A Comparison of the Acquisition of Play Skills Using Instructor-Created Video

Models and Commercially Available Videos

A Thesis Presented

by

Gail Palechka

The Department of Counseling and Applied Educational Psychology

In partial fulfillment of the requirements

for the degree of

Master of Science

in the field of

Applied Behavioral Analysis

Northeastern University

Boston, MA

August 2009

NORTHEASTERN UNIVERSITY

Bouve College of Health Sciences Graduate School

Thesis Title: A Comparison of the Acquisition of Play Skills Using Instructor-Created Video Models and Commercially Available Videos Author: Gail Palechka Department: Counseling and Applied Educational Psychology Approved for Thesis Requirements of Master of Science Degree ______________________________________________________ __________ Rebecca P. F. MacDonald ______________________________________________________ __________ Susan Langer ______________________________________________________ __________ Chata Dickson

A Comparison of the Acquisition of Play Skills Using Instructor-Created Video Models and

Commercially Available Videos

by

Gail Palechka

B.S. Western Michigan University 2005

Submitted in partical fulfillment of the requirements for the degree of Master of Science in Applied Behavior Analysis

in the Bouve College of Health Sciences Graduate School of Northeastern University, August 2009

Acknowledgements

The author expresses gratitude to the New England Center for Children, for its dedication

to research and service to children with disabilities. Grateful recognition is also offered to the

faculty of the MABA program, for sharing their knowledge and experience.

Vast appreciation is extended to Becky MacDonald, the chairperson of the author’s thesis

committee, for her constant support throughout this process. Special thanks are also offered to

Sue Langer and Chata Dickson, members of the author’s thesis committee, for their time and

suggestions and to Lindsay Wilkinson and Cara Grieco for their assistance with data collection.

Table of Contents A Comparison of the Acquisition of Play Skills Using Instructor Created Video Models

and Commercially Available Videos A. Abstract .................................................................................................... 2 B. Introduction

1. Play in Children with Autism .................................................... 3 2. In Vivo Modeling to Teach Play ................................................. 4 3. Video Modeling to Teach Play ................................................... 4 4. Problem Statement and Experimental Question ......................... 8

C. Method 1. Participants .................................................................................... 8 2. Setting and Materials ..................................................................... 9 3. Dependent Variable and Operational Definition ........................... 10

4. Independent Variable .................................................................... 10 5. Experimental Design ..................................................................... 13

6. Procedures ..................................................................................... 13 D. Results ...................................................................................................... 15 E. Discussion ................................................................................................. 17 F. References ................................................................................................. 22 G. Tables ....................................................................................................... 26 H. Figure Captions ........................................................................................ 34 I. Figures ....................................................................................................... 36

Video Modeling 2

Abstract The purpose of the current study was to compare the rate of acquisition of play skills following the viewing of an instructor-created video model to the rate of acquisition of play skills following the viewing of a corresponding commercially available children’s video. The study included 3 children with autism who received educational and clinical

services in a preschool setting. Each participant was exposed to one video of each type and the number of actions and vocalizations was measured. Two children learned more rapidly using the instructor-created video format. Additionally, probe data were taken to further examine the participants’ attending to video and toys across the two video formats.

Video Modeling 3

A Comparison of the Acquisition of Play Skills Using Instructor Created Video Models

and Commercially Available Videos

Basic play skills emerge early in childhood development. Play is considered to

contribute to a child’s social and language development. Children’s play skills are

imitative of things they observe in their daily environment and are practice for skills later

on in life (Lifter, 2000). Many of these symbolic and pretend play skills are lacking in

children with autism spectrum disorders (ASD), which is a concern for their parents in

their further development.

Children with autism (ASD) often have deficits in the developmental area of play

skills. They typically display repetitive and immature play with little or no imitation or

pretend play with a lack of symbolic or social quality; and will play with toys for their

sensory qualities rather than to access the intended function of the toy (Weiss & Harris,

2001). Furthermore, children with ASD may spin or line up toys when playing, in

contrast to typically developing peers of the same age. Possible some variables that may

contribute to this problem of atypical pkay skills in children with autism include: the lack

of spontaneous language, social interactions, and imitation skills all typical for children

with ASD (Weiss & Harris, 2001). It is also possible that children with autism do not

attend to how others manipulate toys. Insensitivity to the behavior of others can produce

deficits in social behavior (Dube, MacDonald, Holcomb, Mansfield, & Ahearn, 2004;

Taylor et al., 2005) including play skills vital to forming social relationships with

typically developing peers.

Stahmer, Ingersoll, and Carter (2003) described many behavior-based teaching

procedures have been used to teach play skills to children with ASD. Among these were

Video Modeling 4

several methods which provide an imitative model. In vivo, or live modeling, has been

shown to be effective in teaching play skills to children with autism. Goldstien and Cisar

(1992) demonstrated the efficacy of in vivo modeling by using it to teach reciprocal

sociodramatic play scripts to children with autism. After observing their typically

developing peers, the learners were able to re-create the scripts with a partner. Training

students using in vivo modeling techniques can also lead to generalization of new

imitation skills to novel settings and instructors (Egel, Richman, & Koegel, 1981).

Additional benefits of in vivo modeling include the emergence of related novel responses

and increases in a child’s social and communication skills (Goldstein & Cisar, 1992).

