Fall 2007, Vol.4, Num. 1

The Effectiveness of Using a Pocket PC as a Video Modeling and Feedback Device for Individuals with Developmental Disabilities in Vocational Settings

Toni Van Laarhoven Northern Illinois University

Traci Van Laarhoven-Myers Waubonsie Valley High School

Leslie M. Zurita Northern Illinois University

Abstract: This study evaluated the Vocational programming and support for effectiveness of using a pocket PC to teach two individuals with developmental disabilities in adolescents with mild and moderate cognitive competitive employment settings has received impairments vocational tasks in competitive, much attention in the last two decades (Lancioni, community-based settings. Participants were O’Reilly, Speedhouse, Furniss, & Cunha, 2000). taught three different tasks in their respective The goal of vocational programming is to assist work sites through video rehearsal and video individuals with performing their job-related feedback strategies. Video files were presented tasks as independently as possible so that they on a pocket PC prior to task engagement and can have the means to support themselves in following repeated errors (video feedback). order to lead productive and self-sufficient lives. Effectiveness of the video-based strategies and Fortunately, there has been a shift away from the the utility of the handheld device were evaluated assumption that individuals with developmental using a multiple probe design across tasks and disabilities will work in sheltered employment participants. Outcomes of the study indicate that settings and more attention and effort has been the introduction of the video-based procedures placed on supporting individuals in community- was associated with significant increases in based vocational settings (Davies, Stock, & independent responding and participants met Wehmeyer, 2002a). As a result, much of the criterion on all three tasks within 3-7 sessions latest research relating to vocational (M=5 sessions). Although there were large programming has focused on strategies for changes in the level of data once the video-based increasing independence in job-related tasks. materials were presented, outcomes are Because there has been a shift from sheltered to somewhat tempered by the fact that some of the community-based settings, it is increasingly baselines for each participant were slightly important to utilize self-management strategies ascending prior to the introduction of the that will promote task completion while reducing independent variable. reliance on outside staff or job coaches (Agran, 1997). Keywords: Video technology, Handheld , Developmental disabilities, To promote independent work-related behaviors Vocational training and to decrease reliance on staff, several researchers have investigated the effectiveness of using supports and prompts such as picture cards and/or booklets, auditory prompts delivered Assistive Technology Outcomes and Benefits / 28

Fall 2007, Vol.4, Num. 1 electronic devices, and visual and auditory were no differences between the two types of prompts delivered on handheld devices to recordings, both types of auditory prompts promote self-directed work behaviors. resulted in a significant increase in the number of independent task changes made by participants. Several researchers have investigated the use of In addition, the results generalized to other pictures to encourage self-directed task initiation settings without additional training. and completion (Cihak, Alberto, Kessler, & Taber, 2004; Copeland & Hughes, 2000; Fisher, Recently, many researchers have combined both 1984; Martin, Mithaug, & Burger, 1990; Martin, visual and auditory prompting systems (i.e., Mithaug, & Frazier, 1992; Wacker & Berg, 1983). photos and auditory cues) delivered on handheld For example, Copeland and Hughes used a two- devices to investigate their utility in promoting part picture prompt strategy to teach two high independent responding among individuals with school students with severe disabilities to developmental disabilities in vocational settings complete work-related skills (i.e., cleaning faculty (Davies, Stock, & Wehmeyer, 2002a; Lancioni, dining room tables and sweeping; cleaning O’Reilly, Seedhouse, Furniss, & Cuhna, 2000; windows and a mirror in hallways of a hotel). Riffel et al., 2005). Davies et al. conducted a pilot The picture prompt strategy involved teaching study to evaluate the effectiveness of a participants to touch a picture to initiate a task program called the Visual Assistant (VA), which and then turn a page in a booklet to indicate task was loaded on a -top . The VA completion. Results indicated that once presented step-by-step pictures of task sequences participants acquired the picture prompt strategy, along with audio instructions on the computer to their independent task initiations (associated with prompt responses. Ten individuals with picture touching) increased; however, task intellectual impairments participated in the study completion (which was associated with page and were taught two vocational tasks that turning) increased for only one of the included a pizza box assembly task and a participants. In a similar study, Wacker and Berg software-packaging task. Participants received used a combination of demonstration, error training with the VA and were then given a correction, and praise to teach five individuals verbal overview of the tasks as well as with moderate to severe disabilities to turn pages demonstrations prior to task engagement. They in a picture booklet to put together pieces of two were then asked to perform each task twice, once different vocational assembly units (i.e., a black with the VA and once without. Results indicated valve and circuit board). Once participants improved accuracy and task independence when acquired the skills, the investigators withdrew all participants used the VA as opposed to when training components except the picture cues and they did not. Riffel et al. extended the research found that the participants generalized their on the use of the VA by teaching instructors to picture cue usage to two additional assembly use the device to assist four students with mild to tasks. moderate intellectual disabilities to perform tasks such as table setting, rolling silverware, and While picture prompts have been effective for laundry tasks. Results indicated that students prompting independent responding, others have increased the percentage of steps completed turned to technology using auditory prompting independently while also reducing prompts from devices to support individuals in employment instructors when the VA was used. In addition, settings. For example, Taber, Alberto, and participants appeared to prefer the device over Frederick (1998) used a self-operated auditory instructor assistance as they increasingly self- prompting device to teach 5 students with selected the VA to assist them in completing the moderate disabilities to independently transition task rather than requesting assistance from from completed vocational tasks to other teachers. vocational tasks. They compared single- and multiple-word recordings, and although there Assistive Technology Outcomes and Benefits / 29

