Making a Writing Plan

Replacing_the_Echolalia_of_Chi.pdf

Home >English homework help >Article writing homework help >Making a Writing Plan

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

Replacing the Echolalia of Children With Autism With Functional Use of Verbal Labeling

Richard M. Foxx,1, 2 Kimberly A. Schreck,1 Jeffrey Garito,1

Angela Smith,1 and Shannon Weisenberger1

We evaluated the direct and generalized effects of cues-pause-point language training procedures on immediate echolalia and correct responding in two children with autism. The overall goal was to teach the children to remain quiet before, during, and briefly after the presentation of questions and then to verbalize on the basis of the cues (pictures) whose labels represented the correct responses. A multiple baseline design across question/response pairs demonstrated that echolalia was rapidly replaced by correct responding on the trained stimuli and there were clear improvements in one child’s respond- ing to untrained stimuli. These results replicate and extend previous research with adults with mental retardation and autism which demonstrated that cues- pause-point procedures can be effective in teaching individuals displaying echolalia to use their verbal labeling repertoires functionally.

KEY WORDS: echolalia; autism; language training; cues-pause-point; generalization.

Echolalia can be regarded as an inappropriate language strategy (Carr et al., 1975) which interferes with language training efforts even though it may be functional for some individuals (Prizant and Duchan, 1981; Schuler, 1979). Historically, behavioral attempts to decrease echolalic responding to questions have focused on either replacing echolalia with stimulus specific responses (Carr et al., 1975; Risley and Wolf, 1967) or a generalized verbal response (Schreibman and Carr, 1978) through the use of such procedures as imitation training, stimulus fading, differential reinforcement, and verbal prompts.

1Psychology Program, Penn State Harrisburg, Middletown, Pennsylvania. 2To whom correspondence should be addressed at Penn State Harrisburg, 777 W. Harrisburg Pike, Middletown, Pennsylvania 17057-4898; e-mail: [email protected].

307

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

308 Foxx, Schreck, Garito, Smith, and Weisenberger

A third treatment option takes advantage of individuals’ verbal label- ing skills in order to teach them a nonecholalic strategy (Foxx, 1999; Foxx et al., 1988a; Foxx and Faw, 1990; McMorrow et al., 1986). This training approach involves labeling relevant cues, a response delay requirement, at- tention training (Foxx, 1977), suppressing off task behavior, and provisions for the transfer of stimulus control from an object to the questioner and question.

The specific training was accomplished through the use of “cues-pause- point” procedures that taught individuals displaying echolalia to (a) remain quiet when the trainer held up his index finger before, during, and briefly after the presentation of targeted questions, and then (b) use a pretrained verbal label as the correct response when the trainer pointed to the ap- propriate environmental cue. Several studies (Foxx et al., 1988a; McMorrow et al., 1987; McMorrow and Foxx, 1986) demonstrated that the procedures re- placed echolalia with correct responses and that adults and adolescents with varying degrees of mental retardation used the trained responses: (a) when they were presented by individuals who had not been involved in training; (b) in settings where no training had been conducted; (c) when no prompts, cues, feedback, or consequences were used; and (d) sometimes several years following training (Foxx and Faw, 1990).

More importantly, the procedures produced generalized improvements in some individuals’ responding to untrained verbal stimuli by teaching them to use environmental cues (e.g., pictures) when they were asked questions. That is, they seemed to learn a set of nonverbal behaviors to use when they were presented with novel verbal input. These nonverbal behaviors seemed to be a by-product of the “cues-pause-point” procedural sequence, because over time trained individuals would: (a) remain silent when spoken to (i.e., attend to the stimulus presentation); (b) pause briefly after being spoken to; and (c) visually scan an array of physical cues which were in the environment before selecting one and verbalizing its label. This scanning or searching was facilitated by ensuring that the individuals could see the correct cue/object when the question that pertained to it was asked.

These nonverbal behaviors combined with the generalized improve- ments in question answering and reductions in echolalia suggested that the individuals learned a nonecholalic strategy over time wherein they would re- spond to questions on the basis of either the cue (e.g., pictures or objects) that were present or their established repertoires of labeling responses. In fact, individuals were much less likely to use their echolalic strategy even though their responses were not always correct because they used object labels from other settings as incorrect responses. Such incorrect responding appeared to represent a bridge between their echolalic strategy and following one where they visually searched their environment for the correct response.

