USING VISUAL ACTIVITY SCHEDULES EMBEDDED WITH VIDEO MODELS TO 
INCREASE SOCIAL INTERACTION IN CHILDREN WITH AUTISM 

By 

Kelsey Hetherington 

A THESIS 

Submitted to 
Michigan State University 
in partial fulfillment of the requirements  
for the degree of 

Applied Behavior Analysis – Master of Arts 

2025 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
ABSTRACT 

Individuals with Autism Spectrum Disorder (ASD) often struggle with social interaction 

and communication, limiting their ability to form relationships and engage in meaningful 

exchanges. Activity schedules provide structured visual supports that promotes independence 

and social engagement. While research supports their effectiveness, few studies have examined 

the integration of video modeling within these schedules to enhance social interactions. This 

study evaluated the effectiveness of embedding video models into activity schedules to increase 

social interactions to children with ASD using a multiple probe across participants design in an 

Early Intensive Behavioral Intervention (EIBI) clinic. Results showed Noah demonstrated an 

increase in independently performed social interaction components, improving from 0 (range 0–

0) at baseline to an average of 4 (range 0–6) during intervention. Morgan showed no change, 

maintaining 0 (range 0–0) throughout all phases. These findings suggest video modeling may 

facilitate social skill acquisition for some but not all learners. Future research should explore 

individual learner characteristics, long-term effectiveness, and generalizability. 

Keywords: photo activity schedules, social interaction, autism, video modeling

 
 
TABLE OF CONTENTS 

INTRODUCTION………………………………………………………………………………...1 

METHOD…………………………………………………………………………………………6 

RESULTS……..………….…………………….……..…….…………………………………...20 

DISCUSSION…...…………………………………………….…………………………………23 

REFERENCES…………………………………………………………………………………..26 

APPENDIX A: IOA DATA SHEET FOR BASELINE AND INTERVENTION...…….……….28 

APPENDIX B: TREATMENT FIDELITY DATA SHEET…….………..………….…………..30 

APPENDIX C: DATA COLLECTION DATA SHEET…….……………..….…………………31 

iii 

 
 
 
 
 
INTRODUCTION 

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by 

challenges in social communication and interaction (American Psychological Association [APA], 

2023). Children and adolescents diagnosed with ASD often experience difficulties in navigating 

social interactions and engaging in age-appropriate activities (APA, 2023). Without appropriate 

intervention, these core deficits are likely to persist and may worsen over time (Osos et al., 

2021). Numerous interventions have been developed to teach new skills to children with ASD, 

aiming to address these challenges and enhance overall functioning (Osos et al., 2021; Wichnick-

Gillis et al., 2016; Brodhead et al., 2014; Betz et al., 2008; Bellini & Akullian, 2007). 

In recent years, there has been an increased focus on the development of effective 

interventions that support individuals with ASD in acquiring critical social and adaptive skills. 

Researchers and practitioners have explored various strategies aimed at addressing these 

challenges while fostering meaningful social engagement and skill development. Teaching social 

skills to individuals with ASD is essential not only for their social and emotional growth but also 

for their ability to integrate into broader social, academic, and community settings (Osos et al., 

2021). Early intervention focused on social skill development enhances communication abilities, 

which in turn helps individuals express their needs, thoughts, and emotions, while also 

promoting relationship-building (McClannahan & Krantz, 2010). Improved social skills 

contribute to a better quality of life, facilitating greater participation in social activities, 

increasing independence, and reducing social anxiety (McClannahan & Krantz, 2010). 

Among the various evidence-based interventions, activity schedules (Bryan et al., 2000; 

Betz et al., 2008; Carlile et al., 2013; Brodhead et al., 2018) and video modeling techniques 

(Bellini & Akullian, 2007; Plavnick et al., 2015; Osos et al., 2021) have emerged as promising 

1 

approaches for promoting social engagement, communication, and adaptive behaviors in children 

and adolescents with ASD. Both activity schedules and video modeling provide structured, 

visually supported frameworks for teaching new skills and facilitating social interactions (Betz et 

al., 2008; Carlile et al., 2013; Osos et al., 2021). Meta-analytic studies have examined the 

effectiveness of these interventions, contributing to a growing body of literature that underscores 

their potential benefits (Bellini & Akullian, 2007; Plavnick et al., 2015). 

Activity Schedules  

Activity schedules are evidence-based practices that can be implemented in various 

formats and adaptations. These schedules serve as written or visual supports designed to help 

learners’ complete tasks independently, engage in social interactions, and promote on-task 

behavior (Bryan et al., 2000; McClannahan & Krantz, 2010). Formats for activity schedules 

include written or typed lists, binders with individual pages containing pictures or words 

representing activities, audio-taped schedules, or video modeling schedules (McClannahan & 

Krantz, 2010). Each schedule incorporates different textual cues, such as pictures, symbols, 

sound cues, or video prompts, to function as a discriminative stimulus for the sequence of 

behaviors necessary to complete an activity (Osos et al., 2021). 