However, limitations to in vivo modeling include the time and maintenance of training

the model (Taylor, 2001), the potential for stimulus overselectivity (Egel et al. (1981),

and the requirement that the therapist provides additional prompting for both the learner

and the model (Taylor, 2001).

A second type of model presentation described by Stahmer, Ingersoll, and Carter

(2003) is known as video modeling. Video modeling is very similar to in vivo modeling

in that it requires imitation by the learner, but they are different in that the model is

presented in a videotaped format. Video modeling is a procedure that has been shown to

teach a wide variety of skills to children with autism (Charlop & Milstein, 1989; Charlop-

Christy, Le, & Freeman, 2000; Haring, Kennedy, Adams, & Pitts-Conway, 1987;

LeBlanc and Coates., 2003; Sherer et al., 2001; Shipley-Benamou, Lutzker, & Taubman,

2002). A typical video model is a videotaped sample of a model engaging in a

predetermined script of actions and/or verbalizations. After viewing the videotaped

model from 1 to 3 times, the individual is provided with an opportunity to perform the

Video Modeling 5

script. When compared to in vivo modeling, video modeling has been shown to produce

more rapid acquisition and greater generalization of skills (Charlop-Christy et al., 2000).

Also, in some cases additional prompting and reinforcement have not been necessary for

script acquisition (Charlop-Christy et al., 2000.; D’Ateno, Mangiapanello, & Taylor,

2003; MacDonald, Clark, Garrigan, & Vangala, 2005). Charlop-Christy et al. (2000)

described a number of advantages of video modeling over in vivo modeling; with video

modeling, the teacher may present a repeated model, edit out distracting or irrelevant

stimuli, eliminate variation in the model, and spend less time modeling and money for

materials and staffing.

Recently, video modeling has become a widely used method for teaching play and

social skills to children with ASD. It has been used to teach reciprocal commenting skills

in conversation (Charlop-Christy et al., 2000), play statements during play with siblings

(Taylor, Levin, & Jasper, 1999) and initiation of play interactions with peers (Nikopoulos

& Keenan, 2003; 2004). Children also have been taught complex chains of pretend,

solitary sociodramtic play, manipulating figurines while providing the dialog for the

characters (D’Ateno, Mangiapanello, & Taylor, 2003; MacDonald et al., 2005;

MacDonald, Sacramone, Mansfield, Wiltz, & Ahearn, 2009 ).

There are additional advantages to using video modeling to teach play. It

decreases the variation in the model presentation. Video modeling allows the trainer to

focus on prompting the child to imitate the model rather than performing the model

(Weiss & Harris, 2001). Additionally, video modeling reduces problems with learning

related to stimulus overselectivity. Children with ASD may commonly focus on

environmental stimuli irrelevant to the task. Through a video presentation of a model, the

Video Modeling 6

trainer is able to edit out distracting auditory or visual stimuli so that only relevant

content is seen. Video permits the accentuation of certain stimulus features and

minimization of distracting and irrelevant features by showing the model up close

(LeBlanc et al., 2003).

There is a variety of model types to choose from when considering video

modeling. The most common type of model used is the adult as a model, which Charlop

and Milstein (1989) examined. No significant learning differences were found based on

the age of the actor. The use of an adult as model is preferred because adults are the

easiest to train. There is also evidence in recent research that the use of peers (Gena et al.,

2005; Nikopoulous & Keenan, 2004) and self (Buggey, Toombs, Gardener, & Cervetti,

1999) as model are effective teaching tools, but it is unclear if either of these model types

have any outstanding benefits. Recently, Hine and Wolery (2006) examined effects of

filming the model from the learner’s point of view. Using videos filmed from this point

of view they taught two sets of play skills to children with autism with good results and

generalization. The effects of using of point of view as opposed to other perspectives is

currently an area of growing research.

Visual media for the purposes of modeling can be separated into two categories:

commercially made and instructor made. Commercially available videos, (specifically

those intended for children) while they may or may not have been made with the intent of

generating specific play skills, many of them do have accompanying toy sets available in

retail stores featuring the characters of the film. The intent of these films can be viewing

them for entertainment or promotional purposes. The second type of model is instructor

made videos. Videos used in the video modeling literature would fall in to this category;

Video Modeling 7

these are videos generated by teachers to target a specific response from a learner. They

are made with the intent to teach a target skill and feature a series of repeated, deliberate

responses, which will later be required of the learner.

Recent video modeling research has used video-based imitation to teach longer

behavior chains and more complex skills than ever before. MacDonald et al. (2009)

trained play scripts of up to 16 actions and 17 vocalizations to children with autism and

their typically developing peers. Until now, researchers have yet to determine the critical

features of the videos used by D’Ateno, Mangiapanello, & Taylor, 2003 and MacDonald

et al., 2005 which made their videos effective teaching tools. If the specific features of

those videos which contributed to the acquisition of sociodramatic play skills in children

with autism can be determined, then video models could be made more effective. A step

forward in this research is to determine in what ways these video models differ from

commercially available videos common to children’s entertainment.

With longer and more complex scripts being generated by video modeling

researchers, generalization of the skill of video modeling is the next step. Video modeling

takes many forms in our daily lives. Commercials and instructional programming are just

some of the media typically developing individuals use to gain information on products,

skills, culture, and fashion. Bridging the gap between the instructor created video models

and current media in our culture can become the next step in the imitative repertoire for

many children with ASD. The problem exists, however, of determining what features of

instructor-created video modeling make it a successful teaching tool.

