Abstract
Children with autism spectrum disorder (ASD) seemingly prefer tablets to other methods of intervention, though professionals posit that a therapist’s presence is beneficial. In this experiment, two unique e-stories were created specifically for children with ASD. Ten children with low- or nonverbal ASD heard each story presented on an iPad; one narrated through the iPad, and one read by a therapist trained in Applied Behavioral Analysis (ABA). The within-participants design enabled researchers to measure the individual benefit of each type of story reading. After hearing each book, participants were tested on their comprehension via multiple choice, sequencing, and open-ended questions. Results indicated that the ideal condition for these young children with low- and nonverbal ASD was the therapist-read condition. These findings suggest that simple stories presented on a rewarding platform, with the minimally intrusive assistance of a therapist, provide young, verbally-challenged children with ASD with a foundation for receiving the benefits of shared storybook reading.
I Introduction
Imagine a parent reading aloud with a typical, curious four-year-old. As they share the story, they are enjoying the development of events in a narrative, coming across new words and ways of describing things in the world or in their imaginations, and are likely engaging in some side-commentary full of questions and answers. All of this helps a typically developing child learn rapidly and deeply about language as well as events in the world through shared storybook reading (e.g. Blewitt et al., 2009; Duursma et al., 2008; Sénéchal and LeFevre, 2002).
Now, let’s imagine how this scenario would work with a parent and a young child with autism. The parent opens a book and begins to read a story to their child. The child may display little interest in the text and lose focus as the parent points to different book features. The child may even become frustrated, overwhelmed, or distracted. The social component (e.g. parents’ extra-textual talk), which is so highly valued for the learning opportunities available and the engagement of a typically developing child (e.g. Walsh and Blewitt, 2006), can actually interfere with the ability of children with autism spectrum disorder (ASD) to learn story content (e.g. Ganz et al., 2014). Could we transform this reading experience, making it more closely resemble the positive and beneficial experience of a neurotypical child if the print book was replaced with a tablet-based e-book, the parental input reduced, and the child’s autonomy increased? Children with autism may find it easier to attend to and be motivated by one source of input with limited extra-textual talk (Khowaja and Salim, 2013; Moore and Calvert, 2000). They might be able to listen more consistently to narration that reads and asks questions in a predictable format. Potentially, autonomous e-reading (or listening to) a story may better prepare a child on the autism spectrum to understand its content. While previous work has focused on extra-textual talk, recent findings suggest that even just a difference in who reads (e.g. an adult or the e-book itself) can affect how the child responds (Ciffone et al., 2016; Dore et al., 2018). For example, Ciffone et al. (2016) found that children attended more to a self-narrated e-book, but were engaged in more interactive dialogue about a story when it was read by a parent.
Why might young children with ASD experience relief from an alternative reading environment? Because children with ASD learn differently than their neurotypical peers. Within the population of children with ASD, many learn language through non-standard methods and at widely variable rates (Lorah et al., 2014). Some children with ASD are nonverbal for most or all of their life, and can learn to communicate through gestures, PECS (Picture Exchange Communication System), or AAC (Augmentative and Alternative Communication) devices (Flores et al., 2014). A nonverbal or low-verbal classification does not necessarily mean that children with ASD cannot learn to use speech, but learning may be less automatic and more deliberate compared to neurotypical language development (Lorah et al., 2014). In addition to basic language challenges such as deficient vocabularies, or difficulties understanding complex sentence structures, many children with low-verbal ASD also display lessened responsiveness to social and nonverbal stimuli (Ganz et al., 2014).
It can be the social nature of shared language experiences that is particularly challenging for children with ASD. Autism has been associated with deficits in theory of mind (Gökçen et al., 2016; Ozonoff et al., 1991), executive function (Craig et al., 2016; Gökçen et al., 2016), and information processing (Foss-Feig et al., 2017; Mazza et al., 2017), all essential for a child trying to track verbal stimuli and comprehend a narrative. Impairments in such processes can create a stressful internal environment that further interferes with the individual’s understanding of and response to the world (Blank and Cull, 2009; Twachtman-Cullen, 2006). Twachtman-Cullen (2006) summarizes this stressful internal environment as a two-way interaction between stress and communication: as children with ASD become more stressed, their ability to communicate decreases. Therefore, the more a parent, teacher, or guardian asks a child a question the more likely the child is to feel overwhelmed and experience difficulty with how/what to respond and might alternatively engage in a challenging behavior.
