=Paper=
{{Paper
|id=Vol-2892/paper-14
|storemode=property
|title=Robot personality for cognitive training
|pdfUrl=https://ceur-ws.org/Vol-2892/paper-14.pdf
|volume=Vol-2892
|authors=Eleonora Zedda
}}
==Robot personality for cognitive training==
https://ceur-ws.org/Vol-2892/paper-14.pdf
Robot personality for cognitive training
Eleonora Zedda1,2
1
ISTI-CNR Pisa, Via Giuseppe Moruzzi, 56124 Pisa PI, Italy
2
University of Pisa, Largo Bruno Pontecorvo, 3, 56127 Pisa PI, Italy
Abstract
Robots are becoming more present in our daily activities and in particular in the health context. To
improve the human-robot interaction in a training session it is important to design and develop social
behavior and personality in robots. Recent studies found that personality is an essential feature for cre-
ating socially assistive robots. For this purpose, I want to investigate if the robot personality (extrovert
or introvert) can improve the user’s cognitive performances in elders with Mild Cognitive Impairments
during one-o-one cognitive training.
Keywords
Human-Robot Interaction, Socially Assistive Robots, Personality,
1. Introduction
In the last decade, the aging of society is occurring worldwide. By 2050, the number of individuals
over the age of 85 is projected to be three times more than today [1]. Aging has a considerable
impact on the health of older adults, and most of them need physical, social, and cognitive
assistance. At the same time robots have become more common in our daily life and they are
no longer only supplementary tools for laboratories. Recently, robots, called socially assistive
robots (SAR) are employed to assist older adults. SAR can aid users through social interaction
to enhance the quality of life and reach optimal results in terms of training tasks. In my Ph.D.
project, I want to use a SAR with older adults affected by Mild Cognitive Impairments (MCI),
which is a progressive cognitive decline between normal aging and dementia [2]. It affects the
completion of complex tasks usually performed with simplicity in normal health conditions,
such as cooking and managing the home. Moreover, every year 10% of elders with MCI have
a high risk to degenerate into dementia. Existing studies on robot personality are limited
because they are based on the assumption that robots are tools to be used rather than social
companions to interact with. For this reason, in my Ph.D. project, I want to investigate the
effects of different robot personalities on elders during cognitive training. In fact, personality
can be a key element for creating SAR that can facilitate human-robot interaction(HRI) and
provide better engagement in the users.
CHItaly 2021 Joint Proceedings of Interactive Experiences and Doctoral Consortium, July 11–13, 2021, Bolzano, Italy
" eleonora.zedda@phd.unipi.it (E. Zedda)
0000-0002-6541-5667 (E. Zedda)
© 2021 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
CEUR
Workshop
Proceedings
http://ceur-ws.org
ISSN 1613-0073
CEUR Workshop Proceedings (CEUR-WS.org)
78
Eleonora Zedda CEUR Workshop Proceedings 78–82
2. Related work
The benefits of SAR in heath context with elderly are investigated in the research by Tanaka et
al.[3]. The researchers compared a speaking humanoid Kabochan Nodding ROBOT with the
same robot but without the communication elements. In the 8-week trials with MCI patients, the
researchers found improvements in cognitive functions. Pino et al. [4] evaluate the effectiveness
of human-robot interaction(HRI) to reinforce therapeutic behavior. In the experiment with
MCI patients, they found improvements in differents cognitive domains and that HRI can
reinforce therapeutic behavior. In literature for social robots performing assistive tasks with
personality, Tapus et al. [5] investigate the role of the robot’s personality in a therapy process for
rehabilitation of post-stroke users. In the study, they used an ActiveMedia Pioneer 2-DX mobile
robot designed to assist, encourage, and socially interact with post-stroke users engaged in
rehabilitation exercises. As result, they found the first evidence for the preference of personality
in the assistive domains.
3. Proposal
In 2019 the HIIS laboratory of CNR-ISTI in conjunction with the Neuroscience Institute of CNR
of Pisa collaborated to investigated how seniors with MCI perceived cognitive training with
a humanoid robot [6]. As result, we found that the robot seems to improve the engagement
in the user, in fact, the older adults considered the humanoid robot as if it had human traits.
Taking into account this experience, my Ph.D. research addresses the limitation in prior work:
personality. The idea behind this project is to design and implement an extrovert and introvert
personality in a robotic system. Research question: Can a social robot showing a personality
improve cognitive performance and engagement in the user?
