pretend play skills for kids with autism

USING VIDEO MODELING TO TEACH RECIPROCAL PRETEND PLAYTO CHILDREN WITH AUTISM

REBECCA MACDONALD AND SHELLY SACRAMONE

NEW ENGLAND CENTER FOR CHILDREN

NORTHEASTERN UNIVERSITY

RENEE MANSFIELD AND KRISTINE WILTZ


AND

WILLIAM H. AHEARN


NORTHEASTERN UNIVERSITY

The purpose of the present study was to use video modeling to teach children with autism toengage in reciprocal pretend play with typically developing peers. Scripted play scenariosinvolving various verbalizations and play actions with adults as models were videotaped. Twochildren with autism were each paired with a typically developing child, and a multiple-probedesign across three play sets was used to evaluate the effects of the video modeling procedure.Results indicated that both children with autism and the typically developing peers acquired thesequences of scripted verbalizations and play actions quickly and maintained this performanceduring follow-up probes. In addition, probes indicated an increase in the mean number ofunscripted verbalizations as well as reciprocal verbal interactions and cooperative play. Thesefindings are discussed as they relate to the development of reciprocal pretend-play repertoires inyoung children with autism.

DESCRIPTORS: autism, pretend play, video modeling

_______________________________________________________________________________

Impairments in reciprocal pretend play arewell documented in children with a diagnosis ofautism spectrum disorder ( Jarrold, 2003; Lifter,2000). The play of children with autism ischaracterized by repetitive behaviors and a lackof symbolic or social quality. Deficits inspontaneous language, imitation, and socialinteractions in general may be critical variablesthat impede the development of play. The lackof the development of reciprocal pretend playmay also be due to the social consequences notbeing as reinforcing for children with autismrelative to their typically developing peers. Thisinsensitivity to social stimuli could directly

produce deficits in social behavior (Dube,MacDonald, Holcomb, Mansfield, & Ahearn,2004; Taylor, Hoch, Potter, Rodriguez, &Kalaigian, 2005).

A variety of behavioral teaching procedureshave been examined to teach play skills tochildren with autism (e.g., Stahmer, Ingersoll,& Carter, 2003). Pivotal response training hasbeen shown to be effective in establishingsociodramatic play skills with adults as playpartners (Stahmer, 1995; Thorp, Stahmer, &Schreibman, 1995). In these studies, childrenacquired play skills, and these skills generalizedacross toys and adults. Pierce and Shreibman(1997) found that using peer-implementedpivotal response training resulted in increasesin the social behavior between children withautism and their typically developing peers.Odom, Hoyson, Jamieson, and Strain (1985)

Address correspondence to Rebecca P. F. MacDonald,New England Center for Children, 33 Turnpike Road,Southborough, Massachusetts 01772 (e-mail: [email protected]).

doi: 10.1901/jaba.2009.42-43

JOURNAL OF APPLIED BEHAVIOR ANALYSIS 2009, 42, 43–55 NUMBER 1 (SPRING 2009)

43

used peer confederates to teach social respond-ing in the context of play and found that playwith peers increased in the context of adults.Odom and Watts (1991) suggested that peer-mediated social reinforcers are necessary tomaintain peer social interactions and requirefurther study.

Modeling using play scripts also has beenshown to be effective in establishing cooperativeplay with peers. Goldstein and Cisar (1992) usedsociodramatic scripts with 3 children, 2 typicallydeveloping children and 1 child with autism,across three toys. They found that subsequent totraining, all children showed increases in theme-related social behavior during play. Jahr, Eldevik,and Eikeseth (2000) compared modeling tomodeling plus verbal rehearsal to teach sustainedcooperative play and found the latter procedureto be more effective in establishing cooperativeplay, both with the training partner and acrossnovel settings and play partners.

Video modeling is a procedure that has beenused to teach a variety of skills to children withautism effectively (Charlop & Milstein, 1989;Charlop-Christy, Le, & Freeman, 2000; Har-ing, Kennedy, Adams, & Pitts-Conway, 1987;LeBlanc et al., 2003; Reeve, Reeve, Townsend,& Poulson, 2007; Sherer et al., 2001; Shipley-Benamou, Lutzker, & Taubman, 2002). Videomodeling typically involves presenting a video-taped sample of models engaged in a specificseries of scripted actions or verbalizations. Thevideotaped model is shown two or three timesand then the individual is provided with anopportunity to perform the scripted behaviorsobserved on the video. Video modeling canproduce more rapid acquisition and greatergeneralization than in vivo modeling (Charlop-Christy et al., 2000), and a number of studieshave shown that prompting (other than thevideo) and reinforcement were not necessary foracquisition to occur (Charlop-Christy et al.,2000; D’Ateno, Mangiapanello, & Taylor,2003; MacDonald, Clark, Garrigan, & Van-gala, 2005).