The use of technology to mediate learning and therapy is often preferred for many children on the autism spectrum as it can help tune out surrounding social interaction and even help in communication (Hong et al., 2017; McEwen, 2014; Mineo et al., 2009; Shane and Albert, 2008). Ganz and colleagues (2014) found that tablet-based visual scripts (e.g. a step-by-step script about hand washing) help improve communication skills for young children on the autism spectrum. They found that more spontaneous speech occurred when children were given visual prompts on an iPad compared to during the non-treatment trials when they received potentially distracting or interfering prompts from another person. Similarly, Lorah et al. (2014) found that within-stimulus prompts were enough to teach picture symbol discrimination to children with ASD without the need for an instructor to intervene or assist.
Despite this support for the idea that some children with ASD may be more comfortable learning from a device than from a live adult, there is debate between professionals about whether the presence of a therapist is more beneficial than a tablet in working on the functional communication abilities of children on the autism spectrum (Couper et al., 2014). While some professionals who work with children with ASD embrace the use of tablets, others remain skeptical, worrying that tablets distract children from communicating with other people. It is important to note, though, that independent iPad use may decrease anxiety for children on the autism spectrum and limit environmental distractions. Not surprisingly, it appears that after independent use, children with ASD may show a decrease in the frequency of their challenging behaviors, an increase in their story retention, and an improvement in their social skills (Couper et al., 2014; Dixon et al., 2015; Hourcade et al., 2011). Further, when language and social skills are honed, researchers have found a marked increase in self-confidence (Campigotto et al., 2013).
Technology is advancing and becoming more accepted in classrooms each day, which supports the learning, communication, and daily lives of children with autism (Campigotto et al., 2013; Dixon et al., 2015; Kagohara et al., 2013; Neely et al., 2013; Shane et al., 2012). Currently, digital tools can combine static, inanimate images with auditory cues which can help remind and cue individuals with ASD to complete important tasks (e.g. personal hygiene, scheduled activities) and learn play skills (Murdock et al., 2013). Even applications commonly found on everyday devices support the communication of children with ASD. McEwen (2014) found that iPod Touch apps (e.g. Applied Behavioral Analysis Basic, iCommunicate, and Proloquo2Go) gave such individuals a viable platform to communicate, which reinforced their desire and ability to engage in a self-initiated manner with peers. Furthermore, researchers posit that mobile technologies increase feelings of awareness, social acceptance, and empowerment for children with ASD (McNaughton and Light, 2013).
Research supports the finding that tablets are popular and viable technological aids for individuals with developmental disabilities and can be effective for children as young as three and four years old (e.g. Dixon et al., 2015; Flores et al., 2012; Ganz et al., 2013; Kagohara et al., 2013; Xin and Leonard, 2014). Some parents believe that, in today’s technologically-driven world, early exposure to tablets benefits children with special needs by giving them more confidence using technology and teaching them necessary, lifelong skills. Additionally, children with autism seem to focus well when using technology (Neely et al., 2013). Educators also seem to prefer tablets for their ease of use, lack of extensive teacher preparation, and their ability to increase the speed at which students can communicate their needs and preferences (Flores et al., 2012).
Tablets, such as iPads, are a common platform for social stories (e.g. stories that focus on emotions, routines, and specific socials skills), and social stories presented on iPad-like devices seem to be especially beneficial for children with autism (Vandermeer et al., 2015). In fact, social stories have been used as a method to teach language, behavioral, and social skills (Hanley-Hochdorfer et al., 2010; Leaf et al., 2012; Schneider and Goldstein, 2010). Social stories are realistic narratives that contain pertinent cues and the perspectives of others while encouraging appropriate behaviors during novel or difficult social settings. They are used to practice challenging routines (e.g. riding on the school bus), decrease challenging behaviors (e.g. tantruming), strengthen prosocial behavior (e.g. having a quiet mouth and calm hands), and reduce anxiety surrounding new situations (e.g. sitting next to a new person). Often, social stories are read multiple times to ensure comprehension and prediction of events and feelings (Leaf et al., 2012; Vandermeer et al., 2015). Shared reading with a parent can also be beneficial for children with ASD when their particular individual needs are considered. Similar to the use of technology aimed to reduce stress and allow learning to be individualized to a particular child with ASD, perhaps parents and therapists can also use shared storybook reading to meet a child where they are and still help to deliver benefits (Boyle et al., 2019; Fleury and Hugh, 2018).
