to the other conditions. P50 to S1, however, remained comparable to baseline levels during the intensity task. Directing attention to S2 during the number task had the unanticipated impact of significantly reducing P50 to S1 relative to baseline levels.

Chronic schizophrenia patients exhibited significant P50 effects for condition, F(2,22)=7.14, p=.008, and stimulus, F(1,11)=38.27, p=.001, and a marginal effect for the condition by stimulus interaction, F(2,22)=3.66, p=.06. Given a priori hypotheses, planned comparisons were conducted to determine the influence of attention during each experimental condition. Directing attention to the initial stimulus during the intensity task served to augment P50 amplitude to S1 relative to baseline levels as hypothesized but failed to significantly impact P50 to S2. During the number task, voluntary attention to the second stimulus resulted in significant increases in P50 to S1 as compared with baseline and to S2 relative to response levels elicited during baseline and the intensity condition.

Discussion

Although P50 suppression has been regarded primarily as a preattentive process, the present results demonstrate that voluntary attention can exert a modulatory influence, thereby providing further evidence for the potential malleability of the P50 deficit in schizophrenia. As might be expected of intact inhibitory processes in healthy participants, there was no appreciable impact on the P50 suppression ratio when voluntary attention was directed towards the initial stimulus. In recent-onset schizophrenia patients, however, directing attention towards S1 suppressed P50 to S2 while exerting no significant impact on the initial response, resulting in a mean P50 ratio score at a level typically observed only in healthy populations. These data therefore demonstrate that P50 suppression deficits among recent-onset patients can be transiently normalized without the aid of pharmacologic agents.

Improved P50 ratio scores also were observed in chronic schizophrenia patients during the intensity task although gains were more modest and the amplitude data were less clearly indicative of suppressed responsiveness to S2. Instead, P50 to S1 increased to a mean level comparable to that of the healthy subjects while P50 to S2 did not change from baseline levels. Attempts to evaluate the pattern and significance of these data underscore the difficulty associated with inferring an inhibitory response from paired stimuli (Smith, Boutros, & Schwarzkopf, 1994). Specifically, changes in the ratio score resulting from alterations in P50 to S1 do not necessarily support the engagement of inhibitory processes; the improved ratio score in chronic schizophrenia patients could be attributed to an enhanced response to S1 alone, rather than any true change in the relationship between responses. A plausible alternative interpretation is that inhibition of the S2 response in chronic patients was improved by attention to S1, with inhibitory processes leading to no appreciable change in P50 to S2 despite augmentation of P50 to S1.

A similar situation is encountered in recent-onset schizophrenia patients when attention was directed towards S2 during the number task, with changes in P50 to S1 contributing to increases in the P50 ratio. Given the complex pattern of findings obtained in the present study and continued reliance upon the ratio measure of P50 suppression, future research should attempt to not only consider factors that might influence S1 or S2 amplitudes but to ascertain the extent to which these determinants influence P50 suppression. Beyond conceptual implications, discriminating between actual inhibition of the P50 and other alterations to the P50 ratio measure is a methodological necessity (White & Yee, 2006).

In chronic schizophrenia patients, direction of attention towards the second stimulus more definitively resulted in poor suppression with P50 to S2 exceeding baseline levels. It appears, therefore, that voluntary attention may have served to override or prevent activation of the inhibitory mechanism during the number task. This pattern of effects may be adaptive given the enhanced salience of S2 when participants are instructed to attend to a second stimulus, rather than eliciting an obligatory inhibition to a subsequent stimulus. Results from the healthy group during performance of the number task also revealed similar interference effects on the P50 ratio and contrast with prior data obtained with the same experimental manipulation in which voluntary attention had no significant impact (White & Yee, 1997). Inspection of the earlier data suggests that a similar pattern of disruption was present but not statistically significant, possibly due to a smaller sample of thirteen participants coupled with a modest experimental effect size among non-psychiatric subjects. Thus, similar to psychosocial stress (White & Yee, 1997; Yee & White, 2001), it appears that voluntary attention can interfere with P50 and its suppression even in healthy individuals (Guterman et al., 1992).

