Using Discrete Trial Teaching within a public preschool program to facilitate skill development in students with developmental disabilities.
There is a great need to identify specific instructional methods that effectively promote positive skill development in young children with developmental disabilities. One method that has received strong empirical support with children with autism is Discrete Trial Teaching (DTT); however, the effectiveness of DTT has not been extensively evaluated with children who have developmental disabilities other than autism. This project was an initial investigation evaluating the practicality and effectiveness of providing DTT instruction to children with a wide range of developmental disabilities within an existing public preschool program. Participants were randomly assigned to receive DTT or individual attention in a control condition. The project evaluated the effects of providing DTT on the participants' cognitive, language, behavioral, and social-emotional functioning. Results generally indicated positive changes in adaptive behavior development and social-emotional functioning for students who received DTT. Implications and directions for future research are discussed.
For the past three decades educators and researchers have sought to identify methods that positively impact the developmental and future educational outcomes of young children with developmental disabilities. In order to summarize the knowledge gained through those efforts the Division of Early Childhood (DEC) of the Council for Exceptional Children (CEC) conducted a federally funded review of the relevant research literature (B. J. Smith et al., 2002). Based on their review, the DEC generated and published the DEC Recommended Practices in early intervention/early childhood special education (EI/ECSE) for assessment; child-focused interventions; family-based practices interdisciplinary models; technology applications; personnel preparation; and policies, procedures, and systems change (Sandall, McLean, & Smith, 2000).
The DEC Recommended Practices for child-focused interventions provide educators with general teaching guidelines based on practices that have shown evidence of effectiveness in EI/ECSE. For example, the current guidelines recommend that a variety of instructional procedures such as incidental teaching, mand-model procedure, naturalistic time delay, peer-mediated strategies, prompting and prompt fading be implemented within and across students' activities "with sufficient fidelity, consistency, frequency and intensity" (Sandall, Hemmeter, Smith, & McLean, 2005, p. 92). Despite these guidelines, research indicates that EI/ECSE educators, including many paraprofessionals, may experience difficulty taking the DEC recommendations and implementing them in the classroom as effective instructional practices. This is due in part to the fact that the majority of preschool educators have little to no formal training in instructional methods that can be used with young children who have disabilities (Bricker, 1995; Schepis, Reid, Ownbey, & Parsons, 2001). Complicating matters further, there remains a lack of consensus among professionals regarding the best way to provide instruction to children with developmental disabilities in EI/ECSE settings. For example, some researchers advocate child-directed and embedded-teaching approaches (Greenspan & Wieder, 1999; Schepis et al., 2001), while others emphasize the importance of using teacher-directed instruction (Engelmann & Osborn, 1970; Green, 1996; Lovaas, 1987, 2003), especially when teaching new skills (Losardo & Bricker, 1994; T. Smith, 2001). Another complication is the striking diversity one finds when examining the population of children who receive EI/ECSE services, which suggests that instructional practices that are effective with any given student may not be effective with others. As a result of the philosophical differences, student diversity, and training issues, developmental and educational outcomes for many children who have disabilities may be less than optimal dependent on the empirical and philosophical rule sets that determine the curricular and instructional decisions of EI/ECSE educators.
In addition to the general guidelines published by the DEC, there clearly is a need to identify and empirically validate simple and specific effective instructional methods that lead to significant gains for students with disabilities. These instructional methods should have clear and efficient training procedures and be relatively easily implemented by EI/ECSE educators (teachers, paraeducators, and caregivers) across a wide range of preschool settings. It is of paramount importance that the instructional methods used lead to tangible and measurable developmental and educational gains for children with developmental disabilities. Additionally, researchers must identify which instructional methods are most effective for students with certain developmental and diagnostic profiles. Methods that are cost-effective, simple, practical, flexible, and that generalize across a variety of settings and children will likely be most useful as educators work to improve the long-term developmental and educational outcomes of children with disabilities.
One instructional method that has been used to successfully improve the developmental and educational outcomes of a specific group of young children (those with autism and developmental delay) is Discrete Trial Teaching (DTT) (Lovaas, 1987, 2003; T. Smith, 1999, 2001). DTT is a specific type of teacher-directed instruction that is empirically grounded in the experimental analysis of behavior and simplifies instruction to improve children's learning. The procedure entails a discrete onset and offset of each instructional interaction that bookend a discrete, defined response by the learner. In addition, the effectiveness of DTT to bring about a correct response is enhanced by the use of response prompting and pacing strategies, and the strategic use of massed, expanded, and distributed instructional trials.
It has been reported (Lovaas, 1987; McEachin, Smith, & Lovaas, 1993) that 47% of children with autism taught through an organized, early intensive curriculum employing DTT instruction achieved normal (i.e., average or above) levels of intellectual and academic functioning after two to three years of treatment and that those gains were maintained over several years. This success rate is remarkable given that other instructional methods used with children with autism resulted in only one to two percent of the children achieving a normal level of functioning (Klinger & Dawson, 1997; Howard, Sparkman, Cohen, Green, & Stanislaw, 2005), a rate equaling that of spontaneous (no treatment) recovery.
The DTT instructional method has several advantages. Perhaps most important is the five-step DTT instructional procedure (described in the methods section), which can be used to teach young children a wide range of skills and has proven particularly effective in teaching children with autism new behaviors and skills. For example, children who were previously completely non-verbal have been successfully taught to speak (Young, Krantz, McClannahan, & Poulson, 1994). Other research has shown that DTT is an effective instructional method that can be used to teach children with autism and developmental delay generalized imitation (Coe, Matson, Fee, Manikam, & Lanarello, 1990; Young et al., 1994), receptive language (Lovaas, 1977), expressive language (Howlin, 1981), conversational skills (Krantz & McClannahan, 1981), grammar and syntax (Risley, Hart, & Doke, 1972), play skills (Coe et al., 1990), and social-emotional skills (Downs & Smith, 2004).