Though the literature is clear that tablets and similar technologies benefit children with autism, it is unclear just how technology compares to a live human interaction which can, hypothetically, better facilitate the child’s interaction with an e-book. Less is known about children on the spectrum with low verbal skills as much of the extant literature focused on the use of technology in therapy has involved participants with higher functioning skills (Hong et al., 2017). Because of the potential for the use of iPads to reduce children’s stress and still present rich information through stories, our study was designed to investigate whether custom, researcher-created, simple e-stories could be an effective teaching tool for children with low- and/or nonverbal autism. Further, we aimed to test whether such books, when read by a therapist trained in Applied Behavioral Analysis (ABA), were potentially more distracting to children with ASD than the same stories narrated by an iPad alone would be, or whether the therapist’s support and interaction was necessary.
II Method
The experimental protocol was approved by the Santa Clara University Institutional Review Board for Social Behavioral Educational research (protocol ID: 15-01-613). Signed consent was obtained from a parent or guardian of each participant and verbal assent was given by each participant before any information was collected and before the e-stories were read.
1 Participants
Ten children participated in this study; six males and four females. Participants’ ages ranged from 3.5 to 10 years, with a mean age of 5.5 years (SD = 2 years). Researchers recruited children ages 3–10 with nonverbal to low-verbal autism whose parent reported their diagnosis to the research team, who relied at least partially on a communication device (e.g. PECS, Proloquo2Go), and received Applied Behavioral Analysis (ABA) therapy weekly. Commonly, children with autism receive ABA therapy weekly to target and improve their behaviors, communication, and social skills. ABA is a household regular in the United States to many nonverbal and low-verbal children with ASD; because of this, the present study modeled reading sessions after ABA sessions to decrease anxiety and increase familiarity for study participants. All participants received ABA therapy at the time of the study, ranging from five to 21 hours per week, with a mean of 12.95 hours (SD = 6). Four participants were categorized as low-verbal because they were able to independently say a few words or phrases (e.g. ‘Cookie’ rather than, ‘May I please have a cookie?’), while six participants were categorized as nonverbal because they relied on alternative sources of communication (see Table 1). Participants were recruited by advertising to parents of children on the autism spectrum through flyers and word of mouth that researchers were seeking volunteers to participate in a study targeting children with nonverbal or low-verbal autism. The recruitment process required continued community outreach; researchers recruited at various locations in a large, metropolitan area in the western United States. Mainly targeted were schools, Applied Behavioral Analysis agencies, pediatric clinics, a local Autism Speaks walk, and autism support groups on Facebook. All of the participants lived in a large metropolitan region of the western United States. Additionally, all participants had prior experience with Apple brand tablets (iPads) similar to the iPad used in the study (for detailed participant information, see Table 1).
Participant details.
Note. PECS = Picture Exchange Communication System.
2 Materials
The two e-books used in the present study, Matt goes to school and Beth goes to the grocery store (for sample pages, see Appendix 1), were researcher-created specifically for this study and are described in detail below. We created a simple Powerpoint with the images, text, and a pre-recorded narration that would automatically play on each page in order to present the children with a 10-page e-book that either they or the researcher could swipe through. The narration was recorded with a clear, male, child-friendly, and appropriately-paced voice.
The e-books created for this study had purposeful and unique features different from those in e-books designed for typically developing children in the same age range. These books featured short, literal sentences, storylines taking place in short time frames, pictures of real people performing the actions described in the text on the same page (e.g. Buckleitner, 2010), and an off-white background that reduced visual overstimulation (e.g. Leaf et al., 2012; Vandermeer et al., 2015). The books described everyday events (getting on the school bus and getting into the car), which are often challenging for children with ASD.