The purpose of the present study was to compare the effectiveness of instructor-

created video models using the techniques described in the research of MacDonald et al.

Video Modeling 8

(2005; 2009), with the effectiveness of corresponding commercially available children’s

videos to teach children with ASD to engage in extended sequences of pretend play. The

targeted play included scripted actions and vocalizations across two sets of toys. The

commercially available video (CAV) was a professionally produced children’s video

edited for length and content by the researcher. The instructor-created video (ICV) was a

film recreating the script from the commercially available video using corresponding toys

and play sets.

Method

Participants

Two male and one female student attending a preschool program at a center for

children with autism participated in this study. All students came to the center with a

diagnosis of autism spectrum disorder. Andrew was 5 years old at the time of the study

(Peabody Picture Vocabulary Test- Third Edition Form IIIB (PPVT-IIIB): 4 years, 7

months; Autism Diagnostic Observation Schedule-General (ADOS): Autism; Mullens:

40.25) and participated in integration for 3.5 hours of the school day. Piers was 5 years

old at the time of the study (PPVT-IIIB: 5 years 7 months; ADOS: Autism; Mullens:

60.75) and was integrated for 3.75 hours of his day. Elaine was 4 years old at the time of

the study (PPVT-IIIB: 4 years 5 months; ADOS: Autism; Mullens: 47.75) and participated

in integration for 3.5 hours of her day. All participants were verbal, and spoke in full

sentences. Participants were selected based on reports of their lack of appropriate

pretend-play skills with figurines.

Video Modeling 9

Setting

All training and probe sessions were run in a therapy room (2.7m x 4.3m) at the

participants’ preschool program separate from the classroom and free from distractions.

In the therapy room, there was a table and chair to view the video and all play specific

materials were set up on the floor behind the participant. A video camera and additional

materials specific to running the sessions and collecting data were also present in the

room. Sessions were run daily and all sessions were videotaped for later scoring.

Materials

Materials used during baseline and training included a portable DVD player, and

two Fisher Price Little People® play sets featured in the selected episodes. The CAVs

were taken from two episodes of the Fisher Price Little People stop motion animated

series. These episodes were then edited so that each had a similar running time and a

similar number of actions and vocalizations by the characters. The ICV used during

training included an adult acting script taken from the corresponding CAV and play set.

Every effort was made to recreate the episode, however sound effects and sound track

were not included.

The Sonya Lee and the Super Sundae episode contained 28 actions, 29

vocalizations, 9 characters, and had a running time of 2 minutes and 41 seconds (Table

1). The Faster than a Speeding Frog episode contained 28 actions, 31 vocalizations, and

had a running time of 2 minutes and 39 seconds (Table 2). All toys were purchased to

recreate the original CAV set, and items that could not be purchased were recreated.

Independent Variable

Video Modeling 10

The independent variable was the format of the training video. A video of each

model type was created for each toy set. The ICV showed an adult modeling the

behavior chain of completing the script using the corresponding play set. The CAV

showed a stop motion clay animation of the script. Each participant was exposed to one

model type of the video for each play set (see Table 3).

Dependent Variables

Script Completion. The dependent variable was the number of actions and vocals

independently completed in the response chain for the toy set. The steps completed did

not need to be performed in the order specified in the response chain, but the completed

toy structure needed to match the picture model. The measurement method was

occurrence of each scripted action or vocalization. Due to the complexity of the scripts,

credit was awarded for partial completion or paraphrasing of vocals and definitions were

created for each action in the script (see Tables 4 and 5).

Attending to Video. The second dependent variable was the amount of time the

participant attended to the video. Attending was defined as any instance of head oriented

and eyes directed toward the portable DVD screen. Attending to video and attending to

toys were measured by duration. A real-time measurement method (Miltenberger, Rapp,

& Long, 1999), which consisted of second-by-second recording, was used to assess

duration of attending to video and toys. Percent attending was calculated by adding the

number of seconds in which the participant was attending dividing by the total seconds of

the session and multiplying by 100.

Video Modeling 11

Attending to Toys. The third dependent variable was the amount of time the

participant attended to the play set while the video was playing. Attending to toys was

defined as any instance of head oriented and eyes directed toward the toys

Interobserver Agreement

A secondary observer independently scored the sessions from video so that

collect interobserver agreement could be measured, . This observer recorded occurrences

of scripted actions and vocalizations. . Interobserver agreement was calculated by

dividing the number of agreements by the total number of agreements plus disagreements

and multiplying by 100%.

Interobserver agreement was conducted for baseline sessions. Interobserver

agreement was calculated in 50% of the pre-assessments for Andrew , 30% for Piers, and

40% for Elaine. Interobserver agreement for Andrew was 93% for actions and 100% for

vocalizations. Interobserver agreement for Piers was 100% for actions and 100% for

vocalizations. Interobserver agreement for Elaine was 98% for actions (range, 93-100%)

and 100% for vocalizations.

Interobserver agreement was calculated in 33% of ICV sessions for Andrew.

Interobserver agreement was 94% for actions (range, 93-96%) and 94% for vocalizations

(range, 90-100%). Interobserver agreement was calculated in 43% of ICV sessions for

Piers. Interobserver agreement was 98% for actions (range, 93-100%) and 99% for

vocalizations (96-100%). Interobserver agreement was calculated in 38% of ICV

sessions for Elaine. Interobserver agreement was 97% for actions (range, 94-100%) and

100% for vocalizations.

