=Paper=
{{Paper
|id=Vol-1388/demo_paper7
|storemode=property
|title=Adaptive Applications to Assist Students with Autism in Succeeding in Higher Education
|pdfUrl=https://ceur-ws.org/Vol-1388/demo_paper7.pdf
|volume=Vol-1388
|dblpUrl=https://dblp.org/rec/conf/um/GarciaSB15
}}
==Adaptive Applications to Assist Students with Autism in Succeeding in Higher Education==
https://ceur-ws.org/Vol-1388/demo_paper7.pdf
Adaptive Applications to Assist Students with
Autism in Succeeding in Higher Education
Alejandro Montes Garcı́a, Natalia Stash, and Paul De Bra
Eindhoven University of Technology (TU/e), Eindhoven, The Netherlands
{a.montes.garcia, n.v.stash, p.m.e.d.bra}@tue.nl
Abstract. In this demo we discuss a few possible scenarios showing
adaptation of presentation and information to assist autistic students in
succeeding in higher education. These students not only have specific
information need, they are also more concerned about their privacy. We
use WiBAF (Within Browser Adaptation Framework) for user modeling
and adaptation to give users control over the sharing of their data.
1 Introduction
Adaptive frameworks have been used to take into account various features of the
users when presenting information. One of the dimensions that can be considered
for adaptation is individual traits – the features that together define the user
as an individual: personality traits, cognitive factors, cognitive/learning styles
(CS/LS). If we consider user’s context, among other factors of environment that
are not really “user” features, we can distinguish affective state such as mo-
tivation, frustration, engagement. In the European project Autism&Uni1 the
transition of young adults on the autism spectrum to a university environment
is being studied as well as ways to help them navigating through the university
system and succeeding in their studies in the best possible way. Like other in-
dividual traits autism has everything to do with the personality. Having autism
can also influence the affective state such as getting anxiety in certain situations
which are not problematic for people without autism. Although apparently all
of us have autism to a certain degree.
Despite the existence of a large number of tools to support autistic people,
especially kids to develop certain skills, to our knowledge there are no tools to
support students with autism during their studies in higher education. We also
need to take into account that autistic people tend to be more concerned about
their privacy than other people. We think that this is the area where adap-
tive technologies could help and more specifically we consider the new generic
adaptive framework that we are developing at the TU/e called WiBAF [6].
WiBAF is radically different from the previous platforms AHA! [3] and its
successor GALE [7] in that in WiBAF the user modeling and adaptation can both
be performed completely within the browser. This property makes it possible to
1
http://www.autism-uni.org/
offer end-users the power of making decisions in the Privacy vs. Personaliza-
tion trade-off described by Kobsa [5]. It also makes it easier to add adaptive
functionality to existing websites without the need for major changes or soft-
ware installation. WiBAF has been used successfully by TU/e students in an
“Adaptive Web-Based Systems” course already, to develop a variety of adaptive
applications with client-side user modeling and adaptation.
This demo paper is the first attempt of the authors within the scope of the
Autism&Uni project to show possible ways of adapting information to autistic
students. We identify two different ways in which WiBAF can be useful:
1. customising the web page. At the moment the adaptation decisions are based
on the information about autism collected through a number of workshops at
TU/e that involved several experts in the domain (see Acknowledgements).
2. allowing students to make decisions about what is done with their data.
Those are discussed in sections 3 and 4 respectively. But first, in section 2, we
talk a bit more about autism and briefly describe the situations when assisting
students with autism would be helpful.
2 Use scenarios for assisting autistic students
One of the major problems with autism can be described as follows. Non-autistic
people are good at making semantic connections between topics that they study.
On the other hand, autistic people are good at “seeing trees in the forest”, they
like details, sometimes maybe too much, but they have difficulties setting up
semantic connections between the details thus with “seeing the whole forest”
[2]. It feels like autistic people have their own ways of processing and analysing
information, their own ways of learning and as such can be linked to different
CS/LS. Furthermore, typically the social skills of autistic people are underde-
veloped which results in their preference for individual rather than group work.
Within Autism&Uni project about 20 scenarios were collected describing the
situations when students with autism need support. Because of the lack of space
we discuss only two typical scenarios:
1. Lectures, group work and other study activities differ between high school
and university. Students are assumed to have a higher level of independence,
lectures move from small classrooms with tens of students to lecture halls
with hundreds of students, etc. It takes more effort for students with autism
to adapt to these new ways of working. In order to reduce their anxiety we
can show the students what situations they will face in a customised way,
according to their “CS/LS”.
2. People with autism find it difficult to find their way around big areas such as
a university campus and suffer from anxiety when they do not know where
to go. They need to know when and where the next lesson is going to be,
how to reach the library, the cafeteria, registration offices, etc. They need
very concrete pointers, not a site overview to explore.