Research has demonstrated that activity schedules are effective in enhancing on-task 

behavior (Bryan et al., 2000), improving on-schedule behavior (Bryan et al., 2000), and 

promoting independence in children with autism (Bryan et al., 2000; Osos et al., 2021). They 

have been successfully used to teach routines, leisure skills (Carlile et al., 2013), facilitate varied 

applications (Brodhead et al., 2018), and encourage peer engagement (Betz et al., 2008). 

Additionally, activity schedules have been shown to support social interaction among 

children with ASD (Betz et al., 2008; Wichnick-Gillis et al., 2016; Osos et al., 2021). For 

2 

example, Betz et al. (2008) explored the use of joint activity schedules to foster engagement 

between students with autism during interactive play. Unlike traditional activity schedules, joint 

activity schedules require two participants to follow different, yet synchronized schedules. This 

method was shown to increase peer engagement and the number of games completed for two out 

of three pairs of participants, with skills maintained during generalization (Betz et al., 2008). 

In another study, Wichnick-Gillis et al. (2016) incorporated scripts into photographic 

activity schedules to teach young children with autism key social interaction skills. Using a 

script-fading procedure (Brown et al., 2008), the study focused on teaching students how to 

approach peers, initiate conversations, orient to speakers, wait while others spoke, and respond 

reciprocally. The results indicated that embedding social interaction cues within the natural 

environment—through visual supports and prompts—enhanced the success of teaching these 

skills. Additionally, script fading was found to be effective in transferring these learned behaviors 

to real-world social settings (Wichnick-Gillis et al., 2016). 

Video Modeling 

Video modeling is another effective tool for teaching social behavior to children with 

ASD (Bellini & Akullian, 2007). Video modeling involves the learner watching a video that 

demonstrates the target behavior and then having the opportunity to perform the behavior 

themselves (Osos et al., 2021). Bellini and Akullian (2007) conducted a meta-analysis 

synthesizing findings from multiple studies on video modeling and self-modeling interventions 

for children and adolescents with ASD, highlighting significant improvements in various 

domains, including social skills. Osos et al. (2021) further noted that video modeling offers 

several advantages in educational settings, such as fostering creativity, ensuring consistency, and, 

in some cases, promoting generalized skills. 

3 

Technological advancements have also led to the integration of mobile devices in video 

modeling and activity schedules. Carlile et al. (2013) explored the use of iPod Touch devices to 

implement activity schedules for teaching leisure skills to children with ASD. Their findings 

demonstrated the feasibility and effectiveness of using technology-based interventions to 

promote skill acquisition and engagement, emphasizing the potential of mobile devices as 

portable and personalized tools for intervention. 

The rise of technology also facilitated the development of technology-based activity 

schedules (Plavnick et al., 2015; Osos et al., 2021), which incorporate digital elements such as 

pictures, text, and video clips to aid in task completion. These schedules have been shown to 

increase independent engagement in leisure activities in classroom settings (Carlile et al., 2013). 

Osos et al. (2021) examined the use of tablet technology and video-enhanced activity 

schedules to promote social interactions in children with ASD. Their study found that video-

enhanced schedules were effective in teaching preschoolers to self-initiate social interactions 

with peers when combined with least-to-most prompting. In this study, Osos et al. (2021) sought 

to determine whether integrating video modeling into activity schedules would accelerate the 

acquisition and generalization of social skills compared to using static images alone. The study 

involved four participants, all diagnosed with ASD, who were enrolled in an early intensive 

behavioral intervention (EIBI) program. The researchers employed electronic activity schedules 

created using Apple Keynote applications (Brodhead et al., 2018; Osos et al., 2021), with an 

alternating treatment design to compare the effects of video-enhanced schedules versus 

electronic schedules featuring static images (Osos et al., 2021). 

The electronic activity schedule in the study featured pre-determined slides with static 

images on a light blue background, whereas the video-enhanced schedule used video models as 

4 

visual prompts on a red background (Osos et al., 2021). Osos et al. (2021) assessed five 

conditions: baseline, activity schedule probe, electronic activity schedule, video-enhanced 

activity schedule, and transfer of training. Results indicated that video-enhanced activity 

schedules, when paired with video modeling, significantly improved social interactions in 

children with ASD. While electronic schedules alone benefited some participants, video 

modeling was found to be particularly effective in teaching children to initiate social interactions 

(Osos et al., 2021). 

Despite these positive outcomes, Osos et al. (2021) acknowledged limitations, including 

the small sample size, variability in participant outcomes, and the lack of improvements in social 

competence. To address these issues, they recommended increasing ecological validity by 

teaching varied social skills across multiple contexts. The current study seeks to extend these 

findings by targeting three different types of social interactions, rather than just one. Rather than 

focusing on teaching isolated social skills, this study emphasizes increasing the occurrence and 

variety of social interactions among young children with ASD. By expanding on the work of 

Osos et al. (2021), this study aims to contribute to ongoing efforts to refine the effectiveness and 

applicability of video modeling paired with activity schedules in increasing social interactions in 

children with ASD. 

