One lesson from contemporary emotions research is that, unfortunately, there is no fail-safe or direct way to elicit a given emotion reliably in all people. People cannot simply will themselves to feel a particular emotion, nor can clinicians or experimenters directly instill emotions in people. As such, all emotion induction techniques are by necessity indirect, often focusing on one component of the more complex, multicomponent emotion system. Because emotions typically unfold from individuals’ appraisals of the personal meaning of particular circumstances, the most ecologically valid way to induce emotions is to shape people’s appraisals of situations. Yet because meaning assessments are the products of unique personal histories and goals, they often are difficult to control. Instead, the focus often is placed on creating or recalling situations that, for most people, tend to elicit certain meanings, and therefore certain emotions. Other emotion induction techniques depart from ecological validity and aim to “jump-start” the emotion system by activating one or more downstream emotional responses: a facial or bodily muscle configuration, a physiological state, or a mode of thinking. Laboratory experiments have shown that cultivating these downstream components of an emotion often (but not always) can initiate or jump-start the entire, multicomponent emotion process (e.g., Levenson, Ekman, & Friesen, 1990; Schachter & Singer, 1962; Stepper & Strack, 1993; Velton, 1968). I use these perspectives on emotion induction to reexamine intervention strategies based on relaxation and increasing pleasant activities.