Design of a Personalized Affective Exergame to Increase Motivation in

the Elderly

Fenja T. Bruns

and Frank Wallhoff

Institute of Technical Assistance Systems (ITAS), Jade University of Applied Sciences,

Ofener Str. 16/19, 26121 Oldenburg, Germany

Keywords:

Affective Computing, Exergame, Elderly People, Personalization, Sensors.

Abstract:

Older people are among the most physically inactive. Game-based training programs (exergames) can motivate

this group to exercise more. However, for long term beneﬁt it is important to maintain engagement and

motivation. Therefore, the emotions of the player should also be taken into account. This paper ﬁrst gathers

requirements to develop a motivating exergame for elderly people. This includes speciﬁcs regarding the target

group as well as the inclusion of emotion theories. Based on these considerations, a concept for a framework is

proposed how these requirements can be implemented with the help of personalization and sensor technology

to create a successful and motivating exergame. A sample application demonstrates the ﬂexibility of the

presented framework.

1 INTRODUCTION

In the coming years, the proportion of older adults

will increase. The elderly are among the physically

inactive (Mathews et al., 2010). This, in turn, can lead

to increased falls and fear of falling, which are two

of the main reasons for hospitalizations. Falls result

from incorrectly executed steps and lack of balance,

among other things. But inactivity can also cause hy-

pertension or diabetes, placing inactivity among the

fourth most common risk factors for mortality (World

Health Organization, 2010).

Sports exercises and balance training can counter-

act this and also take away the fear of falling. They

help improving movement patterns, gait and posture

(World Health Organization, 2010) (Leveille et al.,

1999). It has been shown that an important factor in

predicting fall risk is the ability to adapt gait (Caetano

et al., 2016).

It is important for effective training that the given

trainig program is followed. Daily repetitions are

frustrating and do not arouse the interest of partici-

pants, resulting in approximately 65% of patients not

adhering to the program and dropping out of physical

activity (Bassett and Phty, 2003). Also, complicated

and incomprehensible exercises lead to not perform-

ing the exercises (Dobson et al., 2016).

https://orcid.org/0000-0003-0395-3854

https://orcid.org/0000-0002-7791-3225

A promising and inexpensive way of balance

training are so-called exergames. The term exergame

is a composition of the words exercise and game

and describes computer games that are controlled by

movement and thus promote the physical ﬁtness and

effort of the player (Oh and Yang, 2010). The goal

here is to maintain and increase motivation in physical

activity through fun (Lee et al., 2017). Exergames are

available for both light and strenuous tasks and can

be used in the home environment. Components of ex-

ergames can include step training, yoga, or strength

(Taylor et al., 2011). Game consoles that detect

sensor-based motion, such as the Nintendo Switch

or Microsoft Kinect, can be used to capture physi-

cal activity (Vox and Wallhoff, 2017). Exergames

have been shown to have positive effects on bal-

ance and strength training, improve mental health,

and strengthen social relationships (Alhagbani and

Williams, 2021) (Chen et al., 2018).

To maintain engagement while playing a com-

puter game, the player should be in the so-called ﬂow

zone. In this zone, the person is completely absorbed

in his or her activity and forgets the sense of time. In

the ﬂow zone, the player is neither over- nor under-

challenged by the exergame and experiences feelings

of happiness. This cognitive state inﬂuences the per-

son’s performance (Csikszentmihalyi, 2014). One

way to maintain the ﬂow state is by regularly adjust-

ing the difﬁculty level based on the player’s perfor-

Bruns, F. and Wallhoff, F.

Design of a Personalized Affective Exergame to Increase Motivation in the Elderly.

DOI: 10.5220/0010892300003123

In Proceedings of the 15th International Joint Conference on Biomedical Engineering Systems and Technologies (BIOSTEC 2022) - Volume 5: HEALTHINF, pages 657-663

ISBN: 978-989-758-552-4; ISSN: 2184-4305

657

mance. In this way, the game settings are adjusted

to the player’s performance and abilities. It has been

shown that an individualization of exercises increases

the adherence of the training (Jordan et al., 2010).

For another way to adapt exercises to the user and

keep the user in the ﬂow zone, affective computing

can be used. This is an area that deals with the emo-

tions and affects of the user. In real time, this involves

adapting the application to the user’s emotional state.

