-
Psychological Effects of a Self-Sufficiency Model
Paper:
Psychological Effects of a Self-Sufficiency ModelBased on Urge
System
Teruaki Ando and Masayoshi KanohSchool of Information Science
and Technology, Chukyo University
101 Tokodachi, Kaizu-cho, Toyota 470-0393, JapanE-mail:
[email protected]
[Received April 17, 2010; accepted July 8, 2010]
In recent years, robots to coexist with humans havebeen
developed. Their ability to communicate is indis-pensable for their
coexistence with humans, so stud-ies on the interaction between
humans and robots areimportant. This paper proposes a model of the
self-sufficiency system of a robot, in which we apply theurge
system to the autonomous system of emotion. Inthis model, a robot
expresses its changing psycholog-ical and physiological conditions
(physiological loadcondition) and conveys them sensitively to the
user.This is expected to result in a mental interaction ef-fect
between the user and the agent. We carry outsimulation experiments
on this model and verify thepsychological interaction between the
software robot(agent) and the user. As a result of these
experiments,it is recognized that the agents with the ability to
prop-erly express physiological load among those with thismodel
implemented have a tendency to receive higherevaluations from their
users.
Keywords: urge system, self-sufficiency, human robotinteraction,
human symbiosis systems, Ifbot
1. Introduction
In recent years, studies have been done on the interac-tion
between humans and robots or software agents thathave artificial
emotion implemented. Shibata et al. [1, 2]have developed a pet
robot that acts in such a way thatit appears to have the autonomy
or emotions to interactphysiologically with humans. They have shown
that thepet robot may have the spiritual effects of giving peo-ple
pleasure or peace of mind through this interaction.Takeuchi et al.
[3] have proposed an emotion genera-tion model based on the
likability rating of dialogists in arobot. This model leads to the
implementation of a robotthat changes its behavior depending on the
reactions of thedialogist. Most of these studies focus mainly on
entertain-ment, or on producing feelings of pleasure, peace of
mind,or interest in human subjects. In almost all cases,
there-fore, the system in which emotion is implemented dealswith
how to express or represent it. However, if the pur-pose of adding
emotion artificially to agents is their co-
existence with humans and their ability to communicatewith them,
studies on systems that autonomously activatethe emotion function
in an adaptive way to environmentscan be considered important.
Toda [4, 5] considers that emotion is an inherently
mo-tivational program in which appropriate motives functionin
emotional situations, and Toda proposed the urge sys-tem based on
that assumption. This study also pays atten-tion to autonomous
systems possessed of emotion, and itaims at the development of an
autonomous activity sys-tem for an agent, a system based on
artificial emotion. Itfocuses specifically on the self-sufficiency
function of theautonomous activity system, which supplies its own
en-ergy. This paper verifies what psychological effects workon
humans in their interaction with an agent that
“satisfiesself-sufficiency by others.”
Self-sufficiency is the ability of an agent to maintain it-self
for a long period of time, and it is one of five charac-teristic
concepts (situatedness, autonomy, self-sufficiency,embodiment, and
adaptivity) relative to complete au-tonomous systems [6, 7]. For a
robot, for example, fuelis supplied to maintain a battery level.
Self-maintenanceis often thought to be done intuitively in this
self man-ner, but there are times when this is done “in terms
ofother persons” in the real world. One of these cases isthe
self-sufficiency of a human baby, which is absolutelydependent on
its mother and satisfies its self-sufficiencythrough its mother. In
other words, a baby expresses theinstability of its psychological
or physiological conditionor the dissatisfaction of its physical
condition with its fa-cial expressions or physical motions. This
results in let-ting its mother know what conditions need
improvement.This paper is based on this concept and proposes a
self-sufficiency model to which an urge system is applied.
Inexperiments, we build up a limited environment in whichthe urge
system specific to an agent works in virtual space(Fig. 1). We do
this to carry out simulation experimentson this model, and we
verify the effects of the psycholog-ical interaction between the
agent and the user.
