Despite experimental findings and some treatment research supporting the use of cues as a means to induce and extinguish cravings, interventions using cue exposure have not been well integrated into contemporary substance abuse treatments. A primary problem with exposure-based interventions for addiction is that after learning not to use substances in the presence of addiction cues inside the clinic (i.e., extinction), stimuli in the naturalistic setting outside the clinic may continue to elicit craving, drug use, or other maladaptive conditioned responses. For exposure-based substance use interventions to be efficacious, new approaches are needed that can prevent relapse by directly generalizing learning from the therapeutic setting into naturalistic settings associated with a high risk for relapse. Basic research suggests that extinction reminders (ERs) can be paired with the context of learning new and more adaptive conditioned responses to substance abuse cues in exposure therapies for addiction. Using mobile phones and automated dialing and data collection software, ERs can be delivered in everyday high-risk settings to inhibit conditioned responses to substance-use-related stimuli. In this review, we describe how associative learning mechanisms (e.g., conditioned inhibition) can inform how ERs are conceptualized, learned, and implemented to prevent substance use when delivered via mobile phones. This approach, exposure with portable reminders of extinction, is introduced as an adjunctive intervention that uses brief automated ERs between clinic visits when individuals are in high-risk settings for drug use.

The medial prefrontal cortex integrates information about salience and valence of stimuli, including rewarding solutions like alcohol and sucrose, and regulates aspects of alcohol seeking and consumption. However, our understanding of how cortical outputs encode alcohol consumption is limited. Using fiber photometry to measure calcium activity in putative prelimbic (PrL) glutamatergic projection neurons, we show similar but distinct patterns of activity during the peri-consummatory phase in response to consumption of water (non-deprived conditions), ethanol (20% v/v), or sucrose (1% w/v). PrL population activity appears to track hedonic value, as GCaMP6f signals ramped immediately preceding bouts for water, ethanol, and sucrose, and the signal scaled with presumed hedonic value, i.e., water<ethanol<sucrose. Further, using machine learning, population activity of PrL neurons prior to consumption was sufficient to predict both consumption of and distinguish between these different solutions. To assess valence encoding, we adulterated the ethanol solution with quinine, a bitter tastant. In non-dependent mice, calcium activity surrounding drinking bouts was reduced, paralleling the decreased consumption of quinine-adulterated ethanol. This effect was not present in ethanol dependent mice, suggesting altered hedonic value of the adulterated solution either due to reduced sensitivity to the aversiveness of quinine or increased sensitivity to the reinforcing value of ethanol. The global population of PrL glutamatergic neurons also display sustained GCaMP6f 'up-states' that last tens to hundreds of seconds, which were longer and larger when consummatory bouts occurred. Overall, our results demonstrate a functional signature in PrL neurons that differs across solutions and is disrupted by ethanol dependence.

Dopamine release in the nucleus accumbens is classically linked to associative learning, signaling relationships between predictive cues and outcomes. Yet, dopamine is also strongly modulated by novelty, a nonassociative factor that has received comparatively little attention. Here, we used optical dopamine sensors in awake, behaving male and female mice to define how novelty alters the temporal dynamics of dopamine release during aversive learning. We manipulated novelty in three ways: (1) omitting expected footshocks, (2) introducing novel neutral cues concurrently with shock-predictive stimuli, and (3) presenting novel stimuli in an unpaired fashion within a context. Across all conditions, manipulations robustly increased dopamine release and in some cases altered the directionality of cue-evoked dopamine responses. Notably, these effects extended beyond the immediate stimulus window, altering subsequent responses to both conditioned cues and footshocks. Together, these findings demonstrate that changes in the environment that extend beyond prediction-based learning can exert a powerful and sustained influence on dopamine signaling, reshaping how aversive cues and outcomes are represented in the brain.