Stage of illness also was implicated in the basic P50 deficit and its malleability. A pattern of increasing disruption was observed during the passive paradigm from healthy to recent-onset to chronic groups and may be related to course of illness. Based upon consideration of the effect sizes, there was some suggestion that patients in the initial stage of schizophrenia achieved more substantial gains in improving P50 suppression than those in the chronic phase. Additionally, when directing attention towards the second stimulus the magnitude of disruption to P50 suppression may have been diminished in recent-onset patients relative to patients who had progressed to the chronic phase. It is possible that our ability to distinguish statistically between the patient samples was constrained by medication differences and sample size. Two-thirds of chronic schizophrenia patients, as compared with none of the recent-onset patients, were treated with clozapine, which may have a more pronounced effect in ameliorating the P50 deficit than other antipsychotic medications (Adler et al., 2004). Although we did not replicate this finding, clozapine could have exerted a normalizing effect on one of the underlying P50 mechanisms, such that suppression in chronic patients was already enhanced to some degree. Without clozapine, the P50 deficit might have been substantially greater in chronic schizophrenia patients and, therefore, potentially less amenable to improvement and more susceptible to disruption from voluntary attention than in recent-onset patients.

Another aspect of the study that may have influenced our findings was reliance on different laboratories. However, significant efforts were made to calibrate one laboratory against the other and appear to have been successful, given that differences were not detected following multiple attempts. As noted above, one other factor reducing any influence of the laboratory change is that a significant proportion of the statistical analyses relied upon a within-subjects design.

Taken together, results from the present study suggest that reductions in P50 amplitude to the second of two auditory stimuli may reflect preattentive inhibitory mechanisms during a standard passive paradigm. Directing early voluntary attention towards S2, however, was sufficient to alter P50 suppression in healthy individuals. Among schizophrenia patients, early voluntary attention exerted a considerable influence on P50 amplitude and therefore on the P50 ratio score, ranging from enhancing P50 to S1 and decreasing the P50 ratio to augmenting P50 to S2 and increasing the ratio score. Under some circumstances, therefore, it is possible that the P50 ratio may reflect the effects of early voluntary attention, either compensatory or disruptive in manner, rather than a strictly preattentive inhibitory mechanism. In a departure from the prevailing automatic, inhibitory process model of sensory gating in schizophrenia (Freedman et al., 1996), the current data offer support for compensatory and disruptive effects of attention.

It is noteworthy that the magnitude of the P50 deficit in schizophrenia patients has been found to be associated with the degree of attentional difficulties on clinical and neuropsychological measures (Cullum et al., 1993; Erwin, Turetsky, Moberg, Gur, & Gur, 1998; C.M Yee et al., 1998). One implication of these findings is that although directing early voluntary attention to S1 has the potential to improve P50 suppression, attentional processes may be compromised to such a degree that the automatic call of attention to the first of two identical stimuli is no longer functioning for some schizophrenia patients, particularly those in the chronic phase. If attention is sufficiently spared, however, voluntary direction of early attention to task-congruent stimuli may provide an opportunity to override or circumvent certain sensory gating deficits. It is possible that failures to detect the P50 deficit in some studies of schizophrenia may result when relatively loud and, therefore, attention-enhancing stimuli are utilized.

An important next step towards deciphering the contribution of these various processes will be to characterize the neuronal mechanisms that underlie P50 suppression and its attentional modulation in schizophrenia. The pattern of effects obtained in the present study is consistent with results of recent research examining the neural network associated with P50 gating during the standard paired-stimulus paradigm. Applying EEG source localization, we and others determined that a distributed neural network, involving the superior temporal gyrus, hippocampus, thalamus and dorsolateral prefrontal cortex (DLPFC), is involved in the generation of P50 in schizophrenia patients and healthy individuals (Tregellas et al., 2007; Williams, Yee, Nuechterlein, & Subotnik, 2009). The groups may differ, however, in the neural generators associated with the gating response. In healthy individuals, we observed hippocampal and, to a lesser degree, DLPFC involvement while in schizophrenia patients, only DLPFC activity was detected. Evidence for a modulatory influence of attention on P50 suppression further underscores potential interactions between the prefrontal cortex and other cortical or subcortical systems in sensory gating, particularly in schizophrenia patients.

Acknowledgments

This study was supported in part by grants MH57322, MH37705, and Center grant P50 MH066286 from the National Institute of Mental Health, Bethesda, MD. We thank Marilyn Kesler/West, Ph.D. for assistance with study procedures.

Footnotes

¹Because clozapine can improve P50 suppression (Adler et al., 2004; Light et al., 2000), associations with medication type were examined and revealed poorer suppression in chronic schizophrenia patients receiving clozapine (n=8; M=.75, SD=.32) than those on other atypicals (n=4; M=.32, SD=.30), F(1,11)=5.12, p=.047, possibly reflecting selection of clozapine for patients with a more severe course of illness.