2. Urge System
Urge theory is an emotion theory that extends con-ventional
emotional concepts. This theory assumes that
Vol.14 No.7, 2010 Journal of Advanced Computational Intelligence
877and Intelligent Informatics
-
Ando, T. and Kanoh, M.
Fig. 1. Experimental environment overview.
emotion has an advanced wilderness rationality and is
anenvironmentally adaptive selection system (mental soft-ware) that
is limited by the basic framework geneticallygiven. This mental
software controls the mental andphysical abilities, such as the
physiological activation oflearning, cognition, memory,
consciousness, and physicalfunctions, along with the purpose unique
to the activatedemotion. An emotion started with a “purpose” in
mind iscalled an “urge,” that with no purpose in mind is called
a“mood-state,”1 and the mental software is called the
“urgesystem.”
2.1. UrgesThe urge system divides conventional concepts of
emo-
tion into urges and mood-states. An urge is an “angryurge”
generated when “your possessions are intentionallydestroyed by
another person.” In addition to such an urgebased on a general
emotional concept, an internal signalurge, such as “hunger,” is
understood as one caused by theindividual function of the mind and
a “physiological urge”assumed to be generated. The urge system is
largely char-acterized by dealing with the physiological functions
thatcannot generally be considered as emotional functions inthis
way.
2.2. Mood-StatesA mood-state is a function of the mind that
doesn’t
have the purpose of its own activity, in contrast to an
urge,which is a function of emotion or the heart with its ownunique
purpose. For example, the function of “joy” or“sadness” becomes a
mood-state. An action plan, such asthe “punishment action plan for
assaulters,” is always gen-erated in an angry urge, but such an
action plan does notexist in joy or sadness. In other words, the
joy or sadnessurge does not exist, and it is due to the
“demonstrationurge.” On the other hand, a mood-state has an
importantrole and causes a specific urge to be easy or difficult
toactivate. That is, it acts to encourage the activation or
in-activation of the start of an urge.
1. In this paper, we do not consider emotional attitude, which
is the thirdclass of emotional concepts after urges and
mood-states.
Urges
Post-mortemphase
Decision-making phase
Action-controlphase
Physiological Anger Surprise DemonstrationFear
desirable
User
Increasing joy
Increasing sadness
Choosing action plan
Yes
No
Helping (doing something)
Activation phase Situation cognition
Facial expressions
BehaviorVocalizing
Fig. 2. Overview of proposed model.
2.3. Four Phases of Urge ActivitiesThe urge system has four
phases; activation phase,
decision-making phase, action-control phase, and post-mortem
phase. All urges function in these phases and areprocessed in a
step-by-step manner. The following is a de-scription of the
processing in each phase. The activationphase is in charge of the
situation cognition process. Thedecision-making phase is in charge
of determining actionplans generated by urge activation. The
action-controlphase is in charge of the starting and ending of the
imple-mentation of a determined action plan. The post-mortemphase
is in charge of the starting of urges to execute re-consideration
or learning. In this paper, we model thesefour phases of urge
activities and implement them in thecomputer agent.
3. Self-Sufficiency Model Based onUrge System
Subsection 3.1 gives an overview of the proposedmodel and
Subsection 3.2 describes the specific imple-mentation content.
3.1. OverviewThe proposed model is overviewed in Fig. 2.
This
model was based on four phases of the urge function tobuild up a
series of flow for sequential processing on astep-by-step basis. It
recognizes information from an en-vironment (situation cognition),
activates an urge in re-sponse to a situation, and then selects an
action plan thatcan be taken based on the urge activated in the
decision-making phase for a specific action (facial
expression,etc.). The activity of this action-control phase causes
thepsychological condition of an agent to be represented,which
makes it possible for the user to observe a changein the agent.