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

Replacing Echolalia 309

Because of the successful application of cues-pause-point procedures with adult populations diagnosed with mental retardation and autism, it seemed reasonable to assume that the same procedures might benefit young children diagnosed with autism spectrum disorders. Accordingly, the present study was designed to replicate and extend previous research involving cues- pause-point procedures (e.g., Foxx et al., 1988a) by applying the procedures to young children diagnosed with autism who exhibit echolalia and evaluat- ing the direct and generalized effects of the procedures.

METHOD

Participants

Karl, a 5-year-old male, was diagnosed with autism. Psychometric test- ing by the authors supported this diagnosis (see Table I). Karl displayed immediate echolalia. He was able to use some brief phrases (i.e., 2–4 words) appropriately. Karl was enrolled in a nonstructured educational program and received occupational therapy.

Gerry, a 6-year-old boy, was diagnosed with autism by the second author approximately 6 months before the study (see Table I). Gerry frequently echoed phrases, questions, and object labels. During the study, Gerry re- ceived home- and school-based applied behavior analysis intervention.

Experimental Design

A three-leg multiple baseline across picture sets design was used.

Settings

Karl’s baseline and training trials were conducted at the University in a 2 m × 4 m room which contained a one-way vision mirror, a table, and

Table I. Participants’ Demographic Information and Psychometric Scores

Language Autism

Child Agea Receptiveb Expressivec Clinicald Parente

Karl 5 40 55 37 97 Gerry 5 40 58 40 83 a In years. bPeabody Picture Vocabulary Test—III (standard score). c Expressive One Word Vocabulary Test (standard score). d Childhood Autism Rating Scale Total Score (completed by researchers). e Gilliam Autism Rating Scale Autism Quotient (completed by parents).

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

310 Foxx, Schreck, Garito, Smith, and Weisenberger

chairs. The generalization sessions were conducted in a room across the hall. Gerry’s sessions were conducted in his home in a 4 m × 4 m room. His gen- eralization sessions were conducted in the family’s living room. Each setting contained a variety of irrelevant stimuli.

Target Behaviors and Recording

In each experiment, the first word or sequence of words that followed the initiation of a question was scored in one of three, mutually exclusive cat- egories; echolalia, incorrect, or correct. Echolalia was defined as a repetition of one or more of the words contained in the question regardless of whether other verbalizations occurred. An incorrect response was scored when a ver- balization contained a stimulus-irrelevant word(s), regardless of whether a correct response was included. A correct response was scored when a verbal- ization either matched the trained response or provided a different appropri- ate answer to the question. Using these definitions, a response that contained any combination of echolalic, correct, or incorrect verbiage was scored as an echo and a response that included any combination of correct and incorrect verbiage was scored incorrect. Responses were scored by the trainer imme- diately after each question was presented. If no response occurred, it was scored as incorrect. All sessions were either video- or audiotaped.

Reliability

Using randomly selected tapes, one of four independent raters tran- scribed and scored responses from at least two trials in each condition. To limit the possibility that their scoring would be influenced by the trainer’s feedback, the raters were instructed to stop the tapes immediately after each response (i.e. before the feedback was given). Interobserver agreement was computed by dividing agreements on the occurrence of each behavior by agreements plus disagreements, and multiplying by 100. Excluding the few taped verbalizations that could not be scored because they were inaudible, percent agreement on all target behaviors averaged 98% for Karl, and 90% for Gerry.

Trained Stimuli and Responses

In each experiment, stimulus (i.e. question) and response (i.e. picture card) pairs were developed by identifying pictures of objects or animals that the participants were likely to be able to label and creating a ques- tion that pertained to each. Question/response pairs were then arbitrarily separated into three sets. The trained responses (i.e. picture labels) were

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

Replacing Echolalia 311

one- to three-syllable words such as giraffe, gummy bears, and bed. Commer- cially available picture cards were used (Language Builder Picture Cards, Stages Learning Materials, Chico, CA). The questions were between four and six words long and never contained the response or part of a response. Examples included: “What animal has a long neck?” and “What do you sleep on?” The participant’s responses to the questions had not been assessed prior to the baseline conditions.