The primary research question guiding this study is: To what extent does video modeling 

embedded in photo-based activity schedules increase social interactions among young children 

with ASD? 

5 

 
 
 
Participants 

METHOD 

Participants included two individuals with a medical diagnosis of ASD. Participants were 

recruited through the help of a Board-Certified Behavior Analyst (BCBA) and staff at a 

university-based early intensive behavioral intervention (EIBI) clinic for children with ASD. 

Two additional participants were initially enrolled in the study but were removed by the 

experimenter due to engaging in behaviors that interfered with conducting research session. 

Participants had a mean age of three and a half, and ages ranged from three to four. All 

participants attended an EIBI clinic where they received 30 hours of applied behavior analysis 

(ABA) therapy weekly.  

All participants met a series of inclusion criteria: (a) had previous imitation skills, (b) 

ability to follow a three-to-four-page activity schedule, (c) ability to follow a video model and 

(d) had a form of functional communication modality in English (verbal, Picture Exchange 

Communication, AAC device, etc.). 

Consent was obtained from all participant’s caregivers. To protect the privacy of 

participants, they will be given alias names.  

Participant 1 was Noah, a three-year-old male diagnosed with ASD. Based on his most 

recent Verbal Behavior Milestones Assessment and Placement Program (VB-MAPP) assessment, 

he achieved a total score of 131. His strongest skill areas included manding, tacting, play, social 

skills, and listener responding. 

Participant 2 was Morgan, a four-year-old female diagnosed with ASD. According to her 

most recent VB-MAPP assessment, she obtained a total score of 63. Her highest-performing 

areas were reading, math, and play. 

6 

Participant 3 was Michael, a four-year-old boy diagnosed with ASD. Michael was 

withdrawn from the study due to the emergence of behaviors that suggested he was distressed 

upon seeing the video camera. When the camera was directed toward him, Michael would 

immediately begin to cry, scream, and drop to the floor. These behaviors occurred reliably during 

baseline, which led the researchers withdrawing Michael from the study. 

Participant 4 was Mara, a four-year-old girl also diagnosed with ASD. Mara was 

withdrawn from the study due to engaging in screaming, crying, and hitting her behavior 

technician when corrected for an error in completing the activity schedule. These behaviors 

occurred reliably during baseline sessions which led to the researchers withdrawing Mara from 

the research study.  

Setting and Materials  

This study was conducted in a classroom within an Early Intensive Behavioral 

Intervention (EIBI) clinic housed in a public-school building. The classroom supported eight 

children diagnosed with ASD, each receiving one-on-one instruction from one of eight behavior 

technicians (BTs) under the supervision of a Board-Certified Behavior Analyst (BCBA). BTs 

were assigned to work with one child for half the day and a different child for the second half. 

The classroom environment was structured to promote individualized learning and 

engagement. It featured an open play area, a reading corner, two large group tables, a bookshelf, 

and a dedicated shelf for activity schedule materials. Each child and their assigned BT had a 

designated learning area equipped with a table, two chairs, and a three-tier cart containing 

program materials and reinforcers. This setup ensured accessibility to necessary resources while 

facilitating structured instruction. 

7 

A handheld camera was used to record each session for later analysis of treatment fidelity 

and reliability. Additionally, the researcher collected real-time data using a self-constructed 

datasheet and a writing utensil. The Photo Activity Schedule (PAS) datasheet (Appendix C) was 

used to track participants’ progression through their activity schedules and video modeling tasks. 

The PAS recorded specific behaviors such as opening or turning pages, looking, or pointing, 

obtaining items, completing activities, and putting materials away. Additionally, social 

interaction components, including approaching a partner, orienting the body, emitting responses, 

and ending interactions, were documented and scored. 

Each session began with a preference assessment to identify motivating reinforcers for 

participants, which varied between edible and tangible items. Discrete trial activities 

incorporated into the study included a ring stacker, shape sorter, puzzles (ranging from four to 

twelve pieces), sorting bears, and stringing beads. These materials were stored on a designated 

shelf in the classroom for easy access. 

Activity Schedule  

The electronic activity schedules used in this study were created on an Apple® iPad, with 

individual slides designed using Microsoft PowerPoint. Each slide featured a white background 

to maintain a consistent and visually clear format. The first slide of the activity schedule 

displayed the participant’s name in black text, ensuring personalization and ease of identification. 

Participants could navigate through the schedule using an arrow located in the bottom right 

corner of each slide. Upon advancing to the next slide, either a discrete trial activity or a social 

interaction with a peer would be presented in video format against the white background. The 

specific activities included in the schedule were selected using a random number generator, 

where each discrete trial activity was assigned a number between one and five. The researcher 

8 

used an online random number generator to determine which activity would be included on a 

given slide. 