This allows the user to dynamically approachthe agent. An approach
from the user is capable of chang-ing the situation cognition of
the agent and is used tomodel the post-mortem phase. Specifically,
if a change
878 Journal of Advanced Computational Intelligence Vol.14 No.7,
2010and Intelligent Informatics
-
Psychological Effects of a Self-Sufficiency Model
induced by the user approach is desirable for the agent,the joy
mood-state is increased, and, if not, the sadnessmood-state is
increased. Through the flow outlined above,it is expected that the
agent can achieve self-sufficiency ina manner dependent upon
another person (user).
3.2. Implementation of Urge SystemIt is first necessary to
determine the situation cognition
to start various urges for the agent.
3.2.1. Situation CognitionThis model divides the input system of
the agent into
the internal and external cognition systems. The internalsystem
is considered to consist of two parameters: one isits own
physiological load p(t) at time t and the other thefavorability
rating f (t) for a certain target of attention.2The values of these
parameters are accumulated throughthe following equations.
P(t) = αpP(t −1)+ p(t), . . . . . . . (1)F (t) = α f F (t −1)+ f
(t), . . . . . . . (2)
where αp and α f are decay parameters, and 0 < αp,α f < 1.
Note that the physiological load types exist forthe number of
internal signals. For example, consid-ering a robot agent, a
physiology urge can be consid-ered that is activated when the CPU
temperature risesor when the amount of energy remaining in a
batterydrops. Thus, it is necessary to prepare p(t) and P(t)for the
relevant physiological load. That is, the physiol-ogy load
parameter becomes the vector information, suchas p(t) = (p1(t),
p2(t)...), P(t) = (P1(t),P2(t)...). (Thesuffix is an identifier of
the physiology load.)
Variable o(t) was given by the following equation thatdetermined
whether or not the user existed as the externalinput system.
o(t) ={
1 (User exists)0 (User does not exist). . . . . (3)
Processing of the activation phase is determined in ac-cordance
with the situation cognition as described above.
3.2.2. Activation PhaseAttention is paid to the emotions of a
human baby as
an urge needed for self-sufficiency. The emotions of thehuman
baby are developed and specialized from content-ment, interest, and
distress at the time of birth to joy,surprise, sadness, disgust,
anger, and fear (these nineemotions are called the “primary
emotions”) [8]. As inIzard [9], this model takes the position that
“emotion =representation” and considers the correspondence of
theprimary emotions to Ekman’s six basic emotions (anger,surprise,
fear, sadness, disgust, and joy) [10] for easy rep-resentation by
facial expressions. In other words, there
2. A target of attention is a candidate for selection. For
example, if onewants a branch that fits within the palm of the
hand, the targets of atten-tion are the branches appropriate to
that condition (one does not observebigger branches). All targets
of attention at some point in time are col-lectively called the
target system.
are three urges, i.e., anger, surprise, and fear, that can
cor-respond to the general emotions among those placed inthe
activation phase. Joy and sadness are represented bythe
demonstration urge. The physiology urge that dealswith the internal
factors (physiological and physical sit-uation variables) specific
to the agent is considered tobe activated when the physiological
condition degrades,and it corresponds to the representation of
disgust. Werespond to contentment, interest, and distress as
below,which are not made correspondent to the six basic emo-tions.
Contentment is made correspondent to joy whenself-sufficiency is
successful. Interest is excluded becauseit has no relation to
self-sufficiency. Distress is made cor-respondent to disgust.
The activation phase is the one in which various urgesare
activated by situation cognition, but the agent is placedin a
condition that continues to momentarily receive envi-ronmental
inputs and to activate an urge at any time.
The following is a description of the activation condi-tions of
various urges.
(a) Physiology UrgeThe physiology urge is an activity plan that
dependsmainly on internal signals, such as pain, appetite,and
sexuality. The physiology urge correspondingto physiological load i
was decided to be activatedby the following equation:
Pi(t)> θphys, . . . . . . . . . . (4)
where θphys is the threshold of the physiology urge.