In addition, questions assessing the participants’ comprehension and memory for each story were incorporated into each e-book at the end of the narrative. These questions were also designedly geared towards a population of children with autism who have verbal limitations, which meant gesture answers were possible and also frequently given by participants. A variety of question types were developed to assess retention as well as story comprehension (e.g. Kunda and Goel, 2010). First, three multiple-choice questions served as story-retention questions. Multiple choice questions focused on actions and objects, such as, ‘Who did Beth go to the grocery store with?’ with answer options of, ‘her mom’, ‘her dad’, or ‘her teacher’ (e.g. Buckleitner, 2010). Next, a simple, three-step sequencing task was included to measure comprehension and memory of the story, without relying on a participant’s verbal vocabulary. This type of task has been shown to be more accessible for children with low-verbal ASD (Allen et al., 2015; Kunda and Goel, 2010; Twachtman-Cullen, 2006). Finally, participants were asked two open-ended questions which related the storylines to their lives. The first question simply required a ‘yes’ or ‘no’ response (e.g. ‘Do you ride on the school bus?’) and the second question challenged the child to make a personal connection to the story, requiring a certain amount of self-awareness (e.g. ‘What do you buy at the grocery store?’). These questions allowed the participants to demonstrate understanding and recall of the story in alternative ways as well as to bridge their personal experiences to the story plot.
3 Procedure
The study used a within-participants design in which each participant heard two stories, one in each of two conditions (‘e-book’ or ‘therapist’) during one session. Both order of condition and the two stories were counterbalanced by randomly assigning children to an order (either e-book or therapist first, and either Matt goes to school or Beth goes to the grocery store first). During the session, a researcher sat with the participants somewhere in their home or school, in a place where they often participated in ABA therapy.
In the e-book condition, the session began with five minutes of warm-up time in which the researchers did not interact with the participant but let him/her play in the same room to feel more comfortable in the researcher’s presence. Then, the child and researcher sat at a table or on the ground as the iPad narrated the story. The child was able to hear the story and interact with the device. The researcher did not verbally engage with the child, and nonverbally intervened only if the child was engaging in challenging behaviors, such as eloping or self-injurious behaviors. After the story was narrated by the iPad, it automatically read the three multiple choice questions, the instructions for the sequencing task, and the two open-ended questions while allowing ample time for the child to respond to each question. The researchers silently observed and recorded the child’s behaviors during the story as well as their answers to the questions; in the iPad condition, no researcher intervention was used to redirect the participants back to the story if they became distracted. Similar to other studies (e.g. Buckleitner, 2010), researchers measured both correct and incorrect gestural (e.g. pointing, head-nodding) and verbal responses to all questions as well as echolalia and challenging behaviors throughout the duration of the storybook readings.
In the therapist condition, five minutes were timed during which the participant could play and interact with the researcher in an effort to become more familiar with one another. Then, the ABA therapy-trained researcher read the e-book story aloud to the participant, swiping each page manually, and then asked him/her all of the retention questions verbally, giving time for responses in between questions while simultaneously tracking responses, challenging behaviors, and echolalia. In the therapist condition, gentle and noninvasive methods (e.g. guiding the iPad back to the participant or saying ‘the story is almost over, then you can play’) were used to redirect participants back to the story if they became distracted.
In both conditions, a second ABA therapy-trained researcher inconspicuously double-recorded answers to questions, distracting behaviors, and echolalia. The two coders met and spoke directly after each session to corroborate their observations and ensure that each session was recorded correctly. No discrepancies in coding were found.
4 Measures
a Retention measures
Retention of both stories was measured by examining the children’s scores on the multiple-choice questions, the sequencing task, and the two open-ended questions. Points were given for each of these three retention tests when a child indicated a correct response either verbally or gesturally, resulting in a total possible score for ‘multiple choice responses’ out of 3, for ‘sequencing’ out of 3, and for ‘open-ended responses’ out of 2. For open-ended responses, half points were given for verbal responses that were not relevant but still indicated attempts to communicate a response to the question (e.g. echolalic sounds, repetitive noises) for each child in each condition.
