Investigating the role of expectancy in caffeine withdrawal using the balanced placebo design

Objective: This study investigated psychological influences on drug withdrawal symptomatology using a caffeine‐based model. Methods: Using the 2 × 2 balanced placebo design caffeine dose (given caffeinated vs decaffeinated coffee) was crossed with dose expectancy (told caffeine vs. decaf) among 87 (16‐hr abstinent) regular coffee consumers in a 2‐day study. Results: There were effects of expectancy and pharmacology that differed depending on the measure. Those told decaf reported greater caffeine cravings than those told caffeine 45 min and 8 hr postmanipulation. There were no expectancy effects on withdrawal symptoms or cognitive performance. There were pharmacological effects on all measures. Those given decaf reported greater withdrawal symptoms and showed poorer cognitive performance 45 min and 8 hr postmanipulation, with effects for headache and flu‐like symptoms first emerging 8 hr postmanipulation (i.e., 24 hr abstinence in given decaf conditions). Caffeine readministration alleviated all withdrawal symptoms and cognitive decrements within 45 min. No drug by expectancy interactions were observed. Conclusions: These findings confirm a strong pharmacological basis for caffeine withdrawal and an important role of cognition in drug craving. Future research should investigate the role of expectancy in drug withdrawal and craving and the potential use of expectancy manipulations in symptom prevention and management.