Introduction

Autism spectrum disorder (ASD) is a developmental disability that affects 1 in 68 children in the United States (The Center for Disease Control and Prevention 2012). These children usually exhibit deficits in social reciprocity and language and communication skills, but demonstrate repetitive and stereotypical behaviors (DSM-V, 5th ed., 2013). In addition to repetitive behaviors, students with ASD often exhibit poor self-management and inadequate attending skills that negatively affect their school performance (Holifield et al. 2010; Wilkinson 2008). Thus, individually designed interventions are suggested for these students to learn self-management and attending skills in order to be successful in school (Soares et al. 2009).

Self-monitoring is a procedure for students to systematically monitor their own behaviors (Turnbull and Turnbull 2001). The procedure has two steps: (1) teaching the student to recognize if the target behavior has occurred or not, and (2) teaching the student to self-record his or her own behavior (Blood et al. 2011). Self-monitoring has been shown to increase on-task behaviors as well as improve students’ motivation, which has been attributed to taking responsibilities for their own behaviors (Wilkinson 2008). Self-monitoring, especially when used to address attention issues, has been reported as being relatively easy for students with disabilities to implement in class, (Soares et al. 2009) and practical for teachers (Holifield et al. 2010).

Visual prompts are often used to teach students with ASD to manage their own behavior during a given social situation such as those described in a social story (Gray 2000). SMART technology, whereby an interactive white board was used, can also be utilized to teach these students social stories to learn appropriate behaviors and social skills (Xin and Sutman 2011). In addition, recording a student performing an appropriate behavior saved as a video image to be reviewed by the student prior to performing a task to cue himself/herself, either verbally or silently, has been found to benefit these student with ASD (Gelbar et al. 2012). The video can be repeatedly presented as a visual model for the learner to practice with a clear and detailed display of an appropriate behavior that is available to view and imitate (Corbett 2003). It has been found that such video modeling is particularly effective for learners with ASD because many of these students are visual learners, have selective attention, and present behaviors that are of a repetitive nature (Wilkinson 2008).

Apple iPad is a handheld device with the capacity to display visual models on a portable instrument that can provide a unique means for displaying picture prompts (Blood et al. 2011). Its multi-touch functions create an avenue for an individual student to use video self-monitoring privately without interrupting other students in class (Blood et al. 2011). The program applications or “apps” are easily obtainable, affordable, customizable, and user friendly. It has been found that this mobile device may reduce teacher prompting and, as a result, teachers can spend more time on engaging students in instructional activities and less time on managing behaviors (Cihak et al. 2010). In addition, the iPad’s sensory component, direct screen touch instead of a mouse, small size and lightweight design are of high interest for students with ASD (Cihak et al. 2010). This new technology may provide an opportunity for these students to learn self-monitoring skills using video self-modeling. Although several studies have demonstrated the effectiveness of visual modeling to improve the self-management of students with ASD (Wilkinson 2008; Gul and Vuran 2010), little research has been conducted using an iPad. Only one study was found to examine the use of an iPod touch for video modeling and self-monitoring of an elementary student with emotional disabilities (Blood et al. 2011). In their study, the student was taught to touch an icon on the iPod Touch to watch a video segment presenting peer or his on-task behavior, then record his own on- or- off task behaviors by marking on a sheet. The results showed that the student responded positively to both the video modeling and self-monitoring by increased on-task behaviors, such as following directions and engaging in class activities, however, the student’s own behavior recording was the same as the traditional way of using a paper sheet, even though a timer was used for time accuracy. This current study used iPads presenting visual images of self-modeling of on-task behaviors as well as a self-recording screen for individual students to manage their own behaviors.

Previous research (Williamson et al. 2009) assumed that the demonstration of on-task behaviors could potentially increase students’ actual work completion. However, data collection of student performance has been lacking, and this assumption could not be supported. It could be possible that students appear on-task, but they might not be engaged in meaningful academic activities. To date, the literature has presented mixed results concerning the relation between students’ on-task behaviors and their academic achievement (Gul and Vuran 2010; Soares et al. 2009; Wilkinson 2008). Thus, in addition to observing and analyzing students’ behaviors, this study evaluated their performance on weekly quizzes in language class when the iPad was used in self-monitoring, to explore if student’s behavior change would impact their academic performance. It attempted to increase on-task behaviors of students with autism in class by learning self-monitoring skills using an iPad for visual prompts, and evaluate their performance in both behavioral and academic areas.