DTT procedures are readily adaptable to individualized, developmentally appropriate curricula and facilitate ongoing data collection that allows for a continuous formative assessment of student progress. The very short instructional units used in DTT can lead to rapid skill acquisition and are particularly appropriate for young students with limited participation in instructional tasks and formats. An additional benefit of the short DTT instructional units is that the method can be used flexibly depending on the needs and resources of each student and EI/ECSE program. For example, DTT can be used as an adjunct to regular EI/ECSE programming (as speech, physical, and occupational therapy often are) in time periods ranging anywhere from 10-120 minutes per day. Alternatively, although often thought of as occurring outside of students' typical routines, DTT instruction can be embedded throughout the ongoing daily activities of the preschool. Clear and efficient training procedures ensure that EI/ECSE teachers, paraprofessionals, tutors, and caregivers can be taught to implement the DTT instructional procedures across environments (Koegel, Russo, & Rincover, 1977). Finally, DTT procedures are consistent with DEC recommended practices for child-focused interventions (Sandall et al., 2000; Sandall et al., 2005)
Because DTT procedures emphasize the establishment of discriminated operants (i.e., specific learner responses that are reliably given in the presence of specific instructional cues) some children with autism require formal generalization and maintenance programming to show learned behaviors outside of the one-to-one teaching situation (T. Smith, 2001). DTT can also become highly labor intensive as it is often utilized to engineer rates of up to 180-200 overt learner responses per hour during 25-40 hours of instructional programming per week (T. Smith, 2001). Finally, some educators and professionals have voiced philosophical and theoretical concerns about using applied behavior analytic approaches to educate young students with autism (Greenspan, 1992; Kimball, 2002). These concerns largely center on the teacher-directed nature of behavioral approaches and a bias that such techniques emphasize skill building but neglect children's social and emotional needs.
Despite its challenges, DTT has proven to be an extremely useful instructional technique for improving the educational outcomes and minimizing the future costs of providing services for children with autism and developmental delay (Lovaas, 1987; McEachin et al., 1993). Indeed, according to Jacobson, Mulnick, and Green (1998) early intensive behavioral interventions for children with autism can lead to cost savings of approximately $200,000 per child between the ages of 2- and 22-years, and from $656,000 to $1,082,000 per child between the ages of 3- and 55-years. Although these results are impressive, the effectiveness of DTT has not been thoroughly investigated when used to facilitate skill development in children presenting other developmental disabilities in public school settings. The remarkable success of DTT with children with autism and developmental delay suggests that this method should be widely utilized to promote the academic growth of children with other development disabilities. As noted by Lovaas (2003) children with developmental disabilities other than autism often have many characteristics that are similar to those exhibited by children with autism (e.g., poor social skills, attentional problems, language and cognitive delays, and behavioral problems). Because many children with developmental disabilities often are less delayed in their development than children with autism, one would expect that such children may make even more remarkable gains when taught through DTT procedures; however, that contention needs to be evaluated.
Further, it is important to determine whether DTT instruction can be implemented within existing public preschool programs in a practical, time-limited, and cost-effective manner. It was the goal of this project to conduct an initial investigation of whether DTT could be used within an existing public preschool setting to significantly and positively impact the cognitive, adaptive, communication, social-emotional, and/or behavioral functioning of children with developmental disabilities.
Participants were twelve children (seven boys and five girls) who were enrolled in a publicly funded developmental preschool located in a rural community in Washington State and whose parents elected to have them participate in the research project. Students were referred to and enrolled in the preschool when they demonstrated significant developmental delay (i.e., 2 or more standard deviations below the mean) in one or more areas of functioning (communication, motor skills, social/adaptive behavior, and/or cognition). Some of the students had a specific diagnosis such as autism; however, most had not yet received any clinical diagnosis and were simply demonstrating developmental delay in one or more domains, thus placing them at serious risk for poor educational, behavioral, and psychosocial outcomes. Students were randomly assigned to the experimental or control groups at the beginning of the 2004-2005 academic year. Table 1 presents information for each student in the two groups regarding age, gender, and diagnostic/developmental status. The mean age at baseline of students in the experimental and control groups were 48.83 months and 47.50 months respectively and were not significantly different. Nine of the children were White, three were of mixed ethnicity, and all came from homes where English was the primary language.
DTT Instruction. Each DTT unit of instruction lasts for approximately 3-10 seconds and consists of five parts (see Lovaas, 2003 and T. Smith, 2001 for a more detailed description of DTT procedures):
1. Discriminative Stimulus (Cue): The instructor presents a brief instruction or question to the student (e.g., "Do this" or "Give me blue").
2. Prompt: In conjunction with or immediately after the cue, the instructor assists the student to make the correct response (e.g., the teacher models the response or takes the student's hand and guides him/her to perform the response). As instruction progresses the prompt is gradually faded and eventually eliminated so that the student responds to the cue alone.
3. Response: What the student does or says following the cue.
4. Consequence: Correct responses are immediately reinforced with verbal praise, access to toys, or other things the student enjoys. If an incorrect response occurs the instructor withholds positive reinforcement and, following the inter-trial interval (see below), implements an "error correction" procedure.
5. Inter-trial interval: After the consequence, the instructor pauses for 1-5 seconds before presenting the cue for the next trial.
Students typically spend 15 minutes or less in each DTT session and take an approximately 5- to 15-minute play break in between each session during which time the instructor prepares materials for the next session and checks the just collected data for accuracy and completeness. The data collection consists of a documentation of each learning trial that includes information on the items targeted, student responses, and the level and type of any prompt that was used. This ongoing data is used to continuously modify instruction and goals to ensure maximum benefits and efficacy.
DTT Training. The lead experimenter, a clinical psychologist with extensive training and experience in DTT, provided training in DTT procedures to the preschool staff, caregivers, and undergraduate and graduate student research assistants at the beginning of the academic year. Research assistants assigned to provide DTT (hereafter referred to as instructors) received between 15 and 20 hours of didactic training, modeling, and practice with corrective feedback provided by the lead experimenter and a graduate student who also had several years experience providing DTT. The two primary trainers assessed instructor competence in correct use of DTT instructional procedures and recording of student responses (assessment) in vivo following the training period using an adaptation of definitions and procedures outlined by Koegel et al. (1977). Inter-observer reliability was calculated for 25% of the instructor competence ratings with agreement ranging from 90 to 100%. Competence of individual instructors following training ranged from 79 to 96% correct use of DTT instructional and assessment procedures.
Intervention Group. All students in both the intervention and control groups attended the preschool for two hours per day an average of three days per week (6 hours total per week). Students in the intervention group were "pulled out" of their regular preschool programming in 10- to 15-minute blocks of time in order to receive the DTT instruction on a one-to-one basis in an adjacent classroom or outside on the playground. Each student received between 30 and 42 hours of DTT over the course of the 27 weeks of the intervention at an average of 1.30 to 1.58 hours of DTT per week.