General Procedures

Response Identification Training

Prior to training the participants were taught to verbally label each picture card that would be used. Labeling training consisted of the trainer (a) displaying the picture cards from each set on a table; (b) verbally (“what’s this?”) and/or gesturally (i.e. pointing to or tapping the picture) prompting the participant to identify the objects; (c) providing feedback (i.e. “good an- swer” for a correct response, “no, that’s not right” for an incorrect response, a “no” for an echo, and “you didn’t answer” for no response); (d) saying the correct label when the participant failed to do so and then prompting them to label correctly; and (e) giving intermittent edible reinforcers for correct responses. This training continued until each picture was correctly labeled during three consecutive trials when the trainer simply pointed to it. Training typically progressed quickly because the participants already could label the majority of the pictures used.

Baseline

The baselines were conducted after the participants had been trained to label correctly. Nevertheless, each trial during conditions where pictures were available was started by labeling the pictures which were displayed. After this labeling, the trainer said, “I’m going to ask you some questions and I want you to answer them the best you can.” The trainer then asked each of the set questions in a random order and provided response-specific feedback following the participant’s verbalization. The feedback was the same as in the response identification training. Participants received verbal praise and an edible or sip of a favored drink for each correct response.

Cue-Pause-Point (Uncovered)

All of the baseline procedures remained in effect. During this condition the trainer prompted the participant to remain silent before, during, and

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

312 Foxx, Schreck, Garito, Smith, and Weisenberger

briefly after she presented the questions and then to label one of the pictures (i.e. correct response) that was present on a table in front of the child. This was done by having the trainer (a) hold up her right index finger at eye level midway between the participant and herself whenever silence was desired (i.e. during the instructions, questions, and for approximately 1 s following the question) and say “no” or “shh” whenever a verbalization occurred (i.e. the pause prompt); (b) move her finger so that it touched the correct cue (picture card) approximately 2 s after the question was completed (recall that at the end of response identification training each participant was responding correctly to the point prompt only); and (c) use the response identification training prompts (e.g. tapping the object, “what’s this?”) if necessary to ensure that the labeling response occurred.

Cue-Pause-Point (Covered)

In this condition all of the uncovered procedures remained in effect, except now the picture was covered with a folder. This was done by having the trainer (a) cover the picture with a folder immediately after the un- covered trial, pause prompt again, restate the question, and move her right index finger so that it touched the folder when a correct response was de- sired (i.e. the trainer prompted again and used the response identification prompts if a response did not occur even though the picture was covered; and (b) provide the same verbal feedback and consequences which were used in the uncovered condition for the first verbalization that occurred. Because the above training sequence relies heavily on manual prompts, verbal mediation was kept to a minimum.

Pause Only

During this condition, the pretrial labeling was discontinued, the pic- tures were removed from the training setting, and no point prompts were used. The trainer simply used the pause prompt as she presented the ques- tion and then withdrew her hand so that it was closed and in contact with her chest when a response was desired. The feedback and consequences were the same as before.

Baseline II

This condition was identical to the initial baseline except that the pictures were not present. In other words, the trainers simply asked the questions in a random order, and provided feedback and consequences as they always had.

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

Replacing Echolalia 313

Programming Generalization and Maintenance

This condition consisted of several phases. Initially, sessions were con- ducted in the original and new settings by the original and a new trainer. Next, all 15 questions were randomized and presented by the original and new trainers in the original setting. Finally, all feedback and consequences were faded by the original trainer. These trials are described in more detail below.

RESULTS

Karl

Figure 1 shows that Karl echoed in response to between 0 and 90% of the set one questions during baseline even though the picture cards were present and feedback was provided after each response. He answered between 0 and 38% correctly. During the first six set two trials, he echoed in response to between 40 and 90% of the questions and answered between 0 and 22% correctly whereas, in set three, he echoed between 0 and 80% while correct answering was low and only occurred in two trials. The addition of the pause and point prompts in the uncovered condition in set one decreased echoes to zero after three trials while correct responses increased to 100%. In sets two and three, echoes trended downward whereas correct responses continued upward.

