Implicit and explicit alcohol cognitions and observed alcohol consumption: three studies in (semi)naturalistic drinking settings.

AIMS   Dual-process models imply that alcohol use is related to implicit as well as explicit cognitive processes. Few studies have tested whether both types of processes are related to ad libitum drinking. In a series of three studies, we tested whether both implicit and explicit alcohol-related cognitions predicted the amount of alcohol consumed in an ad libitum (semi)naturalistic drinking situation. DESIGN   Two experimental studies used trained confederates (same-sex peers) who consumed either alcoholic or non-alcoholic beverages, while observing participants' drinking behaviour in a 30-minute session. The third study involved observations of participants' alcohol use during a 45-minute session in which participants spent time with five to seven friends. SETTING   A (semi)naturalistic drinking setting, a laboratory bar. PARTICIPANTS   Participants were undergraduates recruited at Radboud University (study 1: n = 115; study 2: n = 121; study 3: n = 200). MEASUREMENTS   We used coding of drinking behaviour from observations, questionnaire data on positive alcohol expectancies and alcohol use patterns and implicit association tests to assess alcohol associations. FINDINGS   Implicit associations were not related to observed alcohol use, whereas explicit positive expectancies were related positively to observed alcohol use in study 1 and study 2. CONCLUSIONS   Among undergraduate students in (semi)naturalistic drinking settings with peers, implicit alcohol-related cognitions do not predict the amount of alcohol consumed.