The following manuscript is the final accepted manuscript. It has not been subjected to the final copyediting, fact-checking, and proofreading required for formal publication. It is not the definitive, publisher-authenticated version. The American Psychological Association and its Council of Editors disclaim any responsibility or liabilities for errors or omissions of this manuscript version, any version derived from this manuscript by NIH, or other third parties. The published version is available at www.apa.org/pubs/journals/abn.

References

Adler LE, Hoffer LD, Wiser A, Freedman R. Normalization of auditory physiology by cigarette smoking in schizophrenic patients. Am J Psychiatry. 1993;150(12):1856–1861. [PubMed]
Adler LE, Olincy A, Cawthra EM, McRae KA, Harris JG, Nagamoto HT, et al. Varied effects of atypical neuroleptics on P50 auditory gating in schizophrenia patients. Am J Psychiatry. 2004;161(10):1822–1828. [PubMed]
Braff DL, Light GA. Preattentional and attentional cognitive deficits as targets for treating schizophrenia. Psychopharmacology (Berl) 2004;174(1):75–85. [PubMed]
Bramon E, Rabe-Hesketh S, Sham P, Murray RM, Frangou S. Meta-analysis of the P300 and P50 waveforms in schizophrenia. Schizophr Res. 2004;70(2-3):315–329. [PubMed]
Brockhaus-Dumke A, Schultze-Lutter F, Mueller R, Tendolkar I, Bechdolf A, Pukrop R, et al. Sensory gating in schizophrenia: P50 and N100 gating in antipsychotic-free subjects at risk, first-episode, and chronic patients. Biol Psychiatry. 2008;64(5):376–384. [PubMed]
Cardenas VA, Gill P, Fein G. Human P50 suppression is not affected by variations in wakeful alertness. Biol Psychiatry. 1997;41(8):891–901. [PubMed]
Cullum CM, Harris JG, Waldo MC, Smernoff E, Madison A, Nagamoto HT, et al. Neurophysiological and neuropsychological evidence for attentional dysfunction in schizophrenia. Schizophrenia Research. 1993;10:131–141. [PubMed]
Erwin RJ, Turetsky BI, Moberg P, Gur RC, Gur RE. P50 abnormalities in schizophrenia: relationship to clinical and neuropsychological indices of attention. Schizophr Res. 1998;33(3):157–167. [PubMed]
First MB, Spitzer RL, GIbbon M, Williams JBW. Structured Clinical Interview for DSM-IV Axis I Disorders - Patient Edition (SCID-I/P) 2001
Freedman R, Adams CE, Adler LE, Bickford PC, Gault J, Harris JG, et al. Inhibitory neurophysiological deficit as a phenotype for genetic investigation of schizophrenia. Am J Med Genet. 2000;97(1):58–64. [PubMed]
Freedman R, Adler LE, Gerhardt GA, Waldo MC, Baker N, Rose GM, et al. Neurobiological studies of sensory gating in schizophrenia. Schizophrenia Bulletin. 1987;13:669–678. [PubMed]
Freedman R, Adler LE, Myles-Worsley M, Nagamoto HT, Miller C, Kisley M, et al. Inhibitory gating of an evoked response to repeated auditory stimuli in schizophrenic and normal subjects: Human recordings, computer simulation, and an animal model. Archives of General Psychiatry. 1996;53:1114–1121. [PubMed]
Gold JM, Fuller RL, Robinson BM, Braun EL, Luck SJ. Impaired top- down control of visual search in schizophrenia. Schizophr Res. 2007;94(1-3):148–155. [PMC free article] [PubMed]
Gur RE, Turetsky BI, Loughead J, Snyder W, Kohler C, Elliott M, et al. Visual attention circuitry in schizophrenia investigated with oddball event-related functional magnetic resonance imaging. Am J Psychiatry. 2007;164(3):442–449. [PubMed]
Guterman Y, Josiassen RC, Bashore TR. Attentional influence on the P50 component of the auditory event-related potential. International Journal of Psychophysiology. 1992;12:197–209. [PubMed]
Heinrichs RW. In Search of Madness: Schizophrenia and Neuroscience 2001
Hillyard SA, Vogel EK, Luck SJ. Sensory gain control (amplification) as a mechanism of selective attention: electrophysiological and neuroimaging evidence. Philos Trans R Soc Lond B Biol Sci. 1998;353(1373):1257–1270. [PMC free article] [PubMed]
