=Paper= {{Paper |id=Vol-2730/paper33 |storemode=property |title=LI-AR: an integration of technology and ABA methodology to improve communicative behavior in autism |pdfUrl=https://ceur-ws.org/Vol-2730/paper33.pdf |volume=Vol-2730 |authors=Roberta Simeoli,Luigi Iovino,Angelo Rega,Davide Marocco |dblpUrl=https://dblp.org/rec/conf/psychobit/SimeoliIRM20 }} ==LI-AR: an integration of technology and ABA methodology to improve communicative behavior in autism== https://ceur-ws.org/Vol-2730/paper33.pdf

LI-AR: An integration of technology and ABA methodol-
ogy to improve communicative behavior in autism
Roberta Simeoli1, Luigi Iovino2, Angelo Rega1, Davide Marocco1
1
University of Naples Federico II, Department of Humanistic Studies, Naples, 80138, Italy
2
Neapolisanit S.R.L. Rehabilitation Center, Ottaviano NA 80044, Italy
roberta.simeoli@unina.it

Abstract. Previous research on the relative benefits of Speech Generating Device
(SGD) as Augmentative and Alternative Communication Systems (AACs) has
indicated controversial results regarding the effectiveness on improving commu-
nication skills and vocal production and on decreasing problem behaviors for
children with autism spectrum disorder (ASD). The Language Interface for AAC
Rehabilitation (LI-AR), is a new type of SGD. Its software creates innovations
for teaching social interaction skills. Unlike the most popular SGDs, during the
training with LI-AR, the vocal output is fully managed by a communication part-
ner/therapist. The software is based on Applied Behavior Analysis methodology,
and it allows to personalize teaching procedure in an automated way. The study
presents the changes in communication skills and speech production reached by
a child with ASD, after the LI-AR training. Results show a significant increase
in communicative behaviors and a statistically significant improvement in the
quality of speech production.

Keywords: Speech Generating Device, Augmentative and Alternative Commu-
nication, Autism.

1 Introduction

According to the National Institute on Deafness and Other Communication Disorders
(NIDCD, 2010), about 25% of people with ASD are unable to use language naturally
to communicate.
AAC tools offer a communication forms that may make it easier for children with ASD
to communicate their needs. Three forms of AAC used with children with ASD include
unaided approaches (e.g., signs and gestures), low-tech image-based systems (e.g.,
PECS), and "hi-tech" speech generating device systems (SGDs) [1].
The effectiveness of an AAC method depends on the ability to customize the tool with
which the child will interact.
SGDs are devices designed to produce recorded or synthesized vocal outputs. They are
designed to accommodate a variety of configurations that allow for customization and
individualization, which is particularly helpful for students with severely limited lan-
guage skills [2].

Copyright © 2020 for this paper by its authors. Use permitted under Creative Com-
mons License Attribution 4.0 International (CC BY 4.0).
2

Many SGD studies demonstrated an increase in manding skills [3,4,5,6] while a few
studies demonstrated a more general increase in vocal production [4] or other commu-
nicative and social interaction skills [7].
LI-AR is a new type of SGD. Unlike the most popular SGDs, during the training with
LI-AR, the vocal output is fully managed by a communication partner (therapist).
Through a Bluetooth device, the therapist decides to activate the vocal output only if
the student has successfully completed the communication task, approaching the ther-
apist and delivering him the tablet, as for the Exchange Phase II of PECS [8]. The LI-
AR procedure focuses on “exchange” and “interaction” with the communication part-
ner as pivotal behaviors to communicate. Using this procedure, the communication
does not take place without interaction. The LI-AR procedure is innovative in the field
of SGDs as it requires the child to approach his communicative partner and start an
interaction with him before being able to issue the communication act.
The theoretical background that inspired LI-AR is based on the Applied Behavior Anal-
ysis methodology and on the concept of “Mand” as proposed by Skinner in his Verbal
Behavior theory (1957) [9].
For the first time, with LI-AR, these methodologies merge the new technologies capa-
ble of increasing the motivation to communicate and interact with the external environ-
ment (eg audio-video technology, NFC, Bluetooth, wi-fi, Augmented Reality, Virtual
Reality).
The aim of the present study is to observe the effects of the LI-AR training on commu-
nication skills and vocal production in a child with ASD and communication impair-
ment.

