Are Negative Symptoms Merely the “Real World” Consequences of Deficits in Social Cognition?

Abstract

Many investigations have demonstrated that negative symptoms and social cognitive deficits in schizophrenia play a large role in determining functional outcomes and ultimately long-term prognosis. Given this, there is increasing interest in understanding the relationship between these two symptom domains, particularly since studies have consistently found moderate to large associations between them. This shared variance raises a key question: to what degree do these two categories of symptoms arise from overlapping or identical changes in brain function? In other words, do some or all negative symptoms represent merely the downstream effects of social cognition deficits on daily functioning? In this commentary, the evidence for and against this possibility, limitations of currently validated empirical measurements of these symptoms, and directions for further investigation of this hypothesis are discussed. Understanding the shared and distinct mechanisms of these disabling deficits will have important implications for the design of novel, personalized treatments for psychotic illness.

Introduction

The negative symptoms of schizophrenia consist of deficits in the initiation and maintenance of day-to-day behavior, emotional experience and/or expression, verbal and nonverbal communication, and social motivation. Although these symptoms have been well-described since the earliest accounts of schizophrenia¹ and are reliably measured using a number of well-validated research instruments,^2–5 their cognitive and neural mechanisms remain incompletely understood.^6–13 However, new clues about mechanisms have emerged from evidence linking negative symptoms with the group of operations of the brain that support social functioning, referred to collectively as “social cognition.”^14,15 Given that negative symptoms are associated with impairments in social functioning and are highly correlated with disability in schizophrenia,^16,17 a natural question that arises is: how much do negative symptoms reflect the “real world” expression of deficits in social cognition? In this commentary, we selectively review some of the existing literature that addresses aspects of this question and propose some new approaches for further research.

The Question

To examine the relationship between these two domains, one must first review how negative symptoms and social cognition are currently defined and measured. Several recent factor and network analyses of negative symptom profiles, measured using standardized research interviews, have indicated that there are five separable negative symptoms: anhedonia (diminished intensity/frequency of pleasure), avolition (diminished initiation and persistence in goal-directed activity), asociality (reduced desire for/frequency of social interaction), blunted affect (diminished facial, vocal, and body expression of emotion), and alogia (reduced quantity of speech).^8,18–21 The term “social cognition” refers to a constellation of psychological operations, typically measured using objectively defined behavioral responses, that involve other people and ourselves.^15,22 This heterogeneous category of processes includes the perception of social cues (including the perception of auditory cues such as prosodic information in speech and visual cues such as facial emotional expressions), sharing of emotional experience with others (e.g., recognizing the experience of a particular emotion in another and sharing/simulating that experience), mentalization (the ability to interpret others’ thoughts, intentions, and emotions), and managing emotions.^23,24 Studies show that anywhere from 40% to 90% of individuals with schizophrenia have some negative symptoms^25,26 and that approximately 90% exhibit impaired social cognition.²⁷

A relationship between these two domains has been proposed based on the associations found between both negative symptoms and social cognition deficits and impairments in real-world functioning, subjective well-being and quality of life,^6,16,28–32 as well as correlations found between negative symptoms and social cognition deficits^33–36 (but also see refs. ³⁷ and ³⁸). For example, studies using the Scale for the Assessment of Negative Symptoms (SANS)³⁹ and/or the Positive and Negative Syndrome Scale (PANSS)⁴⁰ to measure negative symptoms have found small to large correlations (r = 0.20–0.54) between levels of social cognition deficits and negative symptoms.^33,34,41 More recent investigations using newer negative symptom measures (e.g., the Brief Negative Symptom Scale [BNSS]^3,42) have revealed larger effect sizes linking the two domains, including moderate to large associations between facial affect recognition deficits and all five negative symptom domains (r = 0.41–0.75).³⁶ Mentalization deficits have also been linked to BNSS-defined negative symptoms, showing large associations with blunted affect and alogia (e.g., r = 0.49–0.58).³⁶

In contrast, one principal component analysis found that four types of deficits in social cognition (in emotion recognition, managing emotions, social attribution, and a measure of reality testing that showed strong relationships with social outcomes, but not theory of mind) loaded onto a factor that was distinct from a negative symptoms factor.³⁴ Another study using structural equation modeling (SEM) showed that separating negative symptoms and social cognition into distinct constructs was the most statistically robust model for representing the data.³³ Thus, taken together, the existing data suggest that negative symptoms and social cognition are related, yet distinct constructs. However, factor analyses of cross-sectional data cannot exclude the possibility that, over time, deficits in social cognition manifest behaviorally as negative symptoms.