b Challenging behaviors
Throughout the reading and questioning period of each story, separate instances of challenging behaviors were recorded by the researchers. These were defined as behaviors that intruded on the reading of the stories or the questions, such as a participant getting up from his/her chair and leaving the room, or a participant requesting to engage in an alternative task. Additional behaviors included: yelling, tantruming, stereotypy, pica, eloping, and aggression (e.g. Buckleitner, 2010). Across both conditions, the number of such instances was tallied into a raw count of each child’s challenging behaviors.
c Echolalia
Echolalia was also tracked during the reading of the e-books and during the question periods, as it can indicate engagement and verbal language development (Charlop, 1983; McDaniel et al., 2018). Echolalia was recorded in ‘instances’ and not duration; one instance was defined as 15 seconds or more of the echolalic behavior; a break for three or more seconds was required before another instance could be counted. Across both conditions, echolalia was tallied into a raw count for each child.
d Grouping of participants by verbal ability
We categorized children as nonverbal if their parents reported that their main source of communication was a pictorial communication system (e.g. PECS), sign language, or gesturing and the researchers witnessed no verbal utterances. Children were classified as low-verbal if their parents mentioned that they use a pictorial communication system, but the child was able to verbally answer questions the therapist asked with or without prompting or if the parents reported that the child’s main source of communication was verbal.
e Grouping of participants by level of focus
We also categorized each child’s level of focus in each condition into one of three groups by qualitatively examining their challenging behaviors (e.g. eloping, tantruming) and the foci of their attention (e.g. on the iPad or on the researcher), along with prosocial behaviors such as sitting, listening, and echolalia. Children who were attending to the story and exhibiting no distracting challenging behaviors were assigned 3’s, children with some moderate distraction (e.g. stereotypy, eloping, pica) were assigned 2’s, and children displaying a significant number of challenging behaviors, distraction, and an absence of prosocial behaviors were assigned 1’s. With these categorizations, we were able to broadly evaluate the level of focus each child displayed in the e-book or the therapist condition.
III Results
1 Retention in e-book vs. therapist conditions
In our first round of analyses, we compared children’s performance in the ‘e-book’ condition versus the ‘therapist’ condition on the three retention variables: multiple choice questions, sequencing, and open-ended questions. In all three of these retention variables, children’s mean scores were greater in the therapist condition than in the e-book condition. However, using dependent samples t-tests, we found that the only retention variable in which children performed significantly better in the therapist condition was open-ended questions (Mtherapist = .75, SD = .63 compared to Me-book = .25, SD = .42), t(9) = 3.35, p = .008, with a large effect size, d = −1.162. Thus, overall the 10 participants showed better retention by our measures in the therapist condition than in the e-book condition (see Table 2).
Means and standard deviations (SD) of raw score for each measure.
2 Analysis based on verbal ability
In an effort to explore our findings more deeply, we also conducted sub-analyses of the differences between children’s responses in each condition based on their categorically assigned level of verbal ability (either ‘nonverbal’ or ‘low-verbal’) as described above. The results from the analyses described below should be considered preliminary due to the small sample size of the present study. Using repeated measures analyses of variance (ANOVA), we found there were significant main effects of condition for the multiple choice questions (F(1,8) = 14.26, p = .005, η2 = .25) as well as for open-ended questions (F(1,8) = 78.40, p < .001, η2 = .62) such that, as above, children scored higher on these two retention tests in the therapist condition than in the e-book condition. There were also significant interactions of condition with verbal ability for both multiple choice answers (F(1,8) = 35.20, p < .001, η2 = .61) and open-ended responses (F(1,8) = 40.00, p < .001, η2 = .32), such that children classified as nonverbal performed equally or better when reading and testing with an e-book alone compared to with a therapist-read e-book. But, the pattern was strikingly different for children categorized as low-verbal, who responded more often and more accurately with the therapist than with the e-book alone.
3 Challenging behaviors, echolalia, and level of focus during the stories
As described above, for each participant in each condition, instances of challenging behaviors were counted. These total counts did not show any significant differences by condition; the mean number of challenging behaviors across participants was 5.6 in both the e-book condition and the therapist condition. There was the appearance of a pattern in which nonverbal children had a higher average number of challenging behaviors in the therapist condition (M = 8.0) compared to the e-book condition (M = 6.3) while low-verbal children showed the reverse, with fewer challenging behaviors in the therapist condition (M = 2.0) compared to the e-book condition (M = 4.5). However, none of these differences, nor the interaction between condition and verbal ability, were statistically significant.