Fall 2007, Vol.4, Num. 1

Similarly, Lancioni et al. (2000) conducted two individuals without disabilities, which makes their studies that also investigated the effectiveness of use socially acceptable. Although handheld using a palm-based computer system for teaching devices appear to be useful for instructing vocational tasks to individuals with severe individuals with disabilities, research to date has developmental disabilities. In the first focused primarily on the presentation of photos, experiment, the authors compared the auditory prompts, and cueing systems (vibration effectiveness of a palm-top computer that or alarms) with these devices. To our knowledge, presented line drawings (in conjunction with a there has not been any research conducted on special auditory device or vibratory mechanism the use of video-based materials presented on placed under participants’ belts) with a card handheld devices to teach skills to persons with system that was a booklet containing 25-31 disabilities. pictures of the steps required in the skill sequences. Six adults participated and were Video technology is emerging as an effective taught different sets of tasks that involved medium for teaching life skills to individuals with cleaning and food preparation. Results indicated developmental disabilities. For example, it has that participants not only had higher percentages been used to teach complex skills such as of correct responding with the computer system, purchasing items (Cihak, Alberto, Kessler, Taber- but also preferred it to the card system. Three of Doughty, & Gama, 2006), vocational tasks the participants who had acquired a high degree (Martin et al., 1992; Morgan & Salzberg, 1992), of correct responding in the first study also community skills (Alberto, Cihak, & Gama, 2005; participated in the second study. In the second Branham, Collins, Schuster, & Kleinert, 1999), study, the researchers taught the same tasks as in grocery shopping (Ayres & Langone, 2002; study one and presented all stimulus materials Mechling, 2004; Mechling & Gast, 2003), social using the computer-based system, but compared skills (Goldsworthy, Barab, & Goldsworthy, variations of how the pictures were presented 2000; Nikopoulos & Keenan, 2004; Simpson, (i.e., the stimulus materials were altered Langone, & Ayres, 2004), and daily somewhat so that the participants would not living/domestic skills (Bidwell & Rehfeldt, 2004; have to return to the computer as often during Graves, Collins, Schuster, & Kleinert, 2005; task engagement). Results indicated that clustered Lasater & Brady, 1995; Norman, Collins, & presentations were more effective in maintaining Schuster, 2001; Rehfeldt, Dahman, Young, correct task performance and participants Cherry, & Davis, 2003; Shipley-Benamou, required less prompting, or instructional Lutzker, & Taubman, 2002; Sigafoos et al., 2005; opportunities, delivered from the device. Van Laarhoven & Van Laarhoven-Myers, 2006).

In essence, it appears that handheld computers, Research that has been conducted on the use of when used as prompting systems, are effective video technology within the vocational domain for promoting correct independent responding has had mixed results. For example, Morgan and among individuals with moderate to severe Salzberg (1992) used video-assisted instruction to disabilities. In addition, handheld devices appear teach employment-related problem-solving skills to be effective at reducing external prompting to adults with disabilities. They showed from staff during task engagement and have also participants videos of positive and negative been used to encourage time management and examples and asked a series of questions to scheduling among individuals with mental provide discrimination training prior to assessing retardation (Davies, Stock, & Wehmeyer, 2002b) participants in actual work settings. Results as well as initiation and completion of daily tasks indicated that effects did not generalize until among individuals with ADHD (Epstein, Willis, behavioral rehearsal was introduced for 2 of the Conners, & Johnson, 2001). Handhelds are also 3 participants. Martin, Mithaug, & Burger (1992) desirable because they are portable, relatively compared several different instructional inexpensive, and used frequently among strategies for teaching assembly skills to Assistive Technology Outcomes and Benefits / 30

Fall 2007, Vol.4, Num. 1 secondary students with moderate disabilities in to as video rehearsal and will refer to the entire an unused classroom. These strategies included: video sequence being presented prior to task (a) photographs (a photo of completed piece of engagement. The video feedback component will furniture); (b) sequenced pictures (comprised of refer to having the participant view a video clip line drawings); (c) sequenced pictures plus (displaying correct performance) following errors modeling (same as previous, only the that occurred during task engagement. experimenters also modeled how to perform assembly); (d) picture referencing (used in The instructional methods that were compared conjuction with sequenced pictures, the are built on existing research that demonstrates experimenter also pointed back to picture when the success of video modeling and video student made an error or no response); (e) video feedback while employing the latest technology modeling (participant was shown video clip of and theory of computer-assisted instruction. step prior to performing step; also called video prompting); and (f) video referencing (used in Method conjunction with previous, only video clip was played again following errors or no response; also Participant Selection called video feedback) The researchers found picture referencing to be more effective than Participants were recruited from high school video modeling and video referencing. However, programs located in the suburbs of Chicago. To video referencing (e.g., video feedback) was recruit, a description of the study was e-mailed to almost as effective as picture referencing, and teachers in several school districts. Of those who video modeling became more robust as students responded, a follow-up questionnaire was sent had exposure to it. to:(a) identify students who were in the process of obtaining new jobs in community To our knowledge, there has not been any environments, (b) obtain personal information research done investigating the utility of using a for the participant to obtain informed consent handheld device to present video-based materials and assent, and (c) obtain information regarding to promote independent responding among potential job sites in order to request permission individuals with mild to moderate disabilities. from the employers to conduct research in their The purpose of this research was to determine if establishments. Participants were then selected video modeling and video feedback, when from the pool of respondents based on the presented on a portable handheld device, would criteria listed above. increase independent responding of two individuals who were employed in community- Participants based environments. The current study extends available research in a couple of ways. First, this Two young men with mild and moderate study differs from the others in that it evaluated cognitive impairments, who attended public the effectiveness of using video-based school programs, participated in the study. Both instructional materials presented on a handheld were recently hired at two different restaurants in device, whereas prior research using this the community and neither had any prior technology evaluated the effectiveness of using exposure to their assigned tasks. picture-based materials. Second, a feedback component was also used to provide error Devon was an 18-year-old student enrolled in a correction. And third, the research was large suburban high school where his educational conducted in community-based settings with goals were met through inclusive programming non-disabled coworkers. and practices. His full scale score on the Wechsler Abbreviated Scale of Intelligence For the purposes of this paper, the video (WASI; Psychological Corporation, 1999) was 78 modeling/rehearsal component will be referred Assistive Technology Outcomes and Benefits / 31