(b) Anger UrgeThe nature of anger can be seen from the action
todefend the domain of animals. That is, if another an-imal
intrudes on one’s authorized domain, the occu-pant of authority
activates the emotion of anger anddirects it at the intruder. In
this study, it is consid-ered that the so-called original emotional
function ofa wild animal can be fully referenced. This paper
as-sumes the activation condition of the anger urge tobe an
abnormal physical value in response to an ex-ternal input
(user).
o(t) = 1 and ∃i(pi(t)> θang), . . . (5)
where θang is the threshold of anger urge activation.
(c) Surprise UrgeThe activation condition of the surprise urge
can beconsidered a situation without any expected inputfrom the
anticipated target system3 or with an inputfrom the target among
general targets of attention4that does not belong to this
anticipated target sys-tem. One example is a surprise urge that is
activatedeven if a large sound is expected, but no sound isheard.
The activation condition of the surprise urge
3. The target system fitting established expected conditions.4.
Targets of attention with respect to circumstances highly related
to sur-
vival.
Vol.14 No.7, 2010 Journal of Advanced Computational Intelligence
879and Intelligent Informatics
-
Ando, T. and Kanoh, M.
was decided to be “an input value outside the ex-pected range or
an input from the unexpected inputsystem.”
∃i(pi(t)− pi(t −1))> θsup or o(t) = 1and ∃ j(P j(t)> φsup)
and
∃k(pk(t)> ρsup), ( j �= k), . . . . . (6)
where θsup, φsup and ρsup are the thresholds of sur-prise urge
activation.
(d) Fear UrgeA situation in which the fear urge is activated is
de-termined including situations in which “if you do nottake any
action, you will be hurt” or in which “youhave an extremely small
possibility of controlling thedanger.” The activation condition of
the fear urgewas decided to be the “situation in which no rest
isallowed in case of a high level of danger.” A condi-tion with a
high level of danger can be represented byusing accumulation P(t)
in the internal input sys-tem. A situation in which no rest is
allowed is one inwhich the action of the physiology urge is
restrainedby external input control. From the above, fear
isactivated by the following equation.
o(t) = 1 and ∃i(Pi(t)> θ f ear), . . . (7)
where θ f ear is the threshold of the fear urge activa-tion.
(e) Demonstration UrgeThe demonstration urge is intended for the
socialcognition of improvements in its condition. The ac-tivation
condition of this urge was determined as be-low by using an
increase or decrease in the joy orsadness mood-state depending on
the success of thehelp from the user.
F (t)< θsad or θ joy < F (t), . . . . (8)
where θsad and θ joy are the thresholds of the demon-stration
activation and θsad < θ joy. Note that an ac-tivated urge is the
demonstration urge, but the actionplan varies depending on the
value of F (t).
3.2.3. Decision-Making PhaseThere is not always only one urge
that is activated in
the activation phase. Some urges may be activated at thesame
time depending on situation cognition (in almost allcases). Even if
the same urge is activated, the equiva-lent action plan is not
always determined. However, thedecision-making phase with many
factors that are closelyinterconnected is very difficult to
implement. This paperemploys a decision-making method based on the
priorityranking of urges.
Specifically, the decision-making priority is based thenature of
emotional “interrupt” or “survival.” Each urgeof fear, anger,
physiology, or demonstration is closely
related to the survival action. The priority of thesefour urges
can be determined based on the sequence offear > anger >
physiology > demonstration in terms ofthe avoidance of their own
danger. The “surprise” urge isone that has a strong interrupt
nature rather than survivalnature and is activated to direct the
target of attention toothers when another urge is activated. It is
not activatedinstantaneously at the same time attention is changed.
Inthis way, the surprise urge is the most important urgein view of
changing the target system of attention forsurvival. In
consideration of the above, the priority bystrength of urge was
defined as in the equation below.
surprise urge > fear urge > anger urge> physiology urge
> demonstration urge. . (9)
3.2.4. Action-Control PhaseThe action-control phase is the one
that starts or termi-
nates the actual action in the action plan determined by
thedecision-making phase. For actual processing, when theaction
plan is determined by the decision-making phase inactual
processing, its corresponding action is generated.
