Chapter 6

The Case for Generalization

“Begin with the end in mind.”

—Stephen R. Covey, The 7 Habits of Highly Effective People

In his acclaimed book The 7 Habits of Highly Effective People, Stephen Covey (1989) listed as one of the now-famous seven habits the importance of beginning “with the end in mind” (p. 95). We attach the same degree of importance to this practice in the delivery of appropriate educational services. In our opinion, beginning with the end in mind means that we must not only pay careful attention to the generalization of skills but that we must do so at the beginning of the teaching process, not just at the end. The importance of generalization cannot be overstated, because skills that can only be performed in specific settings, or with only one person, or in one type of activity, are neither functional nor truly mastered.

The term generalization refers to the transference of the acquired skill or skills to other settings, activities, and people. Students with ASD have well-known problems with generalization. Hence, specific attention to the generalization of skills is critical for this population. That said, attention to generalization is also important for students with ADD/ADHD, NLD, S/LI, and ED.

The paucity of attention given to generalization belies its importance. For example, how functional would it be for a student to be able to request food items only from his teacher or only while seated at the horseshoe-shaped table in the classroom during snack time? In the autism community, these examples are by no means far-fetched, because students with ASD have a situation-specific, rigid learning style that does not easily accommodate the generalization of skills across people, settings, and activities. Similarly, how functional would it be for a student with ADHD or social-emotional difficulty to apply problem-solving strategies only in school, or for those with NLD or language-learning difficulties to be able to appreciate paralinguistic cues only in the therapy room? Clearly, although specific attention to generalization is imperative for students with ASD, programming for generalization should occur for all students with special needs in order to ensure skill mastery and functional performance across people, activities, and settings.

Generalization Differences Between ASD and Other Conditions

A tremendous amount of research documents the profound generalization needs of students with ASD (Gaylord-Ross, Haring, Breen, and Pitts-Conway, 1984; Gena, Krantz, McClannahan, and Poulson, 1996; Koegel, Koegel, and O'Neill, 1989; Ihrig and Wolchik, 1988; Taylor and Harris, 1995). In fact, problems in generalization are considered endemic in this population. This is not to say that generalization of skills should be assumed for students with other disabilities, for clearly that is not the case. However, the problems with generalization seen in students on the autism spectrum are not only different from but also go well beyond the scope of those of students with other disabilities. Specifically, where students with ADD/ADHD, NLD, S/LI, and ED may have problems generalizing certain skills, their transference problems appear to be rooted in more “superficial” circumstances—for example, in external conditions such as inattention, forgetfulness, distractibility, or working-memory problems. In contrast, students with ASD have deep-seated systemic problems with generalization that appear to be “wired in” to the disability. There are deficits inherent in the disorder itself that can actually preclude such individuals from independently connecting up disparate pieces of information in order to form a generalized concept. For example, stimulus overselectivity is a well-recognized feature of autism in which the student overfocuses on one small feature or component of an object or event. The item of focus is usually irrelevant and often related to the student's idiosyncratic interests. Such a narrow focus interferes with the establishment of meaning, because the relevant cues that enable understanding are ignored. Stimulus overselectivity can interfere with the generalization of skills to an unfamiliar environment if the student, overfocusing on an irrelevant detail, ignores the relevant cues that are intended to prompt the learned behavior in the new setting. There are many other reasons for the generalization problems that occur in ASD, including difficulty making appropriate connections and problems recognizing the need for the new behavior in a different environment or activity.

Generalization difficulty is not confined only to traditional academic subjects but also extends to such areas of functioning as executive function, communication and language, social skills, critical thinking, and other developmental domains, many of which loom large for individuals with the conditions covered in this book. Olley and Stevenson (1989) talk about the serious limitations in social skills progress in preschool children with autism that result from “their failure to generalize” (p. 356). These authors distinguish between the following two different types of generalization:

Most typically social behavior learned in one setting does not occur in other settings. Skills learned in the presence of certain children or adults are not used in the presence of other people. Behavior learned at one time seems lost or forgotten a short time later if the exact conditions of training are not present. All of these are examples of failures in stimulus generalization. Response generalization is a similar, difficult problem. A student may learn one response and use it consistently, but when a somewhat different response is called for, generalization does not occur. [p. 356]

