Introduction

Motivational impairment has been linked to schizophrenia since Kraepelin’s [1] early descriptions of a “weakening of the wellsprings of volition” in this disorder. Motivational disturbances figure prominently in contemporary descriptions of schizophrenia, particularly of negative symptoms such as anhedonia, avolition, and asociality, which are strongly associated with poor treatment outcome and functional impairment [2]. Despite the theoretical and functional significance of motivational disturbances in schizophrenia, little is known about their neurophysiological and psychological bases [3], [4]. The current study addresses this issue by examining an influential model of approach and withdrawal motivational systems assessed with electrophysiology (EEG) and self-report measures in schizophrenia.

Considerable evidence supports the motivational direction model of asymmetric frontal brain activity [5]–[10]. According to this model, two fundamental and distinct motivational systems, one supporting approach behaviors and another supporting withdrawal behaviors, are associated with separate neural circuits that involve different regions of the frontal cortex. The approach motivational system is thought to modulate planning for and reactions to appetitive or rewarding stimuli associated with emotions (e.g., enthusiasm, pleasure, anger) and behaviors that prompt us approach desired goals and rewards. The avoidance motivational system, in contrast, is thought to modulate processing of aversive stimuli and to generate corresponding withdrawal/avoidance (e.g., fear, anxiety) emotional and behavioral responses to avert threats and punishment. The neural correlates of these systems have primarily been examined through resting EEG assessments of lateralized frontal alpha band (8–12 Hz) power, which traditionally has been interpreted as inversely related to regional activity. Left frontal activity (i.e., lower left alpha power) is believed to reflect approach motivation, whereas right frontal activity (i.e., lower right alpha power) is associated with withdrawal motivation. These asymmetrical frontal EEG findings are supported by lesion studies, though hemodynamic studies have been less consistent [9],[11]–[13].

Frontal asymmetries assessed in the resting EEG are believed to relate to stable dispositional motivational tendencies. Consistent with this notion, several studies demonstrate associations between degree of lateralization of activation and scores on conceptually related self-report motivational trait measures. For example, in studies using the BIS/BAS scales [14], which were designed to assess sensitivities of the behavioral inhibition and activation systems as originally described by Gray [15], higher BAS scores correlate with higher relative left frontal activity, and higher BIS scores often, though not always, correlate with higher relative right frontal activity [7]. A large number of studies also demonstrate that frontal asymmetries are related to various clinical conditions [8], [16]. For example, elevated right frontal lateralization is associated with certain types of anxiety disorders, reduced left lateralization is associated with vulnerability to anhedonic depression, and elevated left is associated with vulnerability to mania [16], [17]. Hemodynamic neuroimaging studies have provided support for some of the EEG findings [18]–[20].

Frontal asymmetry and motivational tendencies have been examined explicitly in schizophrenia in only two prior studies. One found that recent-onset patients showed more right frontal lateralization (using a single unconventional pair of electrodes) than healthy controls during a two-minute recording with eyes closed [21], a somewhat shorter epoch than recommended [22], [23]. The other examined chronically ill outpatients with schizophrenia spectrum disorders (no comparison group) and found that asymmetry scores showed good stability over 36 months and that leftward lateralization correlated with higher positive symptoms [24], [25]. It is also worth noting that two studies compared schizophrenia and healthy control groups on the BIS/BAS scales without EEG; both found that patients reported significantly elevated BIS scores, which is consistent with greater right frontal lateralization, but not differences in BAS scores [26], [27].

The current study assessed both resting frontal EEG asymmetries and self-reported approach/withdrawal tendencies in schizophrenia. The prior literature, though very limited, led to the prediction that patients would show relatively more right frontal lateralization and higher BIS scores than matched healthy controls. We also explored whether frontal asymmetries relate to individual differences on the BIS/BAS and to symptom levels among patients.

Methods

Results

Demographic and clinical characteristics

Demographic information for both groups and clinical data for the schizophrenia group are presented in Table 1. The groups did not significantly differ in sex or ethnicity. Although the patients had lower personal education levels than controls, the groups did not differ in parental education. There was a significant age difference between the groups; patients were older than controls. To evaluate the potential impact of age on the results, we computed correlations between age and all of EEG, self-report, and symptom (for patients) variables separately within each group. There were no significant correlations involving age and we therefore did not account for age in the primary data analyses. There were also no significant correlations between CPZ equivalents and any other study measure.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic and Clinical Data.

https://doi.org/10.1371/journal.pone.0110007.t001

The schizophrenia group had a typical age of onset and was chronically ill. They showed mild to moderate levels of clinical symptoms at the time of testing that are comparable to prior studies of stabilized outpatients [40], [41].

