So what is the difference between stereotypy in typical and atypical development? A colleague of mine, Becky MacDonald (MacDonald et al., 2007), conducted a study in which young children were given the opportunity to play (and also observed when interacting with a familiar teacher) and the level of stereotypy was recorded. Children of various ages were studied and there were some children with and some without an ASD. The children with an ASD that participated were just entering early intensive behavioral intervention. One of the interesting observations was that two year olds with an ASD engaged in about as much stereotypy as their typically developing peers. However, three and four year olds with an ASD had much higher levels of stereotypy. This implies that differences in how much stereotypic behavior occurs may partly be a function of age. It might also be the case that as the child with autism ages, they learn to engage in additional forms of stereotypy and, perhaps, the motivation to engage in stereotypy increases.

Why is stereotypy, at least sometimes, problematic? When stereotypy occurs during academic instruction it is thought to interfere with learning (e.g., Dunlap, Dyer, & Koegel, 1983; Morrison & Rosales-Ruiz, 1997). It should be noted that a few studies suggest that it doesn't always interfere and can sometimes be used as motivation to enhance learning (e.g., Hanley et al., 2000). That said, there is little question that stereotypy can have a negative impact on social learning (e.g., Jones, Wint, & Ellis, 1990; Wolery, Kirk, & Gast, 1985) but why does it occur?

It has been widely presumed that stereotypy has no purpose and serves no function (e.g., Bodfish et al., 2000; Matson et al., 1997). Behavioral research has led to a different interepretation. A number of behavioral researchers and clinicians have suggested that stereotypy occurs because the sensory consequences that engaging in it produce are reinforcing (e.g., Iwata, 1999; Lovass, Newsom, & Hickman, 1987; Rincover, 1978). This is sometimes referred to as automatic reinforcement and numerous experimental analyses of stereotypy generally indicate that it is more likely that stereotypic behavior is either automatically reinforced by sensory consequences emanating from engaging in the behavior or is controlled by multiple sources of reinforcement, including the sensory consequences of the stereotypic behavior (e.g., Ahearn et al., 2007; Iwata, 1999).

Vaughan and Michael (1982) describe automatically reinforced responding as operant behavior maintained through the consequences directly produced while emitting the behavior. That is, the response produces consequences that the individual perceives or consumes. This reinforcement maintains the behavior in the absence of social consequences. A focus of the Vaughan and Michael paper was to describe the role that automatic reinforcement can have in the development and maintenance of complex, socially meaningful behavior such as language. Certainly social mediation is not involved in all instrumental behavior, and stereotypic behavior maintained by sensory consequences seems to fit the perceiving and producing categories of automatic reinforcement described by Vaughan and Michael. Problem behavior maintained by automatic reinforcement is of particular importance because it presents a significant obstacle to the development of effective intervention, at least partly due to the inaccessibility of the functional consequence (Vollmer, 1994; Vollmer & Smith, 1996). That is, the clinician can't directly alter the consequences for responding. Furthermore, with this type of behavior, capitilizing on these consequences might be crucial for producing adaptive behavior like functional language and play skills.  I'll have more on this topic in a future post.

There are many effective interventions for stereotypy and I will briefly mention a few here.  However, the utlimate goal with stereotypy is not suppressing behavior.  The ultimate goal is producing a varied repertoire of more adaptive behavior that occurs in the situations in which stereotypy may be problematic, like during academic instruction and social interaction.  One approach to treatment for automatically reinforced behavior has been to identify the specific source(s) of stimulation that maintains such responding. It has long been speculated that identifying a specific source of stimulation that maintains behavior can be translated into establishing other means of accessing similar sensory stimulation that can effectively complete with the less desirable behavior (e.g., Favell, McGimsey, & Shell, 1982; Vollmer, 1994). This hypothesis has spurred productive intervention research that has confirmed that providing alternative access to the sensory stimulation maintaining behavior (e.g., Piazza et al., 2000; Goh et al., 1995) can lead to higher levels of more (socially acceptable) appropriate behavior and lower levels of stereotypic behavior. However, it should be noted that access to dissimilar forms of sensory stimulation (e.g., Ahearn et al., 2005; Vollmer, Marcus, & LeBlanc, 1994) can also lead to lower levels of stereotypy (and higher levels of appropriate behavior).  These types of procedures, however, don't always work.

One study I published on this topic describes an alternative treatment, response interruption and redirection that involves explicitly redirecting stereotypic behavior (Ahearn et al., 2007).  Our study involved a therapist issuing a series of vocal demands, such as social questions the child could readily answer and that the child readily complied with during their regular academic programming. Vocal demands were presented contingent upon the occurrence of vocal stereotypy and were continuously presented until the child complied with three consecutively issued demands without emitting vocal stereotypy. For each child, treatment produced levels of vocal stereotypy substantially lower than those observed in baseline. For three of the four children, a substantial increase in appropriate communication was also observed. Follow-up probes were also conducted for three of the children in their classrooms and showed that these gains were maintained.

These two types of approaches are geared towards altering the level of stereotypic behavior but procedures that directly teach or foster other more socially acceptable behavior are a critical component to treatment.  In a future post I will discuss teaching procedures such as video modeling to promote play and social skills.  One last point I'd like to make is that though focused intervention is often very important in working with children with ASDs, I'm also a firm believer in assuring that there is some time available for these responses to be freely engaged in.  If behavior is not harmful and is something the person enjoys, then there should be some time for it.

Again, the references in this post are numerous and if you'd like them, let me know.  I'll pop them in the comments.