Mission | Strategic Plan | Org. Structure | Newsletters | Code of Ethics | Diversity Policy | Position Statements | Terms of Use

txtTitle Portal
txtTitle SABA
txtTitle Facebook
txtTitle ABAI Hotline
txtTitle ABAI FAQs

2008, Fall

A Message from the ABAI President

ABAI's Commitment to Licensing of Applied Behavior Analysts

Strategy for Governmental Affairs

Interview with Ann Poppenga: Behavior Analyst Elected to Town School Board

Dissemination of Behavior Analysis in the Treatment of Autism

Dissemination of Behavior Analysis in Education

Updates from ABAI Chapters and SIGs

Upcoming Conferences

2008 SABA Donors

Happy Birthday, Dr. Sidney Bijou!

ABAI Membership Information

ABAI Membership Form

2009 Convention Registration Form

2009 Convention Hotel Information

Autism Conference Registration

Donate to SABA

Newsletter

Volume 31 | 2008 | Number 3

Comprehensive ABA Programming: Matching Learner Needs with Instruction Strategies

By Dr. Mary Jane Weiss, BCBA, Rutgers University

Introduction

Parents and educators of children with autism spectrum disorders are faced with a wide array of choices of instructional approaches.  Most approaches are not empirically validated and are sorely lacking in evidence of their effectiveness.  ABA strategies have been empirically validated to a much greater extent than any other treatment approach.  A vast amount of evidence regarding the effectiveness of ABA has been accrued.  However, within ABA, there is variability in the amount of evidence that exists for particular strategies.  At times, clinical practice trends result in the widespread application of techniques that are not fully verified. 

In this article, we will review the techniques within ABA that have the strongest data base, as well as some of the clinical trends that are very promising.  Strategies for informing consumers and adding to scientific understanding will be explored.

Discrete Trial Teaching

Historically, within ABA, Discrete Trial Training (DTT) has been a primary instructional approach.  Discrete trial training (DTT) uses repetition and sequenced instruction to build skills in students with autism (Lovaas, 1981; Lovaas, Koegel, Simmons, & Long, 1973; Smith, 1993).  It has been successful in teaching a wide variety of skills in a structured, formalized context.  DTT provides a very large number of learning opportunities, which most certainly contributes to its success as a teaching methodology.  Elements of its effective use include errorless learning procedures (e.g., Etzel & LeBlanc, 1979; Lancioni & Smeets, 1986; Terrace, 1963; Touchette & Howard, 1984) and task variation and interspersal (e.g., Dunlap, 1984; Mace, Hock, Lalli, West, Belfiore, Pinter, & Brown, 1988; Winterling, Dunlap, & O’Neill, 1987; Zarcone, Iwata, Hughes, & Vollmer, 1993).  These strategies diverge from some historical applications of DTI, which often utilized blocks of identical target trials and procedures which allowed for repeated errors.  In addition, in clinical application, DTT has changed in other important ways, including the use of a naturalistic tone of instruction, the integration of strategies to promote generalization, and the shortening of inter-trial intervals. 

Discrete trial instruction is still very useful for teaching skills to children with autism, and its utility has not been eliminated with the emphasis on more naturalistic approaches.  DTT is well suited to teaching skills requiring repetition, to teaching skills that are not intrinsically motivating, and to building solid repertoires of receptive and expressive language skills (e.g. Sundberg & Partington, 1998; 1999). 

Naturalistic Teaching

Over the past 25 years, there has been a strong focus in ABA on the development of naturalistic teaching methodologies to meet the needs of learners with autism. Incidental teaching emphasizes getting an elaborated response from the individual, following their expression of interest in an item or a topic (Hart & Risley, 1982).  Incidental teaching has been demonstrated to be an effective approach for increasing initiation skills and for building a wide variety of language and conversation skills (e.g., Farmer-Dougan, 1994; McGee, Krantz, & McClannahan, 1985; 1986).  Importantly, incidental teaching procedures have been shown to have substantial generalization advantages, compared to discrete trial teaching (McGee, Krantz, & McClannahan, 1985).  These are significant advantages, as the strength of DTT is in building responsivity, and relative weaknesses include failure to build initiation skills or generalization in the absence of additional training.