2 Method

2.1 Participants
The study was attended by Dario, a 4-years-old boy with an autism spectrum disorder
diagnosis. He did not use language in a functional way, he presented some echolalia of
portions of words and generally monosyllabic vocalizations; a sign language training
had been introduced, but it was unsuccessful due to Dario’s propensity for scrolling and
was subsequently discontinued. Dario had never used and SGD before the study. Com-
munication and fine-motor skills were previously assessed thought the Vineland test.
Results showed a value of ≤ 2.0 years for communication skills and a value ≥ 1.0 years
for the fine-motor motor skills.

2.2 Materials and Settings

The study was conducted in a room 5 m by 3 m, in the presence of a, therapist and an
assistant as prompter. The room contained a table, and two chairs positioned one in
front of the other. The LI-AR software was installed on an Android Tablet 6.0. The
tablet was always present on the table during the sessions.
3

2.3 Procedure
At the beginning of the study a preference assessment was conducted according to the
multiple stimulus-without-replacement assessment procedure [10].
An A-B-A single subject design [11] was used to compare performance before, during
and after the training with LI-AR.
The baseline sessions lasted 10 min during which the therapist kept the student engaged
in work or play activities and created 5 opportunities to request preferred items, show-
ing the child the most motivating objects, arranged on a tray, positioned to be visible
but not reachable by the child.
During the training session the student was thought to use the tablet to request what he
wanted. The tool provides five distinct phases, in this study, the student attended three
of these phases and a fourth phase adapted for the study, as follow:
Phase I The child is in presence of his major reinforcement. When his motivational
operation (MO) is revealed, the therapist prompted the student to drag the image corre-
sponding to the desired item towards the box positioned at the top of the screen (Fig.
1). This action will produce the vocal output and the student obtain his reinforcer.
Phase II In this phase the student learned to “exchange”. The vocal output was produced
by an external action of the therapist, only after the child has given the tablet to the
therapist. During this phase, the therapist works on distance and persistence, gradually
moving himself and the tablet away from the child.
Phase III In this phase pictures showed on the screen were managed by the therapist.
Some distractors were randomly presented on the screen (Fig. 2) in order to teach the
discrimination skill.
Phase IV In this phase pictures on the screen were categorically divided in different
sheets attainable by swiping from right to left and vice versa on the tablet screen (Fig.
3). The student was completely free to choose his motivation. Each complete exchange
was considered a functional communication behavior.

Fig. 1 Phase I/II layout Fig 2. Phase III layout Fig 3. Phase IV layout

Follow-up: Two months after the last observation, three probe sessions were proposed
to observe the maintenance of the acquisitions. These sessions were arranged as for the
baseline.
To observe changes in communicative behaviors, we recorded the percentage of occur-
rence of request and other communicative behaviors. Vocal production was recorded
during 10 min sessions divided into 20 intervals of 30 sec. We calculated an index of
“quality of speech” for each request occasion: 0 points were attributed if no vocalization
occurred; 0.5 points for non-functional and not understandable vocalization; 1 points
for each vocalization that was functional and understandable, but not completely pro-
nounced (e.g., at least half of the syllables with which the original word is composed);
2 points for understandable and functional vocalization. (Figure 2).
4

2.4 Results
During the baseline (Fig. 4), Dario did not show functional communicative behaviors.
He presented some not understandable vocalizations and few approximations of words.
He reached the 100% of communicative behavior after 4 probe sessions during the
training. A statistical analysis confirmed the results of the visual inspection. It was pos-
sible to observe a statistically significant change in the trend of communicative behav-
ior, with a p-value <.001. These data were analyzed by a second independent observer
for the 50% of the sessions. An Interval by Interval IOA method showed an accordance
of 95%.
Baseline Training Follow-up
Communicative behaviors

Sessions

Fig. 4. Percentage of intervals in which a communicative behavior appears (understandable
vocalization, word production, pointing, sign, use of LI-AR.

The graph in Figure 5 shows the trend of Dario's speech productions. The visual in-
spection revealed that the vocal production increased in terms of quality, from an aver-
age score of 0.4 during the baseline to 1 point during the follow-up, confirming an
improvement in the “quality of speech”. The quantity of attempts was stable with a
slight decrease at the beginning of treatment. At the middle of the treatment Dario
showed, for the first time, understandable and functional vocalization. These data were
analyzed by a second independent observer for the 50% of the sessions. An Interval by
Interval IOA method shown an accordance of 95%.