Video modeling has been used to increaseplay statements during play with siblings(Taylor, Levin, & Jasper, 1999) and to teachchildren to initiate play with others (Nikopou-los & Keenan, 2003, 2004). In addition,Nikopoulos and Keenan have shown that videomodeling was effective in extending length ofplay between adults and children with autism.D’Ateno et al. (2003) found that they couldestablish complex pretend-play sequences withindividual children with autism. They taughtchildren to play with three play sets (e.g., a teaset), incorporating both verbalizations and playactions using toys (e.g., drinking from ateacup). MacDonald et al. (2005) used videomodeling to teach children similar pretend-playsequences, using figurines or characters in theplay set that required the child to talk and actfor figurines. Both of these studies found rapidacquisition of these complex sequences of play.

Video modeling has been shown to result inrapid acquisition of long chains of solitary playin children with autism without the use ofresponse prompting and reinforcement. Thepurpose of the present study was to assess theeffects of video modeling to teach children withautism to engage in long sequences of reciprocalpretend play with typically developing peers.The targeted play included reciprocal verbalinteraction chains and cooperative play acrossthree sets of toys.

METHOD

Participants

Two pairs of children participated in thisstudy. Each pair consisted of a child with autismand a typically developing child. Both childrenwith autism were enrolled in a preschool thatprovided early intensive behavioral interventionthat offered individualized programming 5 daysper week, 6 hr per day, using the principles ofapplied behavior analysis. A portion of theschool day was spent in an on-site integratedpreschool classroom that consisted of 15 typi-cally developing children.

44 REBECCA MACDONALD et al.

The first participant, Colin, was a 7-year-oldboy with a diagnosis of autism. At the startof this study, he had received 27 months ofintensive intervention. He communicated usingfull sentences but required prompting to initiaterequests. He demonstrated two-step imitationand picture-to-object matching. Video model-ing had been used to teach Colin solitary play,across six play activities, during the year prior tothe study. Prior exposure to video modeling didnot include the videos used in the current study.Colin was integrated into a local kindergartenclassroom 5 mornings a week with the supportof a therapist and performed at the kindergartengrade level on academic skills. He spent hisafternoons in the specialized preschool pro-gram. Colin’s partner, Helen, was a 5-year-oldtypically developing girl who was enrolled in theintegrated preschool classroom.

The second participant, Alden, was a 5-year-old boy who also had a diagnosis of autism. Hehad received 16 months of intensive interven-tion when he began the study. He haddeveloped a repertoire of play skills with anumber of toys, but his play was solitary. Hecommunicated using full sentences, which heused to describe events in his environment andto make requests. He had well-developingimitation and matching skills. Video modelingalso had been used with Alden to teach solitaryplay, both in the classroom and at home withhis mother. He had learned to play successfullywith over a dozen toys using video modeling.Prior exposure to video modeling did notinclude the videos used in the current study.Alden was integrated into a preschool classroomfor 1 hr a day, with the support of a therapist.His partner, Gerry, was a 5-year-old typicallydeveloping boy who was enrolled in theintegrated classroom.

Both typically developing peers were selectedto participate as peer tutors because theydemonstrated many of the peer-selection crite-ria mentioned by Taylor (2001). They werecooperative in following adult instructions,

assertive in play sessions, socially competent,and capable of keeping an interest in activities.Both peers had been taught previously to serveas peer tutors for children with autism (but notthose in this study). Peer tutoring occurredinformally across the school year and includedprompts from teachers to share materials or tooffer play directives. They had no priorexperience with video modeling.

Setting

Baseline and video modeling sessions wereconducted in a small testing room to control forvariables such as noise and visual distracterspresent in the child’s classroom. The roomcontained two child-sized tables, two child-sizedchairs, and one toy shelf with various toys on it.The materials necessary for the play activitybeing trained were placed in an open area onthe floor. The video model was presented on atelevision with a VCR, which was placed on atable. One session per participant was conduct-ed daily. All sessions were videotaped for laterscoring. The video camera was held by theexperimenter at a distance of approximately1.5 m from the children.

Materials

Three play sets were used to teach reciprocalpretend play. The play sets included a Fisher-Price Little People airport, a Fisher-Price LittlePeople Animal Sounds zoo, and a Playskoolgrill. Scripts were developed for each play set.Each play set had a base structure (airport, zoo,or grill) and seven characters or objectspertaining to the theme of the structure thatwere used in the video modeling scripts. Allthree play sets were capable of making soundsand movements; however, these features weredisabled in the present study.