Additionally, video models were incorporated into the activity schedules to demonstrate 

target behaviors. These video models were created using Microsoft PowerPoint and consisted of 

slides displaying the participant’s name, two leisure activities (see Table 1.1 for definitions), and 

a corresponding video model. Each slide featured a white background with black text at the top 

indicating the activity. For example, a slide labeled “Puzzle” would include the activity name at 

the top, followed by an image of the puzzle beneath it. This structured and visually supported 

format ensured clarity and ease of use for participants while reinforcing the learning objectives 

of the study. 

Table 1: Lesisure Activities for Data Collection.  

Activity  

Ring Stacker 

Definition/ Description  

Recommended age: 12 months and older. 

White, long, plastic center piece that five 

colorful (red, orange, yellow, green, blue) 

rings that are placed on by size (largest to 

smallest). Individuals are expected to take the 

rings and stack them in size order.  

9 

 
 
 
 
 
Table 1 (cont’d) 

Shape Sorter  

Recommended age: 18 months and older.  

Rectangle shaped box with removeable lid. 

Built into the lid are five different shape cut 

outs (triangle, square, circle, star, cross). The 

individual sorts, stacks, and drops shapes 

through the shape-sorter lid. Individuals are  

expected to take the shape and sort them into 

the corresponding shape. 

Puzzles (4 to 12 pieces) 

Recommended age: two and up.  

Individuals are expected to obtain the puzzle 

from activity schedule shelf, take the pieces 

off the puzzles, place the pieces in the correct 

spots on the puzzle board, continue putting 

pieces on until they are all on, put the 

completed puzzle back on the shelf, and walk 

back to table.  

10 

 
 
 
 
 
 
Table 1 (cont’d) 

Sorting Bears 

Recommended age: 6 months- 3 years old. 

Six cups in distinct colors (red, orange, 

yellow, green, blue, purple) that come with up 

to ten bears in the same colors as the cups. 

Individuals are expected to take the bears 

provided and sort them into the 

coordinating/matching color cup.  

Stringing Beads 

Recommended age: 1 to 3 years old. 

Colorful beads (red, orange, yellow, green, 

blue purple) in varying shapes (circle, square, 

hexagon) that have holes at either end. 

Individuals are expected to take the beads and 

put them on the string through the two holes.  

Video Models 

The video models for the activity schedule were recorded on an Apple® iPad using the 

camera application, with each video averaging seven seconds in length. All recordings took place 

in an empty classroom within the EIBI clinic attended by all participants. To maintain 

consistency and clarity, the researcher ensured minimal background noise, eliminated potential 

distractions, and verified that the video quality was appropriate for instructional use. The 

controlled environment helped standardize the videos, ensuring participants could focus on the 

modeled behaviors without external interference. 

11 

 
Each video model for a social interaction depicted the researcher approaching a peer, 

emitting a request for attention, engaging in the action, and concluding the interaction. The social 

interaction served as the thumbnail image displayed on the slide, providing a visual cue for the 

participant. For example, if the target social interaction was a high five, the thumbnail captured a 

still frame of the researcher and a social partner at the exact moment their hands met. When the 

participant pressed the play button, the video demonstrated the full sequence of events, showing 

the researcher locating the social partner and engaging in the designated interaction. 

The video models for social interactions varied across different actions, including giving 

a fist bump, requesting a tangible item from a peer, and greeting a social partner. These 

interactions were carefully selected to promote fundamental social skills essential for peer 

engagement. The use of short, structured video models provided clear, repeatable demonstrations 

of the target behaviors, reinforcing learning and facilitating skill acquisition among participants. 

Dependent Measures 

Dependent Variables 

The primary dependent variable in this study was the number of independently performed 

social interaction components while using a photo activity schedule embedded with video 

models. Social interactions were operationally defined as reciprocal exchanges between at least 

two peers, including giving a fist bump, greeting a peer (e.g., saying "hello" or waving), or 

requesting a tangible item from a peer. 

To ensure objective measurement, social interaction components were broken down into 

four distinct, measurable components: : (1) approach partner, (2) orienting body toward partner, 

(3) emitting request and (4) ending the interaction. Participants received a correct score when 

12 

they completed all components independently. If prompting was required or an error occurred, 

the response was scored as incorrect. 

A structured scoring system was used to evaluate each component of the social 

interaction, with scores ranging from 0 to 4 per video model. A score of 0 indicated the 

component was missing or not completed, while a score of 4 reflected that the component was 

fully met and executed correctly. At the end of each session, the researcher calculated a total 

score by summing the number of components the participant performed independently. The total 

possible score increased with the number of video models shown, as each model was scored 

individually on the 0–4 scale. For example, if one video model was presented, the total score 

ranged from 0 to 4; with two models, the range was 0 to 8; and with three models, the range 

extended to 0 to 12. Table 2 outlines the operational definitions for each social interaction 

component. 

Table 2: Operational Definitions of Social Skills. 

Term  

Fist Bump 

Operational Definition  

Approach a peer, orient body toward social 

partner, use a closed hand fist and reach arm 

toward partner to lightly hit against the social 

partner close hand fist then end the interaction 

(saying a variation of bye, see you later, high 

fiving, etc.)   

