The reinforcer pathology model proposes that alcohol use disorder (AUD) arises, in part, attributed to the excessive valuation of alcohol and increased preference for immediate rewards (Bickel et al., 2014). In the first domain, researchers have used techniques from behavioral economics to measure the relative reinforcing value (i.e., demand) of alcohol with simulated alcohol purchase tasks (APT; Kaplan et al., 2018; Murphy and MacKillop, 2006). These tasks measure self-reported alcohol consumption across a range of prices yielding an alcohol demand curve. Several indices of alcohol demand are generated from the APT, including intensity (i.e., consumption at free price), breakpoint (i.e., the price that suppresses consumption to zero), and Omax (i.e., maximum expenditure). Demand curve modeling (Hursh and Silberberg, 2008; Koffarnus et al., 2015) is also used to derive α, or the rate of change in elasticity across increased price. These indices are robustly associated with drinking levels and AUD severity (Martínez-Loredo et al., 2021) and predict alcohol treatment outcomes (Gex et al., 2022; Murphy et al., 2015).

While most behavioral economics research has considered alcohol demand to be a ‘trait-like’ feature, the reinforcer pathology model was recently expanded to incorporate contextual influences (Acuff et al., 2023). The updated model argues that the likelihood of alcohol consumption is influenced by the environmental choice context, including the presence of stimuli that may alter the reinforcing value of addictive substances. Thus, the degree to which a person values alcohol can vary substantially in different contexts, and the presence of alcohol-related cues may be one contextual determinant of alcohol demand. Experimental analyses of these effects typically use a cue reactivity approach (Reynolds and Monti, 2013) which has consistently demonstrated that exposure to alcohol-related cues—such as a person's preferred alcoholic beverage, alcohol bottles, alcohol artwork and advertisements—increases subjective craving for alcohol (Carter and Tiffany, 1999). A growing number of behavioral economic studies have investigated whether alcohol valuation also increases in the presence of alcohol cues (Amlung et al., 2012a; MacKillop et al., 2010; Owens et al., 2015; Rose et al., 2018; Spelman and Simons, 2018). A meta-analysis of this literature indicated significant alcohol cue effects across studies, including increases in intensity (Cohen's d = .34), breakpoint (d = .39), and Omax (d = .51), and decreases in elasticity (d = .77) (Acuff et al., 2020).

These studies suggest that alcohol valuation is increased in the presence of alcohol-related cues; however, the brain processes underlying this increase remain largely unclear. Addressing this question falls under the purview of neuroeconomics which involves using non-invasive neuroimaging techniques to better understand how the brain supports decision making (Glimcher et al., 2009). To this end, MacKillop et al., (2014) conducted the first functional magnetic resonance imaging (fMRI) study on alcohol demand in male heavy drinkers. Decisions to purchase alcohol were associated with frontoparietal activity, reflecting executive functioning, deliberation, and prospective thinking. Choices in the intermediate phase of the demand curve (i.e., elastic demand decisions) elicited robust significant frontostriatal and anterior insula activity, presumably reflecting the interplay among subjective reward value in the ventral striatum, interoceptive visceral processing via the anterior insula, and the external contingencies in dorsolateral prefrontal cortex (DLPFC). This study provided proof-of-concept that alcohol demand can be characterized using fMRI and paralleled research using other fMRI paradigms for purchases of cannabis (Bedi et al., 2015) and tobacco (Gray et al., 2017; Lawn et al., 2020).

In a separate line of research, numerous fMRI studies have characterized the neural underpinnings of alcohol cue reactivity in drinkers with a range of AUD severity (for meta-analyses, see Kühn and Gallinat, 2011; Schacht et al., 2013; Zeng et al., 2021). Across studies, alcohol cues elicited robust activation in mesocorticolimbic regions, including ventral striatum, anterior cingulate and ventromedial and dorsolateral prefrontal cortices. Studies comparing people with AUD and controls have consistently found AUD is associated with greater activation of parietal and temporal regions (Schacht et al., 2013) as well as medial prefrontal cortex and anterior cingulate cortex (Zeng et al., 2021). Associations between cue reactivity and subjective craving are less consistent across studies. Ventral striatum was the most consistent correlate in the Schacht et al. (2013) meta-analysis, but non-significant associations across studies have also been reported (see Zeng et al., 2021). Taken together, fMRI research using alcohol cue reactivity paradigms has revealed the neural correlates of responses to alcohol cues, but these responses are not always indicative of subjective alcohol craving, perhaps suggesting alternative appetitive or avoidance mechanisms (e.g., Chen and Li, 2023).

Despite this robust cue reactivity literature, no studies have examined the neural correlates of alcohol cue effects on demand. In addition, our previous neuroeconomics study included a relatively small sample of only male participants (MacKillop et al., 2014). To replicate and extend our prior research, the current study combined our validated fMRI APT paradigm with an alcohol cue reactivity protocol to investigate the neural correlates of alcohol demand decisions following exposure to neutral and alcohol cues in a gender-balanced sample of adult heavy drinkers. We hypothesized that alcohol cues would significantly increase alcohol demand and craving relative to neutral cues. An empirical region-of-interest (ROI) approach was used to test our hypotheses that increased alcohol demand following alcohol cue exposure would be associated with increased fMRI activation in brain regions related to reward salience (i.e., ventral striatum), craving and interoception (i.e., anterior insula), and awareness/introspection (i.e., posterior cingulate).