The airport base structure had an airportterminal with a snack bar and an attachedrunway with a place to get gas and park theairplane. The characters included a boy passen-ger and a pilot. The additional objects includeda taxi, a transport vehicle, a suitcase, an

RECIPROCAL PRETEND PLAY 45

umbrella, and an airplane. The zoo base was theanimal house. The characters included a boy, azookeeper, and three animals (a bird, a polarbear, and a lion). The objects included a truckand a bucket of fish. The grill consisted of a toygas grill with a lid and places for utensils. It alsohad knobs used to start and adjust the heat onthe grill. The additional objects included abottle of ketchup, a spatula, a fork, a cheese-burger, a hot dog and bun, and two paperplates.

Videos

Two adults were videotaped acting out thesequence of pretend play. Prior research hasshown that adults are easy to train and veryeffective as models to teach play to children(MacDonald et al., 2005). Two male modelswere used for the pair of male children (Aldenand Gerry), and the models for the boy-and-girlpair (Colin and Helen) were a male and afemale. The models were matched for genderbecause we believed this would increase thelikelihood of imitation.

Each video was related to a designated playactivity and contained 14 to 17 scriptedverbalizations and 14 to 17 actions associatedwith the play activities. These scripts were basedon observations of typically developing childrenplaying with these toys. Typical peers werevideotaped playing with toys and an observertranscribed their verbalizations and actions.Samples from these transcripts were used tocreate the scripts for each toy. (Copies of thecomplete scripts can be obtained from the firstauthor.) Scripts were recorded on a VHS-Ccamcorder and shown on a portable TV with a13-in. screen and a VCR player. The TV/VCRwas placed on a table separate from the area inwhich the children played with the toy set.

Airport and zoo. Adults on the videomanipulated the characters, spoke for thecharacters, and manipulated the materialsthrough the characters (doll as agent; Lifter,2000). For example, in the airport script thetypically developing child’s character says ‘‘First

we need gas’’ and puts the nozzle in the gas tankof the plane, then the child with autism’scharacter says ‘‘I’ll get my suitcase’’ and puts thesuitcase in the compartment under the plane. Inthe zoo script, the typically developing child’scharacter says ‘‘Can I feed him?’’ in the presenceof the polar bear, and the child with autism’scharacter says ‘‘Sure, he loves fish.’’ Then thetypically developing child’s character throwsfish into the polar bear’s den.

Grill. Adults on the video manipulated thematerials and spoke to each other (child asagent; Lifter, 2000). For example, the childwith autism says ‘‘Let’s check our food,’’ andthe typically developing child opens the grill lidand says ‘‘I think it’s ready.’’ They both useutensils to take food off the grill.

Independent Variable

The presentation of video models thatdepicted scripted play interactions for the threesets of play materials was the independentvariable. No reinforcement or prompting wasprovided to the participants with autism.

Dependent Measures

All sessions were videotaped and scored laterfor the occurrence of the following responses:(a) scripted verbalizations, (b) scripted playactions, (c) unscripted verbalizations, (d) un-scripted play actions, (e) cooperative play, and(f ) reciprocal verbal interaction chains.

Data were scored from videotapes of all ofthe 4-min play sessions (baseline, training, andprobes). Data for scripted behaviors wererecorded for each child on his or her part ofthe script. For example, in the airport script thechild with autism took the role of the passenger,and the typically developing child took the roleof the pilot. Data were recorded on the numberof scripted actions and verbalizations only fortheir assigned character. The number of scriptedactions and verbalizations ranged from 14 to 17for each child, depending on the script. Aseparate coding system was developed to scoreunscripted behaviors. Unscripted behaviors


were measured during baseline and probesessions only to assess changes in unscriptedplay as a result of video modeling.

To assess cooperative play and reciprocalverbal interaction chains, samples of thesebehaviors were scored from videotaped sessions.These behaviors were measured during the first90 s of the two initial baseline sessions and twomastery probe sessions. A real-time measure-ment method (Miltenberger, Rapp, & Long,1999), which required second-by-second re-cording, was used to assess percentage ofintervals of cooperative play and duration(number of seconds) of reciprocal verbalinteraction chains. Mean duration of reciprocalchains was calculated by adding the number ofconsecutive seconds (during each interactionchain) in which each child was talking to theother and dividing by the total number ofreciprocal verbal interaction chains.

Scripted verbalizations. Scripted verbalizationswere defined as vocal statements that matchedthe statement of the video model. In addition,statements that were similar to the modeledresponse but not identical also were scored. Thisincluded the substitution or omission of a word.Repetitions of prior verbalizations used in thesame session were not scored as scripted. Thenumber of scripted verbalizations ranged be-tween 14 to 17, depending on the script. Eachchild was scored on the occurrence of scriptedverbalizations for his or her own portion of thescript.

