Abstract
Social Personalized Human-Machine Interaction for People with Autism: A close look to Proprioceptive and Kinematic Cues
Nowadays, research in socially assistive robotics (SAR) [1] is in expansion. One of the target populations is people suffering of
Autistic Spectrum Disorders (ASD). Individuals with ASD have impaired skills in communication, interaction, emotions recognition,
joint attention, and imitation [2], [3], [4], [5]. Many studies show that children with ASD have a great affinity with robots, computers
and mechanical components [6]. Moreover, in SAR robots have been used several times as tools in socialization therapies for children
with ASD [7], [8], [9], [10]. Because the cognitive and functional profile of people with ASD persist with development at an individual
level, a personalized therapy plan needs to be proposed. This aspect is mainly examined in our work. We work in collaboration with
two Associations (see ) dedicated to people suffering of ASD and we have access to two age groups: children and young adults.
Visual and sensory motor impairments are also present in ASD (see [11], [12]). It is suggested that individuals with ASD
manifested an exacerbated reliance on proprioceptive feedback, and a deficient use of kinematic visual cues. A link between the
individual integration of proprioceptive and/or kinematic feedback and communication, interactions skills, and emotions recognition
have already been discussed in [13], [14], [15], [16]. In our work, in order to define individual profiles among the autistic participants,
we make the hypothesis that the individual’s responses to proprioceptive and kinematic cues affect the interactions between the human
and the robot. We posit that an individual strongly dependent on proprioceptive information and independent on the visual field has
more difficulties to have successful interactions.
In our research work, we investigate and evaluate the participants’ profiles with the help of two methodologies. First, The
Adolescent and Adult Sensory Profile (AASP) developed by Brown and Dunn [17], which allows to assess an individual’s sensory
processing preferences. Secondly, with the help of an experimental setup that we designed and that measures the effect of a swaying
virtual room for different difficulties of balance on the variations of the Center of Pressure (CoP) in children and adults with ASD.
Participants were asked to stand in front of a virtual scene rolling with a frequency of 0.25 Hz in their line of sight with an inclination
of ±10 ◦
. The first results are promising.