DTT procedures were used to teach the students skills in several developmental areas including receptive and expressive language (e.g., identification of objects, behaviors, emotions, colors, shapes), socialization (e.g., conversational skills, turn-taking), pre-academics (e.g., letters, numbers, counting), imitation (e.g., gross and fine motor), daily living skills (e.g., following directions), and fine motor skills (e.g., drawing, cutting). Following baseline assessments and consultation with caregivers and the regular preschool teacher, an individualized curriculum was developed for each student in the DTT group using materials adapted from Lovaas (2003) and Leaf and McEachin (1999). Due to the varying developmental levels presented in the intervention group not every student received instruction in every domain (e.g., some children did not receive instruction in the expressive identification of numbers and letters). Whenever possible, instruction was balanced across developmental areas for each student in the intervention group.
Procedural integrity. To ensure treatment fidelity, the lead experimenter and/or a graduate assistant with extensive training and experience implementing DTT supervised all sessions, and daily feedback was given to each instructor orally and via a structured checklist. The checklist was developed using definitions provided by Koegel, Russo, and Rincover (1977) and was utilized daily until the instructor achieved at least 90% competence in all skill areas of instruction and assessment. After achieving at least 90% competence, instructors were rated on the checklist and provided feedback a minimum of once per week, while also continuing to receive daily oral feedback. Inter-rater reliability for the structured checklist was assessed periodically throughout the school year (i.e., every 3 weeks), with agreement ranging from 87 to 100%. Daily data also was collected to assess the amount of intervention provided and the extent to which instruction was provided across all programs in each student's curriculum.
Caregivers of students in the intervention group were provided training in DTT and the availability of ongoing supervision, and were encouraged to provide additional in-home instruction for their child. However, none of the caregivers of students in the intervention group reported providing in-home DTT to their children, with each citing time constraints as the major barrier to doing so. Thus, only the designated DTT instructors provided DTT instruction to students in the intervention group. Additionally, it is important to note that none of the students in either the intervention or control group were reported by their caregivers as receiving any services in the home or community other than those provided at the preschool.
Control group. Students in the control group received their regular preschool programming, which followed a child-centered philosophy stressing social and emotional development, modeling, and incidental teaching of skills. Although not following a specific curriculum, the regular preschool programming sought to facilitate students' growth across the same developmental areas targeted by the DTT instruction (e.g., language, socialization, pre-academics, daily living skills, motor skills). In order to address the possibility that students in the intervention group could make gains simply due to receiving individual attention, a sufficient number of research volunteers were included in the project to raise the number of staff in the preschool so that students in the control group received the same amount of individual attention as students in the intervention group. Daily observations of the preschool were conducted to ensure that students in the control condition received one-to-one attention at rates comparable to the students in the intervention group.
Due to the consistency in the developmental areas targeted by the regular preschool programming and the DTT instruction it is possible that the regular classroom instruction may have supported or maintained the skills targeted by the DTT instruction. However, the regular preschool educators were not informed of the details of the intervention group students' curricula so that the educators could not explicitly seek to reinforce the learning that was occurring in response to the DTT instruction. This lack of coordination, although clearly not desirable in real-world settings, was instituted to help reduce the possibility that the experiences of intervention and control group students in the regular preschool environment would systematically differ in ways other than receiving or not receiving a proportion of their instruction via DTT.
Assessment. All participants in the study received a comprehensive evaluation within two weeks of the beginning and end of the academic year, with the evaluations separated by eight months. The evaluations assessed each student's cognitive, language, adaptive, behavioral and social-emotional functioning. The evaluations were designed to measure a wide variety of skills and behaviors that are important indicators of positive development, and many of which are predictive of later educational, social-emotional, and behavioral success. Advanced graduate students who were blind to conditions and supervised by the lead experimenter conducted all evaluations.
Baseline and Post-Intervention Measures.
Cognitive Functioning. Each student's level of cognitive functioning was assessed using the Wechsler Preschool and Primary Scale of Intelligence-Third Edition (WPPSI-III) (Wechsler, 2002). The WPPSI-III has demonstrated excellent internal consistency (.86-.97) and test-retest (.84-.92) reliability and validity for both children with and without developmental delay and strongly correlates with other measures of cognitive functioning (Wechsler, 2002).
Language Functioning. Each student's receptive vocabulary skills was assessed with the Peabody Picture Vocabulary Test-Third Edition (PPVT-III) (Dunn & Dunn, 1997). The reliability and validity of the PPVT-III have been extensively assessed, and it illustrates excellent internal consistency (.94) and test-retest (.91-.94) reliability and correlates well with other cognitive measures of verbal ability (Dunn & Dunn, 1997).
Adaptive Functioning. The Vineland Adaptive Behavior Scales: Interview Edition (VABS) (Sparrow, Balla, & Cicchetti, 1984) was administered with each student's caregivers to determine the student's level of adaptive functioning in the areas of communication, daily living skills, socialization, and motor skills. The VABS is a widely used measure that has demonstrated good internal consistency (.70-.95) and test-retest (.77-.98) reliability, as well as adequate content, construct, and criterion-related validity (Sparrow et al., 1984).
Behavioral Functioning. The Behavioral Development Questionnaire (BDQ) (Castelloe & Dawson, 1991) was given to caregivers in order to assess each student's social interaction behavior. The BDQ has adequate internal consistency (.63-.85) and inter-rater reliability (.60-.81) (O'Brien, 1996) and correlates well with clinician ratings and behavior observations (Castelloe & Dawson, 1991). Behavioral and emotional functioning also were assessed via caregiver report with the Social Skills Rating System (SSRS) (Gresham & Elliot, 1990) and the Behavior Assessment System for Children (BASC) (Reynolds & Kamphaus, 1992). The SSRS has good internal consistency (.73-.95) and test-retest reliability (.65-.93) (Gresham & Elliot, 1990). The BASC also demonstrates good internal consistency (.84) and test-retest reliability (.83-.95) and correlates moderately to strongly with five other measures of emotional and behavioral functioning (Reynolds & Kamphaus, 1992).
Social-Emotional Functioning. Each student's ability to interpret the emotions of others was assessed with the Emotional Understanding Test (EUT) (Howlin, Baron-Cohen, & Hadwin, 1999). The EUT has demonstrated good internal consistency (.86) and test-retest reliability (.61 after a 6-month inter-test period) (Downs & Strand, in press).
DTT Programs and Item Mastery
Table 2 summarizes data from the experimental group regarding the number of items that each student did not know at baseline and mastered by the end of the academic year in the broad curriculum areas of receptive language, expressive language, and imitation. Items were considered mastered if the student responded correctly at least 80% of the time over at least two days, with two different instructors. The number of specific programs across those three broad areas in each student's curriculum ranged from a low of eight for student #3, to a high of 15 for student #4, over the course of the academic year. Because control group participants were not directly exposed to the DTT programs and ongoing data collection there were no comparison data available regarding item mastery.