This can improve the usability of the system (Picard,

1999).

In the following, we will ﬁrst present work in

which exergames were adapted to the user using

different methods. Subsequently, requirements that

are important for creating an affective exergame are

worked out on the basis of a literature research. Both

requirements for emotion recognition and require-

ments for older adults are described. Afterwards, a

concept for a framework will be presented that can

be used to implement these ascertained requirements

when creating an exergame for the elderly. The con-

ceptual framework consists of several components,

which will be explained in more detail. The paper

is concluded by a short summary and outlook.

2 RELATED WORK

The most important goal of a game is to entertain the

user. However, people amuse themselves in different

ways.

There are already different works in which games

have been adapted affectively. Besides presenting

games for education (Bontchev and Vassileva, 2016),

there are also works that discuss the design of affec-

tive games and game engines (Hudlicka, 2009). Us-

ing built-in pressure sensors in a controller, measur-

ing the speed and uniformity of a user’s movement,

the difﬁculty level of the game was adjusted to the

patient’s performance (Sucar et al., 2014). Hossain

et al. developed an exergame that uses speech analy-

sis to analyze user emotion and adapt the game. The

player gets feedback on his emotional state through

vibration (Hossain et al., 2018).

Games in which the difﬁculty level is adjusted are

also applied in rehabilitation. For example, the dif-

ﬁculty can be adapted based on the ratio between the

number of hits and the number of trials (Pezzera et al.,

2019). Also papers that use facial expressions to de-

tect emotions and change the difﬁculty have been pre-

sented (Aranha et al., 2017).

In (Rodriguez-Guerrero et al., 2017), (Erdogan

et al., 2018), and (Darzi and Novak, 2019), the difﬁ-

culty of rehabilitation games was adjusted using fea-

tures from physiological signals of the players. How-

ever, these games are limited to upper extremity re-

habilitation in seated exercises. Moreover, these sys-

tems are often tested with patients of a younger age

group and not with seniors.

There are still many challenges in the area of af-

fective computing because due to the variety of emo-

tions and the difﬁculty to distinguish them from each

other. In addition, the context in which the system is

to be used, the variety of emotions to be recognized,

as well as the target group must also be taken into

account. Despite considerable progress, the full po-

tential has not yet been realized. Partial solutions can

help to deal with this complex problem.

3 REQUIREMENTS

To create an affective exergame, requirements can be

divided into two areas: Requirements for the Elderly

and Requirements for Emotion Recognition. It is nec-

essary to combine these two areas in order to develop

an optimal exergame. In the case of requirements for

older people, the special needs of this target group

must be taken into account. This includes impair-

ments in hearing and seeing, as well as (senso-) mo-

toric limitations. For the requirements of an affective

game, existing emotion theories have to be consid-

ered. In the following, the requirements of the two

areas will be presented in more detail.

3.1 Requirements for the Elderly

As already mentioned in the beginning, older people

often suffer from impairments such as a poor sense of

hearing or vision. The increased risk of falling must

also be given special consideration when designing

an exergame. However, commercial exergames were

not designed with such aspects in mind, making them

mostly unplayable for older adults. (Konstantinidis

et al., 2015) (Gerling and Masuch, 2011) Therefore,

these games must be designed so that seniors can play

them as well. This includes having exercises in both

standing and sitting positions. In addition, the exer-

cises should be adapted to the ability of the players,

with a tolerance range in the accuracy of the execu-

tion of the movements (Brox et al., 2017).

Due to age-related illnesses, older individuals of-

ten experience social isolation. Also during the

COVID-19 lockdown, the elderly were the worst af-

fected by social isolation (Privor-Dumm et al., 2021).

Exergames can be used to improve psychosocial well-

being (Alhagbani and Williams, 2021). This can

be done, among other things, through a multiplayer

HEALTHINF 2022 - 15th International Conference on Health Informatics

658

mode or through competitions by counting points

(De Schutter and Vanden Abeele, 2008). In compe-

titions, it must be noted that the points scored are cal-

culated depending on the skills of the player. If the

performances are compared with an average player

(without any handycaps), a negative feeling of failing

the game may arise (Gerling and Masuch, 2011).