Scripted play actions. Scripted actions weredefined as motor responses that matched theactions of the video model and resulted in thesame change in the environment as seen in themodel. If the child engaged in the completescripted sequence, a play action was scored. Forexample, getting into the airplane was scoredonly if the character was put in the airplane andleft there, or feeding fish to the bear at the zoowas counted only if the character was used tothrow the fish to the bear. The number ofscripted actions ranged between 14 and 17,

depending on the script. Each child was scoredon the occurrence of scripted play actions forhis or her own portion of the script.

Unscripted verbalizations. Novel unscriptedverbalizations included verbalizations that werenot modeled in the video script but wereappropriate to the context of the toy (e.g.,talking about arriving at grandma’s house in theairport script or talking for animals that do nottalk in the script). These behaviors were scoredusing a data sheet that had a variety of possibletypes of vocalizations that were not part of thescript as well as a column for recording otherunscripted vocalizations that the child emittedbut that were not listed on the data sheet. Forexample, in the airport play set the script endedwith the plane flying away, but if the childcontinued to play with the plane by landing itand said ‘‘Here we are at grandma’s house,’’ thiswas scored as an unscripted verbalization.Unscripted verbalizations were scored onlyduring baseline and mastery probe sessionsbecause the goal was to assess changes inunscripted play as a result of the videomodeling.

Unscripted play actions. Novel unscripted playincluded a play action that was not modeled inthe video script but that was appropriate to thecontext of the toy. For example, again using theairplane play set, landing the plane and gettingout of the plane were scored as unscriptedactions. These behaviors were again scoredusing a data sheet similar to the one describedfor unscripted verbalizations.

Cooperative play. The percentage of intervalsof cooperative play were measured duringbaseline and mastery probes. Cooperative playwas defined as the child being within 0.33 m ofthe peer and engaged in the same activity ininterdependent or shared play. Interdependentplay included handing materials to the peer(e.g., giving the peer ketchup when preparingfood that was cooked on the grill or having thecharacters ride in the same vehicle). It alsoincluded participating in the same activity and


talking about the same scripted or unscriptedtopic (e.g., each child cooking a different foodon the same grill and talking about what theywere cooking).

Reciprocal verbal interaction chain. The dura-tion of reciprocal verbal interaction chains wasmeasured during baseline and mastery probes.The chain was defined as a sequence of two ormore verbalizations between a participant andpeer. A chain began when 1 child respondedwith a contextually appropriate verbalization tothe other child’s statement or request and endedwhen no verbalizations occurred for 2 s. Anexample of a verbal interaction chain was 1child saying, ‘‘May I have the ketchup?’’ and thepeer responding by saying ‘‘sure,’’ and the 1stchild then saying ‘‘Thanks, this hamburger isgood.’’ These interaction chains includedscripted and unscripted verbalizations.

Experimental Design

A multiple-probe design across play sets wasused to assess the effects of the video modelingintervention on reciprocal pretend play. The playsets were taught in the same order for each pair ofchildren starting with the airport, then the zoo,and finally the grill. Both children in each pairwere required to perform at or above masterylevels on scripted verbalizations and scripted playactions before training could begin on the nextplay set. Baseline sessions were conducted priorto the introduction of training for each play set,and mastery probes were conducted subsequentto acquisition of each new play set.

Procedure

Baseline. During baseline sessions, the playsets were arranged on the floor prior to the pairof children entering the room. For the airportand zoo play sets, the characters were positionedto the left and right of the play area. When thechildren arrived, they were prompted to sit onthe floor in front of the play set. The child withautism was always directed to sit on the left andthe typical peer to sit on the right, because thesepositions coincided with the positions of the

characters they would be assuming during videomodeling training. Once the children wereseated, the experimenter gave the instruction‘‘It’s time to play.’’ The children were allowed4 min to play with the toys. An adult stood justbehind the children and did not give anyadditional instructions or physical cues toensure that the toy rather than the adultcontrolled reciprocal play, while the experi-menter video taped each session. If a child spoketo the experimenter or tried to leave the areabefore the end of 4 min, the experimenter said‘‘Play with your toys’’ and directed the childback to the play set.