3.2.5. Post-Mortem PhaseThe post-mortem phase evaluates whether
or not the
urge action just terminated was successful and what wentwrong in
case of failure, and then it makes a learning cor-rection in the
same type of urge activity. This model feedsback success or failure
in the user’s help to the joy andsadness mode-states. In other
words, when the user’s helpsucceeds, the joy mood-state increases;
when it fails, thesadness mode-state increases. To implement this
matter,the favorability rating should change as follows.
f (t) ={
1 (Help success)−1 (Help failure). . . . . . . (10)
4. Simulation Experiment
This section describes an evaluation experiment on
theself-sufficiency model to which the urge system is ap-plied. In
this experiment, the validity of this model isverified by making a
qualitative analysis of the effectof the mental interaction between
an agent and a personby means of qualitative analysis through a
questionnairebased on the Semantic Differential method (SD
method).Ifbot is used as the agent in this experiment (Fig. 3).
4.1. IfbotAn overview of the Ifbot is shown Fig. 3. Ifbot has
a
length of 45 cm and a weight of 9.5 kg. It has two armsand moves
on wheels. Fig. 4 shows an overview of the fa-cial expression
mechanism of the Ifbot. Ifbot is equippedwith 10 motors and 101
LEDs for its facial expressions.These motors operate the neck on
two axes (θN1, θN2 inFig. 4), the right and left eyes on two axes
(θ (L)E1 , θ
(L)E2 ;
θ (R)E1 , θ(R)E2 ), and the right and left eyelids on two
axes
880 Journal of Advanced Computational Intelligence Vol.14 No.7,
2010and Intelligent Informatics
-
Psychological Effects of a Self-Sufficiency Model
Fig. 3. Ifbot overview.
Fig. 4. Facial expression mechanisms.
(θ (L)L1 , θ(L)L2 ; θ
(R)L1 , θ
(R)L2 ). LEDs are placed in the head (LH ),
mouth (LM), eyes (LE ), cheeks (LC), tears (LT ), and ears.They
generate three head colors (orange, green, and red),one mouth color
(orange), three eye colors (green, red,and blue), one cheek color
(red), one tear color (blue),and one ear color (orange). These
mechanisms enable theIfbot to communicate through a variety of
facial expres-sions.
This paper used the parameters θ (L)E1 , θ(L)E2 , θ
(R)E1 , θ
(R)E2 ,
θ (L)L1 , θ(L)L2 , θ
(R)L1 , θ
(R)L2 , LM , and LT to represent the facial
expressions of the agent.
4.2. ConfigurationIn this paper, we made an application (Fig. 1)
to carry
out a simulation experiment. This application is built upas a
kind of game that images the coexisting task betweena human subject
and a machine while the user helps theagent and aims at achieving
the goal within a limited pe-riod of time. The main window displays
the facial expres-sion of the agent in Fig. 1. The right part of
the mainwindow is the user activation window and the lower partis
the TASK VIEW window. Their extended figures are inFigs. 5 and 6,
respectively.
The user sets up the amount of work with each button:“normal
work,” “rather hard work,” or “hard work.” As
Normal work
Rather hard work
Hard work
Fig. 5. User activation window.
Fig. 6. TASK VIEW window.
the amount of work increases, so do the user’s movementsand
physiological load. The amount of movement is dis-played in the
TASK VIEW window. In this experiment,we assumed that there were
three kinds of physiologi-cal load, and that one of them increased
at random whenthe button was pressed to determine the amount of
work.Thus, the user must decide the condition of the agent fromthe
facial expression displayed in the main window.
The user observes the facial expression of the agent andhelps
the agent with each of the A, B, and C buttons in theuser
activation window. This help causes the cognitionsituation to
change and newly activates a variety of physi-ology, anger, fear,
surprise, and demonstration urges. Theuser looks at these changes
to set up the amount of workand helps the agent again.