Based on these considerations, not only the exer-

cises themselves and the points scored, but also the

difﬁculty of the exercises should be adapted to the

player’s abilities. Thus, more active players need to

be challenged on a higher lebel compared to rather in-

active ones (Brox et al., 2017). The goal should be to

achieve a ﬂow state in which the player focuses only

on the game (Csikszentmihalyi, 2014). Therefore, the

goals of the game must be achievable to maintain mo-

tivation, and should be adjusted as needed.

3.2 Requirements for Emotion

Recognition

To maintain motivation, the approach of affective

computing can be used. For this purpose, the emo-

tions of the players have to be recognized. Thereby,

not the basic emotions according to Ekman (Ekman,

1992), like disgust or surprise shall be identiﬁed, but

learning-centered emotions (Sann and Preiser, 2017),

like joy, confusion or boredom, or also moods. Be-

cause emotions are often felt individually and sit-

uationally, the emotion recognition system must be

trained to work with context-dependent data.

In order to react individually to the emotions, the

affective states, intensities and triggers can be stored

in user proﬁles. Such a proﬁle serves as a basis for

adjustments, e.g. of the difﬁculty. Regular updating is

essential to capture new triggers and emotions. With

the user proﬁle, appropriate adaptation strategies for

the game can be found to ensure player engagement

(Hudlicka, 2009).

Different sensors can be deployed to measure the

data needed for emotion recognition. In addition

to recognition based on facial expressions, emotions

have also been analyzed using health data. For this, it

is important that the sensors are used non-invasively

and that continuous real-time measurement is possi-

ble (Hudlicka, 2009). The sensors should be comfort-

able to ensure acceptable usability. For this purpose, it

is important that no attachment of electrodes is neces-

sary. In addition, cables can distract subjects and pre-

vent them from completing the task, so these should

also be avoided. Accordingly, the hardware should be

robust with regard to a sufﬁcient signal quality. Fur-

thermore, direct and immediate data access must be

possible in order to continuously adapt the system to

the user’s condition. Therefore, no proprietary soft-

ware of the manufacturer should be necessary, where

the collected data can only be further used after an

export (Peter et al., 2005).

4 PROPOSED APPROACH

The requirements described above were taken into ac-

count in the development of the conceptual frame-

work. In addition, three main objectives representing

the individual needs had to be considered:

• Assist the person in completing the training

• Increase the person’s motivation by making the

tasks more fun

• Increase the person’s sense of competence

Figure 1 shows the proposed framework for an af-

fective exergame. The individual components are de-

scribed in more detail below.

Portal

Game Setting

Motion

Detection

Webcam

Database

Customization

Affect

Detection

Wristband/

Smartwatch

Data

Motions

Data

Affects

Difficulty

Game & Player

Information

Game &

Player

Information

Game & Player

Information

Data

Figure 1: Framework for an affective exergame.

The overall concept of the framework is modular.

Flexibility is provided in terms of extensibility and

the combination of different games with sensors and

methods for customization.

Design of a Personalized Affective Exergame to Increase Motivation in the Elderly

659

4.1 Motion Detection

The crucial component for the game to be an ex-

ergame is that the game is typically controlled by

movements. A camera, ideally with depth informa-

tion, is used to see whether the required exercises are

performed. With the help of the image information,

the exercises can be recognized. By using a camera,

the game charackter may be controlled and placed in

an augmented reality scene. As a result, the player

ﬁnds himself and his surroundings reﬂected in the

game.

The camera could be a depth camera. This has the

advantage that additionally the accuracy of the exe-

cution of the movement can be evaluated. Another

advantage of a depth camera is that skeletal data can

be used directly. This eliminates the need for com-

plex image data, which can result in computationally

expensive computations. Depth information may also

lead to an increased acceptance of video recordings.

In contrast to the depth camera, a commercially

available webcam might alternatively also be used.

This is an inexpensive alternative and is already avail-

able in many end devices such as laptops or tablets.

Webcams can also be used without much experience.

However, the previously described advantages of a

depth camera are not available here, which means that

recognition accuary and maybe acceptance problems

may be expected.

