Neurobiological Basis of Individual Variation in Stimulus-Reward Learning

Abstract

Cues in the environment can guide behavior in adaptive ways, leading one towards valuable resources such as food, water, or a potential mate. However, cues in the environment may also serve as powerful motivators that lead to maladaptive patterns of behavior, such as addiction. Importantly, and central to this article, there is considerable individual variation in the extent to which reward cues gain motivational control over behavior. Here we describe an animal model that captures this individual variation, allowing us to better understand the psychological and neurobiological processes that contribute to cue-evoked behaviors. When a discrete cue is paired with a food reward in a Pavlovian manner it acquires greater control over motivated behavior in some rats ("sign-trackers, STs) than in others ("goal-trackers", GTs). We review studies that have exploited this animal model to parse the neurobiological mechanisms involved in learning associations between stimuli vs. those involved in attributing incentive salience to those same stimuli. The latter seems to be dependent on dopamine and subcortical circuits, whereas the former may engage more cortical "top-down" mechanisms.

Figure 1. Characterization of a “sign-tracker” and “goal-tracker”

A) Sign-trackers (STs) approach and manipulate the lever (conditional stimulus, CS), reflective of incentive salience attribution. B) Goaltrackers (GTs) approach the location of food reward (unconditional stimulus, US) delivery upon presentation of the lever-CS. Images adapted from [58]. C) Mean +/− SEM Pavlovian conditioned Approach Index, a composite score used to assess the propensity of an individual rat to approach the lever-CS vs. the food cup (see [59]), is shown across 5 conditioning sessions for GTs (n=1867), those in the intermediate group (IG, n=2296) that vacillate between the two responses, and STs (n=1934). An Approach Index of −1 indicates behavior directly solely towards the food cup, whereas that of +1 indicates that behavior is directly solely towards the lever-CS. D) A histogram illustrating the population distribution of the propensity to attribute incentive salience to a reward cue in 6097 rats (the same rats used for panel C). Phenotype classification is based on an Approach Index of −1 to −0.5 for GTs, −0.5 to 0.5 for IG and 0.5 to 1 for STs. The large population of rats used for C and D has come from a database generated using Sprague-Dawley rats that have been screened for Pavlovian conditioned approach behavior in the labs of Drs. Shelly Flagel, Jonathan Morrow and Terry Robinson at the University of Michigan.

Figure 2. Sagittal schematic of the brain areas that are involved in reward processing for incentive stimuli

Brain regions highlighted in gray are those that show enhanced cue-induced activation in response to either a food- or drug-cue that has been attributed with incentive salience, but not both; those highlighted in yellow show enhanced activation to both food- and drug-associated cues that have been attributed with incentive salience; and those highlighted in red show enhanced activation in response to a food- or drug-cue, and have also been implicated in incentive salience attribution using other methods (i.e., lesions (PVT, BLA, lateral habenula), chemogenetics (VP), electrophysiology (VP)). Regions that are identified, but not highlighted are those that are believed to be involved in incentive salience attribution, but published data is currently lacking. Abbreviations: BLA, basolateral nucleus of the amygdala; CeA, central nucleus of the amygdala; CeM, central medial nucleus of the thalamus; HC, hippocampus; IMD, intermediodorsal nucleus of the thalamus; LH, lateral hypothalamus; MeA, medial nucleus of the amygdala; PVT, paraventricular nucleus of the thalamus; VP, ventral pallidum Image adapted from [27].