13 

 
 
 
Table 2 (cont’d) 

Asking for a Tangible Item  

Approach peer, oriented body towards the 

social partner, emit the request for attention 

(Communication modality, or vocal language 

to ask for tangible item), and say something 

to end the interaction (say thank you, high 

fiving, saying bye etc.).  

Greeting a Peer  

Approach peer, oriented body towards the 

social partner, emit the request for attention 

(waving or communication modality or vocal 

language to say a variation of hi) and 

something to end the interaction (saying a 

variation of bye, high fiving, etc.). 

Interobserver Agreement  

To assess the reliability of data collection, interobserver agreement (IOA) was measured 

for 33% of sessions across all conditions and participants. IOA was calculated using a point-by-

point agreement method, with an agreement defined as both observers assigning identical scores 

across all components. The IOA percentage was computed by dividing the number of agreements 

by the total number of agreements plus disagreements, then multiplying by 100. 

Prior to formal data collection, the secondary observer completed training that included a 

PowerPoint presentation outlining the study overview, operational definitions, data sheet 

procedures, and scoring criteria. Training also included guided practice with pre-recorded video 

14 

 
samples, followed by independent scoring of additional videos to ensure consistency and 

accuracy. 

All IOA data were collected via video recordings of study sessions to maintain 

standardization and allow for detailed review. If IOA for any session fell below 80%, the primary 

researcher reviewed discrepancies with the observer and recalibrated the scoring procedures 

before continuing further data collection. 

IOA data were collected across all phases of the study. For Participant 1, IOA was 100% 

during baseline, and intervention, indicating perfect agreement between observers throughout. 

Similarly, for Participant 2, IOA remained at 100% during both baseline and intervention phases. 

These consistently high IOA scores reflect strong procedural fidelity and high observer accuracy, 

supporting the internal validity of the study and the reliability of the recorded data. 

Experimental Design  

A multiple probe across participants design was employed for this study (Ledford & 

Gast, 2018; Osos et al., 2021). This design allowed for the evaluation of the effects of video 

models embedded within an activity schedule on participants' social interaction components 

while maintaining experimental control. By staggering the introduction of the intervention across 

participants, the design helped control demonstration effects and ensured that changes in 

behavior could be attributed to the intervention rather than external factors (Ledford & Gast, 

2018). 

The order in which participants received baseline and intervention conditions was 

determined based on the stability of their baseline performance. A multiple probe design, rather 

than a traditional multiple baseline design, was selected because participants were expected to 

demonstrate zero or low levels of target behaviors prior to the intervention. It was anticipated 

15 

that participants would only be able to effectively use the photo activity schedule with embedded 

video models once explicit instruction was provided. Additionally, by reducing the number of 

sessions, this design helped minimize the potential for frustration and problem behaviors that 

could arise from repeated unsuccessful attempts at completing tasks independently. 

Procedures 

Baseline  

The purpose of the baseline condition was to assess participants' current skill levels 

independently following an activity schedule and engaging in social interaction components. 

Each session began with the researcher sitting near the participant and conducting a brief 

Multiple Stimulus Without Replacement (MSWO) preference assessment. This assessment 

involved placing five items in front of the participant and allowing them to select one as a 

reinforcer. 

Once the reinforcer was identified, the researcher initiated the session by approaching the 

participant in their designated learning area. The iPad containing the activity schedule was 

placed in front of the participant. The researcher then provided the vocal instruction, “Follow 

your schedule.” After delivering this initial instruction, the researcher moved approximately four 

feet away to allow the participant to complete the activity schedule independently. 

The activity schedule used during the baseline phase did not include video models; 

instead, it contained only static images. The thumbnails for social interactions were presented as 

static pictures, and each activity schedule consisted of two discrete trial activities and one static 

image representing a social interaction with a peer. These variables remained consistent across 

all baseline sessions to establish a clear comparison with the intervention phase and determine a 

functional relationship between conditions. 

16 

Participants were expected to complete the entire activity schedule, progressing through 

all pages independently. A session was terminated if the participant failed to initiate the first 

scheduled activity within 10 to 15 seconds of the instruction or if they disengaged from the 

schedule for 10 to 15 seconds. No additional prompts, reinforcements, or corrective feedback 

were provided during baseline to accurately measure the participants’ independent performance 

levels. 

Throughout the session, the researcher recorded data using a datasheet (Appendix C) 

while maintaining four feet from the participant. The same datasheet was used across all 

conditions to ensure consistency. The researcher documented performance by marking a (+) for 

correctly completed steps and a (-) for incorrect or skipped steps. If a participant made an error, 

no prompts or corrections were provided. 

All baseline sessions were recorded using a handheld camera to ensure accurate data 

collection and allow for later review. 

Intervention 

Intervention sessions were identical to baseline sessions in structure, location, and 

procedures—with one key modification: video models were embedded in place of static images 

for the social interaction slides. Discrete trial activity pages remained unchanged and continued 

to use static images. The intervention aimed to increase the number of social interaction 

components completed by incorporating short video clips demonstrating specific behaviors, such 

as giving a fist bump, requesting a tangible item, or greeting a peer (e.g., waving or saying 

“hello”). 