3 Possible customisation
Regarding the adaptation to the study environment of a large lecture hall (first
scenario), autistic students need to be prepared before actually going into a
noisy hall, going through crowded corridors, etc.. Such introductory information
to reduce anxiety is not needed for non-autistic students. For presenting this
information we need to distinguish types of students depending on the order
they prefer to learn dealing with new situations and on the type of content that
is more effective.
As discussed in the previous section we consider that the needs of autistic
students can be linked to different CS/LS. In [8] we showed possible ways of
inferring and adapting to three dimensions of CS/LS: global vs analytic, verbal vs
visual, active vs reflective. Concentration on details characterises autistic learner
as having analytic style. As such (s)he can be guided in depth-first order through
the domain hierarchy. Need for structure, clear goal, help with understanding
of the semantic connections between the topics suggest that the student may
need graphical representation of the domain. This need can be linked to visual
style and could be helpful in developing global style. For students with autism
every new situation feels like new because they have difficulties with establishing
semantic connections between similar situations. But for one specific situation
we could have a set of instructions on how to deal with it and then the student
can turn this knowledge into real action. This can be linked to reflective style.
We could therefore consider autistic student as analytic, visual and reflective
who prefers to read more details about specific situations, one by one, see exam-
ples on how to deal with these situations, watch a video with an overview of the
university life to get a broader picture and then turn this knowledge into action.
Not all autistic students however are the same. Opposite global, verbal and ac-
tive preferences or other combinations of those could also be possible. For this
reason our domain experts suggest that as discussed in [8] it might be interest-
ing to try inferring different patterns by observing student’s browsing behaviour
while interacting with the application and to adapt information correspondingly.
For the second scenario, the student can be helped by entering the schedule
into a system that will then present him/her a step by step procedure. This is
again related to the fact that autistic students like to concentrate on one task at
a time. Overall, in this day and age young people are used to a very busy (or even
fancy) interface style on informational websites. According to the domain experts
autistic students on the contrary need a simple interface (without distractions)
in which only the most relevant information is shown, with a very limited number
of options. Throughout the process of acquiring information they may also need
frequent reminders of the essential information.
In the demo session we will show the scenarios’ implementation in WiBAF.
4 Taking decisions about what is done with the data
According to our domain experts, people with autism tend to have less trust
in other people. Therefore it is expected that they will also not trust services
that store and use their data. People in general are becoming more concerned
about privacy issues [1], so this does not only apply to autistic users. To develop
trust, it is important for services to allow users to scrutinize their user profiles
[4]. WiBAF not only allows to inspect, modify and delete parts or the whole
user model, but also allows the user to define rules about where the user data
is stored, in what level of detail and for which usage. Moreover, it allows them
to specify situations in which they should or should not be tracked. In exchange
for data the user discloses, the service should offer better customisation and
recommendations as described in Section 3.
5 Conclusions and Future Work
In this paper we presented our first ideas of how adaptive technologies, in particu-
lar WiBAF, could help students with autism reduce their anxiety when studying
at the university. To do so we looked at two specific scenarios. More scenarios and
more types of adaptation to support autistic students will be developed in the
future. Even though the current demo has been prepared using guidelines given
by the domain experts, evaluation of these and future scenarios with autistic
students is yet to be done by the universities participating in the project.
Acknowledgments. This research is supported by the project WiBAF and the
Lifelong Learning Programme (project no. 10018300 Authew 539031-LLP-2013-
1-UK). The authors thank the Autism&Uni project team, TU/e psychologists
specialising in autism, TU/e students with autism who participated in interviews
with us, Stichting Handicap+Studie 2 and Autastic! 3 for their valuable input.
References
[1] TRUSTe 2014 GB Consumer Confidence Privacy Report: Consumer Opinion
and Business Impact. Technical report, TRUSTe, 2014.
[2] T. Attwood. The Complete Guide to Asperger’s Syndrome. 2000.
[3] P. D. Bra, D. Smits, and N. Stash. The Design of AHA! In Proc. of the 17th
ACM Hypertext Conference, 2006, Odense, Denmark, pages 133–134, 2006.
[4] J. Kay. Scrutable Adaptation: Because We Can and Must. In Adaptive
Hypermedia and Adaptive Web-Based Systems, volume 4018 of Lecture Notes
in Computer Science, pages 11–19. Springer Berlin Heidelberg, 2006.
[5] A. Kobsa. Privacy-Enhanced Web Personalization. In The adaptive web,
pages 628–670. Springer, 2007.
[6] A. Montes Garcı́a. WiBAF: Within Browser Adaptation Framework. In
Extended Proc. of UMAP Conference. CEUR-WS, 2014.
[7] D. Smits. Towards a Generic Distributed Adaptive Hypermedia Environment.
2012.
[8] N. Stash. Incorporating Cognitive/Learning Styles in a General-Purpose
Adaptive Hypermedia System. 2007.
2
http://www.handicap-studie.nl/
3
http://royhoutkamp.nl/