In addition, there were no overall differences between conditions in average amounts of echolalia measured during the e-book reading. In an analysis of variance (ANOVA) of echolalia with both condition and verbal ability as independent variables, only a main effect of verbal ability was found, such that children who were classified as nonverbal produced more instances of echolalia across conditions (M = 4.58) than did children who were considered low-verbal ((M = .63), F(1,8) = 9.40, p = .048, η2 = .40). However, there was no interaction between verbal ability and condition in regard to amount of echolalia.
Despite these null condition effects in challenging behaviors and echolalia, we were still interested in whether the more qualitative description of children’s level of focus as they heard the e-books read could have had an impact on their retention in each condition. As described above, each child was given a qualitative categorization of level of focus on a scale of 1–3 (with 1 being seriously distracted and 3 being fully attentive). The children’s relative levels of focus between conditions separated them into four groups as follows: two children were fully focused (3’s) in both conditions; one child was seriously distracted (1’s) in both conditions; three children were more focused with the therapist reading the e-book than with the e-book alone; and four children were more focused in the e-book condition than the therapist condition (see Figure 1).
Level of focus of each participant by condition.
Interestingly, a child’s level of focus and grouping was uncorrelated with their level of verbal ability. And secondly, each of these subgroups demonstrated the same pattern of better performance on the retention measures in the therapist condition compared to the e-book condition. For the two children who were fully focused in both conditions, they scored higher on multiple choice questions and open-ended questions in the therapist condition, as did the child who was distracted for both conditions. For the three children who were more focused in the therapist condition, they scored higher on all three retention measures in the therapist condition than in the e-book condition. Most surprisingly, even the four children in the sample who were more focused in the e-book condition while the story was being read still scored higher on all three retention measures in the therapist condition.
This first, small-scale study of its kind found that, counter to our original hypothesis, children with low- and nonverbal ASD performed better on retention measures of a simple e-story in a condition where an ABA therapy-trained therapist read the e-book to them as compared to listening to the story autonomously in an iPad-only condition. The trend towards better performance with the therapist was present in all three retention measures, but only statistically significant with the most challenging open-ended questions. Additionally, when children were grouped based on their low- or nonverbal status, we found significantly higher performance by low-verbal children in the therapist condition for two of the three retention and comprehension measures (multiple choice and open-ended questions), but for nonverbal children there were no differences in retention, which was already quite low, between conditions. Finally, we also found that the children’s level of focus during reading demonstrated no preference for one condition over the other and was entirely uncorrelated with the children’s generally better performance on retention in the therapist condition. These results show that even when the participants seemed to be more comfortable or engaged when they had the iPad all to themselves, they actually did better when they had a therapist present to guide and scaffold them with minimally invasive and gentle redirection.
IV General discussion
E-books can be effective media for exposing children on the autism spectrum to simple social stories and potentially teaching them about language and the social world. However, these preliminary results lend support to the idea that the therapist is still a useful mediator of e-stories for children with verbal challenges. This finding is congruent with recent research; Hong and colleagues (2017) conducted a meta-analysis and found that tablet-based interventions are more effective for children who are on the higher-functioning end of the autism spectrum. Thus, it is likely that the tablet-based stories in the present intervention still required therapist scaffolding due to the nature of the population, as participants were either nonverbal or low-verbal and also quite young. It appears that once children with autism reach the stage where they can manage an iPad alone, they may be less likely to be distracted by the behaviors of themselves or others. With the help of a minimally involved therapist, children with low-verbal ASD can potentially focus relatively well on the iPad and answer retention questions that indicate some comprehension of the story and how it relates to their lives. This study adds to what we already know about how technology can benefit children with ASD (e.g. Flores et al., 2014; Lorah et al., 2014), by reminding us that learning from stories is multifaceted and depends on a child’s readiness: An e-version of a story on an iPad may not simply be a ‘one size fits all’ solution, and as researchers and practitioners in the field we should be cognizant of when a child is ready for autonomous learning and when they may still need some scaffolding.