Video Modeling 12

Interobserver agreement was calculated in 33% of CAV sessions for Andrew.

Interobserver agreement was 91% for actions (range, 82-96%) and 96% for vocalizations

(range, 87-100%). Interobserver agreement was calculated in 38% of CAV sessions for

Piers. Interobserver agreement was 85% for actions (range, 76-92%) and 98% for

vocalizations (93-100%). Interobserver agreement was calculated in 31% of CAV

sessions for Elaine. Interobserver agreement was 96% for actions (range, 86-100%) and

100% for vocalizations.

Interobserver agreement was conducted for initial mastery probe sessions.

Interobserver agreement was calculated in 50% of sessions for Andrew, 50% for Piers,

and 50% for Elaine. Interobserver agreement for Andrew was 98% for actions (range,

96-100%) and was 97% for vocalizations (range, 97-97%). Interobserver agreement for

Piers was 94% for actions (range, 87-100%) and 100% for vocalizations. Interobserver

agreement for Elaine was 96% for actions (range, 96-96%) and 100% for vocalizations.

Interobserver agreement was calculated for attending to video and attending to toys in

38% of training sessions for Andrew. Agreement for attending to the video was 96%

(range, 95-97%) and agreement for attending to the toys was 97% (range, 95-98%).

Interobserver agreement was calculated for attending to video and attending to toys in

33% of training sessions for Piers. Agreement for attending to video was 98% (range,

96-100%) and agreement for attending to toys was 92% (80-100%). Interobserver

agreement was calculated for attending to video and attending to toys in 33% of training

sessions for Elaine. Agreement for attending to the video was 84% (range, 78-89%) and

agreement for attending to the toys was 84% (range, 82-86%).

Video Modeling 13

Experimental Design

To compare the effectiveness of ICVs models to CAVs a multielement design

within participant and across model types was used. Additionally, a multiple probe

design was used across participants. Two scripts of a similar difficulty were taught using

one video of each format for each participant. Each participant completed mastery probes

for both scripts before training of another participant began. Mastery probes were

conducted prior to the introduction of another participant for all participants.

Procedures

Baseline. Prior to each baseline and training session, play sets were arranged with

all characters and vehicles as they would appear in the first scene of the video before the

participants entered the room. Characters or vehicles that did not appear in the first scene

of the video were arranged in the positions in which they first appear in the video. For

example: in Sonya Lee and the Super Sundae, the cow does not appear until the rest of

the characters arrive at the farm. Therefore, the cow’s starting position in the play set

was at the farm. Baseline sessions began with the student seated in front of the play set.

The experimenter gave the instruction “It’s time to play” and participants had 5 min to

interact with the materials. During all sessions, the experimenter operated the video

camera to record the session while remaining outside of the play set. All sessions were

videotaped and later scored by the experimenter.

Instructor-Created Video. Participants were exposed to one model type for each

toy set (see Table 3). Each session began with the participant sitting at the table with the

portable DVD player in front of them. The experimenter started the video and told the

participant, “It’s time to watch the movie.” The participant viewed the video model of the

Video Modeling 14

episode two consecutive times. Next, the participant was instructed to sit down in front

of the play set. The therapist told the participant “It’s time to play.” At the end of the 5-

min session the participant was told, “Playing is all done.” Initially, sessions were 4

minutes but due to the participants’ inability to finish the script within that time

constraint, all sessions were extended to 5 minutes. After the participant’s data met the

predetermined criterion of 75% total script completion or greater for 3 consecutive

sessions or demonstrated a stable trend with no increase in script acquisition, the

participant was exposed to the mastery probe phase.

Commercially Available Video. Sessions for the CAV were run identically to

those of the ICV. At the beginning of training, the experimenter told the participants

what each of the items in the play set represented, as they were not identical to the items

featured in the video. For example: The cloud and the yo-yo for the Faster than a

Speeding Frog script were not available in stores and were recreated specifically for this

experiment.

Mastery Probes. Mastery probes were identical to the baseline sessions. The

participants were told, “It’s time to play,” and were given 5 min to interact with the toy

sets. Criteria to probe for mastery for all participants were three sessions at 75% or

greater script completion or stability with no increasing trend in the data through visual

inspection. After three consecutive sessions of 75% accuracy with the videos, mastery

probes were conducted without the video. If the participant completed the script with

70% accuracy for two consecutive sessions without the video, they met mastery criteria

for that script and the next participant could begin training.

Video Modeling 15

Results

Total percentage of script completion for each participant is shown in Figure 1.

Andrew met mastery criteria for script completion in 12 sessions with the ICV model and

met mastery criteria for script completion in 12 sessions with the CAV model as well.

Piers met mastery criteria for the ICV script in 4 sessions, and took 8 sessions to reach

mastery criteria for the script taught using the CAV. Elaine met mastery criteria for

script completion in 16 sessions with the ICV model, and never met mastery criteria with

the CAV model.

Script completion of actions and vocals for Andrew is shown in Figure 2. Andrew

completed almost no scripted actions or vocals during baseline across both play sets.