Fall 2007, Vol.4, Num. 1 and he had a diagnosis of Asperger’s syndrome. computers for the purposes of word processing, Much of his coursework was taught at the basic PowerPoint™, and basic . He could fundamental level by general education teachers, complete basic tasks with visual and/or verbal coupled with the assistance and support of prompts. His previous work experience involved special education teachers and/or special light custodial and some office work. He was education teaching assistants. He had been hired at Applebee’s, another large restaurant enrolled in a basic skill computer class and was chain, for nine hours per week and this is where fairly adept at utilizing a PC computer for basic baseline and instructional sessions took place. word processing, PowerPoint™, internet His primary duties involved portioning food for searches, and email activities. Devon also various recipes and cleaning and sanitizing his received speech therapy, occupational therapy workspace. and adaptive physical education services. He demonstrated a significant “aversion” to any type Setting of job or employment opportunity and initially refused to participate in any capacity. After one Baseline and instructional sessions were year of career counseling, he agreed to conducted in the participants’ employment “sampling” job tasks for periods extending no settings; Devon was at Red Robin, and Marcus longer than 45 minutes per week. Some of these was at Applebee’s. Devon worked primarily in an experiences involved light office duties and some area off the kitchen at Red Robin and was food preparation. With some coaxing, he agreed scheduled to work on Friday and Saturday to increase his work hours to eight hours per evenings for four-hour shifts. This area could be week and was hired at Red Robin, a large described as an open area or large hallway that restaurant chain where the study took place. He was near the refrigerator and back exit. Most of seemed very motivated to keep this particular job his tasks were completed at a tall stainless steel and engaged in frequent discussions regarding his rolling table (e.g., rolling and sorting silverware, attitude and what types of dispositions were and sanitizing the rolling table), however, he also important for employees to possess. His primary had to go to the front of the restaurant to replace job responsibilities involved sorting and completed silverware rolls and to clock in and sanitizing silverware and rolling silverware. out. Other aspects of his job required him to also go to the dishwashing area to sanitize sorted Marcus was an 18-year-old student enrolled in an silverware or to obtain trays of rinsed silverware inclusive high school. Most of his instructional that needed to be sorted. This setting was often programming, however, was conducted in a self- quite busy, i.e., loud music was in the contained, life skills classroom taught by special background and coworkers were rushing around education teachers. He had Down syndrome and while engaged in work activities. Staff members his full scale IQ score on the Wechsler Adult were very outgoing and friendly and the Intelligence Scale – 3rd Edition (WAIS-III; managerial employees were very supportive and Wechsler, 1997) was 47. Marcus required always encouraged him to be part of the team. extensive prompting to complete most tasks and was highly cue dependent. Just before he entered Marcus worked in an area behind the kitchen at the high school environment, he experienced Applebee’s. He was scheduled to work three some type of social/emotional setback and was mornings a week before they opened for lunch unable to complete academic and work tasks to and his primary responsibilities involved the degree that he had previously performed portioning food for various recipes and cleaning them. He also demonstrated a great drop in and sanitizing his work area. Most of his tasks receptive and expressive communication, social were completed while standing at a tall stainless participation, and moving throughout his steel counter. He also had to place trays of environment without direction. Previously, portioned food in the walk-in refrigerator, bring Marcus received instruction accessing PC Assistive Technology Outcomes and Benefits / 32

Fall 2007, Vol.4, Num. 1 empty vegetable bins to the dishwashing area, get task sequence was videotaped and then all of the ice cubes from an ice machine in the kitchen, and steps in the sequence were combined together clock in and out at the front of the restaurant. through the use of transition “swipes” to create a Staff members at Applebee’s were very step-by-step video of the task. Once the video supportive of Marcus. They adapted recipe sheets sequences were edited and rendered, they were for him, had an assistant kitchen supervisor then compressed to Windows Media Video assigned to train and assist him, and often set up (.wmv) file formats using the free download of his station before he arrived. Easy AVI/VCD/DVD/MPEG Converter™, version 1.1.8 software (8864soft.com, 2005). This Tasks was necessary, or the videos would not play on the Pocket PC. Devon had three tasks that were targeted for instruction. These included: (a) rolling silverware, Once the videos were compressed, each (b) sorting and sanitizing silverware, and (c) sequence was then placed on a presentation slide clocking in and out. Marcus also had three tasks using Pocket Slides™, version 4.0.100.1190 targeted for instruction including: (a) portioning presentation software (Conduits Technologies, recipes (i.e., 4 oz house salads, 8 oz salads, 8 oz Inc., 2005). This presentation software was used stir fry veggies, & 5 oz side veggies); (b) clocking to make it easier for the participants to select the in and out; and (c) cleaning and sanitizing his correct video that corresponded with the tasks work space. Task analyses of each task are they were to perform at work. The slides also had available from the first author. text above the pictures to describe the content of the video file (e.g., rolling silverware). Instructional Materials Unfortunately, we were unable to find An HP iPAQ hq2700 series Pocket PC™ (that presentation software that would allow for videos operated with the Pocket PC 2003, to be embedded within the slideshows. Unlike 2nd edition software™) was used as the PowerPoint™, which can embed videos within prompting device. Videos were taped in each the presentation, the software for handheld participants’ vocational site and were comprised devices at the time the study was conducted did of a combination of “self” models (the not have the capabilities to do so. In order to participant performing the task), and “other” view the videos, the software had to open models (both male and female adults that were ™, and then that had to either one of the first two authors or another be closed to return to the slideshow. Originally, employee). Videos were edited using Pinnacle we intended to use video prompting to present Studio 8™ (Pinnacle Systems, 2002) and each clusters of steps (two to three steps) within the task was edited to show the entire sequence from skill sequence; however, that would have start to finish (e.g., clocking in sequence). Video involved frequent opening and closing of the segments were comprised of a combination of media player, thereby making it more difficult to wide angle (full view of the model in context) use. We felt that it would be easier to use the and zoom shots (showing the hands of the device for presenting full sequences of the task model). A photo of the most salient feature of prior to task engagement (video rehearsal) and the sequence (e.g., sliding employee card in also chose to use video rehearsal strategies due to computer) was “grabbed” out of the video and the nature of the tasks that were being placed at the beginning of each video segment performed. The participants needed the use of and voice over narration was added to highlight their hands for most aspects of their tasks. critical components of the task. Prior to videotaping each task, task analyses were written In addition to the video-based materials, to ensure that all of the steps would be additional visual supports were provided to assist represented in the video model. Each step in the Marcus with using the scale to portion food Assistive Technology Outcomes and Benefits / 33