Robot system The humanoid robot I will use is called Pepper Robot, which is developed by
Softbank Robotics. Pepper is a 1.2-m-tall wheeled humanoid robot, with 17 joints for expressive
body language. Pepper has multimodal interfaces for interaction: touchscreen, speech, tactile
head, hands, bumper, LEDs, and 20 degrees of freedom for motion in the whole body. It hears
thanks to four directional microphones on his head, which allow it to detect the provenance
of sounds and voices and turn its face in the direction of those who are talking. The robot
speaks or reproduces sounds thanks to two speakers in the ears, and it is also equipped with
4 microphones on the head. For the vision, it is equipped with two identical 2D cameras, a
3D camera, and a stereo camera. To develop personality in the robot I will use the QiSDK for
android, a Java library that offers a set of modules to program the robot.
Behaviour and personality In literature, different personality models are proposed. I con-
sider the Five-factor model[7] because a considerable amount of research indicates that extrover-
sion and introversion are the most observable personalities. For this reason, I want to implement
extrovert and introvert personalities in a humanoid robot during cognitive training with three
serious games. After a preliminary interview with the psychologies and therapist of the clinic
the two personalities have been identified and, before the user test, they will be validated
79
Eleonora Zedda CEUR Workshop Proceedings 78–82
Table 1
Personality Parameters
Extrovert Introvert
Verbal
Intonation: Joyful Neutral
Pitch variation: pitch set at 150 [50-200] pitch set as default 100
Volume: set 90 % maximum [0-100%] set 50% maximum
Speak rate: 170 words per minute 150 words per minute
Rythms variation: Variation rhythm set 2 [0-2] Variation rhythm set 0
Feedbacks: reinforcement feedback neutral feedbacks
Non-verbal
LEDs light: shiny colours, flashing light neutral colors or light colors
no flashing
Gesture: both arms, head, torso movements one arm with small angles,
with big angles, faster responses slower response
Set of animations: animations that convey neutral animations
different emotions
Moving speed: 40% faster than introvert movements slower movements
Sounds: movement associated with sounds no sounds or melody
and melody
Autonomous movements: autonomous movements few autonomous movements
by psychologists and therapists. Table 1 summarizes the verbal and non-verbal cues used to
modulate the robot’s behavior. For the verbal parameters, I identify: modulation in the robot’s
intonation between neutral, joyful, and didactic; variation in the pitch and rhythm; changes
in the volume and speech rate; different set of feedbacks. For the non-verbal cues, I identify:
LEDs with different colors and intensities(e.g flashing light), customized set of animation that
can convey different emotions; creation of complex gesture with different angles; association
of movements with sounds; modulation of the movement’s speed; addition of autonomous
movements (e.g. natural movements of the arms). For each interaction with the users, Pepper
will provide a different combination of verbal and non-verbal cues according to the personality
performed.
Scenarios In this paragraph, I describe two scenarios in which I propose a possible use of the
robot personality during cognitive training with older adults. Mario is a 75 years old teacher
and since he has a diagnosis of MCI, he regularly attends cognitive training in a cognitive center
near his house. Recently a one-o-one cognitive training with a humanoid robot was integrated
into his program by the clinic. Monday morning, as usual, he prepares and goes to the clinics.
The psychologists welcome and take him to the training room. After the introduction phase in
which the robot introduces the cooking game and the ingredients of the first level:
Extrovert scenario.
1. Pepper says with a joyful intonation and a high-pitched voice "Which is the first ingredient
to cook the chicken curry: chicken, curry, salt or yogurt?" highlighting its LEDs with a
bright blue and moving both its arms over its head.
80
Eleonora Zedda CEUR Workshop Proceedings 78–82
2. When Mario answers right to the question, immediately Pepper says "Good job Mario,
you are doing a great job!". At the same time, it raises both arms two times and nods
its head, the eyes LEDs become bright green and flashing, and it produces a sound that
expresses joy.
3. Pepper says "How many grams of flour are needed for the recipe: 10, 150, 200, 400 gr?".
Pepper highlights its LEDs with a bright blue and opens both arms at 40 degrees outwards
as to encourage the user.
4. For the first wrong answer, Pepper says with a loud voice and an enthusiastic intonation
"You made a mistake, but you can try again!". When Mario answers wrong more than one
time Pepper says, " Mario, stay focus! You know the answer, try again!". At the same time,
it shakes the head and moves the arms up and down, an "oh" sound is vocally provided,
and all the LEDs become bright red and flashing.
5. Before the game end, Pepper combines different gestures to simulate a victory animation
accompanied by sounds of applause, while all the LEDs are flashing and become green
and blue. In the end, Pepper salutes Mario saying "Really good job Mario, it was fun to
play with you! Hope to see you again next time!". During the cognitive session, Pepper
performs autonomous movements (e.g simulate breathing movements).