Following training, scripted vocalizations and actions increased for Andrew. Scripted

vocalizations for the ICV script increased from a baseline level of 0 vocalizations per

session to a level of 27 (out of 29) vocalizations on mastery probes, and scripted actions

for the ICV script increased from a baseline level of 1 action per session to a level of 23

actions during mastery probes (out of 28). Scripted vocalizations for the CAV script

increased from a baseline level of 0 per session to 29 vocalizations per session (out of 31)

on mastery probes, andscripted actions for the CAV script also increased from a baseline

level of 0 actions per session to a level of 21.5 actions during mastery probes (out of 28).

Script completion of actions and vocals for Piers is shown in Figure 3. Piers

completed no scripted actions or vocals during baseline across both play sets. Following

training, Piers showed increases in scripted vocalizations and actions. Scripted

vocalizations for the script taught using the ICV increased from a baseline level of 0

vocalizations per session to a level of 28 (out of 31) vocalizations on mastery probes as

Video Modeling 16

well as actions taught using the ICV increased from a baseline level of 0 actions per

session to a level of 25 (range, 22-28) actions during mastery probes (out of 28). Scripted

vocalizations for the script taught using the CAV increased from a baseline level of 0 per

session to 28.5 vocalizations per session (out of 31) on mastery probes while scripted

actions for the CAV script also increased from a baseline level of 0 vocalizations per

session to a level of 22 actions during mastery probes (out of 28).

Script completion of actions and vocals for Elaine is shown in Figure 4. Elaine

completed very few scripted actions or vocals during baseline across both play sets.

Following training, scripted vocalizations and actions for the script taught using the ICV

increased for Elaine. However, Elaine displayed variable results in the CAV condition.

Scripted vocalizations for the script taught using the ICV increased from a baseline level

of 0 vocalizations per session to a level of 28 (out of 29) vocalizations on mastery probes,

and scripted actions for the ICV script increased from a baseline level of 2 actions per

session to a level of 21 actions during mastery probes (out of 28). Scripted vocalizations

for the script taught using the CAV increased from a baseline level of 0 per session to 3

vocalizations per session (out of 31) on mastery probes, and scripted actions for the script

taught using the CAV also increased from a baseline level of 0 verbalizations per session

to a level of 4.5 actions during mastery probes (out of 28).

Attending data for Andrew are shown in Figure 6 and Table 6. Andrew

consistently attended to either the video (mean, 85% range, 76-96%) or the toys (mean,

13% range, 4-24%) for the duration of the training videos. There was no clear difference

in attending to the video and the toys between the ICV and CAV conditions.

Video Modeling 17

Attending data for Piers are shown in Figure 7 and Table 7. Piers also

consistently attended to either the video (mean, 90% range, 74-100%) or the toys (mean,

7% range, 0-22%) for the duration of the training videos. Piers’ attending to the video

decreased, and attending to the toys increased as script completion increased.

Attending data for Elaine is shown in Figure 8 and Table 8. Elaine’s attending to

the task (video and toys), while not as consistent, was similar in pattern to Pier’s. Elaine

attended to either the video (mean, 80% range, 60-99%) or the toys (mean, 9% range, 0-

15%) for the majority of the presentation of the training videos. Like Piers, Elaine’s

attending to the video also decreased, and attending to the toys increased as script

completion increased. Additionally, this pattern was observed with both play sets even

though Elaine only learned one script.

Discussion

The CAVs may have proved to be too complex a teaching tool for some of the

participants in this study. While the basic content was the same, the differing aspects of

the CAVs and the ICVs yielded different results for each of the participants. In contrast,

the ICVs proved to be a more reliable model type for teaching play skills. Though care

was taken to re-create the timing, content and script of the children’s videos, the content

that was not re-created seemed to make a difference the videos’ potential as a teaching

tool. Additionally, these differences in acquisition may be the lack of prerequisite skills

or the need for a more literal teaching model.

There were many features of the CAV that were extraneous to the targeted task.

Auditory stimuli such as sound tracks and sound effects competed with the character

dialogue. Additionally, there were many visual aspects of the CAVs that could not be re-

Video Modeling 18

created. In both episodes, characters had moveable limbs and facial expressions. Their

eyes shifted and their hands manipulated items. Although they appeared to be similar to

the toys, the videos had many features that their static counterparts did not. Stop-motion

clay animation also allowed for many special effects that the ICV could not re-create. In

the Faster than a Speeding Frog episode, Eddie landed in a cloud that hung unsupported

in the air and Michael did tricks with his yo-yo. In the Sonya Lee and the Super Sundae

episode, the dump truck tunneled underneath a snow pile and the fire truck sprayed liquid

ice cream toppings on to the sundae. In the end, the participants’ data indicates that some

of these studio animation effects may have been more of a hindrance to the video’s

potential as a learning tool than a help. Only 1 participant out of 3 was able to learn

equally well across both video model types.

Conversely, the same special effects that may have hindered skill acquisition had

no effect on other dependent variables. The commercially generated effects and

soundtrack did not affect participants’ attending to the video more than its instructor

created counterpart. This may indicate that more complex videos do not have the added

advantage of increased attending. Also, based on Elaine’s results, increased attending

does not necessarily have the guarantee of increased skill acquisition.

The unique features of the ICV, while not technically impressive, did aid the

model in its use as a teaching tool. Distracters such as sound effects and the music score

did not compete with any of the dialogue. Also, the presence of an actor manipulating

the figurines provided the participants with a more literal translation of the task. The toys

featured in the ICV were also an exact match to those materials used by the participants

to recreate the script. From this study it can be determined that the differing features in

Video Modeling 19

the videos did indeed make a difference in the videos effectiveness as a teaching tool;

furthermore we may be one step closer to determining what features will create the most

effective video possible.