17 

 
The activity schedules, created using Microsoft PowerPoint and presented on an iPad, 

consisted of four to six slides. Each participant began with one social interaction video, which 

was systematically, by one video, increased as mastery criteria were met. 

As in baseline, each session began with the instruction, “Follow your schedule,” and the 

researcher stepped four feet away to allow for independent responding. The researcher recorded 

data using a task analysis sheet, marking a (+) for independent responses and a (−) for incorrect 

or prompted steps. 

If the participant failed to initiate the activity within 5 to 10 seconds or showed signs of 

disengagement during the session, a gestural prompting protocol was introduced as needed. 

Gestural prompts involved using a pointed finger to draw the participant’s attention—starting 

from the participant’s eyes and moving toward the tablet to redirect their focus. Importantly, 

prompts were used solely to support attending to the tablet and were not intended to guide or 

correct schedule completion behavior. Mastery was defined as completing 100% of steps 

independently for two consecutive sessions. 

Sessions were terminated if the participant exhibited distress (e.g., crying, yelling, self-

injury, aggression) or engaged in elopement behaviors. All intervention sessions were recorded 

using a handheld camera for accuracy and subsequent analysis 

Procedural Fidelity  

For all components of the intervention a second researcher collected treatment fidelity 

data using a task analysis checklist developed by the primary researcher (see Appendix B). 

Treatment fidelity data were collected for 33% of Noah’s intervention sessions and 40% of 

Morgan’s intervention sessions. Sessions were randomly selected using an online random 

number generator. For Noah, treatment fidelity ranged from 80% to 100%, with most sessions 

18 

demonstrating high procedural integrity. For Morgan, treatment ranged from 90% to 100%. 

These data indicate that the independent variable was implemented with acceptable fidelity. 

19 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
RESULTS 

Figure 1 depicts Noah’s (top) number of independently performed social interaction 

components during baseline and intervention phases. At baseline, Noah did not engage in any 

independent social interaction components (0, range 0–0). During Phase One of the intervention, 

when one video model was embedded into the photo activity schedule, Noah began 

demonstrating increased independence, performing all four social interaction components by the 

12th session. Upon transitioning into Phase Two, where two video models were embedded, 

Noah’s performance initially maintained at four independently performed components. Then 

Noah’s score rose to six independently completed components, indicating a promising trend. 

However, this increase was short-lived, as his performance dropped to zero during the following 

sessions and remained low for the rest of the intervention phase. 

Figure 1 also depicts Morgan’s (bottom) number of independently performed social 

interaction components across all phases. Like Noah, Morgan engaged in zero independent social 

interaction components at baseline (range 0–0). However, unlike Noah, Morgan’s performance 

remained at zero throughout the intervention phase, showing no measurable change despite the 

introduction of video modeling. 

These results suggest that while video modeling embedded into photo activity schedules 

may support the acquisition of social skills in some children with ASD, such as Noah, it may not 

be universally effective for all learners, as evidenced by Morgan’s lack of progress. 

20 

Figure 1: Noah and Morgan’s number of independently performed social interaction components 

and PAS completion percentage. The closed black circles represent baseline session data points. 

The closed black squares represent intervention sessions. The closed black triangles represent 

total photo activity schedule (PAS) completion in a percentage. 

This graph displays the effect of video modeling on participants' social interaction 

components. Each panel represents an individual participant’s performance across sessions. The 

left Y-axis shows the social interaction component score, while the right Y-axis represents the 

PAS percentage (i.e., the percentage of program activities completed). Session numbers are 

presented along the X-axis. A parallel break on the X-axis indicates a long interruption due to 

winter break. Data points are marked with distinct symbols: filled black circles represent baseline 

21 

sessions, black squares indicate intervention sessions, and black triangles show PAS completion 

data. Dashed vertical lines denote phase changes from baseline to intervention, and a horizontal 

dashed line connects the phase changes between participants to highlight the staggered 

intervention introduction. 

22 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
DISCUSSION 

The purpose of this study was to examine whether embedding video models within a 

photo activity schedule could support increases in social interactions among children with ASD. 

Using a multiple probe across participants design, the study aimed to determine if participants 

could independently engage in targeted social interaction components following exposure to 

video modeling. Both participants demonstrated zero independent social interaction components 

during baseline. However, following intervention, only Noah showed a modest and variable 

increase in performance, while Morgan demonstrated no change—highlighting individual 

differences in response to the intervention. 

Noah’s performance provides limited support for the use of video-enhanced activity 

schedules in increasing independent social interaction components. In Phase One, he 

demonstrated progress with one video model, suggesting an initial increase of targeted social 

skills components was possible under these conditions. Notably, during Phase Two, when the 

number of embedded video models increased to two, Noah achieved his highest performance—

completing six social interaction components independently. However, this gain was short-lived, 

followed by a sharp and sustained decline to zero. Despite no changes to the session structure or 

expectations. This pattern suggests that the increased complexity may have exceeded Noah’s 

instructional tolerance, disrupted stimulus control, or reduced motivation. Alternatively, he may 

have struggled to generalize imitation across multiple models, or the novelty of the video model 

may have worn off. These results raise important considerations about the sustainability and 

generalization of skills acquired through video modeling. 