Video viewing. During video viewing sessions,the materials were set up in the same manner asbaseline, and a TV/VCR was set up on a tablein the corner of the room with two chairs infront of it. The children entered the room andwere directed to sit in the chairs to watch thevideo; again the child with autism was seated onthe left and the typically developing child wasseated on the right. The airport and zoo scriptedroles were determined by the specific charactereach child was assigned to play with, and thechildren were positioned in front of thosecharacters in the context of both the videoand play set. The position of the child alonedefined his or her role for the grill script. Theywatched the video model twice and then weredirected immediately to the play materials andtold ‘‘It’s time to play.’’ As in baseline, thechildren were allowed 4 min to play with thetoys while the adult stood behind them. Noprompts or reinforcement was delivered duringthese sessions. Video viewing sessions continueduntil the child with autism demonstratedmastery level performance on all actions andverbalizations for his portion of the cooperativeplay script. Mastery level was defined asaccuracy on 13 of the 15 actions and 12 ofthe 14 verbalizations for the airport, 12 of the14 actions and 13 of the 17 verbalizations forthe zoo, and 13 out of the 16 actions and 12 ofthe 14 verbalizations for the grill.


Additional coaching for typical peers. Becauseboth typical peers did not initiate the scriptviewed on the video, additional coaching wasneeded. Coaching was provided to Helen priorto Sessions 3 and 4 for the airport script, andfor Gerry coaching was provided prior toSession 5 for the airport and Session 4 for thegrill. The experimenter encouraged them to sayand do everything the model said and did in thevideo for their character or role. They also wereencouraged to ‘‘talk a lot.’’ No additionalprompting was given during the training orprobe sessions.

Mastery probes. Once mastery criteria weremet in the context of video modeling teachingsessions, mastery probes were conducted with-out the video. During these probes, the childrenwere presented with the toy set, instructed to sitto the left or right of the materials, and giventhe instruction to play. If the child with autismdemonstrated an accuracy level of 13 of the 15actions and 12 of the 14 verbalizations for theairport, 12 of the 14 actions and 13 of the 17verbalizations for the zoo, and 13 of the 16actions and 12 of the 14 verbalizations for thegrill for two consecutive sessions without thevideo model, they met mastery criteria for theplay set and training could begin on a new playset.

Follow-up probes. Follow-up probes wereconducted to determine maintenance over timewithout access to the toys or videos. Theseprobes were introduced approximately 1 monthfollowing mastery of the airport and zoo. Threefollow-up probes were conducted on theairport. They were assessed after the zoo wasmastered and then again after the grill wasmastered. Follow-up probes (one for Colin andHelen, two for Alden and Gerry) wereconducted on the zoo after the grill wasmastered. These probe sessions were identicalto baseline and mastery probe sessions.

Interobserver Agreement

Interobserver agreement was calculated in45% of sessions for scripted verbalizations and

actions. Interobserver agreement for play ac-tions was 95% for the airport, 95% for the zoo,and 92% for the grill. Agreement for verbali-zations was 96% for the airport, 96% for thezoo, and 96% for the grill. Agreement data forunscripted behaviors were collected in 33% ofthe sessions. Agreements for unscripted playactions were scored if both observers recordedthe same play action, and disagreements werescored if the observers recorded different playactions. The same procedures were used to scoreagreements and disagreements for unscriptedverbalizations. Agreement for unscripted playactions was 100% for the airport, 91% for thezoo, and 91% for the grill. Agreement forunscripted verbalizations was 96% for theairport, 100% for the zoo, and 98% for thegrill. Agreement data for cooperative play andreciprocal verbal interaction chains were col-lected in 83% of the sessions. Agreements werescored if both observers scored the occurrenceor nonoccurrence of cooperative play and verbalinteraction chains within each 1-s observationinterval. Interobserver agreement was calculatedby dividing the number of intervals of agree-ments by the number of intervals of agreementsplus disagreements, and this ratio was convertedto a percentage. Agreement for cooperative playwas 95%, and agreement for reciprocal verbalinteraction chains was 95% across play sets.

RESULTS

Scripted behaviors for the first pair ofchildren are shown in the left (Colin) and right(Helen) panels of Figure 1. Colin exhibited lowlevels of scripted verbalizations (M 5 0.33 persession; range, 0 to 1) and scripted actions (M5 4.67 per session; range, 4 to 5) duringbaseline for the airport; these increased to 14verbalizations (of 14) and 13.5 actions (range,13 to 14) per session on mastery probes. Heexhibited low levels of scripted verbalizations(M 5 0) and scripted actions (M 5 0.8 persession; range, 0 to 2) during baseline for thezoo; these increased to 16 verbalizations (of 17)


and 12 actions (of 14) per session on masteryprobes. He exhibited low levels of scriptedverbalizations (M 5 0.43 per session; range, 0to 2) and scripted actions (M 5 5.67 per

session; range, 2 to 8) during baseline for thegrill; these increased to 12.5 verbalizations (of14) and 15.5 actions (range, 15 to 16) persession on mastery probes. Helen showed

Figure 1. Number of scripted verbalizations and scripted actions demonstrated by Colin and Helen with the airport,zoo, and grill.


similar increases in scripted verbalizations andactions subsequent to video modeling training.