For some students is was found that although a CAV may be a time and cost

effective alternative to an ICV, they are most likely not the preferred method of teaching.

More research is needed to determine what the prerequisites are in order to learn from a

CAV. Learning from CAVs is the generalization of the skill of video modeling.

Additionally, there is the possibility that it is a higher order skill than learning from an

ICV, with the requirement for exposure to more types of video models before it can be

effective. Teaching individuals to learn from CAVs may be a stepping-stone to teaching

individuals to use generalized video modeling in their daily lives.

There are several limitations to this study that must be considered. First, all

participants in this research had a previous learning history with using video modeling

and the results of the experiment may have been different if the participants had no

previous exposure to this teaching strategy. It should be noted that these data were

collected with high functioning children with autism and may not generalize to a larger

population. The materials used in this study were also limited to the use of clay-mation

as model type. Further investigation would be needed to determine the success of the

intervention using live action or animation. Additionally, there are a number of potential

dependent variables that could be measured in future studies, including the quality of play

(including inflection and intonation) and novel or unscripted play generated by each

model type.

Video Modeling 20

There are many exciting possibilities for future research based on the findings of

this study. The use of different video formats could be tested to investigate the degree to

which the toys are able to deviate from the video model; and whether individuals can

learn using live action film as opposed to the clay-mation, which is a closer match to the

toys. Also, the responses of typically developing individuals have not been assessed and

should be taken as a standard when assessing those with disabilities. The ICVs were

based on the camera angles used in the CAVs, but replications could assess effects of the

point of view of the video model. Additionally, the CAV contained pre-existing cuts

where vital actions to the progression of the plot were left out. For example, the tractor

in the sundae script is never shown being boarded or traveling to the farm, but appears

later pulling up to the farm. Despite these gaps in the script, all the participants were able

to complete the script. Further research could explore the necessity of script elements

and the possibly of programming for unscripted play.

Further possibilities for future research could be related to what we learned about

the participants’ attending during the model presentation. Attending to the model has

always been considered an important factor in an individual’s success with an imitation

task (McCoy, 2007), but attending to the materials may be an important factor not yet

considered. Attending to the materials as well as to the model may make a difference in

task completion or rate of acquisition. Attending to materials may also be an important

correlate of or even a prerequisite for video modeling that has not been considered. For

therapists, a model presented in the presence of the materials may yield significantly

different results than a model presented in the absence of the materials. Finally, if

Video Modeling 21

shifting attention from the model to the materials is observed in some individuals, then

can it be trained in individuals lacking this skill?

Video Modeling 22

References

Buggey, T., Toombs, K., Gardener, P. & Cervetti, M. (1999). Training responding

behaviors in children with autism: Using videotaped self-modeling. Journal of

Positive Behavior Interventions, 1, 205-214.

Charlop, M. H. & Milstein, J. P. (1989). Teaching autistic children conversational

speech using video modeling. Journal of Applied Behavior Analysis, 22, 275-

285.

Charlop-Christy, M. H., Le, L., & Freeman, K.A. (2000). A comparison of video

modeling with in vivo modeling for teaching children with autism. Journal of

Autism and Developmental Disorders, 30, 537-552.

Charlop-Christy, M. H., Le, L., & Daneshvar, S. (2003). Using video modeling to teach

perspective taking to children with autism. Journal of Positive Behavior

Interventions, 5, 12-21.

D’Ateno, P., Mangiapanello, K., & Taylor, B. A. (2003). Using video modeling to teach

complex play sequences to a preschooler with autism. Journal of Positive

Behavior Interventions, 5(1), 5-11.

Dube, W. V., MacDonald, R. P. F., Holcomb, W. L., Mansfield, R. C., & Ahearn, W. H.

(2004). Toward a behavioral analysis of joint attention. The Behavior Analyst, 27,

197-207.

Egel, A.L., Richman, G.S., & Koegel, R.L. (1981). Normal peer models and autistic

children’s learning. Journal of Applied Behavior Analysis, 14, 3-12.

Gena, A., Courloura, S., & Kymissis, E. (2005). Modifying the affective behavior of

Video Modeling 23

preschoolers with autism using in-vivo or video modeling and reinforcement

contingencies. Journal of Autism and Developmental Disorders, 1-12.

Goldstein, H., & Cisar, C. L. (1992). Promoting interaction during sociodramatic play

teaching scripts to typical preschoolers and classmates with disabilities. Journal

of Applied Behavior Analysis, 25, 265-280.

Haring, T. G., Kennedy, C. H., Adams, M. J., & Pitts-Conway, V. (1987). Teaching

generalization of purchasing skills across community settings to autistic youth

using videotape modeling. Journal of Applied Behavior Analysis, 19, 159-171.

LeBlanc, L.A., & Coates, A.M. (2003). Using video modeling and reinforcement to

teach perspective-taking skills to children with autism. Journal of Applied

Behavior Analysis, 36, 253-257.

Lifter, K. (2000). Linking assessment to intervention for children with developmental

disabilities or at-risk for developmental delay: The Developmental Play Assessent

(DPA) Instrument. In K. Gitling-Weiner, A. Sandgrund, & Schaefer (Eds.), Play

diagnosis and assessment (2nd edn, 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.

