How learning influences motivation: the importance of timing (PIT juice)

Timing-dependent valence reversal is the phenomenon for which the valence of a conditioned stimulus can switch from positive to negative (or vice-versa), depending on the temporal relationship with the valenced outcomes it is associated with (Tanimoto et al., 2004). For instance, an odor that consistently follows the end of a painful shock acquires positive value (relief), while the same odor will be perceived as negative if it consistently predicts a painful shock (Andreatta et al., 2012; Yarali et al., 2008). The same mechanism has been observed in the reward domain, whereby cues consistently paired with the termination of a rewarding experience acquire inhibitory effects on subsequent appetitive Pavlovian conditioning (Hellstern et al., 1998), and can acquire negative valence (Delamater et al., 2003). The learned valence can therefore be altered by a simple switch in the temporal relation between the outcome (i.e., reward or punishment) and the conditioned stimulus. The value acquisition of conditioned stimuli is not only interesting because it is a fundamental mechanism but also because such stimuli can have a profound influence on human choices and behavior. Literature on cue-reactivity and Pavlovian-instrumental transfer has demonstrated that conditioned stimuli can influence motivated behavior and choice in humans (Betts et al., 2021; Carter & Tiffany, 1999; Holmes et al., 2010; Talmi et al., 2008). However, associative learning research has focused mainly on forward conditioning, (i.e., on how stimuli that predict a reward or punishment acquire value). This focus inevitably left backward conditioning, or the process by which the removal of a valenced outcome changes the probability of a subsequent stimulus or response, on the sidelines (Bordeaux & Koob, 2023). Also the cue reactivity literature used primarily forward conditioning protocols (i.e., the neutral stimulus predicts reward or punishment). The same is true for literature focusing more specifically on how conditioned stimuli can influence behavior. To our knowledge, no attempt has been made at evaluating such CS-induced motivational effects (i.e. incentive motivation; Ferrario, 2020) in a study that systematically compares forward and backward stimuli in the reward and punishment domains. Therefore, the goal of this project is to investigate: 1. Whether backward stimuli acquire a valence opposite to the outcomes and forward stimuli. 2. Whether and how these stimuli influence responding in a Pavlovian-Instrumental Transfer task. In a first paradigm with auditory outcomes, we found an asymmetry between the reward and punishment domains, as timing-dependent valence reversal was stronger in the negative than the positive domain. This difference might have been due to the fact that punishments were significantly more arousing than rewards, as self-reported by participants. The current project aims to address this limitation by using primary (liquid) outcomes, and a calibration procedure which will allow us to select matched appetitive and aversive stimuli for each participant, individually.