Group comparisons on alpha asymmetry

Descriptive statistics and results of between-group comparisons are summarized in Table 2. For the primary measure of frontal asymmetry, F4–F3, patients had negative scores that were significantly lower than those of controls. This pattern suggests patients showed either more right lateralization or less left lateralization than controls. To further characterize this between-group difference, we conducted a Group X Hemisphere ANOVA, which revealed a significant interaction effect, F(1,48) = 4.56, p<.05, but non-significant effects for group and hemisphere (p’s>.45). Follow up t-tests indicated that controls showed greater left than right activation, t(24) = 2.47, p<.05, whereas patients did not show a significant difference between left and right activation, t(24) = –1.02, p>.30); there were no between-group differences within either the left or right hemisphere (p’s>.25). Thus, patients failed to show the left lateralization that was present in the control group.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Group Differences on Asymmetry Scores.

https://doi.org/10.1371/journal.pone.0110007.t002

Although patients had numerically lower mean scores on the secondary index (F8–F7) than controls, the groups did not significantly differ. Finally, for the analysis of the P4–P3 electrodes, there was no significant group difference in parietal asymmetry scores.

Group comparisons on self-report motivation

As shown in Table 3, patients reported significantly higher BIS scores than controls, indicating that patients reported greater behavioral inhibition sensitivity. There were no significant group differences on the BAS total scores or on any of the subscale scores. Internal consistency estimates for each subscale were acceptable within the patient and control groups.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Between-Group Comparisons on the BIS/BAS.

https://doi.org/10.1371/journal.pone.0110007.t003

Conclusions

In this first study to evaluate approach and withdrawal motivation in schizophrenia using both self-report and EEG measures, patients showed an atypical pattern of differences from controls. On the resting EEG measure patients failed to show the left lateralized activity that was present in controls, suggesting diminished approach motivation in the schizophrenia group. On the self-report measure, patients reported higher BIS sensitivity than controls, which is typically interpreted as heightened withdrawal motivation. EEG asymmetry scores however, were not significantly correlated with self-reported motivational traits in either group. The overall pattern of group differences converges to suggest a motivational disturbance in schizophrenia characterized by elements of both diminished approach tendencies and elevated withdrawal tendencies.

The current between-group motivational differences confirm and extend the few prior studies using either EEG or self-report measure alone. Regarding EEG, the two groups showed opposite asymmetry patterns at mid-frontal sites, the most commonly examined sites in the frontal asymmetry literature [7], [16], with a left bias in controls versus a right bias in patients. This finding is broadly similar with an earlier report of right frontal lateralization in recent-onset schizophrenia patients [21]. The current study had several methodological strengths compared to the earlier report, including data from a much longer recording epoch, analyses on a commonly-examined set of frontal electrode sites, and well-established data analysis procedures. It also extends the earlier report by demonstrating that the patients’ abnormal asymmetry score appears to be more a reflection of diminished left lateralization rather than greater right lateralization. Diminished left lateralization is strongly associated with depression, particularly anhedonic depression, as well as vulnerability to depression, and is thought to reflect reduced approach motivation and reward sensitivity [9], [10], [16]. A similar motivational disturbance is believed to underlie certain negative symptoms of schizophrenia, such as anhedonia and amotivation, though asymmetry scores were not associated with negative symptoms in the current study.

The patients’ elevated self-reported scores on the BIS scale, but normal scores on the BAS scales, also converge with the few prior relevant studies in schizophrenia [26], [27]. Elevated BIS scores reflect hyperactivity of the withdrawal motivational system and are associated with heightened sensitivity to threat, physiological arousal, and experience of negative emotions such as fear and distress [5]–[8], [10]. This pattern fits well with several lines of research in schizophrenia indicating that patients frequently show elevated trait negative affectivity, elevated reactivity to negative stimuli/stress and punishment feedback, and difficulty regulating negative emotions [42]–[44]. The heightened self-reported BIS (but normal BAS) sensitivity seen in schizophrenia patients is comparable to the pattern seen in certain anxiety disorders, such as social anxiety and panic disorders [9], [10], [16].

Overall, the current findings suggest that when it comes to understanding motivational factors that hold people with schizophrenia back from pursuing personally relevant goals, it is important to consider multiple processes. Diminished initiation and persistence in goal-directed activities may stem from a lack of desire to pursue potentially rewarding outcomes and/or a strong drive to avoid potentially unpleasant emotions and cognitions associated with efforts to pursue desired outcomes.