Cognitive and Language Functioning
Due to the small number of participants, results are presented for each student enrolled in the study. Table 3 presents the students' cognitive and language scores at baseline and at the end of the academic year. Mann-Whitney U tests revealed that there were no significant differences between the DTT and control groups on WPPSI-III Verbal IQ (VIQ), Performance IQ (PIQ), Full-Scale IQ (FSIQ), or PPVT-III performance at baseline or post-intervention. Additional Wilcoxon signed rank tests indicated that neither the WPPSI-III nor the PPVT-III scores of students in the DTT nor the control group changed significantly over the course of the academic year. Given the small number of participants and the large variance in cognitive and language scores within both groups these results were expected.
Looking more closely at the performance of individual students in the DTT and control groups (see Table 3) we see that in the control group the WPPSI-III VIQ, PIQ and FSIQ of students 7, 8, 9, and 12 remained extremely stable over the course of the academic year, with standard score differences ranging from 0 to 8 points. Student 10 showed a standard score decrease of 14 points in VIQ, 22 in PIQ, and 21 in FSIQ, while student 11 showed a standard score increase of 25 points in PIQ, and 16 in FSIQ. PPVT-III scores also remained largely consistent throughout the year as the standard scores of students 9, 10, 11, and 12 showed increases of 0 to 5 points, while student 7 showed a 10 point increase and student 8 received a standard score of 57 at post-test after obtaining the minimum score of 40 at baseline.
In the DTT group, the WPPSI-III VIQ, PIQ, and FSIQ standard scores of students 2, 3, and 5 remained stable from baseline to post-intervention, with standard score differences ranging from 2 to 7 points. Student 1 showed a standard score decrease of 13 points in VIQ, 19 points in PIQ, and 15 points in FSIQ. On the other hand, student 4 showed a standard score increase of 14 in PIQ and 14 in FSIQ, while student 6 showed a standard score increase of 14 in VIQ, 18 in PIQ, and 13 in FSIQ. Post-intervention PPVT-III standard scores in the DTT group remained within 8 points of baseline performance for students 1, 2, 3, 4, and 5, while student 6 showed a 10-point increase post-intervention.
Several analyses were conducted examining the students' adaptive behavior in the areas of communication, socialization, daily living skills, and motor skills as measured by the Vineland Adaptive Behavior Scales (VABS). Table 4 presents each student's standard scores on the VABS, and Table 5 presents the development rate for each student at baseline and over the course of the intervention. Following procedures recommended by Snyder-McLean (1987) each student's baseline development rate was calculated by dividing their age-equivalent score on each domain of the VABS by the student's chronological age. The development rate during intervention was calculated by subtracting the baseline age-equivalent score from the post-intervention age equivalent score within each domain and dividing the result by the number of months between the baseline and post-intervention assessments (eight months).
Mann-Whitney U tests indicated that there were no significant differences in adaptive behavior standard scores between the DTT and control groups at baseline or post-intervention. However, Wilcoxon signed ranks tests indicated several significant changes in adaptive behavior from baseline to post-intervention for students in the DTT group. Specifically, students in the DTT group were rated by their caregivers as demonstrating significantly higher levels of adaptive behavior at post-test in the domains of daily living skills z = 2.00, p < .05 and socialization z = 2.21, p < .05, and marginally higher levels of adaptive behavior in the domain of communication z = 1.90, p = .06. Additionally, students in the DTT group were rated by their caregivers as showing significantly higher levels of overall adaptive behavior (VABS-adaptive behavior composite) z = 2.20, p < .05 at post-test. There were no significant changes in adaptive behavior scores from baseline to post-test for the children in the control group.
Regarding adaptive behavior development rate, between-group analyses indicated that the rate of development both before and during intervention was not significantly different for the DTT and control groups. Within-group analyses indicated that for children in the DTT group the development rate during intervention was significantly higher than the baseline rate of development in the domains of daily living skills z = 2.20, p < .05, and socialization z = 2.20, p < .05. There were no significant differences in baseline rate of development and development rate during intervention for students in the control group.
Examining baseline and intervention rates of development for individual students, we see that the adaptive behavior of four out of six students in the DTT group developed at least somewhat faster during intervention than at baseline across all four domains (communication, daily living skills, socialization, and motor skills). One student in the DTT group showed a higher development rate during intervention in all domains except motor skills, while the remaining student showed a higher development rate during intervention in socialization, a lower rate in communication and motor skills, and a virtually equivalent rate in daily living skills.
Individual examination of baseline and intervention development rates of students in the control group indicated that only one student developed at least somewhat faster during intervention than at baseline across all four adaptive behavior domains. Two students showed a higher development rate during intervention in three out of four domains, one student demonstrated a higher intervention development rate in two domains with a lower rate in two domains, and one student showed a higher intervention development rate in one domain and lower rate in three domains. The remaining control group student showed a decrease in development rate across all four domains during the intervention.
Social-Emotional and Behavioral Functioning
Table 6 summarizes baseline and post-test scores for the DTT and control groups on the Social Skills Rating Scale (SSRS), Behavior Assessment System for Children (BASC), Behavioral Development Questionnaire (BDQ), and Emotion Understanding Test (EUT).
Social Skills Rating Scale (SSRS)
Mann-Whitney U tests indicated that there were no significant differences between students in the DTT group and the control group in caregiver ratings of social skills or problem behaviors at baseline or post-intervention as measured by the SSRS. However, within-group analyses indicated that students in the DTT group were rated by their caregivers as having marginally higher levels of social skills z = 1.78, p = .08 and problem behaviors z = 1.83, p = .07 at post-test compared to baseline. There were no significant differences between baseline and post-test social skills or problem behavior scores for students in the control group.
Behavior Assessment System for Children (BASC)
Analyses indicated that there were no significant differences between caregiver ratings of student behavior at baseline on the BASC. At post-test students in the DTT group had marginally higher adaptive skills composite T-scores z = 1.77, p = .08 than students in the control group. Within-group analyses indicated that students in the DTT group were rated by their caregivers as having significantly higher social skills T-scores z = 2.02, p < .05, anxiety T-scores z = 2.22, p < .05, and adaptive skills composite T-scores z = 2.20, p < .05, as well as marginally lower hyperactivity T-scores z = 1.68, p = .09 at post-test compared to baseline. Students in the control group were rated by their caregivers as having marginally lower hyperactivity T-scores z = 1.78, p = .08 at post-test compared to baseline.
