Patients with schizophrenia spectrum disorder and their relatives may have a general reduced sensitivity to reward, indicating that reward learning may be an important familial marker in schizophrenia spectrum disorder, according to study results published in Schizophrenia Research.
Results from previous studies have shown that patients with schizophrenia spectrum disorder demonstrate impairments in reward learning that may be related to dysfunctions in the dopaminergic system. This deficit may impede social decision making and social reward learning; therefore, understanding how social and nonsocial reward processing can influence decision making may help reveal the underlying mechanisms of social dysfunction in schizophrenia spectrum disorder. The objective of this study was to investigate social and nonsocial reward learning in patients with schizophrenia spectrum disorder, unaffected first-degree relatives, and healthy controls.
In this study, researchers evaluated 50 patients with schizophrenia spectrum disorder, 20 unaffected first-degree relatives, and 49 healthy controls who had no personal or family history of schizophrenia spectrum disorder. A trust game consisting of 20 trials was used to measure social reward learning, in which patients played the role of investor against a computer programmed to behave like a benevolent human player. A lottery game also consisting of 20 trials was used to assess nonsocial reward learning and was identical to the trust game with the exception that patients were informed they were playing against a computer, not a human. Both games were programmed to increase the likelihood of higher payouts in response to increased investments. Researchers used multilevel regression analyses to examine reward learning across trials.
Results revealed that healthy controls increased investments over trials (b=.03, 95% CI, 0.003-0.05], P =.03) and invested significantly less in the trust game than in the lottery game (b=-.32, 95% CI, -0.51 to -0.12], P =.001). Patients with schizophrenia spectrum disorder and relatives did not demonstrate any reward learning (P =.44 and P =.86, respectively). Patients with schizophrenia spectrum disorder invested significantly less in the trust game than in the lottery game (b=-.31, 95% CI, -0.53 to -0.09], P =.005), while their relatives invested significantly more in the trust game than in the lottery game (b=.42, 95% CI, 0.09-0.74, P =.01).
Compared with relatives and controls, patients with schizophrenia spectrum disorder increased investments less often and invested more equal amounts. Patients with lower negative symptoms (Positive and Negative Syndrome Scale score ≤14) made higher investments in the trust game vs the lottery game (b=.37, 95% CI, 0.05-0.69, P =.03), while patients with higher symptom severity (Positive and Negative Syndrome Scale score >14) made higher investments in the lottery game vs the trust game (b=-.98, 95% CI, -1.28 to -0.68, P <.001).
Overall, healthy controls demonstrated social reward learning, but patients with schizophrenia spectrum disorder and relatives demonstrated no learning in social and nonsocial tasks — indicating there may be a generalized insensitivity to reward cues in those with familial liability to schizophrenia spectrum disorder. There was no difference in the degree of reward learning impairments identified between the social or nonsocial context among patients with schizophrenia spectrum disorder.
This study had several limitations. First, the number of unaffected first-degree relatives was relatively small and gender distribution among this group was significantly different than that among the patient and control groups. Second, this study did not include participants at ultra-high risk. Third, researchers used a basic representation of a human counterpart in both games to keep the social and nonsocial task condition as similar as possible. Lastly, this study only evaluated social reward learning on a fundamental level, and did not investigate how these mechanisms may relate to daily and real-life functioning.
The study researchers concluded that reward learning in general may be an important marker for the familial risk to schizophrenia spectrum disorder, given that patients and their relatives both demonstrated a reduced sensitivity to social and nonsocial rewards. They suggest future studies examine “associations between reward sensitivity as indicated by experimental tasks and daily-life (social) motivation by means of experience sampling studies” and “neuroimaging studies directly comparing social and non-social reward learning.”
Hanssen E, Krabbendam L, Robberegt S, Fett AK. Social and non-social reward learning reduced and related to a familial vulnerability in schizophrenia spectrum disorders [published online November 18, 2019]. Schizophr Res. doi: 10.1016/j.schres.2019.10.019