Fall 2007, Vol.4, Num. 1 recipes. These support included small, color- was ineffective for each step in the skill sequence. coded Post-It® flags that were placed on the Maintenance sessions were conducted in the correct number of ounces needed for each recipe same manner as baseline sessions; however, a on Marcus’ scale. The Post-It® flags were added demonstration was not provided. beginning on session seven. This was done to highlight the correct number on the scale and to Instructional Procedures assist him with lining up the arrow within the acceptable “range” of ounces needed (rather than Participants were given the handheld device and aligning the arrow exactly on the number on the shown videos of each instructional sequence scale). prior to engaging in the required task. For example, when they first arrived at work, Design participants were met at the door by the researchers or job coach and were given the A multiple probe design across tasks was used to handheld device and shown the clocking in evaluate the effectiveness of the video-based sequence. Participants independently put on procedures on independent responding and headphones, held the device, and were prompted percentage of prompts and was replicated across to select and play the correct videos using a students (Gast, Skouge, & Tawney, 1984; Horner system-of-least prompts prompting hierarchy. & Baer, 1978). Baseline data were collected for all After viewing the videos, participants gave the three tasks prior to the introduction of the handheld back to the job coach and/or independent variable on the first task. Once the researcher, who then carried the device to the students met criterion on the first task (i.e., 80% location of the next task. When it was time for correct independent responding for three them to switch to a different job task, consecutive sessions), baseline probes were then participants were prompted to get the handheld conducted on the other two tasks in the second and the video sequences prior to engaging and third tier. Baseline probes were again in the next task. They were only given prompts implemented for the third task once participants when necessary. met criterion on the second task. It should be noted that although Marcus had Baseline and Maintenance Procedures four different recipes to portion (and a different video for each), they were all very similar in that Baseline consisted of observing participants they involved weighing food and placing it in engaging in their assigned tasks without access to bags. There were only subtle differences between the handheld device. Data were collected only on the steps required to complete each recipe and the first attempt for each task (e.g., first attempt they varied in terms of the type of food that was at rolling silverware). During initial training, weighed, the number of ounces needed, and the participants were given several demonstrations types of baggies used. The recipes changed from their coworkers on how to perform each across days depending on the needs of the task in their job description (and videotaping was restaurant and he was often required to portion conducted during this time). Participants were two or three different recipes within one work then expected to attempt the tasks themselves session. Each recipe constituted a session and he and were given assistance from the researchers or may have had two or three recipe sessions in one job coach and this is when baseline data were day (on most days, he would portion only one or collected. During this time, participants were first two recipes). Similarly, Devon needed to shift given an opportunity to attempt the task back and forth between rolling and sorting and independently. If there was no attempt within sanitizing silverware throughout his shift. five seconds or an error was made, they were Depending on the level of customer volume, he given a verbal prompt, followed by a may have had three to four sessions of each of gestural/physical prompt if the verbal prompt these tasks within one shift. There were also only Assistive Technology Outcomes and Benefits / 34

Fall 2007, Vol.4, Num. 1 subtle differences between clocking in and Error correction. A two-level prompting clocking out. Therefore, clocking in constituted hierarchy was used during both baseline and one session and clocking out constituted another. instructional phases. In the event of an error or Participants were shown videos prior to no attempt within 5 seconds of the natural engagement with each task as they changed discriminative stimulus, participants were given a during their shifts. After viewing the video, the verbal prompt to respond. If the verbal prompt participants immediately engaged in the task and was not sufficient to prompt a correct response, were given prompts as needed. a gestural or physical prompt was provided (depending on what was necessary for the Training Participants to Use Technology/Photos particular step) to ensure correct responding. During instructional phases, participants were Prior to engaging in instructional sessions, each also shown a positive video model of the skill participant was given instruction on operating sequence following every fifth error in that the handheld device. One of the researchers met specific skill sequence. For example, when with participants individually and demonstrated portioning vegetables for the recipe task, if how to use the device using a model-lead-test Marcus made an initial error when putting on format. A video that was not used in the study gloves, and then made an error on the weighing (i.e., how to set an alarm clock) was used for step for the next four bags of vegetables, he was instruction on device operation; however, neither shown the video for that task. In essence, participant met criterion (which was set at 80% participants were given video feedback once independent correct responding for three there were five errors within a skill sequence, consecutive sessions), and both needed assistance however, verbal, gestural, or physical prompts in operating the device throughout the study. were given during the intervening errors; this was Unfortunately, there was very little time to done to reduce the number of times the provide instructional sessions on device participants needed to stop their work to watch operation prior to the start of the study. Parent the video. Five errors was arbitrarily selected as a permission slips were returned at the end of the point to provide video feedback and this was regular school year and students were out of decided prior to intervention. school for a couple of weeks prior to the beginning of the summer session, which is when Independent Variables the study took place. Both participants were hired during the summer session and began work Video rehearsal. Participants viewed a within two days of being hired. As a result, only video-based multimedia sequence on the three 20-min instructional sessions could be handheld device prior to engaging in the task. scheduled for each participant prior to intervention. Due to the relatively short period of Video feedback. Participants were shown a time between their being hired and beginning video of the skill sequence following every fifth work, neither participant met criterion on device error in the skill sequence. Participants were operation. Marcus needed prompting on how to shown the entire sequence and given positive use the device throughout the study. Devon verbal feedback from the researchers on steps learned to operate it independently by session 15, that were performed correctly and constructive but needed assistance if he accidentally selected feedback on steps that were performed the incorrect video or if he accidentally opened incorrectly. The only exception was that a special the wrong application. Although the timing of feedback video was created for Marcus that the study prevented sufficient instruction on showed adding or removing vegetables from the device operation, we are confident that both bag on the scale during the recipe sequences. He participants could have met criterion had there made several errors on that particular step, and been more time for instructional sessions. watching the full sequence (including hand-

Assistive Technology Outcomes and Benefits / 35

Fall 2007, Vol.4, Num. 1 washing, putting on gloves, etc.) seemed step) and multiplying by 100. Video feedback was unnecessary. not included in this total.