Behavioral Development Questionnaire (BDQ)
Mann-Whitney U tests indicated no significant differences in BDQ scores between students in the DTT and control groups at baseline or post-test. Within-group analyses indicated that students in the DTT group were rated by their caregivers as exhibiting significantly higher levels of typically developing social behavior z = 2.20, p < .05 and lower levels of active-but-odd behavior z = 2.00, p < .05 at post-test compared to baseline. There were no significant changes in BDQ scores from baseline to post-test for the students in the control group.
Emotional Understanding Test (EUT)
There were no significant differences in EUT scores between the students in the DTT group and control group at baseline or post-test. Within-group analyses also indicated that there were no significant changes from baseline to post-test for students in either the DTT or control group. Although, as can be seen in Table 6 the mean scores of the DTT group increased (at non-significant levels) in emotion identification, emotional perspective-taking and total emotional understanding, while the mean scores of the control group did not show any increase across those three areas.
Previous research has demonstrated that early intensive DTT can lead to significant developmental gains across domains for children with autism and developmental delay (Downs & Smith, 2004; Lovaas, 1987; McEachin et al., 1993). However, few studies have evaluated the effectiveness of providing DTT to children with developmental disabilities other than autism. While only an initial investigation, the present study indicates that DTT has potential to be used practically and effectively within existing public EI/ECSE programs to positively impact the development of students with disabilities.
This study was limited by the small and extremely diverse sample of students that yielded low statistical power and limited the generalizability of the results. Also, the use of uncorrected analyses increased the probability of obtaining significant findings by chance. However, given the sample size and variance, and the exploratory nature of the study, uncorrected analyses were judged to be appropriate. Another potential limitation is the possibility that gains reported by parents of students in the intervention group were influenced by expectations that their child would show improvements in response to receiving an individualized intervention. The potential for such bias, although present, likely was reduced because all of the students in the study, including those in the control group, received individualized interventions and one-to-one attention throughout the school year.
Even with the low statistical power, several interesting results were found. Students in the DTT group demonstrated significant gains in adaptive behavior over the course of the academic year in the areas of communication, daily living skills, socialization, and overall adaptive behavior. As a group, students in the control group made no such gains. A significant criticism of DTT use with children with autism has been that skills that are learned often do not generalize to other settings (T. Smith, 2001). While not a direct measure of generalization, our results suggest that the skills learned in a small amount of DTT intervention time (1.30-1.58 hours/week) may have been utilized by students to more effectively function both independently and with others in their daily lives in a wide variety of settings, including home and the community. This was true for students with a wide range of disabilities and baseline levels of functioning. In fact, the only student in the DTT group who did not make gains in adaptive behavior was a student with cerebral palsy who had physical limitations that had pervasive effects on his ability to exhibit adaptive behavior across domains and settings.
In the domain of cognitive development we found that the IQ scores of students in the DTT group were not significantly different from those in the control group at baseline or post-intervention. Individual examination of IQ scores and student characteristics revealed that of the three children in the DTT group whose IQ scores changed very little over the course of the intervention, two had autism and cognitive delay and one had significant cognitive and speech delay. Considering that previous research analyzed by the National Research Council (2001) suggests that for significant cognitive gains to occur, children with autism should receive a minimum of 25 hours per week of direct services, these results are not surprising, and they underscore the need for publicly funded EI/SCSE programs to provide a much higher number of hours of service to students with autism if significant gains in cognitive development are to occur. The results also indicate the possibility that students with significant cognitive delay (i.e., IQ below 70) may not make large gains in cognitive development as a result of DTT interventions. Alternatively, such students may require a higher number of hours of intervention in order to realize significant gains. Future studies should thoroughly investigate those possibilities. The student in the DTT group whose IQ scores decreased over the course of the year had physical limitations which significantly affected his ability to respond to test items, thus the results are not thought to be an accurate measure of his level of cognitive functioning.
The two students in the DTT group who made relatively large gains in IQ scores were diagnosed with speech and cognitive delay, and social/adaptive delay respectively and were higher functioning across developmental domains at baseline. Those two students also made the most progress and mastered the most items within the DTT programs. In the past DTT has been used primarily with students with more severe disabilities and developmental delay (Lovaas, 1987; McEachin et al., 1993; Smith, Klevstrand, & Lovaas, 1995). Although the results should be interpreted cautiously, they suggest the possibility that a relatively small amount of DTT (1.5 or less hours/week) may effectively accelerate the cognitive development of students who exhibit less severe delay at baseline. Additional research is needed to further investigate that possibility.
Turning our attention to the students' behavioral and social-emotional functioning we also found some interesting results. Students in the DTT group were rated by their caregivers as exhibiting higher levels of social skills, adaptability, typically developing social behavior, and lower levels of active-but-odd social behavior at post-test compared to baseline. The only positive behavioral change noted by caregivers of children in the control group was a marginally significant decrease in hyperactive behavior, which also was reported by caregivers of the children in the DTT group.
These results again suggest the possibility that the skills learned through DTT facilitated improvements in the students' functioning when interacting with peers and adults. A potential drawback to DTT is that students are not interacting with peers during the intervention, thus their social-emotional development could feasibly suffer. Greenspan (1992) has argued that behavior analytic approaches to intervention such as DDT hinder the social-emotional and behavioral development of young children with developmental disabilities. However, our results indicate that the opposite may be true, as students receiving DTT made gains in social behavior that were not seen in children in the control group.
Another potential drawback to using DTT with young students is that some may experience frustration and/or anxiety as they are pushed by instructors to gain new skills and knowledge that may be difficult for them to learn. Indeed, we found that students in the DTT group were rated by their caregivers as exhibiting higher levels of problem behaviors and anxiety post-intervention, while children in the control group did not display any increase in behavioral or emotional difficulties. Although none of the students in the DTT group were exhibiting emotional or behavioral problems in the clinically significant range, these results suggest the possibility that the relatively intense nature of DTT instruction could lead to emotional and/or behavioral difficulties for some students. Alternatively, due to the preschool teacher's request that DTT not be provided in the regular classroom, students in the DTT group transitioned in and out of the classroom several times per day. It is plausible that the large number of transitions may have contributed to the increase in anxiety and problem behaviors in the DTT group as observations indicated that students in the DTT group exhibited more emotional and behavioral difficulties during transitions into and out of the classroom than they did in either the regular classroom or DTT settings.