In the set one covered condition, Karl only echoed in one trial and correct responding was 100% in all but one trial. In the set two, uncovered condition, Karl only echoed in two trials and correct responses increased to 100%. In set three, echoes reached zero and correct responses 100%.

During the pause only condition in set one, echoes were at zero dur- ing the last 10 trials and correct responding at 100% in 9 of the 10. During this period, in the set two covered condition, no echoes occurred and cor- rect responding was 100%. In set three, few echoes occurred and correct responding typically reached 100%.

In the set one baseline II condition, no echoes occurred and correct responding was 100%. In the set two, pause only condition, Karl echoed in trial 44 and correct responding averaged 84%. In set three, echoes remained at zero and correct responding at 100% except for trial 46.

Generalization tests A through F (trials 48–59) were conducted as fol- lows: (A) the original trainer T1 presented the question sets in a novel room; (B) T1 presented the sets in the original room; (C) T2, a new trainer, pre- sented the sets in the original setting; (D) T2 presented the sets in the novel setting; (E) T1 presented all 15 questions in a random order in the original

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

314 Foxx, Schreck, Garito, Smith, and Weisenberger

Fig. 1. The percentage of echolalic and correct responses on trained (sets one and two) and untrained (set three) questions. GEN A–F represent different generalization tests. MA represents the fading of feedback and reinforcement phase (see text).

setting; and (F) T2 presented all 15 questions in a random order in the original setting. In the maintenance condition (trials 60–65) T1 presented the questions randomly and faded the feedback and consequences. This was accomplished in a few trials by progressively reducing the number of

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

Replacing Echolalia 315

responses that were followed by feedback and consequences until they were eliminated. Figure 1 shows that Karl’s performance on these tests was 100% correct and zero echoing in all but four trials.

Gerry

Figure 2 shows that Gerry echoed between 20 and 70% of the set one questions during baseline. He only answered one question correctly. During the first six set two trials, his echoing ranged between 10 and 40% and no correct responding occurred, whereas in set three, he echoed between 0 and 44% and correctly answered between 0 and 25%.

In the uncovered set one condition (trials 7–16), echoes decreased to zero correct responding increased to 100%. In set two, echoing increased and Gerry only responded to one question correctly. Responding in set three showed little change. During the covered condition in set one (trials 17–21) echoes were zero and correct responding 100%. In the uncovered condition in set two, echoes decreased to 20% and correct responding reached 80%. In set three, echoes increased somewhat and correct responding showed little change.

Throughout the remaining training, generalization, and maintenance trials in set one, echoing never occurred and correct responding fell be- low 100% only four times. During the set two covered condition, echoes were zero except for trial 26 and correct responding ranged between 60 and 100%. Set three correct responding ranged between 11 and 33% and echoing between 11 and 22%.

In the pause only condition in set two, several problems were encoun- tered. First, Gerry’s performance decreased when the pictures were removed in this condition. To solve this problem, Gerry was briefly pulled out of the design between trials 33 and 34 and returned to the covered condition with a slight modification. Recall that in the typical covered condition, the pic- tures were presented first uncovered and then immediately covered. The modification consisted of chaining the pause only condition to the covered condition (uncovered–covered–pause only) and then fading the uncovered and covered conditions so that only the pause only condition remained. When Gerry responded correctly to all the pause only questions, he and the trainer “reentered the design.” Shortly thereafter another problem emerged when Gerry began to consistently responded to one question (“what goes back and forth?”/ “swing”) with another set two picture/label “backpack” (what do you carry books in?). We determined that the word “back” in the “swing” question was cuing Gerry to say “backpack.” We changed the ques- tion to “what do you sit on that goes up and down?” and taught him to say

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

316 Foxx, Schreck, Garito, Smith, and Weisenberger

Fig. 2. The percentage of echolalic and correct responses on three trained question sets. After trial 44, each data point represents the mean of two trials except for the last trial before each condition change (trials 57, 58, 64, and 68). GEN A through E represents different generalization tests and each data point is shown. Each data point also is shown for the MA (Maintenance) phase when reinforcement and feedback were faded.