We can make fewer conclusions based on this work for our nonverbal participants, who still struggled to respond to the story retention questions asked in either condition. However, one important finding was that nonverbal participants engaged in more echolalia during both readings of the stories than did their low-verbal peers. This may indicate an attempt to relate in a proto-verbal way with the story, the therapist, or both (Prizant and Duchan, 1981). We interpret this as a sign that even nonverbal children were still benefiting from these e-book readings.
The initial findings presented here were obtained in a therapist-child situation, in which the trained therapist/researcher read the stories in a straightforward, didactic manner. Our research does not yet speak to whether more natural, parent-read interactions would result in improved retention with a parent present rather than a child autonomously using a tablet to hear a story. We cannot, with the present findings, account for the additional quality and/or stress of extra-textual conversations and questions that might occur with a parent, as so often do between a reader and a neurotypical child (Blewitt et al., 2009; Whitehurst et al., 1988). Research has found that while e-books are effective in aiding neurotypical children’s retention of words and story events, e-books read by a caregiver rather than self-narrated are better at promoting meaningful conversation (e.g. Ciffone et al., 2016; Dore et al., 2018; Fleury and Hugh, 2018). However, since this type of caregiver reading may be quite different from a therapist reading, future research should test this comparison between adult-guided and autonomous reading with children on the autism spectrum.
In addition, the present study tested relatively young children with very limited verbal capacities; this sample allowed for a small snapshot of a particular developmental period. In addition to having diagnoses of autism with verbal challenges, some of the participants we worked with in this study were not yet four years old. However, as time progresses and these children mature, their familiarity with tablet technology as well as their verbal abilities will certainly change. As done here, basing classification on parental report and dividing children into only two categories based on verbal ability limited the power of the analyses. Future research could use finer grain classification of children’s verbal abilities; e.g. children’s verbal Autism Diagnostic Observation Schedule (ADOS) scores. Future research should also investigate children at older ages, with increasing abilities, and their preferences for autonomous e-book reading compared to having the support of an adult. The current study leads us to conclude that young children on the autism spectrum who are low- to nonverbal benefit more from having a therapist as a reader rather than an e-book alone. However, other work, such as a study by Moore and Calvert (2000), has found that children even of a similar age range, but who may be more skilled verbally, actually learned more vocabulary in a computer-only instruction as compared with a live teacher. Thus, further research needs to capture the larger spectrum of verbal abilities within ASD, specifically including children with autism who are low- and nonverbal, in order to further capture and understand these differences. Additionally, a future limitation of the present study may be the use of researcher-created retention measures. Based on the research we conducted to inform our creation of these retention measures (Allen et al., 2015; Buckleitner, 2010; Kunda and Goel, 2010; Twachtman-Cullen, 2006), we believe that the retention measures used were able to accurately capture story retention in these participants. Furthermore, future research could focus on creating a standardized format of retention-based questions use specifically with this population. In doing so, it would be important to include items that can be answered using gestures, perhaps PECS, and also verbally to account for the primary means of communication for participants in this population.
Finally, in the interest of expanding these findings, we would like to consider and test a widening range of story topics, including other narratives of simple everyday behaviors that young children with ASD typically need support with at home or in the classroom (e.g. washing hands, putting on shoes, waiting in line). Because of practical constraints, in the study presented here children only heard one story presented in either condition, and both conditions were part of a single session. As well as introducing possible order effects for each child, this also limited our ability to look at more long-term effects of repeated reading in each condition. Research should, therefore, examine possible behavioral impacts of what can be learned by repeated exposure to the stories either read alone or with a caregiver, and the practical effects that listening to these stories may have on children’s comprehension, retention, and even real-life application of the stories. Investigating the longer-term learning potential from these simple e-stories, with regard to changes in children’s behavior in similar situations as those described, could provide a window into how impactful the right story on the right platform with the right kind of scaffolding could be.
In sum, we have found that simple stories presented on a rewarding platform, with the minimally intrusive assistance of a therapist, seem to provide young children with verbal challenges on the autism spectrum with a foundation for receiving benefits of shared storybook reading. As further research examines factors which affect how children with autism can learn from shared storybook reading, custom e-books may prove to be a valuable asset to therapy and education in myriad ways.