Relying heavily on positive reinforcement, DTT is designed to be enjoyable for students. When implementing DTT great care should be taken to ensure that students experience high levels of success and reinforcement in order to help maintain positive emotional and behavioral health. Because DTT requires strict instructor control of stimuli, prompts, and reinforcers it is critical that attention be paid to providing students the opportunity to exert control whenever doing so is practical. For example, if at all possible, student selected or natural reinforcers should be used for each DTT session. Some students may benefit from selecting the program they would like to work on next, or the setting in which they would like to work. Additionally, it is important for DTT educators to provide frequent and regular child-directed play breaks where few demands are placed on the student, and the student can access reinforcers independent of their instructor. By doing so students learn that their behaviors can lead to desired outcomes outside the presence of educators and educator controlled discriminative stimuli.
Considering the difficulty students appeared to have with transitions in this study, it may be that DTT may most appropriately be provided within the regular preschool environment, thus reducing the number of transitions students need to make and increasing the amount of time students are engaged in both DTT and their regular preschool programming. Perhaps ideally EI/ECSE educators could be trained to embed DTT instruction within regular preschool activities, with future research evaluating the effectiveness of that delivery model. However it is utilized, it is critical that DTT instruction be accompanied by adequate training, supervision, and individual data collection to ensure that students are receiving this research-based intervention in the manner in which it has been proven effective (see Lovaas, 2003 for suggested training and supervision criteria). In addition, DTT should not be used as the sole instructional method with any student. Instead DTT should be used in combination with other instructional methods such as incidental teaching to help build skills not addressed by DTT and to generalize and reinforce those skills that are learned in DTT.
It is interesting to note that the defining elements of DTT are inherent to many forms of instruction already regularly used by educators (e.g., incidental teaching, model-lead-test). Incidental teaching essentially employs DTT procedures in the context of a student's typical activities. For example, if a student is learning to expressively label colors an educator may place desirable toys of various colors within reach of the student. When the student reaches for a toy the educator presents a discriminative stimulus by asking "What color is it?". The educator then waits for a response, and if the student responds correctly, the student is permitted to access the toy (i.e., receives reinforcement). If a correct response is not given, the educator prompts the child as necessary until a correct response is shaped. In model-lead-test instruction the "model" consists of presentation of an instruction (i.e., a discriminative stimulus) followed by presentation of the desired response (i.e., a full prompt). The "lead" consists of instructing the student to perform the desired response (i.e., discriminative stimulus) along with the teacher (i.e., a partial prompt). The "test" entails providing the instruction (i.e., discriminative stimulus), waiting for a response and then telling the student they are correct (i.e., reinforcement) or helping them give the correct response again (i.e., prompting). Thus, although some may view DTT as an unusual instructional approach that is specific to autism intervention, variants of DTT procedures are used within many other forms of instruction commonly found in classrooms today. DTT simply encourages using such procedures in a systematic and carefully structured manner to maximize learning opportunities for the student.
One issue that should be considered by educators seeking to implement effective instructional strategies within their classrooms is whether instruction should target basic skills or more general learning competencies. For example, the Competent Learner Model (Tucci, Hursh, Laitinen, & Lambe, 2005) focuses on teaching students higher-order learning competencies such as problem solving, listening, and observing, while attempting to increase the student's desire to learn. In doing so, students are thought to become more competent and motivated learners who subsequently develop their skills and knowledge across domains. DTT, on the other hand more typically focuses on teaching students basic skills, while also attempting to increase the students' desire to learn. In DTT the targeted skills are carefully sequenced in a way that is thought to lead to the development of higher-order competencies, thus facilitating further skill and knowledge acquisition (Lovaas, 2003). In the future it will be important to evaluate the extent to which skill building, competence building, or combined approaches lead to desired educational outcomes for children with different diagnostic and developmental profiles.
There is a tremendous need to identify specific instructional methods that significantly improve the developmental and educational outcomes of students with disabilities enrolled in EI/ECSE programs. This problem is a significant one considering the large number of students identified with developmental disabilities that receive publicly funded special education services throughout their school years. Many students and educators could benefit from having access to a simple and specific instructional approach designed specifically to help build skills and accelerate development in students with developmental delay. The present results suggest that even small amounts of DTT (i.e., less than 1.5 hours per week) may lead to significant developmental gains across domains and tangible improvements in school readiness for children with various developmental disabilities. Additionally, EI/ECSE educators with no previous experience in DTT can be trained to effectively implement DTT as one part of existing preschool programs.
Although the results of the present study are somewhat promising, additional larger scale studies are needed to replicate the findings and to analyze how DTT and other specific instructional methods may most effectively be implemented with students displaying specific diagnostic and developmental profiles. This analysis should include an examination of the level of service required for students with differing characteristics to realize significant gains across developmental domains. Future studies should also evaluate the long-term outcomes of students with developmental delay and/or disabilities who receive DTT or other specific instructional services in the preschool years and the long-term cost-effectiveness of such instruction.
The authors thank the families and children who participated in the study, as well as the many research assistants who provided instructional services and collected data. This study was partially supported by the Faculty Research Fund of Central Washington University, Ellensburg, Washington.
Bricker, D. (1995). The challenge of inclusion. Journal of Early Intervention, 19, 179-194.
Castelloe, P., & Dawson, G. (1991). Subclassification of children with autism and pervasive developmental disorder: A questionnaire based on Wing's subgrouping scheme. Journal of Autism and Developmental Disorders, 23(2), 229-241.
Coe, D., Matson, J., Fee, V., Manikam, R., & Lanarello, C. (1990). Training nonverbal and verbal play skills to mentally retarded and autistic children. Journal of Autism and Developmental Disorders, 20, 177-187.
Downs, A., & Smith, T. (2004). Emotional understanding, cooperation, and social behavior in high-functioning children with autism. Journal of Autism and Developmental Disorders, 14(6), 625-635.
Downs, A., & Strand, P. S. (in press). Emotion understanding in English- and Spanish-speaking preschoolers enrolled in Head Start. Social Development.
Dunn, L. M., & Dunn, L. M. (1997). Peabody Picture Vocabulary test-Third Edition. Circle Pines, MN: American Guidance Service, Inc.
Engelmann, S., & Osborn, J. (1970). DISTAR: An instructional system. Chicago: Science Research Associates.
Green, G. (1996). Early behavioral intervention for autism: What does research tell us? In C. J. Kestenbaum & D. T. Williams (Eds.), Handbook of clinical assessment of children and adolescents (pp. 29-44). New York: New York University Press.
Greenspan, S. (1992). Reconsidering the diagnosis and treatment of very young children with autistic spectrum or pervasive developmental disorders. Zero to Three, 13, 1-9.