Method

Participants

Four students, 1 boy and 3 girls, classified with ASD following the state code of eligibility, participated in the study. All students exhibited frequent off-task behaviors during instruction, especially in language class. Their off-task behaviors included making noises, looking around, getting out of their seats, and playing with objects. Their academic performance was lagging behind in language arts, which could be found in their achievement test scores that are far below the mean. Each student had an Individual Education Plan (IEP) developed based on their educational needs. Table 1 presents the general information of the participants.

Table 1 General information of the participants
Full size table

Student A, female, was diagnosed with ASD when she was in 1st grade. She was able to pay attention by looking at the teacher and following directions, but at times she would scream, cry, and run around the classroom if she saw other students were doing something without including her. For example, if she went to throw out trash, she would go around to find out what others were doing. It seemed that she was curious about everything her peers were doing, but not engaging in her own task. Student B, female, was diagnosed with ASD when she was in kindergarten. This girl often stared at the walls or ceiling, turning around during the teacher’s lecturing. It always took a few minutes for the teacher to remind her of starting her assignment and guide her to focus on the task. Student C, female, was diagnosed with ASD when she was in 1st grade. During class, student B often played with her fingers or any objects on the desk. She was often redirected to guide her to look at the teacher and establish eye contact. Often times, she looked around without a focus. Following directions was difficult for her. Student D, male, was diagnosed with ASD when he was 4. He was energetic, but restless and impulsive and often out of seat, touching others when he was walking around in the classroom. These four students were placed in a special education classroom to learn Language Arts each day with a special education teacher, because of their needs in this academic area. Thus, during this time period, only these 4 students were present. The special education teacher implemented the lessons and two teacher assistants were trained to observe and record each student’s behaviors using a checklist. Behavioral recording was conducted in 20 min of each class every day, during which students were assigned to practice, or respond to questions when most of their off-task behaviors would occur.

Experimental Design and Procedures

An ABAB reversal design was used to evaluate the effect of self-monitoring with iPads on the participating students’ behaviors. An interval recording method was used to record on-task behaviors of each student. On-task behaviors were defined as sitting in seat, paying attention to the teacher when the teacher is talking (e.g., listening, eyes and face forward), and working on the assignment. During the 20 min of class practice, the teacher would deliver the assignment to each student, and verbally explain the directions by giving an example and asking questions, followed by student working on the assignment. Paying attention to the teacher (facing forward and looking at teacher) was recorded when the teacher was talking or asking questions, while engaging in assignments was recorded when the assignment was required to start. With the understanding that some students would sit quietly without participating in class activities, engagement was emphasized as “working on assignment”.

Baseline (Phase A1)

During the baseline, students’ on-task behaviors were recorded for 5 days. The class started by the teacher’s instruction that was delivered in a lecture format by modeling the skill, giving examples for explanations, and asking questions to check for understanding. This was followed by 20 min of assigned practice during which the assistants observed and recorded student behaviors by marking “+” for an occurrence and “−” for a nonoccurrence in each 2-min interval. Student content knowledge was evaluated via oral questions or a worksheet at the end of each week.

Intervention (Phase B1)

The teacher introduced the students to the basic functions of an iPad such as on–off switch, lock and unlock, open and close “app”, and verbally explained each step with modeling. Each student was guided to imitate and practice with an assistance of the teacher assistants individually. The teacher and her assistants checked each student to make sure he/she would use the device appropriately. Then, “Choiceworks” app was downloaded and presented with an icon on each student’s iPad screen. This program was developed to help learners complete daily routines and understand and control their feelings associated with the routines. It allows an image or photo to be incorporated (See an example in Online Appendix A). This program was customized into a self-monitoring tool for the participating students. In class, each student was provided an index card with a list of three on-task behaviors including attending to the assignment, facing forward and sitting in the chair, and listening to the teacher. The teacher modeled each on-task behavior, and each student was required to imitate. Their on-task behavior was videotaped using the camera feature of the iPad for a 30 s segment and saved into the “app” as well as their own voices (e.g. I look at Teacher). At the beginning of the class, each student watched his/her own image posted on the screen and listened to the recorded voice. They were guided to view each target behavior shown as a thumbnail sized image in a list for students to check out (see Online Appendix A). At the end of the class, each student was required to check the images one by one to see if he/she presented all three target behaviors. If one was missed, that student had to talk to the teacher and watch the self-images about this particular behavior again. If three target behaviors presented, a reward, such as a preferred activity, was given as reinforcement. This intervention lasted for 15 days, during which the same observation process was conducted to record student behaviors.