Scripted behaviors for the second pair ofchildren are shown in the left (Alden) and right(Gerry) panels of Figure 2. Alden exhibited lowlevels of scripted verbalizations (M 5 0) andscripted actions (M 5 3.5 per session; range, 3to 4) during baseline for the airport; theseincreased to 12 verbalizations (range, 10 to 14)and 13.6 actions (range, 13 to 15) per sessionon mastery probes. He exhibited low levels ofscripted verbalizations (M 5 0.4 per session;range, 0 to 1) and scripted actions (M 5 2.4 persession; range, 1 to 4) during baseline for thezoo; these increased to 16 verbalizations (range,15 to 17) and 13.5 actions (range, 13 to 14) persession on mastery probes. He exhibited lowlevels of scripted verbalizations (M 5 0.33 persession; range, 0 to 1) and scripted actions (M5 3.5 per session; range, 1 to 5) during baselinefor the grill; these increased to 9.5 verbalizations(range, 8 to 11) and 13.3 actions (range, 12 to15) per session on mastery probes. Gerryshowed similar increases in scripted verbaliza-tions and actions.

Across play sets Colin exhibited a mean of1.5 (grill), 2.5 (zoo), and 3.5 (airport) unscript-ed verbalizations and a mean of 2 to 3unscripted actions (data not shown). Aftervideo modeling, he showed a mean of 2.5(grill) and 4 (airport and zoo) unscriptedverbalizations and a mean of 1 to 3 unscriptedactions across play sets. During baseline, Aldenexhibited few verbalizations across play sets. Heexhibited a mean of 1 to 2.5 unscripted actionsacross play sets. After video modeling, heshowed increases in unscripted verbalizationsto a mean of 7.5 (zoo) and 5.5 (grill). His playactions were primarily scripted during masteryprobes.

Cooperative play (data not shown) for Colinoccurred in 17%, 0.06%, and 15% of intervalsduring baseline for the airport, zoo, and grill,respectively, and increased to 87%, 85%, and90% of intervals on mastery probes for the

airport, zoo, and grill, respectively. Cooperativeplay for Alden occurred in 6%, 0.5%, and 3.5%of intervals during baseline for the airport, zoo,and grill, respectively, and increased to 78%,74.5%, and 67% of intervals on mastery probesfor the airport, zoo, and grill, respectively.

The mean numbers of reciprocal verbalinteraction chains (data not shown) for Colinduring baseline were 0 for the airport and zooand 0.5 for the grill; these increased to a meanof 5 for the airport and 6 for the zoo and grillon mastery probes. The mean durations ofreciprocal verbal interaction chains for Colin inbaseline were 0 s for the airport and zoo and1.5 s for the grill; these increased to a mean of7.5 s, 10 s, and 7 s for the airport, zoo, andgrill, respectively, on mastery probes. The meannumber of reciprocal verbal interaction chainsfor Alden during baseline was 0 for all threeplay sets; these increased to a mean of 3, 7, and4.3 for the airport, zoo, and grill, respectively,on mastery probes. The mean durations ofreciprocal verbal interaction chains for Alden inbaseline was 0 s for all three play sets; theseincreased to 20 s, 8.5 s, and 8.75 s on masteryprobes for the grill.

DISCUSSION

In this study, video modeling producedextended sequences of reciprocal pretend playbetween children with autism and typicallydeveloping peers across three commerciallyavailable play sets. Prior to video instructionthere was little appropriate play between thechildren. This is particularly interesting becausethe typical children had extensive play skills intheir classroom with other typical children.They engaged in pretend play with a variety oftoys and used language in their play, but in thepresence of children with autism, these behav-iors were not evident. Both pairs of childrenexhibited rapid acquisition of verbalizations andplay actions, and this performance was main-tained over time. These acquired chains of playincluded up to 17 actions with verbalizations


embedded into the play scenarios. In addition,these play chains were acquired without the useof response prompting other than the video,and there was no experimenter-delivered rein-forcement.

An analysis of unscripted play revealed slightdifferences between children. Although Colinprimarily acquired the scripted behaviors andrelatively few unscripted behaviors, Aldenshowed increases in unscripted verbalizations

Figure 2. Number of scripted verbalizations and scripted actions demonstrated by Alden and Gerry with the airport,zoo, and grill.


with two of the toys following training. Thiscould have occurred because Alden had a moreextensive history of learning pretend play usingvideo modeling both at home and at school. Ingeneral, the children engaged in more novelverbalizations than actions across scripts. Theemergence of any unscripted play, whetherverbalization or action, was encouraging becausechildren with autism tend to have very limitedrepertoires of play; thus, increasing play in anymanner is significant. Acquiring play skills mayalso make it more likely that natural socialconsequences for interaction will come to exertan influence on the behavior of the children withautism. That is, social consequences that wereineffective may come to have some value forthese children through this training.