Greenspan, S., & Wieder, S. (1999). A functional developmental approach to autism spectrum disorders. Journal of the Association for Persons with Severe Handicaps, 24, 147-161.
Gresham, F. M., & Elliot, S. N. (1990). Social Skills Rating System. Circle Pines, MN: American Guidance System, Inc.
Howard, J. S., Sparkman, C. R., Cohen, H. G., Green, G., & Stanislaw, H. (2005). A comparison of intensive behavior analytic and eclectic treatments for young children with autism. Research in Developmental Disabilities, 26(4), 359-383.
Howlin, P. A. (1981). The effectiveness of operant language training with autistic children. Journal of Autism and Developmental Disorders, 21, 281-290.
Howlin, P. A., Baron-Cohen, S., & Hadwin, J. (1999). Teaching children with autism to mind-read: A practical guide. West Sussex: Wiley.
Jacobson, J. W., Mulnick, J. A., & Green, G. (1998). Cost-benefit estimates for early intensive behavioral intervention for young children with autism: General model and single state case. Behavioral Interventions, 13, 201-226.
Kimball, J. W. (2002). Behavior analytic instruction for children with autism: Philosophy matters. Focus on Autism and Developmental Disabilities, 17(2), 66-75.
Klinger, L. G., & Dawson, G. (1997). Autistic disorder. In E. J. Mash & R. A. Barkley (Eds.), Child psychopathology (pp. 315-339). New York: Guilford.
Koegel, R. L., Russo, D. C., & Rincover, A. (1977). Assessing and training teachers in the generalized use of behavior modification. Journal of Applied Behavior Analysis, 10, 197-205.
Krantz, P. J., & McClannahan, L. E. (1981). Teaching complex language to autistic children. Analyses and Intervention in Developmental Disabilities, 1, 259-297.
Leaf, R., & McEachin, J. (1999). Behavioral strategies for teaching and improving behavior of autistic children. New York: DRL Books.
Losardo, A., & Bricker, D. (1994). Activity-based intervention and direct instruction: A comparison study. American Journal on Mental Retardation, 98(6), 744-765.
Lovaas, O. I. (1977). The autistic child: Language training through behavior modification. New York: Irvington.
Lovaas, O. I. (1987). Behavioral treatment and normal educational and intellectual functioning in young autistic children. Journal of Consulting and Clinical Psychology, 55, 3-9.
Lovaas, O. I. (2003). Teaching individuals with developmental delays: Basic intervention techniques. Austin, TX: PRO-ED.
McEachin, J., Smith, T., & Lovaas, O. I. (1993). Long-term outcome for children with autism who received early intensive behavioral treatment. American Journal on Mental Retardation, 97, 359-372.
National Research Council (2001). Educating students with autism. Committee on Educational Interventions for Children with Autism. Catherine Lord and James P. McGee, eds. Division of Behavioral and Social Sciences and Education. Washington, DC: National Academy Press.
O'Brien, S. K. (1996). The validity and reliability of the Wings subgroups questionnaire. Journal of Autism and Developmental Disorders, 26(3), 321-335.
Reynolds, C. R., & Kamphaus, R. W. (1992). Behavior assessment system for children. Circle Pines, MN: American Guidance Service, Inc.
Risley, T., Hart, B., & Doke, I. (1972). Operant language development: The outline of a therapeutic technology. In R. L. Schiefelbusch (Ed.), Language of the mentally retarded (pp. 107-123). Baltimore: University Park Press.
Sandall, S., Hemmter, M. L., Smith, B. J., & McLean, M. E. (2005). DEC recommended practices: A comprehensive guide. Longmont, CO: Sopris West.
Sandall, S., McLean, M. E., & Smith, B. J. (2000). DEC recommended practices in early intervention/early childhood special education. Longmont, CO: Sopris West.
Schepis, M. N., Reid, D. H., Ownbey, J., & Parsons, M. B. (2001). Training support staff to embed teaching within natural routines of young children with disabilities in an inclusive preschool. Journal of Applied Behavior Analysis, 34(3), 313-327.
Smith, B. J., Strain, P. S., Snyder, P., McLean, M. E., Ramsey, A. B., & Sumi, C. (2002). DEC recommended practices: A review of 9 years of EI/ECSE research literature. Journal of Early Intervention, 25(2), 108-119.
Smith, T. (1999). Outcome of early intervention for children with autism. Clinical Psychology: Research and Practice, 6, 33-49.
Smith, T. (2001). Discrete trial training in the treatment of autism. Focus on Autism and Developmental Disabilities, 16(2), 86-92.
Smith, T., Klevstrand, M., & Lovaas, O. I. (1995). Behavioral treatment in Rett's disorder: Ineffectiveness in three cases. American Journal on Mental Retardation, 100(3), 317-322.
Snow, C. E., Burns, S. M., & Griffin, P. (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press.
Snyder-McLean, L. (1987). Reporting norm-referenced program evaluation data: Some considerations. Journal of the Division of Early Childhood, 11, 254-264.
Sparrow, S., Balla, D., & Cicchetti, D. (1984). Vineland Adaptive Behavior Scales (VABS). Circle Pines, MN: American Guidance Service.
Tucci, V., Hursh, D., Laitinen, R., & Lambe, A. (2005) Competent learner model for individuals with autism/PDD. Exceptionality, 13(1), 55-63.
Wechsler, D. (2002). WPPSI-III: Wechsler Preschool and Primary Scale of Iintelligence (3rd ed.). San Antonio, TX: The Psychological Corporation.
Young, J. M., Krantz, P. J., McClannahan, L. E., & Poulson, C. L. (1994). Generalized imitation and response-class formation in children with autism. Journal of Applied Behavior Analysis, 27, 685-698.
Andrew Downs, Robyn Conley Downs, Michael Johansen, and Michelle Fossum
Central Washington: University
Correspondence to Dr. Andrew Downs, Department of Psychology, Central Washington University, 400 E. University Way, Ellensburg, WA 98926-7575; e-mail: email@example.com. Fax: 509-963-2307.