4.2 Emotion Recognition

During the game, sensors will be used to record the

physiological state of the players. Machine learning

will be used to determine the emotional states from

the data obtained. Different classiﬁers shall be com-

pared with each other to achieve an optimal result.

The classiﬁcation should achieve that the respective

emotion can be named.

For this purpose, data sets have to be recorded be-

forehand, in which not only the physiological data

are recorded, but also the subjectively perceived emo-

tions of the player in the situation. With this self-

assessment, a benchmark database can then be gen-

erated for the comparison of different emotion recog-

nition methods.

Non-invasive sensors that can be used easily shall

be applied to collect the data. Therefore, the phys-

iological parameters of skin conductance, heart rate,

and temperature shall be recorded using a smartwatch

or wristband. Also, wrist acceleration can be used to

detect rapid movements and determine the frustration

level. The Empatica E4 (Empatica Inc, 2021), which

can measure the above parameters, comes into con-

sideration. This type of hardware additionally avoids

cables for attaching the sensors and transmitting the

data, which could distract the subjects. Furthermore,

a wristband or smartwatch is intuitive, simple to use

and usually requires no further explanation. In ad-

dition, image and sound recordings can be used via

the webcam, which is used to control the game. The

emotional state can be determined based on facial ex-

pressions. It should also be possible to integrate other

sensors besides those of the smartwatch or wristband

and use them for further evaluation.

4.3 Customization

Before a player plays the exergame for the ﬁrst time,

an external person, e.g., a caregiver or a trainer,

could enter information about, among other things,

the player’s movement limitations, age or health level.

Alternatively, some exercises could be performed in

advance to automatically determine the player’s abil-

ities, such as mobility and balance. Based on these

prerequisite informations, the difﬁculty of the ex-

ergame can be adjusted. Discrete adjustment of difﬁ-

culty (easy, medium, hard) should be avoided to pre-

vent under- or over-challenging people. In addition,

levels that are too difﬁcult can increase the risk of

falling. Instead, the difﬁculty level of the exergame

should be continuously adjusted to the player’s per-

formance. However, the difﬁculty level should not ex-

ceed the player’s capabilities. Therefore, this player

information must be taken into account during the ad-

justment. In order to maintain motivation, the goals

should be achievable. If a previously deﬁned goal

cannot be achieved, e.g. due to movement restric-

tions, the goal must be adjusted. These adjustments

should be done automatically by the game, so that

users can play it at home without a trainer.

Both the goal of the game and the difﬁculty of the

game must not depend only on the initial player in-

formation, but also on the actual emotional state of

the player. The states from emotion recognition, see

section 4.2, should be used to motivate the player by

keeping him in the ﬂow. For example, if it is detected

that the player is bored, the game can increase the dif-

ﬁculty level, while not exceeding the player’s capabil-

ities.

4.4 Database

The emotions as well as current information about the

game (game name, difﬁculty, speed, score, game du-

ration) shall be stored in a database, in which a sep-

arate entry is created for each player. The player’s

skills and goals shall also be stored in a user proﬁle in

HEALTHINF 2022 - 15th International Conference on Health Informatics

660

the database, which is individual for each person.

4.5 Portal

Players should be able to access their data to see. This

can be their own progress as well as an overview of

the levels in which they have encountered difﬁcul-

ties. Each user shall additionally have the possibility

to share their progress with the other players. In this

way, players can compare themselves with each other,

if they wish. This should increase both the psychoso-

cial well-being and the motivation of the players.

4.6 Game Setting

In the Game Setting the different aspects are bundled

and the game is controlled. In addition, the database

entries are managed and access to the portal is pro-

vided.

There are different types of games that can be im-

plemented within the framework. In the following,

two game variants will be presented, based on the in-

troduction in section 1.

When designing an affective exergame, it is im-

portant to train different motor skills in order to en-

sure a balanced workout. Therefore, it is useful to

include balance and gait training as well as strength

training in the game.

In the area of balance and gait training, walking

ability can be improved and conﬁdence in rapidly ex-

ecuted steps can be increased. This can improve mo-

bility. Such training could be implemented playfully

by letting the player collect objects by walking side-

ways. As the difﬁculty level increases, the user is en-

couraged to move faster. This strengthens balance and

self-conﬁdence. However, the player’s abilities must

be taken into account, see section 4.3.

