For a long period in the history of psychological research, emotion and cognition have been studied independently, as if one were irrelevant to the other. The renewed interest of researchers for the study of the relations between cognition and emotion has led to the development of a range of laboratory methods for inducing temporary mood states. This paper aims to review the main mood induction procedures allowing the induction of a negative mood as well as a positive mood, developed since the pioneer study of Schachter and Singer [Psychol Rev 69 (1962) 379-399] and to account for the usefulness and problems related to the use of such techniques. The first part of this paper deals with the detailed presentation of some of the most popular mood induction procedures according to their type: simple (use of only one mood induction technique) or combined (association of two or more techniques at once). The earliest of the modern techniques is the Velten Mood Induction Procedure [Behav Res Ther 6 (1968) 473-482], which involves reading aloud sixty self-referent statements progressing from relative neutral mood to negative mood or dysphoria. Some researchers have varied the procedure slightly by changing the number of the statements [Behav Res Ther 21 (1983) 233-239, Br J Clin Psychol 21 (1982) 111-117, J Pers Soc Psychol 35 (1977) 625-636]. Various other mood induction procedures have been developed including music induction [Cogn Emotion 11 (1997) 403-432, Br J Med Psychol 55 (1982) 127-138], film clip induction [J Pers Soc Psychol 20 (1971) 37-43, Cogn Emotion 7 (1993) 171-193, Rottenberg J, Ray RR, Gross JJ. Emotion elicitation using films. In: Coan JA, Allen JJB, editors. The handbook of emotion elicitation and assessment. New York: Oxford University Press, 2007], autobiographical recall [J Clin Psychol 36 (1980) 215-226, Jallais C. Effets des humeurs positives et négatives sur les structures de connaissances de type script. Thèse de doctorat non publiée. Université de Nantes, Nantes] or combined inductions [Gilet AL. Etude des effets des humeurs positives et négatives sur l'organisation des connaissances en mémoire sémantique. Thèse de doctorat non publiée, Université de Nantes, Nantes, J Ment Imagery 19 (1995) 133-150]. In music or film clip inductions, subjects are asked to listen or view some mood-suggestive pieces of material determined by the experimenter according to standardized music or film sets [J Ment Imagery 19 (1995) 133-150, Cogn Emotion 7 (1993) 171-193] and selected to elicit target moods. According to many authors, these two mood induction procedures seem to be among the most effective manners to induce moods [Br J Psychol 85 (1994) 55-78, Eur J Soc Psychol 26 (1996) 557-580] in an individual or in a group setting [Jallais C. Effets des humeurs positives et négatives sur les structures de connaissances de type script. Thèse de doctorat non publiée. Université de Nantes, Nantes]. As it is believed that multiple inductions contribute additively to a mood [Am Psychol 36 (1981) 129-148], researchers proposed to combine two or more techniques at the same time. Thus, the Velten Mood Induction Procedure has been successively associated with the hypnosis mood induction procedure [J Pers Soc Psychol 42 (1982) 927-934], the music mood induction procedure [Behav Res Ther 21 (1983) 233-239, J Exp Soc Psychol 26 (1990) 465-480] or the imagination mood induction procedure [Br J Clin Psychol 21 (1982) 111-117]. Successful combinations of inductions usually use a first induction that occupies foreground attention and a second one that contributes to congruent background atmosphere. One of the most successful combined mood induction procedures has been developed by Mayer, Allen and Beauregard [J Ment Imagery 19 (1995) 133-150]. This technique associates guided imagery with music and is supposed to increase effectiveness of the induction. In the second part of this paper the aim is to present the usefulness of mood induction procedures in the study of cognitive processes in depression [Clin Psychol Rev 25 (2005) 487-510], borderline personality disorder [J Behav Ther Exp Psychiatry 36 (2005) 226-239] or associated with brain imaging [Am J Psychiatry 161 (2004) 2245-2256]. Then the inherent problems to the use of experimental mood induction procedures are reconsidered. Doubts have effectively arisen about the effectiveness and validity of the mood induction procedures usually used in research. Some authors questioned whether a sufficient intensity of mood is produced or the possibility that the effects observed are due mainly to demand effects [Br J Psychol 85 (1994) 55-78, Clin Psychol Rev 10 (1990) 669-697, Eur J Soc Psychol 26 (1996) 557-580]. In fact, the various mood induction procedures are not equal with regard to the demand effects observed. The question of demand characteristics with respect to mood induction procedures is still under debate, even if demand effects are supposed to be most likely to occur with self-statement techniques (especially with the Velten mood induction procedure) or when subjects are explicitly instructed to try to enter a specific mood state [Eur J Soc Psychol 26 (1996) 557-580]. Another interrogation relates to the effectiveness of these various procedures of induction and the duration of induced moods. Generally, the various techniques used produce true changes of moods in the majority if not the whole of the subjects. However, certain procedures seem more effective in inducing a mood in particular [Br J Psychol 85 (1994) 55-78, Clin Psychol Rev 10 (1990) 669-697, Eur J Soc Psychol 26 (1996) 557-580]. As for the duration of induced moods this depends at the same time on the procedure used and the mood induced. Nevertheless, mood induction remains fundamental in the study of the effects of mood on the cognitive activities, insofar as it makes it possible to study the effects of negative as well as positive moods.