A lack of consensus exists on the cognitive process of food craving. Thus, the current study aimed to develop a model for food craving by examining the effects of intrusive thought and sensory imagery on craving intensity. Participants (297 adults, aged 18–49 years) completed the Japanese version of the Craving Experience Questionnaire about strength after food craving induction through a guided imagery of eating. Using structural equation modeling, the study evaluated the validity of the cognitive model of craving, in which only intrusive thought influences craving intensity. Furthermore, it examined two models derived from the elaborated intrusion theory, in which sensory image entirely or partially mediated the relationship between intrusive thought and craving intensity. Analysis revealed the best fit of the partial mediation model of the elaborated intrusion theory, which suggests that intrusive thought directly influences craving intensity and indirectly via sensory imagery.