(PS9-64) In Vivo Associations between Anticipatory and Consummatory Reward Responsiveness

Abstract Text: Reward system dysfunction involves a range of clinically relevant processes, including motivation to pursue rewards, hedonic capacity, and reward salience. Reward system functioning is also centrally linked to classical and operant conditioning. Reward system impairments are implicated in various forms of psychopathology including depression, PTSD, and schizophrenia (Pizzagalli et al., 2008; Nawijn et al., 2015; Moran et al., 2018). Reward processing may be parsed into two distinct temporal components; anticipatory (ARR) and consummatory reward processing (CRR; Berridge & Robsinson, 1998). ARR refers to the anticipation of an expected reward (i.e., wanting), and CRR involves response to an experienced reward (i.e., liking). To date, little research has examined the associations between these two distinct components of reward processing, particularly as dynamic processes that may interact in real-time. The current study aimed to fill this gap by exploring in-vivo associations between ARR and CRR using ecological momentary assessment (EMA).

547 undergraduate students completed baseline questionnaires assessing demographic information and symptoms using the depression subscale of Depression, Anxiety, and Stress Scale (DASS-D). Participants then completed EMA questionnaires assessing ARR and CRR reward responsiveness three times per day for seven consecutive days. Multilevel modeling with random intercepts and slopes using maximum likelihood (ML) estimation was conducted to examine lagged associations between average ARR and CRR across timepoints. In this model, each participant had 21 repeated assessments (Level 1) that were nested within participants (Level 2). Baseline DASS-D scores were entered as a Level 2 predictor to control for depression.

ARR significantly predicted subsequent CRR (b = .41 [.38: .44], SE = .02, cr = 23.11, p < .001, pr = .71). Results held bidirectionally, such that CRR significantly predicted ARR at the next timepoint (b = -.05 [-.08: -.03], SE = .02, cr = 3.22, p = .001, pr = .14). When controlling for baseline depression, ARR significantly predicted subsequent CRR (b = .43 [.39: .47], SE = .03, cr = 16.14, p < .001, pr = .57), but CRR no longer predicted subsequent ARR.

Results are among the first to show bidirectional associations between the two distinct temporal phases of reward processing. Higher ARR was related to increased CRR at subsequent timepoints. Surprisingly, there was a negative relationship between CRR and future ARR, such that higher CRR was associated with lower subsequent ARR, suggesting potential impairment in anticipation of rewards following increased hedonic responding. Notably, only associations between ARR and subsequent CRR held when controlling for depression, which has established impacts on the reward system (Forbes, 2009). Overall, findings support in vivo associations between two distinct aspects of reward responsiveness and may point to ARR as a key treatment target with potential downstream impacts on CRR and overall hedonic capacity, particularly among individuals with depression.