In contrast, Morgan did not demonstrate improvement at any point during the 

intervention, despite receiving the same materials and procedures. Her consistently low 

23 

performance suggests that video modeling alone was not sufficient to promote social interaction 

components for her. This underscores the importance of evaluating individual readiness factors, 

such as attending to models, imitation ability, and motivation for social engagement. It is 

possible that Morgan required more explicit instructional supports, such as direct prompting or 

reinforcement, to bridge the gap between observation and performance. Her outcome reinforces 

the notion that video modeling is not universally effective and must be considered as one of 

several potential strategies in an individualized intervention plan. 

These findings provide only limited support for the existing literature suggesting that 

video modeling can increase social skill development (Bellini & Akullian, 2007; Plavnick et al., 

2015). Contrary to prior studies that have demonstrated more consistent and sustained benefits, 

the outcomes in this study were highly variable across participants. While Noah showed brief 

gains, these improvements were not maintained over time, and Morgan showed minimal to no 

response to the intervention. This inconsistency suggests that, for some learners, video modeling 

embedded in structured routines (Brodhead et al., 2018; Osos et al., 2021) may be insufficient on 

its own. These results highlight the need for additional or alternative strategies to support 

children who do not respond to video modeling as expected. 

This study carries several practical implications. First, clinicians and educators should 

assess prerequisite skills before implementing video modeling interventions, including attention, 

engagement, and motivation. Second, ongoing progress monitoring is essential. If a learner does 

not show improvement after a few sessions, the intervention should be adjusted. For example, by 

simplifying the task, adding prompts, or exploring alternative instructional approaches. Third, 

video modeling should not be regarded as a universally effective solution; rather, it is one tool 

24 

within a broader evidence-based framework. For some learners, peer modeling, live 

demonstrations, or more naturalistic teaching strategies may yield better outcomes. 

Several limitations should be acknowledged. The small sample size (two participants) 

limits the generalizability of findings. While single-case designs allow for strong internal 

validity, larger-scale studies are necessary to better understand for whom and under what 

conditions video-enhanced activity schedules are most beneficial. Additionally, although Noah 

initially responded to the intervention, the abrupt decline in performance highlights concerns 

about skill maintenance and generalization—key issues that future research must address. 

Future studies should explore whether learner characteristics—such as verbal ability, 

imitation repertoire, or social motivation—predict responsiveness to video modeling. 

Investigations might also examine the efficacy of combining video models with reinforcement, 

peer-mediated strategies, or live prompting, particularly for learners who do not respond to video 

modeling in isolation. Finally, future research should incorporate follow-up and generalization 

probes to determine whether acquired skills persist over time and transfer to novel contexts or 

interaction partners. 

In conclusion, this study offers limited support for the use of video-enhanced activity 

schedules to increase social skills development in children with ASD. While one participant 

showed initial gains, outcomes were not sustained and were not replicated across both learners. 

These findings reinforce the need for individualized, flexible intervention planning and suggest 

that video modeling, while promising in some contexts, may need to be paired with additional 

supports to achieve meaningful and lasting outcomes. 

25 

 
 
REFERENCES 

American Psychological Association. (2023, November 15). Autism spectrum disorder. 
American Psychological Association. https://www.apa.org/topics/autism-spectrum-
disorder.  

Bellini, S., & Akullian, J. (2007). A Meta-Analysis of Video Modeling and Video Self-Modeling 
Interventions for Children and Adolescents with Autism Spectrum Disorders. Exceptional 
Children, 73(3), 264-287. https://doi.org/10.1177/001440290707300301 

Betz, A., Higbee, T. S., & Reagon, K. A. (2008). Using joint activity schedules to promote peer 

engagement in preschoolers with autism. Journal of applied behavior analysis, 41(2), 
237–241. https://doi.org/10.1901/jaba.2008.41-237  

Brodhead, M. T., Courtney, W. T., & Thaxton, J. R. (2018). Using activity schedules to promote 
varied application use in children with autism. Journal of Applied Behavior Analysis, 51, 
80–86. https://doi.org/10.1002/jaba.435. 

Brodhead, M. T., Higbee, T. S., Pollard, J. S., Akers, J. S., & Gerencser, K. R. (2014). The use of 
linked activity schedules to teach children with autism to play hide-and-seek. Journal of 
Applied Behavior Analysis, 47 (3), 645-650. https://doi.org/10.1002/jaba.145 

Brown, J. L., Krantz, P. J., McClannahan, L. E., Poulson, C. L., (2008). Using script fading to 
promote natural environment stimulus control of verbal interactions among youth with 
autism. ScienceDirect, 2, 480-497. https://doi.org/10.1016/j.rasd.2007.08.006. 

