Inefficient effort allocation and negative symptoms in individuals with schizophrenia

Abstract

Negative symptoms like avolition and anhedonia are thought to involve difficulties with reward processing and motivation. The current study aimed to replicate and extend prior findings that individuals with schizophrenia display reduced willingness to expend effort for rewards and that such reduced effort is associated with negative symptoms, poor functioning, and cognitive impairment. The present study compared the effortful decision making of individuals with schizophrenia (n = 48) and healthy controls (n = 27) on the Effort Expenditure for Rewards Task (EEfRT). Individuals with schizophrenia chose a smaller proportion of hard tasks than healthy controls across all probability and reward levels with the exception of trials with a 12% probability and low or medium reward magnitude wherein both groups chose similarly few hard tasks. Contrary to expectations, in individuals with schizophrenia, greater negative symptoms were associated with making more effortful choices. Effortful decision making was unrelated to positive symptoms, depression, cognition, and functioning in individuals with schizophrenia. Our results are consistent with prior findings that revealed a pattern of inefficient decision making in individuals with schizophrenia relative to healthy controls. However the results did not support the hypothesized association of negative symptoms and reduced effort in schizophrenia and highlight prior inconsistencies in this literature. Future research is needed to understand what factors may be related to diminished effortful decision making in schizophrenia and the clinical significance of such performance deficits.

Introduction

Negative symptoms, such as anhedonia and avolition, are problematic for individuals with schizophrenia, as they are often persistent (Möller, 2007), correspond to worse social functioning (e.g., Milev et al., 2005), and are an unmet treatment need (e.g., Fusar-Poli et al., 2015). In order to improve treatments, it is important to better understand how negative symptoms are related to mechanisms that could serve as targets for interventions such as neural processes associated with reward processing (Dowd and Barch, 2012) or effortful decision making (Gold et al., 2013).

Motivational deficits are central to the definition of negative symptoms (Horan et al. 2011). Many individuals with schizophrenia demonstrate motivational impairments on laboratory tasks relative to healthy controls (e.g., Barch and Dowd, 2010). In understanding these motivational deficits it is useful to consider that acting on motivation by allocating effort requires a cost-benefit assessment of task difficulty and reward magnitude (Gendolla and Krüsken, 2002). Individuals with schizophrenia have demonstrated reduced effort allocation or impaired effort-cost computations (Gold et al., 2013) in laboratory tasks such that, compared to healthy controls, they are less likely to choose high-effort options when probability or magnitude of reward is increasing (Barch et al., 2014, Fervaha et al., 2013, Gold et al., 2013, Reddy et al., 2015, Treadway et al., 2015). These findings suggest that individuals with schizophrenia do not display a complete failure to respond to reward, but that they often have difficulty with anticipatory pleasure and recruiting effort to pursue rewards in trials with the greatest probability of reward or reward magnitude.

There are some important inconsistencies in results and methods across these effort studies (Green et al., 2015). The association between negative symptom severity and reduced effort allocation has been quite variable; some studies have shown this association (Barch et al., 2014, Gold et al., 2013, Reddy et al., 2015), but others have not (Fervaha et al., 2013). Gold et al. (2013) found that the association between negative symptoms and effort allocation was only evident in categorical comparisons (high negative symptom individuals versus controls), but correlational analyses examining negative symptoms dimensionally were not significant. Treadway et al. (2015) found a non-significant trend between negative symptoms and an index of expected value when covarying for medication dosage. Studies have also differed in the exploration of a range of probability of reward. Some studies have examined effortful decision making at only two probability levels (50% and 88% Barch et al., 2014 and Reddy et al., 2015; 50% and 100% Gold et al., 2013), whereas other studies have examined performance across three probability levels to investigate responding in the context of very low reward probability (12%, 50% and 88%) (Fervaha et al., 2013, Treadway et al., 2015). Three studies reported that individuals with schizophrenia displayed less effortful decision making only in the context of high reward value and/or high probability of reward (Barch et al., 2014, Gold et al., 2013, Treadway et al., 2015); one study found that these individuals chose to expend more effort than controls in trials of low probability and reward value (Fervaha et al., 2013). Beyond exploring symptom correlates, only two studies (Barch et al., 2014, Horan et al., 2015) have examined the relationship of laboratory assessed effort allocation and functioning, finding that reduced effort was associated with worse community functioning. The contribution of cognitive impairment to performance on effort-based assessments in schizophrenia is unclear, though people generally display a bias toward avoiding cognitive effort (preserving cognitive resources) until incentives offset such demands (e.g., Kool et al., 2010). Better cognitive ability has been shown to be associated with a greater likelihood of selecting high-effort response options in individuals with schizophrenia in two studies (Gold et al., 2013, Horan et al., 2015) but cognitive functioning was not related to effort in another study (Fervaha et al., 2013). Given these mixed results, further exploration is critical.

The present study examined how individuals with schizophrenia process reward and probability cues and choose to allocate physical effort for monetary reward across the full range of probability levels on the Effort Expenditure for Reward Task. Given the limitations of other negative symptom rating scales (Blanchard et al., 2011) we utilized the Clinical Assessment Interview for Negative Symptoms (CAINS; Blanchard et al., 2011, Horan et al., 2011, Kring et al., 2013). We also assessed the relationship between EEfRT performance, role functioning, and cognition. Based on prior findings (Barch et al., 2014, Fervaha et al., 2013, Gold et al., 2013, Horan et al., 2015, Reddy et al., 2015, Treadway et al., 2015), we hypothesized that 1) individuals with schizophrenia who are presented with reward and probability cues would choose to complete “easy” tasks more frequently than “hard” tasks compared to controls across reward levels, and 2) that, in individuals with schizophrenia, more severe negative symptoms (especially those involving motivation and pleasure) would be related to less effortful choices, and 3) that less effortful decision making would be related to worse social functioning and cognition impairment.