This setup is in such a way that an attempt is madefor an agent
to arrive at the goal while the user helps theagent through a
series of these interactions. The condi-tions for terminating each
attempt are met when the agentarrives at the goal and does so
within the limited periodof time. The goal condition is one in
which the profileof the agent displayed in the TASK VIEW window
ar-rives at the goal from its start. The agent shows a
facialexpression adaptive to the environmental input momen-tarily
given in accordance with the amount of work in-structed by the user
(for example, one-step movement forthe “normal” amount of work or
two-step movement forthe amount of “rather hard” work) in the main
window.
Vol.14 No.7, 2010 Journal of Advanced Computational Intelligence
881and Intelligent Informatics
-
Ando, T. and Kanoh, M.
Disgust A Disgust B Disgust C
Joy A Joy B Joy C
Anger A Anger B Anger C
Fear A Fear B Fear C
Sadness A Sadness B Sadness C
Surprise A Surprise B Surprise C
Neutral A
Fig. 7. Action design.
During the interval that “reaction time now” is displayedon the
screen in the user activation window, button inputs(inputs from the
user) are assumed to be unacceptable.
4.3. Verification MethodThe application built in Subsection 4.2
was imple-
mented using 20 university students as subjects. Eachsubject
evaluated three applications as described below.
Application (A), shown in Fig. 1, uses a total of 19facial
expression groups (Fig. 7) as the actions of anagent. There are
three facial expressions for eachemotion and they correspond to
three physiologicalloads. The facial expression generated by a
physiol-ogy urge was assumed to be the facial expression of
Fig. 8. Experimental environment of Application (C).
Table 1. Questionnaire items.
Comfortable–Uncomfortable Gentle–SevereFriendly–Distant
Warm–Cool
Bright–Dark Soft–HardStern–Kind Pleasant–Unpleasant
Plain–Strong-featured Sophisticated–SimpleFree–Busy
Humorous–Serious
Rational–Emotional Complicated–UncomplicatedBrave–Weak
Motivated–Spiritless
Patient–Impatient Discreet–IndiscreetCute–Nasty
Interesting–Uninteresting
disgust representing the feeling of discomfort. Thefacial
expressions of anger, surprise, fear, sadness,and joy were
generated using the various features ofthe face to convey each
emotion. For example, thefacial expression of anger is
characterized by the cor-ners of the eyes being turned up and the
corners ofthe mouth being turned down. We referred to the
re-lationship between the various features of the faceand each
motion for the generation of facial expres-sion patterns, which are
described in [10, 11].
Application (B) has same appearance of Applica-tion (A) as Fig.
1, but uses only the seven facialexpressions of A among those in
Fig. 7 as actions.Thus, the agent corresponds to a facial
expression ona one-by-one basis. A subject cannot read from
theagent’s facial expression which physiological loadamount of the
agent has been degraded.
In Application (C), shown in Fig. 8, subjects cannotobserve any
facial expression of the agent (only “NOIMAGE” is displayed in the
main window). Thus, asubject needs to generate an interaction from
the cur-rent position and progress of the agent as displayedin the
TASK VIEW window.
We asked subjects to respond to the items in Table 1 forthese
three applications. We also recorded the goal-arrivaltimes.
882 Journal of Advanced Computational Intelligence Vol.14 No.7,
2010and Intelligent Informatics
-
Psychological Effects of a Self-Sufficiency Model
0
1
2
3
4
5
6
7
A
B
C
* **
** ** **
* ***
* **
**
p
-
Ando, T. and Kanoh, M.
description ones. It may be necessary to verify and dis-cuss in
detail the viewpoint of the complete autonomoussystem by using an
internal parameter (physiological loadvalue) of the agent. Themes
of the Human Robot Interac-tion (HRI) other than the theme of
self-sufficiency of theautonomous agent, the theme this paper
treats, will needto be experimentally verified before this paper’s
model isimplemented as in artificial emotion agents or
sensitiverobots.
