A rigorous test of dual process model predictions for problematic alcohol involvement - Abstract
Problematic alcohol involvement (PAI), characterized by alcohol consumption that leads to or increases risk for
acute and/or chronic negative consequences in one or more life domains, poses an enormous public health
burden to drinkers, their families, and society. Heterogeneity in PAI’s etiology has been identified as the single
most important barrier to progress in remediating this burden. More effective characterization of neurobehavioral
traits that increase PAI risk is critical to the development of more effective intervention and prevention efforts.
Cognitive science approaches to addiction etiology have suggested PAI risk reflects an interaction between two
internal cognitive systems: appetitive motivation for alcohol (AMA) impelling alcohol seeking and use, and
executive function (EF) regulating the influence of appetitive drives. This Dual Process Model has been highly
influential, but prior studies have failed to support its main interaction hypothesis. We argue this failure is
attributable to weaknesses in the rigor of prior research, and propose several innovative refinements to the DPM
framework to improve its predictive utility. First, individual differences in AMA and EF have not been adequately
characterized in prior studies, most of which represent these constructs with single indicators. Innovations in
neuroclinical assessment indicate that neurobehavioral trait constructs are better characterized with measures
representing self-report, behavioral, and neurophysiological units of analysis. Second, DPMs assume risk for
PAI reflects only the influence of internal processes, implying no role for context. This is a major limitation, as
strong evidence indicates that both environmental contexts (e.g., legal constraints; alcohol access) and alcohol
exposure (e.g., acute effects of alcohol on EF and AMA) strongly affect the extent to which person-level factors
relate to PAI. Finally, given that PAI behaviors are the result of decisions people make while drinking, accounting
for heterogeneity in drinking-related decision strategies can improve prediction of PAI, including the role of DPM
constructs. We propose a multi-session, within-subject alcohol challenge experiment, combined with online
follow-up assessments, structured to take advantage of the abrupt change in contextual factors that occurs with
transition to the minimum legal drinking age (MLDA; i.e., 21st birthday). This transition broadly increases risk for
PAI within persons, but little is known about the role of specific etiologic factors in this increased risk. Participants
(N=220) will be recruited from rural census tracts in mid-Missouri, identified through state administrative and
driver’s license databases. Such individuals are broadly underrepresented in PAI research, particularly alcohol
challenge research, and tend to experience more chronic (less transient) and severe PAI than their relatively
advantaged, college-attending age peers. This sampling approach also will permit over-sampling of individuals
at higher risk for DUI arrest (e.g., based on parental history of DUI and area characteristics). Thus, this study will
fill important gaps in knowledge concerning PAI etiologic factors in an understudied population.