Methodological Issues

Also, several important, potentially confounding issues must be considered when interpreting these studies. One is the heterogeneity of clinical presentations (e.g., in negative symptom severity and type) across and within cohorts of individuals with schizophrenia that may lead to inconsistent results across studies. For example, an infrequently addressed issue is the difficulty of differentiating primary (i.e., enduring, trait-like) negative symptoms from secondary negative symptoms that are likely due to transient or reversible symptoms or consequences of the illness, such as depression, paranoia, or treatment with antipsychotic medications.⁴³ Variation in IQ or other cognitive abilities could also confound measures of negative symptoms and social cognition and affect the strength of correlations between the two domains. Also, attempts to compare and link negative symptoms and deficits in social cognition are complicated by the fact that the primary method of assessing each domain differs. Negative symptoms are typically measured using interview-based rating scales that focus on outward behavior or collateral reports from clinicians or family members, whereas domains of social cognition are measured using laboratory-administered experimental paradigms that have objective performance-based outcomes. Studies that have relied less on self-report or interview-based scales in favor of ecological momentary assessments^44,45 or behavioral evaluations of negative symptoms^6,46,47 are beginning to address this measurement discrepancy, but these methods have yet to be widely adopted.^46,48 The potential impact of assessment methodology is illustrated by a recent meta-analysis showing a large discrepancy in the magnitudes of alogia and blunted affect that was measured by subjective ratings (d = 3.54) compared with laboratory-based assessments (d = 0.80–0.36).⁴⁹ These methodological differences render comparisons across social cognition and negative symptoms more difficult but also underscore an intuitive but unanswered question—are these two domains of psychopathology in schizophrenia independent, with distinct neurobiological correlates, or do they reflect the same underlying pathophysiology? In other words, when we measure, for example, asociality in schizophrenia using self-report questionnaires or clinician-administered interviews, are we merely measuring subjective reporting of impairments in social cognition?

Negative Symptoms are Deficits in Social Functions

This possibility is supported by the essentially social nature of negative symptoms—the majority of negative symptoms are functionally defined outcomes that are characterized by particular types and levels of social activity. For example, avolition is commonly assessed with respect to a role (i.e., school, work) that is often embedded within a social context. Asociality is by definition the desire for and level of engagement in social activity. Anhedonia is often characterized by diminished intensity or frequency in pleasurable experiences that are embedded in a larger social framework. Attempting to separately assess the experience of pleasure that occurs in response to nonsocial compared with social stimuli or events can be challenging, since many experiences that are labeled nonsocial are often experienced in daily life within a social context or have strong associations with social experiences (i.e., eating, substance use, exercise, and interactions with co-workers).

Similarly, blunted affect may be linked to abnormalities in social behavior, i.e., an impaired ability to mirror or respond appropriately to the facial or bodily expressions of others⁵⁰ or respond nonverbally to social cues.⁵¹ Alogia is a deficit in the production of speech, essentially a social function of the brain that is typically driven by a desire or a perception of a requirement to interact with others. Thus, although newer negative symptom assessment methods distinguish the presence of negative symptoms in a social setting from those occurring in a “role” environment such as school or work,^2,3,52 these ostensibly nonsocial, role-related experiences are often highly embedded within a very socially demanding milieu.

Thus, to determine whether negative symptoms represent the behavioral consequences of altered social cognition, it will be critical to examine negative symptoms within both social and nonsocial contexts. For example, one behavioral study examined in-the-moment participant ratings to varying social interactions (e.g., a confederate scowling, smiling) and showed that individuals with schizophrenia report deficits in pleasure in response to socially rewarding stimuli such as smiles but respond to social punishment scenarios (i.e., social exclusion) similarly to controls.⁵³ Also, Fulford et al examined the impact of social encouragement on task effort (which served as a proxy for avolition).⁵⁴ Interestingly, inconsistent with findings of studies examining effort without a social component, they found no significant differences between individuals with schizophrenia and healthy control subjects in overall effort levels.⁵⁴ Another study measured negative symptoms before and after in-lab social interactions. Although self-reported social affiliation and positive affect increased in both the schizophrenia and control groups after social interaction, higher levels of avolition and anhedonia (both social and nonsocial, as measured by the Clinical Assessment Interview for Negative Symptoms [CAINS²]) were linked to a decreased willingness to interact with another person, whereas higher levels of expressive negative symptoms (i.e., alogia and blunted affect) were linked to an increased willingness.⁵⁵ These studies provide examples of the nuanced information that can be obtained by studying negative symptoms using ecologically valid tasks within a social context.

Also, several studies have modified traditional reward paradigms to investigate avolition and anhedonia involving social rewards.^56,57 in one study, individuals with schizophrenia showed diminished valuation of social rewards (i.e., smiles) compared with unaffected peers; in contrast, no group difference was found in ratings of monetary reward.⁵⁶ However, in another study, both individuals with schizophrenia and control participants showed faster reaction times to social rewards relative to monetary rewards.⁵⁷ Related studies have measured the contribution of altered reinforcement learning to negative symptoms^9,58,59 and the impact of cognitive load on negative symptoms such as alogia⁶⁰ within real-world social settings.

Another more naturalistic approach is to measure quantifiable forms of automatic motor behavior that play a central role in nonverbal social communication and may be altered in individuals with negative symptoms, such as eye gaze^61,62 and social spacing or “personal space.” ^63–65 Enlargements in personal space in schizophrenia have been linked to both negative symptoms and overactivity of the dorsal parietal cortex in response to personal space intrusions of social, but not nonsocial, stimuli.⁶³

In sum, studies that examine negative symptoms in social, more ecologically valid contexts suggest that negative symptoms may specifically emerge or become more prominent in social settings. This evidence supports the hypothesis that negative symptoms represent functional outcomes of altered social cognition.

