While past research has found that implicit measures are good predictors of affectively driven, but not cognitively driven, behavior it has not yet been tested which implicit measures best predict behavior. By implementing a consumer context, in the present experiment, we assessed two explicit measures (i.e. self-reported habit and tastiness) and three implicit measures (i.e. manikin task, affective priming, ID-EAST) in order to test the predictive validity of affectively versus cognitively driven choices. The results indicate that irrespective of whether participants chose affectively or cognitively, both explicit measures, but not the implicit measures, predicted consumer choice very strongly. Likewise, when comparing the predictive validity among all measures, the explicit measures were the best predictors of consumer choice. Theoretical implications and limitations of the study are discussed.