Holzman PS, Levy DL, Proctor LR. Smooth pursuit eye movements, attention, and schizophrenia. Arch Gen Psychiatry. 1976;33(12):1415–1420. [PubMed]
Iacono WG, Tuason VB, Johnson RA. Dissociation of smooth-pursuit and saccadic eye tracking in remitted schizophrenics. An ocular reaction time task that schizophrenic perform well. Arch Gen Psychiatry. 1981;38(9):991–996. [PubMed]
Jerger K, Biggins C, Fein G. P50 suppression is not affected by attentional manipulations. Biological Psychiatry. 1992;31:365–377. [PubMed]
Light GA, Geyer MA, Clementz BA, Cadenhead KS, Braff DL. Normal P50 suppression in schizophrenia patients treated with atypical antipsychotic medications. Am J Psychiatry. 2000;157(5):767–771. [PubMed]
Mathalon DH, Heinks T, Ford JM. Selective attention in schizophrenia: sparing and loss of executive control. Am J Psychiatry. 2004;161(5):872–881. [PubMed]
McGhie A, Chapman J. Disorders of attention and perception in early schizophrenia. Br J Med Psychol. 1961;34:103–116. [PubMed]
Miller GA, Gratton G, Yee CM. Generalized implementation of an eye movement correction procedure. Psychophysiology. 1988;25:241–243.
Olincy A, Harris JG, Johnson LL, Pender V, Kongs S, Allensworth D, et al. Proof-of-concept trial of an alpha7 nicotinic agonist in schizophrenia. Arch Gen Psychiatry. 2006;63(6):630–638. [PubMed]
Patterson JV, Hetrick WP, Boutros NN, Jin Y, Sandman C, Stern H, et al. P50 sensory gating ratios in schizophrenics and controls: a review and data analysis. Psychiatry Res. 2008;158(2):226–247. [PubMed]
Posner MI. Cognitive neuroscience of attention. New York: Guilford Press; 2004.
Smith DA, Boutros NN, Schwarzkopf SB. Reliability of P50 auditory event-related potential indices of sensory gating. Psychophysiology. 1994;31(5):495–502. [PubMed]
Tregellas JR, Davalos DB, Rojas DC, Waldo MC, Gibson L, Wylie K, et al. Increased hemodynamic response in the hippocampus, thalamus and prefrontal cortex during abnormal sensory gating in schizophrenia. Schizophr Res. 2007;92(1-3):262–272. [PMC free article] [PubMed]
Ventura J, Liberman RP, Green MF, Shaner A, Mintz J. Training and quality assurance with the Structured Clinical Interview for DSM-IV(SCID-I/P) Psychiatry Research. 1998;79:163–173. [PubMed]
Ventura J, Lukoff D, Nuechterlein KH, Liberman RP, Green MF, Shaner A. Brief Psychiatric Rating Scale (BPRS), Expanded Version (4.0): Scales, anchor points, and administration manual. International Journal of Methods in Psychiatric Research. 1993;3:227–243.
Waldo MC, Carey G, Myles-Worsley M, Cawthra E, Adler LE, Nagamoto HT, et al. Codistribution of a sensory gating deficit and schizophrenia in multi-affected families. Psychiatry Research. 1991;39:257–268. [PubMed]
White PM, Yee CM. Effects of attentional and stressor manipulations on the P50 gating response. Psychophysiology. 1997;34:703–711. [PubMed]
White PM, Yee CM. P50 sensitivity to physical and psychological state influences. Psychophysiology. 2006;43(3):320–328. [PubMed]
Williams TJ, Yee CM, Nuechterlein KH, Subotnik KL. Distinct Neural Generators of Sensory Gating in Schizophrenia. 2009 Manuscript submitted for publication. [PMC free article] [PubMed]
Woldorff MG, Gallen CC, Hampson SA, Hillyard SA, Pantev C, Sobel D, et al. Modulation of early sensory processing in human auditory cortex during auditory selective attention. Proc Natl Acad Sci U S A. 1993;90(18):8722–8726. [PubMed]
Yee CM, Nuechterlein KH, Dawson ME. A longitudinal analysis of eye tracking dysfunction and attention in recent-onset schizophrenia. Psychophysiology. 1998;35(4):443–451. [PubMed]
Yee CM, Nuechterlein KH, Morris SE, White PM. P50 suppression in recent-onset schizophrenia: Clinical correlates and risperidone effects. Journal of Abnormal Psychology. 1998;107:691–698. [PubMed]
Yee CM, White CM. Experimental modification of P50 suppression. Psychophysiology. 2001;38:531–539. [PubMed]