In incidental teaching, the teacher arranges the environment to create interest on the part of the learner.  The teaching interaction begins with the learner initiating a request or a conversation about a particular item or topic.  The teacher prompts an elaboration of that initiation, and more elaborate communication from the learner results in gaining access to the desired item (Fenske, Krantz, & McClannahan, 2001).  One of the most substantial advantages of an incidental approach over a DTT approach is that the learner leads the teaching interaction.  The learner’s interests control the beginning and the flow of the instruction (Fenske, Krantz, & McClannahan, 2001).  Incidental teaching builds spontaneity, increases initiation, and shapes the complexity and sophistication of communicative responses.

Other naturalistic ABA approaches have also emphasized learner interests. Pivotal Response Training (PRT) and Natural Language Paradigm (NLP) emphasize the use of intrinsically motivating materials, teaching in natural contexts, and using the child’s interests to guide instruction in language (e. g., Koegel, Koegel, & Surrat, 1992; Koegel, O’Dell, & Koegel, 1987; Laski, Charlop,& Schreibman, 1988).  PRT Natural Environment Training (NET; Sundberg & Partington, 1998), similar to NLP and PRT, focuses on the use of intrinsically motivating materials and on following the child’s lead in language instruction.  NET also adds, however, the use of Skinner’s Verbal Behavior language classification system to guide language instruction (Skinner, 1957).  The use of this classification system is a newer clinical direction, and has been combined with both naturalistic strategies (as in NET) and with more formal instruction, such as DTT.  The use of the system is popular, as it helps to ensure comprehensive attention to the functions of language in curricular assessment and in programming.  This comprehensive focus is important, given that the verbal operants identified by Skinner have been shown to have functional independence, supporting the need to program for all of them (Lerman et al., 2005).  In comparison to the methods previously reviewed, there is less abundant literature on NET and on the use of the VB classification system to build language skills in this population.  However, the use of the VB system has excellent theoretical embeddedness and clinical relevance.  In addition, many elements of NET (e.g., initiation training, naturalistic teaching strategies) are well-established.

DTT and naturalistic methods target different deficits within autism spectrum disorders, and each methodology has distinct advantages and unique applications.  DTT is efficient and effective in teaching a wide variety of skills, but there is almost always a need for generalization training procedures.  Additionally, while responsivity improves dramatically within DTT, initiation skills, requesting, and conversation may be best taught through more naturalistic approaches.

Social skills

An area that has received increased emphasis in ABA in recent years is social skills.  A wide variety of strategies are in common clinical use for remediating social skill deficits.  Many of these strategies were developed outside of ABA (e.g., social stories; Gray, 1993; 1994) and most are employed as part of a package of interventions to address social difficulties (Weiss, 2007).   Others have more embeddedness within the discipline of ABA, such as rule cards, social scripts, and video modeling.  Rule cards assist students in following the social rules that are associated with a particular activity (e. g., Weiss & Harris, 2001). While there is not a body of literature on the use of rule cards, they are related to a variety of visual prompting strategies and behavioral rehearsal techniques (e.g., Cooper, Heron, & Heward, 2007; Snell & Brown, 2000).

Scripts can be in the form of sentences, words, or pictures, and they can be applied to a variety of circumstances (e.g., Snell & Janney, 2000). A number of studies document the effectiveness of scripts in facilitating social initiation and social interaction in individuals with autism (e.g., Krantz & McClannahan, 1993, 1998). video modeling has been used increasingly to build a variety of skills, including functional academic skills, community-relevant skills, conversational exchanges, and play skills (e. g., Charlop & Milstein, 1989; Haring, Kennedy, Adams, & Pitts-Conway, 1987; Sherer et al., 2001; Snell & Brown, 2000; Taylor, 2001; Weiss & Harris, 2001).   Unanswered questions in social skills applications that have fueled recent clinical innovation and promising research include: the extent to which ABA strategies can be used to address core social deficits in autism, the ways in which procedures can be combined to increase the effectiveness of intervention, and the analysis of the contribution that each method makes to a packaged approach. 

Rate-Building and Fluency

Fluency has been defined as responding accurately, quickly, and without hesitation (Binder, 1996; Dougherty & Johnston, 1996).  While fluency has been a goal of Precision Teaching, a field within the discipline of ABA instruction that has existed for many years and served many populations (e.g. Lindsley, 1992), it has only recently been focused on as relevant for learners with autism spectrum disorders (Fabrizio & Moors, 2003). 

Rate-building may be well-suited to addressing the skill deficits of learners with autism. For example, many learners on the autism spectrum exhibit motor dysfluencies.  They may still perform the task laboriously, inefficiently, or slowly, despite having met accuracy criteria.  Some individuals with ASD’s demonstrate long latencies to respond to instructions or to social initiations/bids.  Such lags in response times can lead to missed opportunities, especially in social contexts (Weiss, 2001; 2005).  