AcknowledgementsPart of this study was carried out by the
Grant-in-Aid for YoungScientists (A) of the Ministry of Education,
Culture, Sports, Sci-ence and Technology (No. 20680014).
References:[1] P. Marti, M. Bacigalupo, L. Giusti, C. Mennecozzi
and T. Shi-
bata, “Socially Assistive Robotics in the Treatment of
Behaviouraland Psychological Symptoms of Dementia,” IEEE/RAS-EMBS
Int.Conf. on Biomedical Robotics and Biomechatronics 2006, pp.
483-488, 2006.
[2] K. Wada and T. Shibata, “Robot Therapy in a Care House – its
So-ciopsychological and Physiological Effects on the Residents,”
IEEEInt. Conf. on Robotics and Automation 2006, pp. 3966-3971,
2006.
[3] S. Takeuchi, A. Sakai, S. Kato, and H. Itoh, “An Emotion
Genera-tion Model Based on the Dialogist Likability for Sensitivity
Com-munication Robot,” J. of Robotic Society of Japan, Vol.25,
No.7,pp. 1125-1133, 2007. (in Japanese)
[4] M. Toda, “Emotion and Urges,” The Urge Theory of Emotion
andCognition, Chapter 1, SCCS Technical Report, 1993.
[5] M. Toda, “Basic Structure of the Urge Operations,” The Urge
The-ory of Emotion and Cognition, Chapter 2, SCCS Technical
Report,1994.
[6] R. Pfeifer and C. Scheier, “Understanding Intelligence,” MIT
Press,1999.
[7] M. Toda, “Man, Robot, and Society, Models and
Speculations,”Martinus Nijhoff Publishing, 1924.
[8] M. Lewis and J. M. Haviland-Jones, “Handbook of
Emotions,”Guilford Publications, 2000.
[9] C. E. Izard, “The Psychology of Emotions,” Plenum Press,
1991.[10] P. Ekman, “Unmasking the Face, Prentice-Hall,” 1975.[11]
A. Araki and M. Kanoh, “Effective Use of LEDs on Face of Com-
munication Robot “Ifbot”,” Joint 4th Int. Conf. on Soft
Computingand Intelligent Systems and 9th Int. Symposium on advanced
Intel-ligent Systems, in CD-ROM, 2008.
Name:Teruaki Ando
Affiliation:Graduate School of Computer and CognitiveSciences,
Chukyo University
Address:101 Tokodachi, Kaizu-cho, Toyota 470-0393, JapanBrief
Biographical History:2009 Received B.S. degree from Chukyo
University2009- Master Course Student, Chukyo UniversityMain
Works:• “A Self-sufficiency Model Using Urge System,” 2010 IEEE
WorldCongress On Computational Intelligence (IEEE WCCI 2010),
2010.Membership in Academic Societies:• Information Processing
Society of Japan (IPSJ)
Name:Masayoshi Kanoh
Affiliation:Associate Professor, Department of Mechanicsand
Information Technology, School of Informa-tion Science and
Technology, Chukyo University
Address:101 Tokudachi, Kaizu-cho, Toyota 470-0393, JapanBrief
Biographical History:2004 Received Ph.D. degree from Nagoya
Institute of Technology2004- Assistant Professor, Chukyo
University2010- Associate Professor, Chukyo UniversityMain Works:•
“Emotive Facial Expressions of Sensitivity Communication
Robot“Ifbot”,” Kansei Engineering International, Vol.5, No.3, pp.
35-42, 2005.Membership in Academic Societies:• The Robotics Society
of Japan (RSJ)• The Japanese Society of Public Health (JSPH)• The
Japan Society of Kansei Engineering (JSKE)
884 Journal of Advanced Computational Intelligence Vol.14 No.7,
2010and Intelligent Informatics