Abstract This paper describes three studies addressing the effects of enthusiasm (E) versus non-enthusiasm (NE) on the of skills taught to young children with autism.. Study 1 compared the effects of E vs. NE for teaching similar skills with slightly different topographies. Study 2 compared the effects of E vs. NE while teaching exactly the same skill; and study 3 examined the social validity of E vs. NE. Results of the first two studies were variable. E had no effect on skill in three cases; more effect than NE in two cases; and less effect than NE in one case. Study 3 indicated that parents and professionals discriminated the difference between the two conditions and preferred E over NE. Keywords: autism, enthusiasm, skill acquisition, social validity. Introduction Several instructional methods have been shown to be effective when teaching skills to children diagnosed with autism, including incidental teaching (Hart & Risley, 1968), discrete trial teaching (Lovass, 1987), milieu teaching (Kaiser, 1993), and naturalistic teaching strategies (Charlop-Christy & LeBlanc, 1999). Although differences exist among these instructional methods, several procedures are used consistently across each. For example, each method typically includes clear instructions, prompting, a system for prompt fading (e.g., prompt delay), and reinforcement of correct responses. In addition to these commonly used procedures to teach new skills, a recommendation is often made for teachers to be enthusiastic during instruction. In fact, being enthusiastic while teaching has been described as an important component in a number of models proposed for teaching children with autism. For example, in their discussion of the Developmental Social-Pragmatic model Prizant, Wetherby, and Rydell (2000) stated that emotional expression and affect sharing are central to the interactive and learning process (p. 205). Similarly, in their description of peer-implemented pivotal response training, Schreibman, Stahmer, and Pierce (1996) highlighted the importance of positive affect used by peers during the training. They recommended that peers should act in a highly animated fashion both verbally and nonverbally to increase social competence (p. 365). Furthermore, enthusiasm has been described as a good instructional method when providing praise (Anderson, Taras, & O'Malley Cannon, 1996, p. 187) and as crucial in fostering language acquisition (Rappaport, 1996, p. 307). Despite these claims about the importance of enthusiasm when teaching young children with autism, few studies have addressed this issue and most have included only typically developing children or adults. For example, Burts, McKinney and Burts (1985) investigated the effects of three levels of teacher enthusiasm (low, medium, and high) to teach concepts to young typical children. No differences were found between the three levels of enthusiasm on learning the concepts. Teachers reported, however, that students were more attentive, interested, and responsive in the high enthusiasm condition. In another study, Filcheck, McNeil, and Herschell (2001) examined the effects of enthusiastic praise, non-enthusiastic praise, and non-enthusiastic description on compliance and general behavior of 30 children between the ages of 3 and 5 years. The results from this study showed that the children had significantly higher rates of compliance in the non-enthusiastic description condition than in the enthusiastic praise condition. The children, however, had higher rates of general behavior (i.e., marble dropping game) in the enthusiastic praise condition than in the non-enthusiastic description condition. In another study, Lamers and Hall (2003) examined the effects of voice prosody (i.e., enthusiastic, conversational, monotonous) during instruction for children with autism and typical children. In study 1, only 3 of 12 children with autism showed a preference for a particular prosody (preference varied across children). …