Data Collection Procedures Number of sessions to reach criterion. The acquisition criteria for each skill sequence was a During baseline, instructional, and maintenance score of 80% or higher for three consecutive phases, task analytic data were collected with sessions as measured by percentage of correct/incorrect responses and prompt levels independent correct responding following the being reported on each step of the skill sequence. introduction of the video-based materials. A ‘+’ was recorded for independent correct responses, a ‘-’ was recorded for incorrect Data Analysis attempts, an ‘n’ was recorded for no attempt, and a ‘√’ was recorded for each prompt given at each Experimental control was determined primarily step (with a maximum of two per step). In the through visual inspection of the data and through event that one of the steps was completed by a comparisons of means and trends for each phase coworker, an ‘NA’ was recorded and that step and changes in the level of data across phases. was not included in the total number of steps With the multiple probe design, experimental during data summarization. Data were recorded control is demonstrated by a consistent change in on the first attempt of the task immediately level and/or trend of the data from baseline following the video rehearsal procedures (e.g., phases to intervention phases and lack of the initial silverware roll). In addition, tic marks changes in the untreated behaviors (Wolery, were made on the data sheet after the initial Bailey, & Sugai, 1988). attempt of the task to keep track of errors (to know when to provide video feedback). Reliability

Dependent Measures Reliability sessions were conducted on 27% of all sessions for Devon and 45% of all sessions for Percentage of independent correct responses. Marcus (including baseline and instructional Participants were assessed on how independently sequences). The percentage agreement index (i.e., they performed the skills selected for instruction number of agreements divided by number of prior to engaging in the instructional sequences agreements plus disagreements and multiplied by (baseline), during instruction, and following 100) was used to calculate inter-observer instruction (maintenance for Marcus). The score agreement. A research protocol was written and was determined by dividing the number of steps shared with all observers regarding methodology with independent responding by the total and data collection procedures. All observers number of steps in the skill sequence and practiced data collection during practice sessions multiplying by 100. Baseline and maintenance in the restaurant settings until they achieved sessions differed from instructional sessions in scores of 90% agreement or higher for three that participants were expected to perform the consecutive sessions. skill without viewing videos on the handheld device. Agreement for correct responding for Devon across sessions resulted in a mean score of 98% Percentage of prompts. Participants were (range = 92-100) and agreement for prompts assessed on the number of external prompts they resulted in a mean score of 95% (range = 85- needed to complete the skill sequence during all 100). Agreement for correct responding for phases of the study. The score was determined Marcus across sessions resulted in a mean score by dividing the number of prompts given by the of 97% (range = 90-100) and agreement for total number of prompts possible (i.e., two per prompts resulted in a mean score of 95% (range

Assistive Technology Outcomes and Benefits / 36

Fall 2007, Vol.4, Num. 1 Figure 1. Percentage of independent correct responses for Devon.

= 86-100). In addition, the second observer Results collected procedural reliability data (Billingsley, White, & Munson, 1980). These measures The introduction of the video-based materials included the following: (a) checking to ensure appeared to be associated with an increase in that the correct video was shown for the independent responding and a reduction in intended task for each participant, (b) checking prompting for both participants. to determine if video feedback was delivered following five errors, and (c) checking that the Percentage of Independent Correct Responses prompting hierarchy was delivered as intended. Reliability was calculated by dividing number of Figure 1 presents data for Devon’s independent correct measures by total number of assessed correct responses. Although the data were variables and multiplying by 100. Procedural ascending slightly in the second and third tier, the reliability agreement averaged 100%. introduction of the video-based materials were associated with a marked change in level from baseline to instructional phases with a rapid Assistive Technology Outcomes and Benefits / 37

Fall 2007, Vol.4, Num. 1 increase in correct responding following one to had large increases with the mean baseline being three sessions with video instruction. However, 12% as compared to 93% during the there were no dramatic increases with the other instructional phases. Unfortunately, no tasks during this time, which suggests that the maintenance data were collected for Devon presentation of the videos were responsible for because he was laid off for a while and obtained the increase in correct responding. When the a different job before being hired back at Red means of each phase were compared, there were Robin. large differences in the percentage of correct independent responding for each task across Figure 2 presents a graphic representation of data baseline and instructional phases. For the rolling for Marcus’ independent correct responses. silverware task, the baseline mean was 0% as Although the data were ascending slightly in the compared to 88% during the instructional phase, baseline of the second tier, the introduction of while the baseline mean for sorting and sanitizing the video-based materials were associated with a task was 5% as compared to 96% during the marked change in the level of data from baseline instructional phase. The clocking in/out task also to intervention. During the instructional phase Figure 2. Percentage of independent correct responses for Marcus.