Qualitative changes in play, as measured bycooperative and reciprocal verbal interactionchains, were particularly interesting. Theyindicate that the children did not exhibitcooperative play prior to the video modelingprocedure. Following the introduction of videomodeling, the children engaged in cooperativeplay throughout the play sessions. In addition,the children were engaged in reciprocal verbalinteractions either by talking for the charactersor speaking for themselves (as in the activityinvolving the grill). This increase in verbalinteractions and cooperative play was achievedwithout prompting (other than the video) andwithout extrinsic reinforcement. It may be thatthe typically developing child’s scripted verbal-izations and play actions served as discrimina-tive stimuli, thereby occasioning the response ofthe child with autism. Nonetheless, these socialinteractions potentially set the occasion for thechild with autism to contact the naturalcommunity of social reinforcers (Baer & Wolf,1970).

Determining the operant mechanism thatproduced the observed change in behavior inthis study was not the purpose of this research.Although video modeling is often referred to asa procedure that does not require explicit

prompting or extrinsic reinforcement (e.g.,Charlop-Christy, Le, & Daneshvar, 2003;Taylor et al., 1999), it must be noted that thepresentation of the video itself is an explicitprompt. Children with autism often do notlearn typical play skills through the socialsituations they encounter in the natural envi-ronment. One possible reason for the effective-ness of video modeling is that it provides adiscrete opportunity to observe play without themany distractions often present in the naturalenvironment. Those children who readilyimitate observed actions can acquire a repertoirepresented on a videotape when provided thematerials and an opportunity to play. This maybe most likely to occur in children with ahistory of being taught to imitate the actions ofothers. Moreover, if imitation has been regu-larly reinforced, this history of reinforcementmay foster generalization to the videotaped skill(see Nikopoulos & Keenan, 2007). Thus, videomodeling can be an effective prompting strategythat capitalizes on a history of extrinsicreinforcement for imitating others, with littlereliance on physical prompting.

A limitation of this study was the lack ofextended novel play. However, the increase incooperative play was significant. These childrendid not engage in any cooperative play prior tovideo modeling and, subsequent to training,engaged in high levels of cooperative play thatincluded verbal reciprocal interactions. Anotherlimitation of the study was the lack ofgeneralization to other natural play settingsand different play materials using the samethemes. An additional limitation was thepotential that the presence of an adult behindthe child could have established stimuluscontrol and therefore exerted some level ofcontrol over responding. Lastly, future researchshould address the issue of generalization ofthese newly acquired play repertoires to novelsettings, peers, and play materials.

In this study, video modeling was an effectiveand efficient strategy for teaching sequences of


cooperative play. Given the opportunity toobserve videos of social interactions in thecontext of play, these children engaged inreciprocal play interactions with typically de-veloping peers. This represents an importantqualitative change in their play behavior. It ispromising that these interactive play skills wereachieved with relatively short exposure totraining and in the absence of responseprompting and reinforcement.

REFERENCES

Baer, D. M., & Wolf, M. M. (1970). The entry intonatural communities of reinforcement. In R. Ulrich,T. Stachnik, & J. Mabry (Eds.), Control of humanbehavior (Vol. 2, pp. 319–324). Glenview, IL: ScottForesman.

Charlop, M. H., & Milstein, J. P. (1989). Teachingautistic children conversational speech using videomodeling. Journal of Applied Behavior Analysis, 22,275–285.

Charlop-Christy, M. H., Le, L., & Daneshvar, S. (2003).Using video modeling to teach perspective taking tochildren with autism. Journal of Positive BehaviorInterventions, 5, 12–21.

Charlop-Christy, M. H., Le, L., & Freeman, K. A. (2000).A comparison of video modeling with in vivomodeling for teaching children with autism. Journalof Autism and Developmental Disorders, 30, 537–552.

D’Ateno, P., Mangiapanello, K., & Taylor, B. (2003).Using video modeling to teach complex playsequences to a preschooler with autism. Journal ofPositive Behavior Interventions, 5, 5–11.

Dube, W. V., MacDonald, R. P. F., Holcomb, W. L.,Mansfield, R. C., & Ahearn, W. H. (2004). Toward abehavioral analysis of joint attention. The BehaviorAnalyst, 27, 197–207.

Goldstein, H., & Cisar, C. L. (1992). Promotinginteraction during sociodramatic play: Teachingscripts to typical preschoolers and classmates withdisabilities. Journal of Applied Behavior Analysis, 25,265–280.