Bryan, L.C., Gast, D.L. (2000). Teaching On-Task and On-Schedule Behaviors to High-
Functioning Children with Autism Via Picture Activity Schedules. J Autism Dev 
Disord 30, 553–567 https://doi.org/10.1023/A:1005687310346 

Carlile, K. A., Reeve, S. A., Reeve, K. F., & DeBar, R. M. (2013). Using activity schedules on 

the iPod touch to teach leisure skills to children with autism. Education and Treatment of 
Children, 36, 33–57. https://doi.org/10.1353/etc.2013.0015. 

Ledford , J. R., & Gast, D. L. (2018). Single case research methodology (3rd ed.). Routledge.  

McClannahan, L. E., Krantz, P. J., & Harris, S. L. (Ed.). (2010). Activity schedules for children 

with autism: Teaching independent behavior. Topics in Autism. 

Osos, J.A., Plavnick, J.B. & Avendaño, S.M. Assessing Video Enhanced Activity Schedules to 

Teach Social Skills to Children with Autism Spectrum Disorder. J Autism Dev Disord 51, 
3235–3244 (2021). https://doi.org/10.1007/s10803-020-04784-x  

Plavnick, J. B., MacFarland, M. C., & Ferreri, S. J. (2015). Variability in the Effectiveness of a 

Video Modeling Intervention Package for Children with Autism. Journal of Positive 
Behavior Interventions, 17(2), 105-115. https://doi.org/10.1177/1098300714548798 

26 

 
 
 
 
 
 
 
 
 
 
 
Wichnick-Gillis, A. M., Vener, S. M., & Poulson, C. L. (2016). The effect of a script-fading 

procedure on social interactions among young children with autism. ScienceDirect, 26, 1-
9. https://doi.org/10.1016/j.rasd.2016.03.004  

27 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
APPENDIX A: IOA DATA SHEET FOR BASELINE AND INTERVENTION 

Figure 2: Task analysis checklist used to assess interobserver agreement during baseline and 

intervention phases. The checklist outlines each step completed by the researcher, including 

preference assessment, instruction delivery, distancing, prompting, data collection, and session 

termination criteria. 

Baseline Steps 

Overview 

Correct 
Implementation 
(mark +, -, or 
NA) 

1.  Conduct brief MSWO preference assessment.  
2.  Prepare reinforcer after brief MSWO preference assessment. 
3.  Researcher presents the electronic activity schedule from behind the 

participant. 

4.  Researcher gives the verbal instruction “follow your schedule”. 
5.  Researcher moves 3 to 5 feet away from participant. 
6.  Researcher provides brief gestural prompt (if needed after 10-15 

seconds of disengagement). 

7.  Session terminated if crying, aggression, elopement, or yelling is 

displayed by participant.  
8.  No physical prompts provided. 
9.  Researcher collect paper-pencil data while participant is completing the 

electronic activity schedule.  

10. After the electronic activity schedule is completed, the researcher 

provides a reinforcer and thanks the participant. 

28 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Figure 2 (cont’d) 

Intervention Steps 

Overview 

Correct 
Implementation 
(mark +, -, or 
NA) 

1.  Conduct brief MSWO preference assessment.  
2.  Prepare reinforcer after brief MSWO preference assessment. 
3.  Researcher presents the electronic activity schedule from behind the 

participant. 

4.  Researcher gives the verbal instruction “follow your schedule.” 
5.  Researcher moves 3 to 5 feet away from participant. 
6.  Researcher provides brief gestural prompt (if needed after 10-15 

seconds of disengagement). 

7.  Session terminated if participant displays crying, aggression, 

elopement, or yelling.  

8.  Researcher collect paper-pencil data while participant is completing the 

electronic activity schedule.  

9.  After the electronic activity schedule is completed, the researcher 

provides a reinforcer and thanks the participant.  

29 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
APPENDIX B: TREATMENT FIDELITY DATA SHEET 

Figure 3: Photographic Activity Schedule (PAS) procedural fidelity checklist used to assess 

behavior technician implementation during intervention sessions. The checklist includes steps for 

pre-intervention preparation, session implementation, prompting procedures, reinforcement 

delivery, data collection, and graphing. 

30 

 
 
 
 
 
 
 
 
 
 
APPENDIX C: DATA COLLECTION DATA SHEET 

Figure 4: Photographic Activity Schedule (PAS) datasheet used to document participant 

performance. The form includes task analysis steps for activity engagement and social 

interaction components, as well as completion of the full schedule. Data is recorded per step 

across video models and summarized for total independent responses. 

Client Name:  

Date:  

PAS –   

Activity   

Opens/turns page   Points/Looks   

Obtains Item    Completes 

Puts Away   

Activity   

  Approach 
Partner 

  Orients Body 
Toward Partner  

  Emit Response    End Interaction    

Turns last page   

Points to back of 
binder   

Closes binder   

Puts Away   

Video Model 1 

Video Model 2 

Video Model 3 

Finishes 
Schedule   

# Completed:   

Total # of components:   

# completed Ind:   

31