To strengthen the arms, a game could be created

that requires reaching for different objects that are dis-

played on the screen. This game can also improve

hand-eye coordination and arm mobility. This game

could be played both sitting and standing and is there-

fore suitable for everyone.

In both games, there may be additional items that

may not be collected. The items can provide contin-

uous feedback to the player during gameplay. This is

ﬁrstly visual feedback. This can be achieved by no-

ticeably increasing or decreasing the high score or by

the disappearance of the touched objects. Addition-

ally, auditory feedback should be available. While

a positive sound is played when collecting items, a

negative sound should be played when touching items

that are not to be collected. Additionally, it is possi-

ble to provide both visual and auditory feedback on

the player’s performance. This can cheer up and mo-

tivate the player.

5 APPLICATION EXAMPLE

In a student project, the proposed concept was im-

plemented in a modiﬁed form. In this project a 2D

jump’n’run game was developed which is controlled

by facial expressions using the Unity engine (Unity

Technologies, 2021). The facial expressions corre-

sponding to different emotions are used to trigger dif-

ferent actions of the player, such as jumping, ducking

or shooting. An example is shown in Figure 2. It

shows that a happy emotion causes the character to

jump. The game is intended to serve as a therapy-

accompanying training measure for facial paresis in

stroke patients. It is supposed to strengthen the pa-

tient’s facial muscles and thus improve the patient’s

emotional facial expressions.

The concept was adapted so that in the compo-

nent Motion Dectection no movements of the limbs

were detected with a camera, but different facial ex-

pressions were detected, i.e. angry, disgust, scared,

happy, sad, surprised and neutral using the real-time

emotion recognition provided by Omar Ayman (Ay-

man, 2018). In the Affect Detection component, as a

skin conductance sensor the Grove GSR sensor V1.2

(Seeed Technology Co., Ltd., 2021) was used together

with a Raspberry Pi, to determine the stress level of

the player. In the Customization component, depend-

ing on the stress level, the difﬁculty level in the game

was changed. This was expressed by changing the

number of enemies in the game as well as the number

of lives of the player.

This example gave evidence of the ﬂexibility of

this concept. By implementing different games and

using different sensors, the presented framework can

be used for different affective exergames.

Figure 2: The emotion happy causes the character to jump.

Design of a Personalized Affective Exergame to Increase Motivation in the Elderly

661

6 CONCLUSION AND FUTURE

WORK

A concept for a framework was presented that uses

affective computing to adapt exergames to the emo-

tional states of the players. This should increase and

maintain engagement and motivation in the long term

and sustainably. During the development of the con-

cept different requirements were considered. This

has resulted in a conceptual framework that includes

a combination of game-based concepts, mechanisms

for personalization, sensor technology and affective

computing. The information of the latter two directly

inﬂuences the gameplay of the exergame.

Currently, the concept is being implemented with

movements of the limbs and will be evaluated with

elderly people. The aim is to evaluate whether athletic

abilities are increased by using the system. The user

experience of the system will also be evaluated.

To this end, games are currently being imple-

mented that are controlled by movement and are in-

tended to generate different emotions. A benchmark

database of emotions will then be created based on

these games. Subsequently, the artiﬁcial intelligence

can be trained to recognize the different emotions

based on the physiological data. A ﬁnal test will eval-

uate the user experience and check whether the ex-

ergames react appropriately to the emotions.

ACKNOWLEDGEMENT

The authors would like to thank the students Carina

Fischer, Emily Hossfeld, Marie Kutscher and Geral-

dine Sutter for their contributions and the prelimi-

nary programming of a 2D jump’n’run game in their

project report using elements of emotion recognition

and stress detection.

REFERENCES

Alhagbani, A. and Williams, A. (2021). Home-Based Ex-

ergames for Older Adults Balance and Falls Risk: A

Systematic Review. Physical & Occupational Ther-

apy In Geriatrics.

Aranha, R. V., Silva, L. S., Chaim, M. L., and de Lour-

des dos Santos Nunes, F. (2017). Using Affective

Computing to Automatically Adapt Serious Games

for Rehabilitation. In 2017 IEEE 30th Interna-

tional Symposium on Computer-Based Medical Sys-

tems (CBMS).