Table 1 Age, Gender, and Diagnoses of Participants Age in Participant Group Months Gender Diagnosis 1 DDT 46 Male Cerebral Palsy 2 DTT 44 Female Communication/Cognitive/Adaptive Delay 3 DTT 48 Male Autism 4 DTT 49 Male Cognitive Delay 5 DTT 45 Female Autism 6 DTT 61 Male Social/Adaptive Delay 7 Control 60 Male Social/Adaptive Delay 8 Control 63 Male Communication/Cognitive/Adaptive Delay 9 Control 40 Female Autism 10 Control 47 Female Social/Adaptive Delay 11 Control 32 Male Communication/Cognitive Delay 12 Control 43 Female Cognitive Delay Table 2 Hours of DTT and Number of Items Mastered by Students in the DTT Group Total DTT DTT Hours Receptive Expressive Participant Hours Per Week Language Language Imitation Total 1 31.92 1.35 59 0 17 73 2 35.10 1.30 80 11 33 124 3 41.10 1.58 17 0 9 26 4 33.75 1.47 20 107 41 168 5 42.65 1.58 15 8 10 33 6 30.30 1.52 24 134 50 208 Table 3 Student Standard Scores on the WPPSI-III and PPVT-III Pre- and Post-Intervention WWPSI-III Participant Verbal IQ Performance IQ and Group Pre Post Pre Post 1 DTT 77 64 82 63 2 DTT 74 70 60 65 3 DTT 53 58 49 47 4 DTT 77 72 72 96 5 DTT 49 53 51 49 6 DTT 85 91 96 114 All DTT 69.16 68.00 68.33 72.33 7 Control 85 83 73 75 8 Control 51 51 51 47 9 Control 55 53 48 49 10 Control 116 102 123 101 11 Control 55 55 51 76 12 Control 61 61 57 49 All Control 70.50 67.50 67.17 66.17 WWPSI-III Participant Full-Scale IQ PPVT-III and Group Pre Post Pre Post 1 DTT 76 61 59 53 2 DTT 62 64 59 55 3 DTT 49 46 40 40 4 DTT 71 85 88 96 5 DTT 42 49 40 40 6 DTT 89 102 84 94 All DTT 64.83 67.83 61.67 63.00 7 Control 75 76 87 97 8 Control 46 43 40 57 9 Control 43 49 40 40 10 Control 122 101 117 122 11 Control 44 60 40 40 12 Control 52 53 57 57 All Control 63.67 63.67 63.50 68.83 Table 4 Student Standard Scores on the Vineland Adaptive Behavior Scales (VABS) at Baseline and Post-Intervention Communication Daily Living Socialization Group Pre Post Pre Post Pre Post DTT 55 54 47 53 64 77 DTT 60 70 64 81 70 75 DTT 51 52 55 58 52 60 DTT 90 92 87 100 94 102 DTT 55 60 60 59 61 65 DTT 74 76 78 91 68 90 All DTT 64.17 67.33 (a) 65.17 73.67* 68.17 78.17* Control 50 77 57 63 70 70 Control 52 52 50 53 60 66 Control 50 47 54 44 51 51 Control 92 84 54 79 70 86 Control 62 52 58 54 64 54 Control 65 62 59 55 69 64 All Control 61.83 62.33 55.33 58.00 64.00 65.17 Motor Composite Group Pre Post Pre Post DTT 37 38 47 51 DTT 60 72 58 69 DTT 65 67 51 54 DTT 105 96 91 96 DTT 60 59 54 56 DTT 57 104 64 87 All DTT 64.00 72.67 60.83 68.83* Control 57 81 56 67 Control 54 73 50 56 Control 44 40 46 42 Control 66 71 65 74 Control 66 57 57 50 Control 53 55 56 54 All Control 56.67 62.83 55.00 57.17 *p<.05, (a) p = .06 Table 5 Adaptive Skills Developmental Rate at Baseline and During Intervention VABS VABS VABS VABS Participant Communication Daily Living Socialization Motor and Group Pre During Pre During Pre During Pre During 1 DTT .41 -.13 .37 .38 .46 1.50 .33 .13 2 DTT .41 1.38 .55 2.00 .50 1.25 .61 1.25 3 DTT .27 .63 .40 1.00 .19 1.38 .67 .88 4 DTT .84 1.13 .82 2.25 .90 1.88 1.04 .38 5 DTT .33 .78 .44 .63 .36 .75 .60 .63 6 DTT .64 1.25 .75 2.00 .49 3.50 .59 4.38 All DTT .48 .84 .55 1.38* .48 1.71* .64 1.28 7 control .45 2.63 .48 1.25 .53 .50 .60 2.25 8 control .33 .50 .38 1.00 .35 1.50 .54 2.13 9 control .15 .25 .35 -.25 .15 .13 .28 .38 10 control .85 .50 .38 2.88 .49 2.38 .66 1.00 11 control .41 -.38 .41 .13 .41 -.38 .63 .25 12 control .44 .38 .44 .13 .49 .00 .47 .75 All control .44 .65 .41 .86 .40 .69 .53 1.12 *p<.05 Table 6 Mean Student Scores on the SSRS, BASC, BDQ and EUT DTT Group Measure Pre M(sd) Post M(sd) SSRS: Social skills 65.67(13.4) 73.00 (b)(11.3) Problem behaviors 94.17(8.1) 103.67 (a)(10.6) BASC: Externalizing problems 48.00(7.7) 45.33(4.9) Internalizing problems 41.00(8.4) 43.33(9.2) Behavioral symptoms 49.17(6.5) 47.00(6.4) Adaptive skills 32.50(8.1) 41.00*(4.7) BDQ: Typically developing 29.00(6.7) 35.33*(10.4) Passive 29.33(6.0) 31.00(4.9) Active but odd 24.50(4.9) 20.00*(7.3) Aloof 26.00(9.1) 27.33(15.8) EUT: Identification 3.33(3.8) 4.33(4.1) Perspective-taking 1.67(3.2) 3.83(5.6) Total understanding 5.00(6.1) 8.17(8.9) Control Group Measure Pre M(sd) Post M(sd) SSRS: Social skills 67.17(12.3) 72.83(17.0) Problem behaviors 100.17(9.1) 103.33(11.2) BASC: Externalizing problems 52.67(16.6) 50.17(12.2) Internalizing problems 46.17(12.1) 44.17(10.4) Behavioral symptoms 54.00(17.6) 50.17(13.3) Adaptive skills 34.33(6.1) 32.17(8.5) BDQ: Typically developing 35.83(9.2) 34.33(17.9) Passive 32.33(12.8) 27.83(14.1) Active but odd 20.67(11.9) 26.67(11.5) Aloof 33.67(8.2) 28.67(11.0) EUT: Identification 3.17(3.8) 3.17(3.9) Perspective-taking 2.50(3.6) 2.00(3.6) Total understanding 5.67(7.3) 5.17(7.2) *p<.05, (a) p = .07, (b) p = .08