“swing.” The final problem was that echoing was still occurring at a level of 20 to 40% (trials 47–56). In trial 54, we modified the feedback for echoes by saying “no” more firmly and slapping the table approximately 18 in. away

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

Replacing Echolalia 317

from Gerry. This change in strategy was effective and by trial 57 echoes were at zero and correct responding was 100%. At that point the baseline II condition was implemented. In the baseline II condition, correct respond- ing typically remained between 70 and 100% and echolalia between 0 and 20%. However, during the generalization and maintenance trials, correct responding was always 100% and echoes were at zero.

No problems were encountered in set three during training. Gerry’s echoing ranged between 0 to 25% and was absent during the generalization and maintenance trials whereas correct responding ranged between 80 and 100% and was 100% throughout the last 10 baseline II and all generalization and maintenance trials.

Generalization tests A through E were conducted as follows: (A) T1 presented the question sets in a new setting; (B) T2 presented the sets in the original setting; (C) T2 presented the sets in the new setting; (D) T1 presented all 15 questions in a random order in the original setting; and (E) T2 presented all 15 questions randomly. In the maintenance phase (MA), T1 returned and faded the feedback and consequences as described for Karl.

DISCUSSION

This study replicated the results of Foxx et al. (1987, 1988a); McMorrow et al. (1987); McMorrow and Foxx (1986) with young children diagnosed with autism spectrum disorders. The training program rapidly replaced the children’s echolalia with correct responses and these communicative im- provements were maintained during conditions where the prompts, cues, feedback, and consequences were eliminated. The only difference in the present study was that the uncovered and covered training conditions were implemented separately whereas in all of the previous research they were combined. This change was made because the young children in this study did not have the large labeling repertories that the older subjects in previ- ous research possessed. Whether or not this change was necessary will be determined in a future study with young children with autism.

The generalization to untrained stimuli differed between participants. For example, Karl’s improvements in set three began to occur within a few trials after training began on set one. Similar results occurred in set two prior to the institution of training on this set. Furthermore, Karl never echoed and usually responded correctly at 100% on set three questions after trial 26. Gerry’s performance varied considerably longer on the sets even though training was implemented on all three.

The present results also partially replicated McMorrow et al. (1987) in that Karl, who displayed generalized effects had levels of echolalia on the

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

318 Foxx, Schreck, Garito, Smith, and Weisenberger

untrained sets in the middle stages of the study that were at or near zero even though his responses were not always correct. Another aspect of this study that McMorrow et al. (1987) also confronted was the need to remove a participant from the design. Recall that Gerry’s set two pause only con- dition decrease in performance required that the pictures be reintroduced outside the design. In McMorrow et al. (1987) a subject was removed from the design temporarily because she began to respond “no” after virtually every question. The other problems encountered with Gerry in set two also were not unusual (see McMorrow et al., 1986).

In this study, the rationale for cues-pause-point language training was that trainers of language deficient children had to greatly increase the like- lihood that the children would respond appropriately to the trainers’ ver- balizations. Several methods were used during structured training. First, the trainer insured that the child had an appropriate verbal response to what- ever the trainer might say. That is, the child could say whatever the trainer wanted him to say. Second, the child was taught to verbalize the appropri- ate response whenever the trainer manually prompted him to do so thereby establishing control over when a particular verbalization would occur. In other words, before the trainer ever presented a verbal stimulus for which a particular response would be appropriate, the trainer had ensured that the child could produce the appropriate response quickly and whenever the trainer sought it. Third, the interactional situation was arranged so that a relevant physical cue was present. Fourth, when a verbal stimulus for which a particular response would be appropriate was presented, a set of manual prompting procedures were used which (a) increased the likelihood that the child would attend to the trainer’s verbal input; (b) reduced the likeli- hood that competing inappropriate responses would occur (i.e., echolalia); (c) directed the child’s attention to a physical cue (picture) in the interac- tional environment; and (d) indicated when the appropriate response should occur. Finally, once the appropriate response reliably occurred following a particular verbal stimulus, the cues and prompts that produced it were faded.