Fluency has been associated with a number of outcomes of learning, which are said to represent true mastery (Binder, 1996; Fabrizio & Moors, 2003; Haughton, 1980; Johnson & Layng, 1996).  Johnson & Layng (1996) emphasized the outcomes of Stability (capacity to engage in behavior in face of distraction); Endurance (capacity to engage in behavior for extended periods); Application (ability to generalize skills); and Retention (ability to maintain skills). 

There is debate within the field about whether fluency is a product of rate building (Doughty, Chase, & O’Shields, 2004).  There are several potentially confounding variables that may actually be responsible for the effects.  These include practice, as well as rate of reinforcement.  Practice itself has been clearly shown to facilitate learning (Samuels, 2002).  The type and amount of practice opportunities offered to learners definitely effects mastery (Ericsson, Krampe, & Tesch-Romer, 1993).  Furthermore, when learners are exposed to immediate feedback and repetitive trials, both their accuracy and speed improve.  It is possible that the high reinforcement rates used in rate-building are responsible for the positive effects.  Given that the mechanisms of effect are not fully understood, it may be that some of the benefits of rate-building can be achieved without a full implementation of this approach.  It may be that sensitizing staff to fluency building procedures (Binder, 1996) and tracking latency as a qualitative aspect of response may result in similar learner outcomes. 

Future Research Directions

Research is needed in a number of areas.  More research on teaching skills naturalistically, and on the advantages of using a naturalistic approach, would be helpful.  In the social skills arena, component analyses of package interventions would help to elucidate the primary mechanisms of change.  Furthermore, research on strategies to remedy central deficits in ASD’s, such as joint attention and perspective taking, might greatly augment our ability to impact these elusive targets of instruction. 

In the area of verbal behavior, more research is needed in replicating the functional independence of verbal operants, and in identifying instructional approaches that may aid transfer across operants.  In addition, examining the changes in the demonstration of verbal behavior (i.e., emergence of specific operants) over time of intervention might contribute to the development of a useful measure of progress in children with ASD’s.  Finally, examining the transfer of skills to the natural environment is a critical need.

Within rate-building to fluency, there is a need to examine the differential impact of overlearning vs. rate-building.  It would also be interesting to see comparative studies of rate-building vs. other ABA teaching approaches, particularly with respect to the outcomes of learning frequently cited as a result of building skills to fluency (i.e., stability, endurance, application, and retention.) In addition, there is a need to empirically investigate the utility of teaching component skills to increase composite skills. It would also be interesting to explore how rate-building can be used to build social skills.  In particular, the utility of rate-building for addressing problems in latency to respond should be explored. 

Clinical Recommendations

In clinical practice, we must maintain our commitment to the use of empirically validated procedures.  In particular, we must remain vigilant about the intensive application of ABA intervention, and beware of diluted approaches.  When exploring the relevance and impact of newer procedures, we must remain accountable by ensuring that we use procedures that are documented to be effective at the level of the individual.  Furthermore, we should contribute to the existing data bases on such procedures by systematically evaluating the effectiveness of such procedures, and by sharing the results of such applications in professional forums.

Summary

The use of a broad array of behavior analytic instructional methods will result in more comprehensive programming and improved learner outcomes for children with ASD’s (Fabrizio & Moors, 2003; Fenske, Krantz, & McClannahan, 2001; Weiss, 2001; 2005).  Specific advantages to utilizing each of the instructional approaches discussed are substantial.  DTT remains an excellent means of building a wide variety of skills.  Naturalistic instructional approaches of all kinds are better suited (than DTI) to increasing skills in initiating, conversing and relating to others.   Strategies that build rate of response may make responses more available in natural settings. 

Perhaps the greatest service we can provide to our consumers is to accurately represent the state of knowledge within our field.  As ABA professionals, we need to acknowledge the power of empirically validated strategies.  As responsible clinicians, however, we want to be open to expanding the array of strategies we use to meet the needs of learners we encounter.  In this context, we must exercise caution.  We can demonstrate our commitment to empirical validation by contributing to the data base that exists for such procedures.  In addition, we can help consumers sort through the hype to understand the elements of effective intervention, the methods for evaluating the success of intervention, and strategies for separating the wheat from the chaff.

References

Back to Autism Conference Presentations