Baseline/Withdrawal (Phase A2)

Students continued to attend their language class, however, their iPads were taken away (e.g., The teacher informed the class that they would not use the iPad this week). The observation process was continued as the same as that in the intervention.

Intervention (Phase B2)

During this phase, the iPads were returned to the students for self-monitoring as described in the intervention (Phase B1). The behavior observation process was the same as that in the previous phases.

Dependent Variable and Measurement

Data were collected related to the following three dependent variables: on-task behavior, academic achievement, and student satisfaction. A 20-min interval recording method (Wheeler and Richey 2014) was used to measure the on-task behaviors. The frequency of behavior occurrences was calculated into percentages. In addition, student performance in language learning was evaluated by a weekly test developed by the teacher based on the lessons delivered and skills taught during the week. Each test included 5 words to sound out and responses to questions about their meanings, and apply in sentences by circling the correct words based on the week’s instruction. All scores were converted into percentages. At the end of the study, each participating student took a survey including 5 statements such as “I like using my iPad”; “I like to see myself on the iPad screen”; “I like to hear my voice”. A rating scale with four levels, such as, “Strongly Agree,” “Agree,” “Disagree,” and “Strongly Disagree” was used with 4 representing the highest and 1 for the lowest.

Inter-observer Agreements

Two teacher assistants observed and recorded students’ on-task behaviors using the same observation checklist. The inter-observer agreement was calculated following the format (agreement/total intervals × 100) indicated by Zirpoli (2005). If there was a discrepancy between the observers, the teacher was involved in a meeting together to check for accuracy, in order to reach at least 90% of agreement.

Data Analysis

A visual analysis was used to compare the percentage of observed behaviors for both baselines and interventions for each participant in the class. In addition, the percentage of non-overlapping data (PND) procedure described by Scruggs et al. (1987) was used. This type of analysis is commonly applied in single-subject research design and has been shown to detect intervention effects (Campbell 2004). The guideline recommended by Asaro-saddler and Saddler (2010) was adopted. This guideline indicated 90% of the intervention points exceeding the extreme baseline value for a very effective treatment; 70–90%, an effective treatment; 50–69%, a questionable treatment, and less than 50%, an ineffective treatment. Tables 2, 3 and 4 present PND in each intervention of individual students.

Table 2 Means, standard deviations, and PND of the on-task behavior across phases (facing forward)
Full size table
Table 3 Means, standard deviations, and PND of the on-task behavior across phases (engaging in assignment)
Full size table
Table 4 Means, standard deviations, and PND of the on-task behavior across phases (looking at teacher)
Full size table

Results

On-Task Behaviors

Figures 1, 2, and 3 present a visual display of each participating student’s on-task behaviors across the four phases. Tables 2, 3 and 4 show the mean percentages of each on-task behavior, i.e. facing forward, looking at teacher, and working on the assignment, during each phase for each participant respectively. Results indicate that all participants had low mean percentages of on-task behaviors during the initial baseline. During the intervention when an “iPad” was used for self-monitoring, all participants increased their on-task behaviors to a group mean of 84% for three target behaviors (facing forward, looking at teacher, and working on the assignment), only one student had an overlapping data point with the baseline data, presenting 95% of PND. This means that over 95% of the intervention data points exceeded the extreme baseline value indicating a very effective treatment. In the second intervention, all students increased on-task behaviors with a group mean of 86% with PND of 100%, indicating a very effective treatment. During the second baseline, the number of on-task behaviors was reduced compared to that of the intervention with a class mean of 40%. The data of the second intervention were continued to maintain the increase with a class mean of 86%, compared to both baselines, with PND of 95% for one student, the rest was 100%, indicating a very effective treatment.

Fig. 1
figure 1

Facing forward

Full size image
Fig. 2
figure 2

Engaging in assignments

Full size image
Fig. 3
figure 3

Looking at teacher

Full size image

Academic Performance

Each quiz score was converted into percentages and presented in Table 5 to demonstrate student performance in vocabulary learning. Results indicated that all students’ test scores increased compared to those in the baseline, though their scores were not reached to 80% accuracy as their goal.