Assistive Technology Outcomes and Benefits / 38

Fall 2007, Vol.4, Num. 1 for recipes (top tier), Post-It® flags were placed associated with the number of ounces needed on Marcus’ recipe sheet by staff and he was and it was important for him to move quickly. expected to place the flag on the corresponding With the Post-It® flags, he needed to have the number on the scale. This support was added arrow somewhere within the area covered by the beginning with session number 7 because he was flag. In addition, a feedback video was created having a great deal of difficulty judging if he had that provided specific instructive information for “enough” or “not enough” food in his portion that step. These videos demonstrated that food bag and was making frequent errors on this step. needed to be added if the arrow was “before” the We felt that it was necessary to highlight the flag, and that some food needed to be removed if correct number of ounces needed and to give the arrow was “past” the flag. Although the him an acceptable “range” in which to align the addition of the Post-It® flag and special arrow of the scale. There were times when he feedback video reduced subsequent errors on would spend an inordinate amount of time trying that step, prior to their introduction, there was to align the arrow with the exact tic mark still an abrupt change in level of data between the

Figure 3. Percentage of prompts for Devon.

Assistive Technology Outcomes and Benefits / 39

Fall 2007, Vol.4, Num. 1 baseline and intervention phase. Also, the 8% as compared to 91% during the instructional addition of these supports only affected one step phase. The cleaning task also had large increases in the sequence and did not influence the scores with the mean baseline being 10% as compared significantly, suggesting that the video materials to 79% during the instructional phases. In were still responsible for the rapid increase in addition, performance was maintained on all correct responding. When the means of each three tasks when assessed 12 weeks following phase were compared, there were large intervention. differences in the percentage of correct independent responding for each task across Percentage of Prompts baseline and instructional phases. For the recipe task, the baseline mean was 12% as compared to When the percentage of prompts were analyzed, 92% during the instructional phase, while the there was also a rapid reduction in prompts for baseline mean for the clocking in/out task was both participants once the video-based materials

Figure 4. Percentage of prompts for Marcus.

Assistive Technology Outcomes and Benefits / 40

Fall 2007, Vol.4, Num. 1 were presented. Figure 3 presents a graphic reduced when assessed during the 12-wk representation of Devon’s percentage of prompts maintenance probe. during baseline and intervention phases. Although the data were descending somewhat in Sessions to Criterion as Measured by Independent Correct tiers two and three, there were changes in the Responding level of data with each introduction of the independent variable followed by a rapid In order to meet criterion, participants had to reduction in prompts by the second or third score at least 80% or higher as measured by session with the video-based materials. When the independent correct responding for each task. means of each phase were compared, there were Both participants required an average of five large reductions in the percentage of prompts for sessions to reach criterion with the tasks once the each task across baseline and instructional handheld devices were introduced, which phases. For the rolling silverware task, the suggests that the video rehearsal and feedback baseline mean was 80% as compared to 7% were very powerful instructional techniques. during the instructional phase; the mean baseline for sorting and sanitizing task was 65% as Outcomes and Benefits compared to 3% during the instructional phase, and the clocking in/out task also had large The purpose of this research was to determine if decreasees with the mean baseline being 84% as video modeling and video feedback, when compared to 5% during the instructional phases. presented on a portable handheld device, would increase independent responding of two Figure 4 presents a graphic display of Marcus’ individuals who were employed in community- percentage of prompts and like Devon, he also based environments. Results of this investigation had a drastic decrease in the percentage of indicate that the introduction of the video-based prompts needed from baseline to intervention materials was associated with an increase in phases. These results are somewhat tempered by independent responding and a reduction in the fact that the data are gradually decreasing in prompting for both participants. In addition, each baseline prior to the introduction of the both participants met criterion with each of their independent variable; he was obviously relying three tasks very quickly once the handheld was less on staff to perform the skills (i.e., he needed introduced, suggesting that this was a very fewer gestural/physical prompts to complete powerful instructional tool. In terms of social some of the steps). However, there were abrupt validation, both participants informally indicated changes in the level of the data for all three tasks that they liked using the handheld and Devon when the video-based materials were introduced, stated that he would like to participate in future which provides evidence that their introduction studies and both participants indicated that they was responsible for the reduction in prompts liked watching videos. The employers and other needed to perform the skills. When the means of coworkers also communicated that they believed each phase were compared, there were large the handheld device was a very beneficial tool. reductions in the percentage of prompts for each task across baseline and instructional phases. For Although the intervention appeared to be the recipe task, the baseline mean was 83% as effective, there are several limitations to this compared to 11% during the instructional phase; study. First, prior to the introduction of the the mean baseline for the clocking in/out task video-based materials, some of the baselines were was 80% as compared to 6% during the slightly ascending which would indicate that the instructional phase, and the cleaning task also participants were learning to perform the skills had large reductions with the mean baseline with prompting alone. However, once the being 81% as compared to 11% during the intervention was applied to each task, there was a instructional phases. Prompts were further marked change in the level of data from baseline to instructional phases followed by a rapid Assistive Technology Outcomes and Benefits / 41