Haring, T. G., Kennedy, C. H., Adams, M. J., & Pitts-Conway, V. (1987). Teaching generalization ofpurchasing skills across community settings to autisticyouth using videotape modeling. Journal of AppliedBehavior Analysis, 19, 159–171.

Jahr, E., Eldevik, S., & Eikeseth, S. (2000). Teachingchildren with autism to initiate and sustain cooper-ative play. Research in Developmental Disabilities, 21,151–169.

Jarrold, C. (2003). A review of research into pretend playin autism. Autism, 7, 379–390.

LeBlanc, L., Coates, M., Daneshvar, S., Charlop-Christy,M., Morris, C., & Lancaster, B. (2003). Using videomodeling and reinforcement to teach perspective-taking skills to children with autism. Journal ofApplied Behavior Analysis, 36, 253–257.

Lifter, K. (2000). Linking assessment to intervention forchildren with developmental disabilities or at-risk fordevelopmental delay: The development play assess-ment (DPA) instrument. In K. Gitlin-Weiner, A.Sandgrund, & C. Schaefer (Eds.), Play diagnosis andassessment (2nd ed., pp. 228–261). New York: Wiley.

MacDonald, R., Clark, M., Garrigan, E., & Vangala, M.(2005). Increasing play using video modeling.Behavioral Interventions, 20, 225–238.

Miltenberger, R. G., Rapp, J. T., & Long, E. S. (1999). Alow-tech method for conducting real-time recording.Journal of Applied Behavior Analysis, 32, 119–120.

Nikopoulos, C. K., & Keenan, M. (2003). Promotingsocial initiations in children with autism using videomodeling. Behavioral Interventions, 18, 87–108.

Nikopoulos, C. K., & Keenan, M. (2004). Effects of videomodeling on social initiations by children withautism. Journal of Applied Behavior Analysis, 37,93–96.

Nikopoulos, C. K., & Keenan, M. (2007). Using videomodeling to teach complex social sequences tochildren with autism. Journal of Autism and Develop-mental Disorders, 36, 678–693.

Odom, S. L., Hoyson, M., Jamieson, B., & Strain, P. S.(1985). Increasing handicapped preschoolers’ peersocial interactions: Cross-setting and componentanalysis. Journal of Applied Behavior Analysis, 18,3–16.

Odom, S. L., & Watts, E. (1991). Reducing teacherprompts in peer-initiation interventions throughvisual feedback and correspondence training. Journalof Special Education, 25, 26–43.

Pierce, K., & Schreibman, L. (1997). Multiple peer use ofpivotal response training to increase social behaviorsof classmates with autism: Results from trained anduntrained peers. Journal of Applied Behavior Analysis,30, 157–160.

Reeve, S. A., Reeve, K. F., Townsend, D. B., & Poulson,C. L. (2007). Establishing a generalized repertoire ofhelping behavior in children with autism. Journal ofApplied Behavior Analysis, 40, 123–136.

Sherer, M., Pierce, K. L., Paredes, S., Kisacky, K. L.,Ingersoll, B., & Schreibman, L. (2001). Enhancingconversational skills in children with autism via videotechnology: Which is better, ‘‘self’’ or ‘‘other’’ as amodel? Behavior Modification, 25, 140–158.

Shipley-Benamou, R., Lutzker, J., & Taubman, M.(2002). Teaching daily living skills to children withautism through instructional video modeling. Journalof Positive Behavior Interventions, 4, 165–175.

Stahmer, A. C. (1995). Teaching symbolic play skills tochildren with autism using pivotal response training.Journal of Autism and Developmental Disorders, 25,123–141.


Stahmer, A. C., Ingersoll, B., & Carter, C. (2003).Behavioral approaches to promoting play. Autism, 7,401–413.

Taylor, B. A. (2001). Teaching peer social skills to childrenwith autism. In C. Maurice, G. Green, & R. M. Foxx(Eds.), Making a difference: Behavioral intervention forautism (pp. 83–96). Austin, TX: Pro-Ed.

Taylor, B. A., Hoch, H., Potter, B., Rodriguez, A., &Kalaigian, M. (2005). Manipulating establishingoperations to promote initiations toward peers inchildren with autism. Research in DevelopmentalDisabilities, 26, 385–92.

Taylor, B. A., Levin, L., & Jasper, S. (1999). Increasingplay related statements in children with autism towardtheir siblings: Effects of video-modeling. Journal ofDevelopmental and Physical Disabilities, 11, 253–264.

Thorp, D. M., Stahmer, A. C., & Schreibman, L. (1995).Teaching sociodramatic play to children with autismusing pivotal response training. Journal of Autism andDevelopmental Disorders, 25, 265–282.

Received September 14, 2006Final acceptance September 11, 2007Action Editor, Bridget Taylor


pretend play skills for kids with autism

Documents