4. Neural similarity in responses to alcohol cues and dyadic social drinking in college students

• In Event: How adolescents’ behaviors and brain shape and are shaped by their peer social networks

Abstract

Background: For many young adults, college is an important life transition. Starting college can be challenging, however, as students face high social and academic pressures without the support of their families. In addition, young adults have not yet fully developed the skills to manage their emotions, particularly in social contexts (Somerville, 2013). The combination of environmental stressors, heightened sensitivity to social cues, and still-developing regulation skills makes young adults vulnerable to peer pressure to engage in unhealthy behaviors like alcohol consumption. In their first year of college, students show a striking increase in drinking (Schulenberg et al., 2001), at a high cost including academic difficulties, unintentional injuries, or even accidental deaths (Hingson et al., 2017). Given that most students report drinking for social reasons (LaBrie et al., 1994), an important question to ask is how do social relationships impact alcohol use? In this study, we propose that a central mechanism is how similarly people respond to alcohol at neural level. Expending on recent work indicating that friends tend to show similar neural responses to audiovisual movies (Parkinson et al., 2018), we hypothesize that students that drink together have more similar representations of alcohol than those who don’t.
Methods: To answer our research question, we used data from 22 college students (Mage = 20) belonging to a close-knit performing arts group. As part of a larger study, all participants rated how often they drank with each other in a round-robin design (231 dyads). Subsequently, they completed an alcohol¬-cue reactivity fMRI task, in which they saw pictures of alcoholic/non-alcoholic beverages in the scanner. We extracted participants’ average response to alcohol cues and then calculated each dyad’s neural dissimilarity when reacting to alcohol in 215 cortical and subcortical regions. Finally, we estimated the impact of drinking together frequency on neural similarity in a Bayesian multilevel linear regression model with random slopes for each brain region.
Results: Social drinking predicted dissimilarity in response to alcohol stimuli in a network distributed across the entire brain. This network, consisting of 63 subcortical (caudate, putamen, accumbens) and cortical regions in the ventral attention network (parietal and frontal operculum-insula), in the limbic network (OFC, temporal pole), in the default network (precuneus, posterior cingulate, PFC), includes a circuit involved in affective self-generated thought (Fox et al., 2018). Our results suggest that, when presented with alcohol stimuli, young adults spontaneously bring to mind memories of social drinking episodes that inform the affective valuations they make of the stimuli in front of them.
Conclusions: College is an important life transition marked by an increase in alcohol consumption, particularly in social settings. Given the negative consequences associated with drinking, it is important to understand how social relationships impact craving and drinking behavior. In this study, we show that students that drink together often exhibit more similar neural responses to pictures of alcohol beverages than those who don’t. This pattern of results has important implications suggesting that our valuation of alcohol stimuli is shaped by previous social interactions that involve alcohol consumption.

Authors

• Ovidia A. Stanoi, Columbia University
  Presenting Author

• Danielle Cosme, University of Pennsylvania

• Yoona Kang, University of Pennsylvania

• Mia Jovanova, University of Pennsylvania

• Xie He, Dartmouth College

• Niall Bolger, Columbia University

• Danielle S. Bassett, University of Pennslyvania

• David M Lydon-Staley, University of Pennsylvania

• Peter J. Mucha, Dartmouth College

• Emily B. Falk, University of Pennsylvania

• Kevin N. Ochsner, Columbia University