Ayman, O. (2018). Emotion-recognition. GitHub Reposi-

tory. Accessed: Nov. 01, 2021. [Online]. Available:

https://github.com/omar178/Emotion-recognition.

Bassett, S. and Phty, D. (2003). The assessment of pa-

tient adherence to physiotherapy rehabilitation. New

Zealand journal of physiotherapy.

Bontchev, B. and Vassileva, D. (2016). Assessing engage-

ment in an emotionally-adaptive applied game. In

Proceedings of the Fourth International Conference

on Technological Ecosystems for Enhancing Multicul-

turality, New York, NY, USA. Association for Com-

puting Machinery.

Brox, E., Konstantinidis, S. T., and Evertsen, G. (2017).

User-centered design of serious games for older adults

following 3 years of experience with exergames for

seniors: a study design. JMIR serious games,

5(1):e6254.

Caetano, M. J. D., Lord, S. R., Schoene, D., Pelicioni, P.

H. S., Sturnieks, D. L., and Menant, J. C. (2016). Age-

related changes in gait adaptability in response to un-

predictable obstacles and stepping targets. 46:35–41.

Chen, C.-K., Tsai, T.-H., Lin, Y.-C., Lin, C.-C., Hsu, S.-C.,

Chung, C.-Y., Pei, Y.-C., and Wong, A. M. K. (2018).

Acceptance of different design exergames in elders.

PLoS ONE, 13(7):e0200185.

Csikszentmihalyi, M. (2014). Attention and the holistic ap-

proach to behavior. In Flow and the Foundations of

Positive Psychology: The Collected Works of Mihaly

Csikszentmihalyi, pages 1–20. Springer.

Darzi, A. and Novak, D. (2019). Using Physiological Link-

age for Patient State Assessment In a Competitive Re-

habilitation Game. In 2019 IEEE 16th International

Conference on Rehabilitation Robotics (ICORR).

De Schutter, B. and Vanden Abeele, V. (2008). Meaningful

Play in Elderly Life. Annual Meeting of the Interna-

tional Communication Association.

Dobson, F., Bennell, K. L., French, S. D., Nicolson, P. J. A.,

Klaasman, R. N., Holden, M. A., Atkins, L., and Hin-

man, R. S. (2016). Barriers and Facilitators to Ex-

ercise Participation in People with Hip and/or Knee

Osteoarthritis: Synthesis of the Literature Using Be-

havior Change Theory. American journal of physical

medicine & rehabilitation, 95:372–389.

Ekman, P. (1992). An argument for basic emotions. 6:169–

200.

Empatica Inc (2021). E4 wristband. Ac-

cessed: Oct. 15, 2021. [Online]. Available:

https://www.empatica.com/en-eu/research/e4/.

Erdogan, H., Palaska, Y., Masazade, E., Barkana, D. E., and

Ekenel, H. K. (2018). Vision-based game design and

assessment for physical exercise in a robot-assisted re-

habilitation system. IET Computer Vision, 12(1):59–

68.

Gerling, K. and Masuch, M. (2011). When Gaming is

not Suitable for Everyone:Playtesting Wii Games with

Frail Elderly. In 1st Workshop on Game Accessibility.

Hossain, M. S., Muhammad, G., Al-Qurishi, M., Ma-

sud, M., Almogren, A., Abdul, W., and Alamri, A.

(2018). Cloud-oriented emotion feedback-based ex-

ergames framework. Multimedia Tools and Applica-

tions, 77:21861–21877.

Hudlicka, E. (2009). Affective game engines: motivation

and requirements. In Whitehead, J. and Young, R. M.,

HEALTHINF 2022 - 15th International Conference on Health Informatics

662

editors, Proceedings of the 4th International Confer-

ence on Foundations of Digital Games, FDG 2009,

Orlando, Florida, USA, April 26-30, 2009, pages

299–306. ACM.

Jordan, J. L., Holden, M. A., Mason, E. E., and Foster,

N. E. (2010). Interventions to improve adherence to

exercise for chronic musculoskeletal pain in adults.

The Cochrane database of systematic reviews, page

CD005956.

Konstantinidis, E. I., Antoniou, P. E., and Bamidis, P. D.