Fall 2007, Vol.4, Num. 1 increase in correct independent responding and a system was what ultimately prevented the decrease in percentage of prompts required for participants from operating the device each participant across all tasks. This change in independently. Prior to initiating this study, we level followed by rapid changes in behavior only attempted to address this issue by investigating in the instructional phases suggests that the various software applications for handheld application of the independent variable was devices to see if there was presentation software responsible for those changes and experimental that had the capability to embed video files control was established. Second, because there within slideshows to make the navigation easier was no component analysis conducted, it is for the participants. Unfortunately, software that difficult to determine if the video rehearsal or is available for handheld devices is still somewhat video feedback conditions were responsible for limited in terms of its compatibility and changes in behavior, or if the combination of the functionality in operating video files. As a result, two were responsible for the change in student participants were required to view video files in responding. Future research should be done to Windows Media Player™ and then had to close compare the effectiveness of the two procedures the player to return to the slide show, making the to determine if they are equally as effective when navigation somewhat more difficult. However, presented in isolation. Third, although the video- we are confident that with advances in based instructional sequences increased technology, handheld devices will eventually independent responding across tasks, they were function as effectively as or desktop used in conjunction with a prompting system computers and that software will continue to (i.e., verbal and gestural/physical prompts). As a improve, making independent operation of the result of using this combination, it cannot be devices more likely among individuals with stated that the video procedures alone were intellectual impairments. Although the navigation responsible for the changes in student systems may need to be simplified in order to responding. However, prior to the introduction promote independent use of handheld devices, of the video-based materials, neither of the presenting video-based materials within the participants acquired the targeted skills when employment setting appeared to be very taught with prompting alone. Therefore, it seems effective. likely that the video-based materials did have an impact on how quickly participants acquired the In particular, we believe that the video rehearsal skills. And finally, both participants needed and feedback strategies were very beneficial for assistance operating the technology which focusing the participants’ attention on a model ultimately made them dependent on staff. immediately prior to task engagement and following errors, especially when the tasks Unfortunately, there was insufficient training changed frequently within one work session. time with the device due to the brief period of Because this study was conducted in a time between the participants being hired and competitive employment setting, participants their first day of employment. Devon was were required to perform tasks that changed eventually able to operate the device by the 15th based on the immediate needs of the session, but still required assistance if an error employment setting. For example, Marcus had a was made in application selection or if he difficult time remembering the subtle differences selected the incorrect video for the task. Marcus between the recipes and would make frequent had a great deal of difficulty operating the device errors as soon as he switched to the new recipe. and needed assistance throughout the entire His errors decreased considerably after he was study (primarily with selecting the correct given the opportunity to view the videos prior to application and the correct video). Lack of switching the the new recipes and following training time and limited exposure to the use of feedback trials, which ultimately led to an handhelds contributed to these difficulties, but increase in independent responding. we believe that the complexity of the navigation Assistive Technology Outcomes and Benefits / 42

Fall 2007, Vol.4, Num. 1

One of the reasons we believe the videos were so and/or adaptations to the environment and/or effective is because the participants were able to job responsibilities may need to be adjusted. focus their attention on the relevant features of the task in a very controlled manner. By focusing In conclusion, video rehearsal and video their attention on the handheld screen and feedback strategies, when delivered on a listening to the narration through headphones, handheld device, appeared to be effective in we were able to reduce the distractions that were promoting independent responding for the abundant in the employment settings (e.g., loud individuals who participated in this study. music, coworkers). In addition, we believe that Hopefully, with continued research and the way the videos were created also assisted with advancement in technology, presenting video- focusing the participants’ attention on the critical based materials on portable devices will become dimensions of the tasks. For example, when a viable instructional technique. creating the videos, we zoomed in on the salient features of the task (e.g, the arrow on the scale) References to ensure that the students were attending to the correct and relevant stimuli. In doing so, we were 8864soft.com. (2007). Easy able to focus the participants’ attention on the AVI/VCD/DVD/MPEG converter software correct model while also reducing the distracting [Free download computer software]. stimuli in the environment. Using zoom shots Retrieved June 18, 2007, from were also important due to the fact that the http://8864soft.com/avi_dvd_mpeg_conver handheld devices have smaller screens. It was ter/index.html important to make the videos as large as possible Agran, M. (1997). Student-directed learning: Teaching so that the participants could see the modeled self-determination skills. Pacific Grove, CA : tasks clearly. Widescreen shots were only used Brookes/Cole. when the tasks required the learner to move from Alberto, P. A., Cihak, D. F., & Gama, R. I. one location to another and when it was (2005). Use of static picture prompts versus important for them to pay attention to the video modeling during simulation instruction. environmental cues (e.g., carrying the sorted Research in Developmental Disabilities, 26, 327- silverware to the dishwashing station). In 339. addition, we also made the videos as short as Ayres, K. M., & Langone, J. (2002). Acquisition possible to maintain the participants’ attention and generalization of purchasing skills using a and to reduce the amount of time spent using the video enhanced computer-based instructional handheld device. package. Journal of Special Education Technology, 17, 15-28. Even though technology was beneficial, the value Bidwell, M. A., & Rehfeldt, R. A. (2004). Using of natural supports in the environment are also video modeling to teach a domestic skill with critical. Perhaps using a combination of natural an embedded social skill to adults with severe supports as well as technology-enhanced mental retardation. Behavioral Interventions, 19, prompting is the answer. In several work 263-274. environments, it is quite natural for employees to Billingsley, F. F., White, O. R., & Munson, R. work cooperatively with coworkers. It is in the (1980). Procedural reliability: A rationale and best interest for workers with disabilities to an example. Behavioral Assessment, 2, 229-241. establish, build, and access naturally occurring Branham, R. S., Collins, B. C., Schuster, J. W., & supports and relationships within the vocational Kleinert, H. (1999). Teaching community realm because these supports are already skills to students with moderate disabilities: available. Coworkers could conceivably assist in Comparing combined techniques of the operation of the devices if necessary. As in classroom simulation, videotape modeling, most employment settings, minor modifications and community-based instruction. Education and Training in Mental Retardation and Assistive Technology Outcomes and Benefits / 43