Fig. 5. Quality and quantity of vocal production
5

During the baseline Dario obtained a maximum of speech quality score of 0.66, after
training he reached a maximum score of 2, indicating that the child tended to emit more
often words that were both grammatically and functionally correct. Moreover, results
indicated an average score of 2 during the follow-up, confirming the consistency of
these improvements.

3 Discussion

The first aim of the study was to determine if children with a diagnosis of ASD could
be taught to use LI-AR for requesting purposes. The data indicated that LI-AR was an
effective tool for Dario to improve his communicative skills.
The second aim of the study concerned whether children with communication problem
could improve their vocal production using LI-AR. Results revealed that Dario has ob-
tained an improvement in vocal production (Fig. 5). Dario showed vocalization at the
beginning of the study. During the treatment, he acquired the ability to produce whole
words, completely understandable. These results revealed that an AAC training did not
interfere with the development of vocal communication and that instead, it can be ex-
tremely useful to improve the quality of speech production.
These results confirmed the literature according which the digitalization of training and
assessment tools can facilitate learning processes and increase motivation in patients
with ASD [12,13].
Nonetheless, this study extends the existing literature in two ways: it demonstrates that
an adapted ABA protocol can be successfully used to teach communication through an
SGD; children with ASD and few vocal repertoire can improve in speech production
through the use of an AAC tool.

References
1. Ganz J. B.: AAC Interventions for Individuals with Autism Spectrum Disorders: State of
the Science and Future Research Directions. Augmentative and alternative communication
(Baltimore, Md.: 1985), 31(3), 203–214 (2015).
2. Lloyd, L. L., Fuller, D. R., & Arvidson, H. H.: Augmentative and alternative communica-
tion: A handbook of principles and practices. Boston: Allyn and Bacon (1997).
3. Durand V.: Functional communication training using assistive devices: effects on challeng-
ing behavior and affect, Augmentative and Alternative Communication, 9(3), 168-176
(1993).
4. Schlosser, R. W., Sigafoos, J., Luiselli, J. K., Angermeier, K., Harasymowyz, U., Schooley,
K., & Belfiore, P. J.: Effects of synthetic speech output on requesting and natural speech
production in children with autism: A preliminary study. 1, 139–163 (2007).
5. Sigafoos, J., Didden, R., & O'Reilly, M. Effects of Speech Output on Maintenance of Re-
questing and Frequency of Vocalizations in Three Children with Developmental Disabili-
ties. Augmentative and alternative communication (Baltimore, Md.: 1985), 19(1), 37–47
(2003).
6. van der Meer, L., Didden, R., Sutherland, D., O'Reilly, M. F., Lancioni, G. E., & Sigafoos,
J.: Comparing three augmentative and alternative communication modes for children with
6

developmental disabilities. Journal of Developmental and Physical Disabilities, 24(5), 451–
468 (2012).
7. Sigafoos, J., Green, V. A., Payne, D., Son, S. H., O'Reilly, M., & Lancioni, G. E.: A com-
parison of picture exchange and speech-generating devices: acquisition, preference, and ef-
fects on social interaction. Augmentative and alternative communication (Baltimore, Md.:
1985), 25(2), 99–109 (2009).
8. Bondy, A. S., & Frost, L. A.: The picture exchange communication system. Seminars in
speech and language, 19(4), 373–424 (1998).
9. Skinner, B. F.: Verbal behavior. New York: Appleton-Century-Crofts. (1957).
10. DeLeon, I. G., & Iwata, B. A.: Evaluation of a multiple-stimulus presentation format for
assessing reinforcer preferences. Journal of applied behavior analysis, 29(4), 519–533
(1996).
11. Hersen M.: Single-Case Experimental Designs. In: Bellack A.S., Hersen M., Kazdin A.E.
(eds) International Handbook of Behavior Modification and Therapy. Springer, Boston, MA.
(1990)
12. Simeoli, R., Arnucci, M., Rega, A., & Marocco, D.: A comparison between digital and tra-
ditional tools to assess autism: Effects on engagement and performance. CEUR Workshop
Proceedings, 2524. (2019).
13. Simeoli, R., Rega, A., & Marocco, D.: Movement detection software to enhance autism as-
sessment processes. 1–4. (2020).