Table 5 Means of student vocabulary scores by percentages across phases
Full size table

Social Validity

At the end of the study, each participating student was given a survey with 4 statements. They were read each statement by the teacher, then to circle “strongly agree, agree, disagree, or strongly disagree” to determine the social validity of the intervention. This survey was intended to obtain information of the students’ satisfaction with the “iPads” in their learning experiences. Results showed that all of the participants enjoyed using the iPad. Three out of 4 students (75%) strongly agreed that they liked to see their pictures posted on the screen reminding their own behavior and listen to their own voice recorded in the iPad. All students would like to continue to use an iPad to learn on-task behaviors.

Discussion

The purpose of our study was to evaluate the effectiveness of the Apple iPad as a tool for teaching students with ASD self-monitoring skills. The participants modeled on-task behaviors while being recorded using the camera feature of the iPad. Then, students’ individual images and voices were recorded and incorporated into the app “Choiceworks.” The study attempted to determine if this self-monitoring strategy would increase their on-task behaviors.

Results showed that the intervention was effective in increasing the on-task behavior of all four participants. The findings are consistent with those of previous studies on self-management for students with autism (e.g. Callahan and Rademacher 1999; Newman et al. 2000; Wilkinson 2008). However, in those studies, the major instrument used to evaluate the learner’s behavior changes was a paper sheet with a list of questions for individual student’s self-check. One study reported using an iPod to teach social and self-management skills, but the participant was only one child with emotional and behavioral disorders (Blood et al. 2011). Although some studies utilized a combination of self-monitoring and visual images, including video modeling, the participants were always young children (e.g. Cihak et al. 2010; Soares et al. 2009).

Our study expands the previous research by using the iPad with application of self-images to model the appropriate on-task behaviors of teenage students with ASD. Our findings should add information to the research area and provide another avenue for teachers to manage class and individual students with ASD. As these students have increased access to the general education curricular, there is limited research on the effects of self-management in those settings (Lee et al. 2007). At the same time, concerns have been raised if the social skills learned by using technology could be generalized into a real life situation. This has been found in recent research to compare young children with autism using or without using an iPad to learn social communication skills (Fletcher-Watson et al. 2016). The results of non-significant differences between groups of children in skill learning with and without an iPad may be impacted by different variables in their study, but it should be cautiously considered when such a device is provided for interventions targeting social communication skills. For example, follow-up practices in a real world without technology should be provided to enhance skill generalization. Further studies are needed in this area to validate findings.

Our study may provide a new avenue for teachers to use technology, such as iPads in classrooms. Teachers could use this technology to develop other strategies to assist students with ASD and other disabilities in their behavior and academic improvement. School administrators may consider purchasing popular electronic devices for class instruction because of students’ needs. Proper training could be provided to teachers, so that iPads could be incorporated into their daily lessons and behavior management, which would benefit both students and teachers. This device is highly portable and easy to use; and as such could serve as an effective piece of assistive technology for students with ASD.

Although only four students with ASD participated in our study, the findings are positive to support iPad’s usage in classrooms to increase their self-monitoring skills and to reduce inappropriate behaviors by learning their own visual modeling and checking their own behaviors. In our study, each student was provided with an iPad to view his/her own previously recorded behavior and to check if his/her current behavior was matched with the appropriate behavior he/she performed in the visual images. Such a self-image of the individual’s appropriate behavior may serve as a reminder for the student and facilitate his/her motivation to further his/her self-modeling. The convenience is that an iPad can be placed at the corner of a student’s desk for a private use without interrupting the teacher’s instruction, and distracting peers, because no television or regular computer screen is needed. This will also reduce the teacher’s supervision time, thus, supporting teachers to spend more time on instruction and less time on class management. Although the activity was occurred in a special education classroom, it is our hope that further studies explore the use of self-monitoring for students with ASD in general education settings to increase their social opportunities with peers in school. We believe that using a handheld device, such as iPad with visual and verbal prompts, together with positive reinforcement to encourage their learning and maintaining appropriate behaviors will assist these students in their academic and behavior improvement, and become independent in school.

Conclusion

One common characteristic of ASD is poorly developed self-management skills, such as difficulty presenting, controlling, maintaining behaviors required by the class routine (Wilkinson 2008). Self-monitoring is a strategy used to teach students with ASD to be more independent and responsible for their own behaviors. The Apple iPad with self-modeling provides an opportunity for students to view their own appropriate behavior through visual modeling. The access of such a handheld device for students with ASD to learn appropriate skills and to manage their own behaviors can prepare them for the increased behavioral demands of the school environment. We believe that self-monitoring using visual modeling with an iPad will be an emerging and promising technology for students with ASD to learn self-control, and foster their independence in school.