MacDonlad, R., Sacramone, S., Mansfield, R., Wiltz, K., & Ahearn, W. H. (2009).

Using video modeling to teach reciprocal pretend play to children with autism.

Journal of Applied Behavior Analysis, 42, 43-55.

Miltenberger, R.G., Rapp, J.T., & Long, E. S. (1999). A low-tech method for

conducting real-time recording. Journal of Applied Behavior Analysis, 32,

119-120.

Video Modeling 24

Nikopoulos, C. K., & Keenan, M. (2003). Promoting social initiations in children with

autism using video modeling. Behavioral Interventions, 18, 87- 108.

Nikopoulos, C. K., & Keenan, M. (2004). Effects of video modeling on social initiations

by children with autism. Journal of Applied Behavior Analysis, 37, 93-96.

Sherer, M., Pierce, K. L., Paredes, S., Kisacky, K. L., Ingersoll, B., & Schreibman, L.

(2001). Enhancing conversational skills in children with autism via video

technology: Which is better, “Self” or “Other” as a model? Behavior

Modification, 25, 140-158.

Shipley-Benamou, R., Lutzker, J., & Taubman, M. (2002). Teaching daily living skills to

children with autism through instructional video modeling. Journal of Positive

Behavior Interventions, 4, 165-175.

Stahmer, A.C., Ingersoll, B., Carter, C. (2003). Behavioral approaches to promoting

play. Journal of Autism, 7, 401-413.

Taylor, B.A. (2001). Teaching peer social skills to children with autism. In C. Maurice,

G. Green, & R.M. Foxx (Eds.), Making a Difference: Behavioral Intervention for

Autism. (pp. 83-96). Austin, TX: PRO-ED.

Taylor, B. A., Hoch, H., Potter, B., Rodriguez, A., & Kalaigian, M. (2005). Manipulating

establishing operations to promote initiations toward peers in children with

autism. Research in Developmental Disabilities, 26, 385-92.

Taylor, B. A., Levin, L., & Jasper, S. (1999). Increasing Play related statements in

children with autism toward their siblings: Effects of video-modeling. Journal of

Developmental and Physical Disabilities, 11, 253-264.

Video Modeling 25

Weiss, M.J., & Harris, S.L. (2001). Reaching out, joining in: Teaching social skills

to young children with autism. Bethesda, MD: Woodbine House, Inc.

Video Modeling 26

Table 1.

Script for Sonya Lee and the Super Sundae

Video Modeling 27

Table 2.

Script for Faster than a Speeding Frog.

Video Modeling 28

Table 3.

Model types and videos for each participant.

Commercially Available

Video Instructor Created

Video

Participant

Andrew Frog Sundae Piers Sundae Frog

Elaine Frog Sundae

Video Modeling 29

Table 4. Scoring guidelines for Sonya Lee and the Super Sundae

Sonya Lee and the Super Sundae: Action Guidelines General Scoring rules

• If a character walks they must contact the ground, they cannot fly • Actions are scored + if they are with the wrong character • Vehicles must contact the ground, they cannot fly over or be lifted • Actions still count if a character/piece is placed somewhere and then falls off • Actions can occur out of order from the original script

Script Description Student Action

Enter Eddie is standing at the sandbox talking to his friends Riding in truck Sonya is in the driver’s seat of the dump truck

enter The truck pulls up to the park and the sandbox from off set Turns to dump sand Truck turns around so the back side is facing the sandbox

Dumps sand Back of truck tips up and sand pile either falls or is placed on the sandbox

Finishes, drives off Back of truck returns to its upright position and truck turns back around so the front bumper is facing the sandbox and gang

Backs away, sad. Truck backs away from the sandbox

leaves Truck leaves the park scene. Can exit on either side

Turns and walks Sonya Lee walks or glides along the surface of the set towards the dump truck Enters on tractor Farmer in tractor enters Walks to tractor At least one member of the gang walks or glides along the set surface towards the tractor

Walking to truck Sonya Lee walks or glides along the surface of the set towards the dump truck. Can be tapped in place next to the truck

Cows skiing Cow bounces or glides along the mound of cotton balls Cow does a flip Cow does at least one flip

In snow with Eddie Sarah and Eddie stand in the snow pile

Points to milk Gestures towards the milk/barn. Can use the farmer to point or can orient the farmer to the milk/barn.

Walk over to milk At least one character must walk/contact the ground Shows them milk Use character to point to or touch the ice cream

Enter Sonya and truck enter (can fly in but must be placed on the set) Backs up and pushes At least once

Breaks through snow Student can part snow with hands and have the truck drive through or they can use the truck to part the snow

Run to truck At least one character runs (makes contact with the set) to the truck

Stop in the middle of town Truck can fly over to the set, but must drive down the street once there. At least one of the vehicles must be present (truck or fire truck) for +

Sprays chocolate from fire truck Chocolate topping is placed on top of one of the ice creams. May or may not pretend to spray it from the truck. Placement is all that is required for +.

Sprays whipped cream White topping is placed on top of one of the ice creams. May or may not pretend to spray it from the truck. Placement is all that is required for +.

Flies in on helicopter Student must spin or fly the helicopter. Character does not have to be present. Also, + is scored if helicopter is on the ground with the blades spinning or it can be in the air without the blades spinning.

Drops cherry on ice cream Cherry can be dropped or placed. Must hit the ice cream topping. Does not have to stay placed.