What is the Model?

A related, unanswered question is: are specific impairments in social cognition linked to one or some negative symptoms but not others (i.e., is there mechanistic heterogeneity)? Or do broad impairments in basic sensory and cognitive processes (e.g., visual perception, motor output, processing speed, and implicit attention) affect downstream processes that are directly involved in social cognitive operations, which, in turn, result in the development of a range of negative symptoms and functional impairment? In support of the first, “non-unitary” model (Figure 1), associations between specific social cognitive domains and negative symptoms have been reported, with one study showing significant correlations between mentalization deficits and blunted affect and alogia, but not anhedonia, asociality, or avolition.³⁶ Also, some evidence suggests that functional outcomes are more closely linked to avolition, anhedonia, and asociality, compared with blunted affect and alogia.⁶⁶ Thus, it may be that distinct social cognitive processes lead to specific negative symptoms (Figure 1).

Fig. 1.

Schematic illustration of a “non-unitary” model of negative symptoms, based on evidence that mentalization deficits have a strong relationship with expressive negative symptoms only, whereas deficits in facial affect recognition have been linked to all five negative symptoms.^34,36

An alternative, related model is that a common abnormality in a mechanism broadly affecting early sensory and motor processes plays a central role in all negative symptoms. This “unitary” model is based on statistical modeling evidence that changes in early visual perception in schizophrenia lead to impairments in social cognition, amotivation, and poor functional outcome, in that order (Figure 2).⁶⁶ Additional evidence for this model comes from a recent study showing that deficits in auditory processing are linked to impaired neurocognition, then negative symptoms, and finally poor functional outcomes in schizophrenia.⁶⁷ Although social cognition was not directly evaluated in this study,⁶⁷ its findings support the hypothesis that all or most negative symptoms result from a basic change in sensory processing that has widespread, deleterious effects on higher-order functions.

Fig. 2.

Schematic illustration of a “unitary” model of negative symptoms which proposes that a fundamental abnormality in early sensorimotor processes affects social cognition broadly, resulting in negative symptoms.^66,67

Predictive coding models have also investigated how changes in early sensory processes contribute to a wide range of behavioral changes in schizophrenia including negative symptoms. In this framework, discrepancies between expectations and incoming sensory information generate a learned experience of uncertainty regarding the prediction of outcomes.⁶⁸ This learned experience ultimately results in inaction and avolition.^69,70 Similarly, Beck and colleagues⁷¹ have proposed that early failures in relationships and achieving expected milestones generate defeatist beliefs that inhibit goal-directed behavior, resulting in negative symptoms. These models are each consistent with the conceptualization of negative symptoms as outcomes that are downstream from impairments in social cognition, which may emerge from fundamental changes in early sensorimotor processes.

Summary and Future Directions

Based on the available evidence, as well as what is known about the parallel, distributed organization of brain circuits that support related sensorimotor functions,^72–74 we suggest that an integration of the two perspectives outlined here (see Figures 1 and 2) may most accurately represent the data. A model in which disruptions of lower-level processes have a general “bottom-up” effect on social cognition and behavior is compatible with the possibility that specific changes in sensorimotor processes and social cognitive functions drive some negative symptoms more than others. For example, hypothetically, abnormalities in the specific parietofrontal cortical pathway involved in monitoring personal space may affect both the expression and interpretation of social cues, which in turn, may negatively affect mentalization abilities and, as a consequence, social motivation, but have little effect on the production of speech or facial affect. Abnormalities in adjacent parietofrontal circuits (such as those involved in eye movements, gestures, and related actions^75,76) may have other behavioral consequences, contributing to individual negative symptoms to varying extents. Research that focuses on specific sensorimotor processes and measures the precise behavioral consequences of perturbations of those processes can generate evidence to support or refute circuitry-based predictions. We suggest that this “integrated model” can account for both the evidence that all five negative symptoms cluster together clinically and that they likely emerge from changes in distinct, but related sensorimotor processes.

In summary, investigating negative symptoms within social contexts, to simulate real-life scenarios where negative symptoms manifest, will likely give rise to novel, less conventional models of these symptoms (e.g., beyond the a priori hypothesis-driven domains), by uncovering unexpected relationships between these symptoms and abnormalities in sensorimotor processing, social perception, and behavior. In addition, a model of negative symptoms as the behavioral manifestation of altered social cognition has the potential to reveal areas of intact functioning, including domains of social cognition that remain unaffected by the illness, which novel treatments could rely upon to enhance the recovery of individuals with schizophrenia.

Funding

This work was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development at the National Institutes of Health (T32HD040127 [to A.P.-B.]) and the National Institute of Mental Health (R01MH109562 [to D.J.H.]).

Conflict of Interest

The authors have declared that they have no conflicts of interest in relation to the subject of this work.