Introvert scenario.
1. Pepper says with neutral intonation and a low volume "Which is the first ingredient to
cook the chicken curry: chicken, curry, salt or yogurt?". After a few seconds, it slowly
undulates the torso twice to the right and left while the shoulders LEDs become light
gray.
2. When Mario answers right, Pepper says with a neutral intonation and a slow speech rate
"Good". It waits two seconds and slowly it nods its head, while the eye LEDs become light
green without flashing.
3. Pepper says "How many grams of flour are needed for the recipe: 10, 150, 200, 400 gr?" It
gently opens its right arm to 20 degrees outwards right while the eyes LEDs become a
light yellow.
4. When Mario answers wrong, Pepper says "Wrong, try again." It lightly shakes the head
one time, while the eyes and shoulders LEDs become light red.
5. In the end, Pepper moves backward and forwards one arm and it salutes Mario saying
"Good job Mario, see you next time". During the cognitive session, Pepper performs a
restricted set of autonomous movements.
Methods and evaluation I want to insert the adaptation of personality in a robot system
for supporting cognitive training according to the users’ personalities. For the evaluation
study, I define two experimental conditions: an extrovert condition in which the robot per-
forms an extrovert behavior during the training and an introvert condition in which the robot
performs an introvert personality. Before and after the test, the users will be subjected to
the UES questionnaire [8] for evaluating the engagement of the users, the Big Five Inventory
test for evaluating the users’ personalities [9] and the Godspeed Questionnaire Series [10].
During the cognitive training, I will collect data to evaluate the improvement in the memory
81
Eleonora Zedda CEUR Workshop Proceedings 78–82
domains stimulated: reaction time, number of the wrong/right answers, time session, number
of attempts, etc. Learning which robot’s personality better fits the user is important to evaluate
the improvements in attention, and also to improve the HRI. Creating a robotic system capable
of adapting its personality to the users could be a useful tool to stimulate the elders to continue
the therapy, and moreover to achieve progress in their rehabilitation. In conclusion, I want to
evaluate if different robot personalities adapting to the user can create an enjoyable interaction;
engage more the user during the tasks; increase the user’s attention, and consequentially if can
improve the user’s task performance and cognitive functions.
Acknowledgments
I would like to thanks my Ph.D. advisors Prof. Fabio Paternò and Dr. Daniele Mazzei for their
help, support, and guidance.
References
[1] World Health Organization, World report on ageing and health (2015). URL.
[2] N. A. Bishop, T. Lu, B. A. Yankner, Neural mechanisms of ageing and cognitive decline,
Nature 464 (2010) 529–535.
[3] M. Tanaka, A. Ishii, E. Yamano, H. Ogikubo, M. Okazaki, K. Kamimura, Y. Konishi, S. Emoto,
Y. Watanabe, Effect of a human-type communication robot on cognitive function in
elderly women living alone, Medical science monitor: international medical journal of
experimental and clinical research 18 (2012) CR550.
[4] O. Pino, G. Palestra, R. Trevino, B. De Carolis, The humanoid robot nao as trainer in a
memory program for elderly people with mild cognitive impairment, International Journal
of Social Robotics 12 (2020) 21–33.
[5] A. Tapus, C. Ţăpuş, M. J. Matarić, User—robot personality matching and assistive robot
behavior adaptation for post-stroke rehabilitation therapy, Intelligent Service Robotics 1
(2008) 169–183.
[6] M. Manca, F. Paternò, C. Santoro, E. Zedda, C. Braschi, R. Franco, A. Sale, The impact of seri-
ous games with humanoid robots on mild cognitive impairment older adults, International
Journal of Human-Computer Studies 145 (2021) 102509.
[7] J. Costa, T. Paul, of personality theories: Theoretical contexts for the five-factor model,
The five-factor model of personality: Theoretical perspectives 51 (1996).
[8] H. L. O’Brien, P. Cairns, M. Hall, A practical approach to measuring user engagement with
the refined user engagement scale (ues) and new ues short form, International Journal of
Human-Computer Studies 112 (2018) 28–39.
[9] B. Rammstedt, O. P. John, Measuring personality in one minute or less: A 10-item short
version of the big five inventory in english and german, Journal of research in Personality
41 (2007) 203–212.
[10] A. Weiss, C. Bartneck, Meta analysis of the usage of the godspeed questionnaire series, in:
2015 24th IEEE International Symposium on Robot and Human Interactive Communication
(RO-MAN), IEEE, 2015, pp. 381–388.
82