Measuring Human Pavlovian Reward Conditioning and Memory Retention After Consolidation.

Abstract

While a body of literature has addressed the quantification of aversive Pavlovian conditioning in humans, Pavlovian reward conditioning with primary reinforcers and its recall after overnight consolidation remain understudied. In particular, few studies have directly compared different conditioned response types and their retrodictive validity. Here, we sought to fill this gap by investigating heart period responses (HPR), skin conductance responses (SCR), pupil size responses (PSR), and respiration amplitude responses (RAR). We conducted two independent experiments (N₁ = 37, N₂ = 34) with a learning phase and a recall phase 7 days later. A visual conditioned stimulus (CS+) predicted fruit juice reward (unconditioned stimulus, US), while a second CS- predicted US absence. In experiment 1, model-based analysis of HPR distinguished CS+/CS-, both during learning (Hedge's g = 0.56) and recall (g = 0.40). Furthermore, model-based analysis of PSR distinguished CS+/CS- in early trials during recall (g = 0.69). As an out-of-sample generalization test, experiment 2 confirmed the result for HPR during learning (g = 0.78) and recall (g = 0.55), as well as for PSR during recall (g = 0.41). In contrast, peak-scoring analysis of PSR yielded low retrodictive validity. We conclude that in our Pavlovian reward conditioning paradigm, HPR is a valid measure of reward learning, while both HPR and PSR validly index the retention of reward memory.