Berkell (1992) echoes the sentiments of Olley and Stevenson, and makes a strong case for the link between generalization and learning, “Generalization strategies, including teaching skills across settings, materials, and people, are crucial to successful instruction” (italics added; p. 101). Likewise, Klin and Volkmar (2000) list the following as one of six specifications felt to be “positive and necessary” when judging the appropriateness of programming for students with Asperger syndrome: “An important priority in the program is to foster generalization of learned strategies and social skills. … From a programming perspective what is important is to define generalization explicitly as a goal to be achieved, including the various specific strategies to be implemented and the goals in the light of which the success of the program will be measured” (p. 348).

The bottom line when it comes to generalization is that there is clear research evidence that demonstrates that the success of the student's educational program is directly linked to the degree to which he or she has achieved generalization of acquired skills across settings, people, and activities. We think that Powers (1992) put it best when he said, “The powerful instructional technologies that have been developed over the years will be of little long-term value to children with autism [or to children with other disabilities] if skills acquired fail to generalize to untrained environments” (p. 237). Clearly, while the generalization needs are not as well documented in the research literature for students with conditions other than ASD, there is ample anecdotal information that indicates that these students will, nonetheless, be well-served by greater attention to generalization of skills. Therefore, we believe that for all of these students, and for the population of children with autism in particular, generalization strategies and conditions should be expressed in the IEP, and directly addressed, if the goal of appropriateness is to be realized.

If this information is insufficient to convince skeptics of the importance of direct attention to generalization, consider the federal court case Drew P. v. Clarke County School District (1989). In this case the lack of generalization of skills from the school environment to the home was cited as a material reason why the child's parents withdrew their son from the public school and placed him in a private residential program for students with autism. The court, applying the Rowley standard (Board of Education v. Rowley, 1982), decided in favor of the parents that this placement was necessary in order for the student to derive any educational benefit.

We have spent a good deal of time on the issue of generalization in order to make the case for careful attention to it; however, our interpretation of how the concept of generalization dovetails with that of appropriate education, while informed by research and clinical practice, is ours alone. Nevertheless, our arguments in favor of including specific information on generalization in the IEP are made on sound educational grounds.

Nuts and Bolts

A mistake that teachers and clinicians make far too often is to treat generalization, quite literally, as an afterthought—something to be thought of after the student has acquired a particular skill. In many cases, that is too late. Remember the adage noted at the beginning of this chapter, “Begin with the end in mind” (Covey, 1989, p. 95). Specifically, given the profound difficulty that students with ASD have with the generalization of skills, it is necessary to begin this time- and labor-intensive process at the beginning of the teaching cycle, rather than at the end. Moreover, we consider this to be an excellent practice for all students with special needs, as well.

While it is beyond the scope of this book to provide detailed how-to information on generalization, members of the IEP team are encouraged to seek out information on the principles of generalization and the strategies that promote it. Hence, we refer the reader to the work of Stokes and Osnes (1988). For our purposes it is important to highlight four interrelated factors of which educators and clinicians need to be mindful as they set about the task of programming for generalization. First is time. Simply stated, students with special needs require a great deal of “time in” when it comes to skill development across different settings, activities, and people in order to have sufficient practice in these contexts. The second factor is structured opportunities. Mere exposure to different settings, activities, and people is not enough. Instead, the educator should create structured opportunities for the skills to occur in each of the targeted circumstances. The final two factors, while important for all students with special needs, are arguably the most important for students with ASD. Teachers and clinicians need not only assist students in making the connections that they may not be able to make for themselves but also scaffold skill development by providing the necessary supports.

Documenting the Protocol

There are many ways to include generalization information in the IEP. The most direct way is to include it in the body of the annual goal and short-term objectives:

To avoid repetition, generalization criteria can also be specified as a “goal” in an addendum that is separate from the annual goal and short-term objectives. Consider the following:

Generalization goal: Seven to ten structured opportunities for practice will be provided across people, activities, and settings throughout the day until a performance level of 90 percent for each generalization target is reached.

Wrapping Up the Main Points