The trainer provided as much assistance as necessary in the beginning to greatly increase the likelihood that the desired response would occur and then gradually reduced the amount of assistance provided until the child was responding appropriately without help. This feature may be one of the primary differences between cues-pause-point training procedures and other types of language training. Although most other training procedures rely on shaping appropriate speech through a series of steps towards a tar- get response, cues-pause-point procedures are intended to make the child’s initial performances as errorless as possible. Thus, the primary goal is to establish the necessary conditions to produce appropriate verbal responses

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

Replacing Echolalia 319

quickly (i.e., the first time a particular verbal stimulus is presented) so that the major focus of the training can be on maintaining the student verbal behavior through a series of fading steps.

In conclusion, this study adds to the growing evidence that cues-pause- point procedures can affect maladaptive language strategies such as echolalia (McMorrow et al., 1987), perseverative (Foxx et al., 1988a), and delusional speech (Foxx et al., 1988c) as well as teach signing to students who are deaf (Foxx et al., 1988b). Future research should be conducted to determine whether the cues-pause-point procedures can be used to teach young chil- dren with autism to imitate the verbal responses of models (see McMorrow et al., 1986).

ACKNOWLEDGMENT

We thank Donna LeFevre for her assistance.

REFERENCES

Carr, E. G., Schreibman, L., and Lovaas, O. I. (1975). Control of echolalic speech in psychotic children. J. Abnorm. Child Psychol. 3: 331–351.

Foxx, R. M. (1977). Attention training: The use of overcorrection avoidance to increase the eye contact of autistic and retarded children. J. Appl. Behav. Anal. 10: 489–499.

Foxx, R. M. (1999). Long term maintenance of language and social skills. Behav. Interv. 14: 135–146.

Foxx, R. M., and Faw, G. D. (1990). Long-term follow-up of echolalia and question answering. J. Appl. Behav. Anal. 23: 387–396.

Foxx, R. M., Faw, G. D., McMorrow, M. J., Kyle, M. S., and Bittle, R. G. (1988a). Replacing maladaptive speech with verbal labeling responses: An analysis of generalized responding. J. Appl. Behav. Anal. 21: 411–417.

Foxx, R. M., Kyle, M. S., Faw, G. D., and Bittle, R. G. (1988b). Cues-pause-point training and simultaneous communication to teach the use of signed labeling repertoires. Am. J. Ment. Retard. 93: 305–311.

Foxx, R. M., McMorrow, M. J., Davis, L. A., and Bittle, R. G. (1988c). Replacing a chronic schizophrenic man’s delusional speech with stimulus appropriate responses. J. Behav. Ther. Exp. Psychiatry 19: 28–33.

Foxx, R. M., McMorrow, M. J., Faw, G. D., Kyle, M. S., and Bittle, R. G. (1987). Cues-pause- point language training: Structuring trainer statements to provide students with correct answers to questions. Behav. Residential Treat. 2: 103–115.

McMorrow, M. J., and Foxx, R. M. (1986). Some direct and generalized effects of replacing an autistic man’s echolalia with correct responses to questions. J. Appl. Behav. Anal. 19: 289–297.

McMorrow, M. J., Foxx, R. M., Faw, G. D., and Bittle, R. G. (1986). Looking for the Words: Teaching Functional Language Strategies, Research Press, Champaign, IL.

McMorrow, M. J., Foxx, R. M., Faw, G. D., and Bittle, R. G. (1987). Cues-pause-point language training: Teaching echolalics functional use of their verbal labeling repertoires. J. Appl. Behav. Anal. 20: 11–22.

Prizant, B. M., and Duchan, J. F. (1981). The functions of immediate echolalia in autistic children. J. Speech Hear. Disord. 46: 241–249.

P1: KHD

Journal of Developmental and Physical Disabilities [jodd] pp1401-jodd-NY00000688 January 7, 2005 10:36 Style file version June 18th, 2002

320 Foxx, Schreck, Garito, Smith, and Weisenberger

Risley, T., and Wolf, M. (1967). Establishing functional speech in echolalic children. Behav. Res. Theory 5: 73–88.

Schreibman, L., and Carr, E. G. (1978). Elimination of echolalic responding to questions through the training of a generalized verbal response. J. Appl. Behav. Anal. 11: 453–463.

Schuler, A. L. (1979). Echolalia: Issues and clinical applications. J. Speech Hear. Disord. 44: 411–434.

Reproduced with permission of copyright owner. Further reproduction prohibited without permission.