Substance use disorder (SUD) is a chronic neuropsychiatric condition characterized by persistent drug seeking and impaired behavioral control. Dopaminergic signaling has long been recognized as a central regulator of reinforcement learning, motivation, and habit formation. Addictive substances profoundly alter dopamine transmission through multiple mechanisms. These drug-induced changes contribute to the initiation, escalation, and persistence of addictive behaviors. In addition to dopamine, the cholinergic system has emerged as an important modulator of striatal circuit function. Acetylcholine and its receptors interact extensively with dopaminergic pathways, shaping striatal signaling dynamics and influencing learning and action selection, with particularly strong relevance for nicotine dependence. In this review, we discuss how striatal dopamine and acetylcholine contribute to learning, habit formation, and addiction-related behaviors, as well as how these systems interact at the circuit level. By integrating these findings, we propose a framework for understanding how dopamine–acetylcholine interactions may influence behavioral regulation relevant to substance use disorders.

Adaptive behavior requires flexible encoding of emotional valence. Although striatal acetylcholine (ACh) signaling is critical for reinforcement learning, its contribution to aversive learning has remained poorly defined. Here, we demonstrate that ACh release in the dorsolateral striatum (DLS) is selectively biased toward negative valence. Using fiber photometry with a genetically encoded ACh sensor, we found that ACh release robustly increased during threat prediction but decreased in anticipation of rewarding outcomes, revealing a bidirectional and valence-specific signature. Optogenetically stimulating ACh release at cue onset accelerated threat learning, impaired extinction, and shifted behavioral responding toward persistent threat expectancy. Concurrent single-cell calcium imaging and optogenetic manipulation revealed that elevated ACh release dynamically reorganized DLS ensemble activity, increasing both excited and inhibited neurons and producing large-scale state-space divergence during threat cues. During extinction, optogenetically sustained ACh release preserved the organization of threat-predictive DLS ensemble activity despite the absence of shock. These findings identify DLS ACh as a valence-specific neuromodulatory signal that reconfigures striatal network dynamics, primes ensembles for impending threat, and biases learning toward threat persistence.

Value‐based decision‐making is at the core of nearly all motivated behaviors and requires the ability to associate outcomes with specific actions and make adaptive decisions about future behavioral action. At the core of value‐based decision‐making and reinforcement is the nucleus accumbens (NAc), which is integrally involved in learning, selecting, and executing goal‐oriented behaviors. The NAc is a heterogeneous population primarily composed of D1 and D2 medium spiny projection (MSN) neurons that are thought to have opposing roles in behavior, in which D1 MSNs promote reward and D2 MSNs promote aversion. By expressing channelrhodopsin selectively in D1‐ and D2‐ populations (using D1‐Cre and A2A‐Cre mice) in the NAc core, we show that mice will nose poke for optical self‐stimulation of both cell types, suggesting D2‐MSN activity is not inherently aversive. While optogenetic approaches give some information about how cellular activation can modulate behavior, they eliminate the temporally specific neural activity patterns that encode information in behaving animals. To understand how real‐time activity in these populations is linked to behavioral execution, we expressed the genetically encoded calcium indicator (GCaMP6f) within D1 and D2 MSNs coupled with in vivo fiber photometry to record from these cell populations in awake and behaving animals during pavlovian and operant conditioning tasks. Utilizing complex reinforcement schedules that allow dissociation of stimulus value, outcome, cue learning, and action, we show that D1 MSNs respond to the presence and intensity of unconditioned stimuli – regardless of value. Conversely, D2 MSNs respond to a mismatch between what is expected and what is received and thus encode errors in prediction in a value‐independent fashion. To establish a causal link between the information encoded within these neuronal populations and behavioral output, we next expressed an inhibitory opsin (halorhodopsin) selectively in either D1‐ or D2‐MSNs to inhibit these cells at specific time points during a fear conditioning task. We show that inhibition at select time points, either by inhibiting D2‐MSN activity during the predictive cue or D1‐MSN activity during the shock outcome, disrupts learning of the association. We thus provide foundational evidence for the discrete aspects of information that are encoded within these cellular populations.