Results from the current study are not fully consistent with prior studies, in that frontal asymmetry scores did not significantly correlate with symptom levels or with BIS/BAS scores. Regarding symptoms, Jetha et al. [24] found that left frontal lateralization correlated with positive symptoms in a sample with a large proportion of patients with prominent positive symptoms. The lack of symptom correlates in the present stabilized outpatient sample, therefore, suggests that frontal asymmetry scores are not simply a marker of clinical state. As noted above, it was somewhat unexpected that negative symptoms were not correlated with asymmetry as they are conceptually related to diminished BAS activity. This may reflect the use of the BPRS negative symptom scale, which focuses on expressive negative symptoms rather than motivation- and pleasure-related negative symptoms [2].

We also did not find significant correlations between frontal asymmetry and scores on the BIS/BAS scales in either patients or controls. Such correlations have been reported in some studies in clinical and healthy samples, though a number of studies have failed to find significant correlations, particularly for the BIS scale [7]. This absence of significant correlations for the BIS may partly reflect an emerging view that the withdrawal construct associated with rightward lateralization has considerably less overlap with the inhibition construct assessed by the BIS scale than originally suggested [7], [45], [46]. The extent to which these are overlapping versus distinct constructs is currently an active area of research. For example, BIS may correlate differentially with areas of dorsolateral prefrontal cortex that relate specifically to selecting approach goals, selecting withdrawal goals, and selecting goal-pursuit strategies [47]. Another factor that may account for this discrepancy is that resting EEG may not be as powerful to detect individual differences in motivational tendencies as task-activated EEG and fMRI paradigms [48], [49]. Future studies of schizophrenia would benefit from using approach- and withdrawal–related emotional challenge paradigms to obtain a potentially more sensitive index of the capacity to activate these motivational systems.

The current finding of increased withdrawal motivational tendencies in schizophrenia should be interpreted in the context of some limitations. First, this cross-sectional study of chronically ill patients cannot address the question of whether this pattern precedes the onset of schizophrenia or is the consequences of living with a severe mental illness. Although prior research demonstrating the same asymmetry pattern is present in recent-onset patients and that asymmetry scores show good longitudinal stability in schizophrenia is consistent with the notion that elevated right frontal lateralization reflects an enduring trait that is present from at least the early post-onset period [21], [25], further research is needed to address this issue (e.g., in high-risk subjects). A second limitation was reliance on the BIS/BAS as the sole measure of approach and withdrawal motivation. Spielberg et al. ([50], [51] also see [52]) developed a composite measure of approach/withdrawal motivation with better psychometric properties in nonclinical samples than the BIS alone. Third, resting frontal EEG asymmetry can be affected by current emotional state and other factors (e.g., fatigue, hunger) [49] that were not directly assessed in this study. Fourth, the patients were taking medications at clinically determined dosages and the impact on the current findings is unclear. Although CPZ equivalents were not significantly correlated with EEG or self-report measures, and prior research has not found that psychotropic medications impact resting frontal asymmetry [16], research in unmedicated patients is required to directly address this question. Fifth, as mentioned above, we were not able to directly measure the relationship between frontal asymmetry clinically rated motivational deficits since the BPRS assesses the expressive, but not the experiential, component of negative symptoms. Sixth, the samples were not matched for age, although supplementary analyses indicated that age did not relate to the study variables.

Although the approach and withdrawal motivational framework has been extensively studied in internalizing and externalizing disorders, the current findings suggest it may be informative for understanding motivational disturbances in the psychosis dimension of psychopathology as well. The framework has treatment implications for schizophrenia patients with elevated withdrawal motivation. For example, CBT, emotion regulation, and mindfulness interventions that target symptoms such as anxiety and hyper-arousal may be useful for these individuals, and frontal asymmetries have been found to provide an informative biomarker in the context of these types of treatment studies [53]–[55]. The general approach/withdrawal framework has also been applied to more specific investigations of functioning in the social domain [56]. Further investigation of social approach and withdrawal may be useful for isolating factors that contribute to the social difficulties that many individuals with schizophrenia experience.

Supporting Information

Acknowledgments

Support for this study came from a NARSAD Young Investigator Award (William P. Horan, Ph.D.) and NIMH Grants MH065707 and MH43292 (Michael F. Green, PhD). The authors wish to thank Amanda Bender, Michelle Dolinsky, Crystal Gibson, Cory Tripp, and Katherine Weiner for assistance in data collection.