Sonya Lee and the Super Sundae: Vocal Guidelines General Scoring rules:

• Scored as + if it is complete and matches the script. • Scored as + if it is a partial vocal (at least 50%) or a paraphrase of a vocal (example: script says “Help us eat the world biggest ice

cream sundae” and student says “Let’s eat the world’s largest ice cream cone”) • Student does not have to be holding/manipulating character for + • Scored as + if it is with the wrong character • Vocals are only counted once, but do not count against student if they repeat them • Scored as + if they occur out of order from the original script

Video Modeling 30

Table 5. Scoring guidelines for Faster than a Speeding Frog.

Faster Than a Speeding Frog: Action Guidelines General Scoring rules

• If a character walks they must contact the ground, they cannot fly • Actions are scored + if they are with the wrong character • Vehicles must contact the ground, they cannot fly over or be lifted • Actions still count if a character/piece is placed somewhere and then falls off • Actions can occur out of order from the original script

Script Description Student Action

Stops traffic Police officer present in the middle of the intersection Crosses street Feet contact the ground, can float or step

Jumps on Eddie’s head Freddie the frog jumps up, leapfrogs on top of Eddie’s head and then jumps again Leapfrogs Eddie and Freddie jump over each other Leapfrogs Eddie and Freddie jump over each other

Jumps on spring Eddie jumps up, lands on spring and falls

Falls back to earth Comes back to earth, contacts the ground either on his back, side or standing

Picks up spring Eddie picks up any of the springs. Child can hang spring on his hand or can hold it up in front of him

Continues to do flips Continues to jump through the conversation Lands in cloud Can land in cloud head or feet first Pulls himself up Straightens himself to an upright, seated position in cloud

Reaches for springs and knocks them down Springs fall out of the cloud, Eddie may or may not touch them

Tears around the corner Fire truck enters. Turns corner and pulls up to cloud Jumps out At least one fireman gets out

Unwinds ladder to reach Eddie Extend ladder to cloud. Ladder may not extend all the way or child may extend truck Reach to each other One or both characters make an effort

Retracts ladder Ladder/truck pulls away from cloud

Jumps, gains attention Frog makes some sort of attention gaining response to the gang. May differ based on video

Takes yo-yo Yo-yo (string or whole thing) goes to frog Winds up and jumps Jump attempt. Does not reach cloud

Tries again Jump attempt landing in cloud

Lands in cloud Anywhere on cloud, does not have to stay in cloud if child cannot balance/handle all pieces at once

Ties yo-yo to cloud Can be actual or pretend Jumps down holding frog and yo-yo Can be either or both yo-yo and frog with Eddie

Faster Than a Speeding Frog: Vocal Guidelines General Scoring rules:

• Scored as + if it is complete and matches the script. • Scored as + if it is a partial vocal (at least 50%) or a paraphrase of a vocal (example: script says “Help us eat the world biggest ice

cream sundae” and student says “Let’s eat the world’s largest ice cream cone”) • Student does not have to be holding/manipulating character for + • Scored as + if it is with the wrong character • Vocals are only counted once, but do not count against student if they repeat them • Scored as + if they occur out of order from the original script

Video Modeling 31

Table 6.

Summary data for Andrew for attending to video and attending to toys during training.

Commercially Available Video

Viewing

Instructor Created Video

Viewing

Attending to

Video

Attending to

Toys

Attending to

Video

Attending to

Toys

1 83% 12% 80% 13%

2 90% 7% 89% 5%

3 95% 5% 76% 24%

4 70% 24% 96% 4%

Video Modeling 32

Table 7.

Summary data for Piers for attending to video and attending to toys during training.

Commercially Available Video

Viewing

Instructor Created Video

Viewing

Attending to

Video

Attending to

Toys

Attending to

Video

Attending to

Toys

1 100% 0% 97% 1%

2 92% 4% 92% 2%

3 74% 22% 82% 14%

Video Modeling 33

Table 8.

Summary data for Elaine for attending to video and attending to toys during training.

Commercially Available Video

Viewing

Instructor Created Video

Viewing

Attending to

Video

Attending to

Toys

Attending to

Video

Attending to

Toys

1 99% 0% 93% 4%

2 76% 10% 60% 15%

3 79% 14% 75% 13%

Video Modeling 34

Figure Captions

Figure 1. Summary data for total script completion across all three participants. Open

triangles represent the script trained using the instructor-created video and closed squares

represent the script trained using the commercially available video.

Figure 2. Summary data for completion of scripted actions and vocals for Andrew.

Open triangles represent the script trained using the commercially available video and

closed squares represent the script trained using the teacher created video model.

Figure 3. Summary data for completion of scripted actions and vocals for Piers. Open

triangles represent the script trained using the commercially available video and closed

squares represent the script trained using the teacher created video model.

Figure 4. Summary data for completion of scripted actions and vocals for Elaine. Open

triangles represent the script trained using the commercially available video and closed

squares represent the script trained using the teacher created video model.

Figure 5. Summary data representing the percentage of attending to video and attending

to toys per probe session for Andrew.

Figure 6. Summary data representing the percentage of attending to video and attending

to toys per probe session for Piers.

Video Modeling 35

Figure 7. Summary data representing the percentage of attending to video and attending

to toys per probe session for Elaine.

Figure 1.

Figure 2.

Figure 3.

Figure 4.

Figure 5.

Figure 6.

Figure 7.