Fall 2007, Vol.4, Num. 1

Developmental Disabilities, 43, 170-181. research in special education (pp. 226-268). Cihak, D. F., Alberto, P. A., Kessler, K., & Columbus, OH: Merrill. Taber, T. A. (2004). An investigation of Goldsworthy, R. C., Barab, S. A., & instructional scheduling arrangements for Goldsworthy, E. L. (2000). The STAR community-based instruction. Research in project: Enhancing adolescents’ social Developmental Disabilities, 25, 67-88. understanding through video-based, Cihak, D. F., Alberto, P. A., Kessler, K., Taber- multimedia scenarios. Journal of Special Doughty, T., & Gama, R. I. (2006). A Education Technology, 15, 13-26. comparison of static picture prompting and Graves, T. B., Collins, B. C., Schuster, J. W., & video prompting simulation strategies using Kleinert, H. (2005). Using video prompting group instructional procedures. Focus on to teach cooking skills to secondary students Autism and Other Developmental Disabilities, 21, with moderate disabilities. Education and Training 89-99. in Developmental Disabilities, 40, 34-46. Conduits Technologies. (2005). Pocket slides Horner, R. D., & Baer, D. M. (1978). Multiple- presentation software [Computer software]. New probe technique: A variation of the multiple York: Author. baseline design. Journal of Applied Behavior Copeland, S. R., & Hughes, C. (2000). Analysis, 11, 189-196. Acquisition of a picture prompt strategy to Lancioni, G. E., O’Reilly, M. F., Seedhouse, P., increase independent performance. Education Furniss, F., & Cuhna, B. (2000). Promoting and Training in Mental Retardation and independent task performance by persons Developmental Disabilities, 35, 294-305. with severe developmental disabilities Davies, D. M., Stock, S., & Wehmeyer, M. L. through a new computer-aided system. (2002a). Enhancing independent task Behavior Modification, 24, 700-718. performance for individuals with mental Lasater, M. W., & Brady, M. P. (1995). Effects of retardation through use of a handheld self- video self-modeling and feedback on task directed visual and audio prompting system. fluency: A home-based intervention. Education and Training in Mental Retardation and Education and Treatment of Children, 18, Developmental Disabilities, 37, 209-218. 389-407. Davies, D. M., Stock, S., & Wehmeyer, M. L. Martin, J. E., Mithaug, D. E., & Burger, D. L. (2002b). Enhancing independent time (1990). Effects of visual cues upon the management and personal scheduling for vocational task performance of students with individuals with mental retardation through mental retardation. Exceptionality, 1, 41-59. use of a palmtop visual and audio prompting Martin, J. E., Mithaug, D. E., & Frazier, E. S. system. Mental Retardation, 40, 358-365. (1992). Effects of picture referencing on Epstein, J. N., Willis, M. G., Conners, C. K., & PVC chair, loveseat, and settee assemblies by Johnson, D. E. (2001). Use of a technological students with mental retardation. Research in prompting device to aid a student with Developmental Disabilities, 13, 267-286. attention deficit hyperactivity disorder to Mechling, L. C. (2004). Effects of multi-media, initiate and complete daily tasks: An computer-based instruction on grocery exploratory study. Journal of Special Education shopping fluency. Journal of Special Education Technology, 16(1), 19-28. Technology, 19, 23-34. Fisher, M. K. (1984). Vocational assembly skills Mechling, L. C., & Gast, D. L. (2003). Multi- using isometric projection exploded view media instruction to teach grocery word assembly drawing for mentally handicapped associations and store location: A study of students. Education and Training of the Mentally generalization. Education and Training in Retarded, 19, 285-290. Developmental Disabilities, 38, 62-76. Gast, D. L., Skouge, J. R., & Tawney, J. W. Morgan, R. L., & Salzberg, C. L. (1992). Effects (1984). Multiple baseline designs. In J. W. of video-assisted training on employment- Tawney & D. L. Gast (Eds.), Single subject related social skills of adults with severe Assistive Technology Outcomes and Benefits / 44

Fall 2007, Vol.4, Num. 1

mental retardation. Journal of Applied Behavior Taber, T. A., Alberto, P. A. & Frederick, L. D. Analysis, 25, 365-383. (1998). Use of self-operated auditory Nikopoulos, J. M., & Keenan, M. (2004). Effects prompts by workers with moderate mental of video modeling on social initiations by retardation to transition independently children with autism. Journal of Applied through vocational tasks. Research in Behavior Analysis, 37, 93-96. Developmental Disabilities, 19, 327-345. Norman, J. M., Collins, B. C., & Schuster, J. W. Van Laarhoven, T., & Van Laarhoven-Myers, T. (2001). Using an instructional package (2006). A comparison of three video-based including video technology to teach self-help instructional procedures for teaching daily skills to elementary students with mental living skills to persons with developmental disabilities. Journal of Special Education disabilities. Education and Training in Technology, 16, 5-18. Developmental Disabilities, 41, 365-381. Pinnacle Systems. (2002). Pinnacle Studio 8 Wacker, D. P., & Berg, W. K. (1983). Effects of [Computer Software]. Tewksbury, MA: picture prompts on the acquisition of Author. complex vocational tasks by mentally Psychological Corporation. (1999). Wechsler retarded adolescents. Journal of Applied abbreviated scale of intelligence. San Behavior Analysis, 16, 417-433. Antonio, TX: Author. Wechsler, D. (1997). Wechsler adult intelligence scale– Rehfeldt, R. A., Dahman, D., Young, A., Cherry, Third edition. San Antonio, TX: Psychological H., & Davis, P. (2003). Teaching a simple Corporation. meal preparation skill to adults with Wolery, M., Bailey, D. B., & Sugai, G. M. (1988). moderate and severe mental retardation using Effective teaching: Principles and procedures of video modeling. Behavioral Interventions, 18, applied behavior analysis with exceptional students. 209-218. Needham, MA: Allyn and Bacon. Riffel, L. A., Wehmeyer, M. L., Turnbull, A. P., Lattimore, J., Davies, D., Stock, S., et al. (2005). Promoting independent performance of transition-related tasks using a -based self-directed visual and auditory prompting system. Journal of Special Education Technology, 20(2), 5-14. Shipley-Benamou, R., Lutzker, J. R., & Taubman, M. (2002). Teaching daily living skills to children with autism through instructional video modeling. Journal of Positive Behavior Interventions, 4, 165-175. Sigafoos, J., O’Reilly, M., Cannella, H., Upadhyaya, M., Edrisinha, C., Lancioni, G. E., Handley, A., Andrews, A., Garver, C., & Young, D. (2005). Computer-presented video prompting for teaching microwave oven use to three adults with developmental disabilities. Journal of Behavioral Education, 14, 189-201. Simpson, A., Langone, J., & Ayres, K. M. (2004). Embedded video and computer based instruction to improve social skills for students with autism. Education and Training in Developmental Disabilities 39, 240-252. Assistive Technology Outcomes and Benefits / 45