(2015). Exergames for assessment in active and

healthy aging-emerging trends and potentialities. In

International Conference on Information and Com-

munication Technologies for Ageing Well and e-

Health, volume 2, pages 325–330. SCITEPRESS.

Lee, S., Kim, W., Park, T., and Peng, W. (2017). The Psy-

chological Effects of Playing Exergames: A System-

atic Review. Cyberpsychology, behavior and social

networking, 20(9):513–532.

Leveille, S. G., Guralnik, J. M., Ferrucci, L., and Langlois,

J. A. (1999). Aging successfully until death in old age:

opportunities for increasing active life expectancy.

American Journal of Epidemiology, 149(7):654–664.

Mathews, A. E., Laditka, S. B., Laditka, J. N., Wilcox, S.,

Corwin, S. J., Liu, R., Friedman, D. B., Hunter, R.,

Tseng, W., and Logsdon, R. G. (2010). Older adults’

perceived physical activity enablers and barriers: a

multicultural perspective. Journal of aging and phys-

ical activity, 18:119–140.

Oh, Y. and Yang, S. (2010). Deﬁning exergames & ex-

ergaming. Proceedings of meaningful play.

Peter, C., Ebert, E., and Beikirch, H. (2005). A Wear-

able Multi-sensor System for Mobile Acquisition of

Emotion-Related Physiological Data. In Affective

Computing and Intelligent Interaction, pages 691–

698.

Pezzera, M., Tironi, A., Essenziale, J., Mainetti, R., and

Borghese, N. A. (2019). Approaches for increasing

patient’s engagement and motivation in exer-games-

based autonomous telerehabilitation. In 2019 IEEE

7th International Conference on Serious Games and

Applications for Health (SeGAH).

Picard, R. W. (1999). Affective Computing for HCI. In HCI

(1), pages 829–833. Citeseer.

Privor-Dumm, L. A., Poland, G. A., Barratt, J., Durrheim,

D. N., Deloria Knoll, M., Vasudevan, P., Jit, M., Bon-

veh

ı, P. E., Bonanni, P., and on Adult Immunization,

I. C. (2021). A global agenda for older adult immu-

nization in the COVID-19 era: A roadmap for action.

Vaccine, 39:5240–5250.

Rodriguez-Guerrero, C., Knaepen, K., Knaepen, K., Fraile-

Marinero, J. C., Perez-Turiel, J., de Garibay, V. G.,

and Lefeber, D. (2017). Improving Challenge/Skill

Ratio in a Multimodal Interface by Simultaneously

Adapting Game Difﬁculty and Haptic Assistance

through Psychophysiological and Performance Feed-

back. Frontiers in Neuroscience, 11.

Sann, U. and Preiser, S. (2017). Emotion und Motivation

in der Lehrer-Sch

uler-Interaktion. In Lehrer-Sch

uler-

Interaktion, pages 213–232. Schweer, Martin K.W.

Seeed Technology Co., Ltd. (2021). Grove - GSR Sen-

sor. Accessed: Dec. 15, 2021. [Online]. Available:

wiki.seeedstudio.com/Grove-GSR Sensor/.

Sucar, L. E., Orihuela-Espina, F., Velazquez, R. L.,

Reinkensmeyer, D. J., Leder, R., and Hern

andez-

Franco, J. (2014). Gesture Therapy: An Upper Limb

Virtual Reality-Based Motor Rehabilitation Platform.

IEEE Transactions on Neural Systems and Rehabili-

tation Engineering, 22(3):634–643.

Taylor, M. J. D., McCormick, D., Shawis, T., Impson, R.,

and Grifﬁn, M. (2011). Activity-promoting gaming

systems in exercise and rehabilitation. 48:1171.

Unity Technologies (2021). Unity. Accessed: Oct. 15,

2021. [Online]. Available: https://www.unity.com.

Vox, J. P. and Wallhoff, F. (2017). Recognition of human

motion exercises using skeleton data and SVM for

rehabilitative purposes. In 2017 IEEE Life Sciences

Conference (LSC), pages 266–269.

World Health Organization (2010). Global recommenda-

tions on physical activity for health. World Health

Organization.

Design of a Personalized Affective Exergame to Increase Motivation in the Elderly

663