=Paper=
{{Paper
|id=Vol-2359/paper12
|storemode=property
|title=Non-player character as a companion in cognitive rehabilitation for adults – Characteristics and representation
|pdfUrl=https://ceur-ws.org/Vol-2359/paper12.pdf
|volume=Vol-2359
|authors=Mareike Gabele,Andrea Thoms,Julian Alpers,Steffi Hußlein,Christian Hansen
|dblpUrl=https://dblp.org/rec/conf/gamifin/GabeleTAH019a
}}
==Non-player character as a companion in cognitive rehabilitation for adults – Characteristics and representation==
https://ceur-ws.org/Vol-2359/paper12.pdf
Non-player character as a companion
in cognitive rehabilitation for adults -
Characteristics and representation
Mareike Gabele1,2, Andrea Thoms3, Julian Alpers1,
Steffi Hußlein2 and Christian Hansen1
1 Otto von Guericke University Magdeburg, Germany
2 Magdeburg-Stendal University of Applied Sciences, Germany
3 HASOMED GmbH, Paul-Ecke-Straße 1, 39114 Magdeburg, Germany
mareike@isg.cs.uni-magdeburg.de
Abstract. A lack of social support reduces the chances of successful rehabilita-
tion. However, the social environment changes considerably during this time.
Therefore, fostering consistent social contacts is highly relevant. To address so-
cial relatedness in software-based cognitive therapy during rehabilitation we in-
tend to use a non-player character as a companion. In this work we analyze pos-
sible forms of representation of the companion based on required characteristics,
age and gender to achieve this goal. These were set in relation to age and gender
of the user. Three female and three male companions in three age groups were
created and subsequently tested in an explorative feasibility study with 40 partic-
ipants. 50% of participants preferred a female middle-aged companion, 25% a
younger male. Older companions were chosen only by women. Regarding the
gender, 62.5% chose a female companion. We present an orientation for devel-
opment of non-player characters as companion in software-based training for
cognitive rehabilitation.
Keywords: Non-player Character, Cognitive Rehabilitation, Social Relatedness
1 Introduction
Software-based training is used in cognitive rehabilitation as a successful therapy for
brain damage [1]. Depending on the damaged region of the brain, cognitive deficits
such as in working memory, logical thinking, visual processing or attention may result.
In addition, motoric restrictions, speech and language disorders can occur. Strokes are
the main cause of acquired disability in adulthood [2]. Accidents with brain damage
can also be a cause. Frequent and regular training in inpatient and subsequently in out-
patient rehabilitation and / or home training is important for successful rehabilitation
[3]. The success may take months to years and cannot be guaranteed. A high level of
social support leads to a faster and better rehabilitation after a stroke. In contrast, social
isolation is a particular risk of poor rehabilitation outcomes [4] or higher mortality rate
[5]. Especially in eHealth, dropping out before completing the goal is high [6]. In home
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 130
training, patients need a high level of motivation in order to carry out the training, as
the training is no longer supervised by therapists at every session, as in the clinic.
Existing social contacts can be reduced considerably by the stay in the clinic, the
temporary or complete leaving of the workplace and motor and cognitive restrictions.
To overcome this problem, we want to address the human needs for consistent and
motivating social support in software-based cognitive training during rehabilitation.
Therefore, we propose integrating a non-player character (NPC) as a companion within
the training software, who will accompany the patient during the whole time of reha-
bilitation (in the following text this NPC is only called companion). The companion
should neither replace the therapist, nor real social interaction, but offer a stable accom-
paniment.
The target group are patients with acquired brain damage who are undergoing cog-
nitive rehabilitation. There is a large number of studies that deal with the relationship
between therapist and patient and deal with different effectiveness factors. This in-
cludes, among other things, the therapist's empathy and appreciation [7, 8]. Similar
characteristics may be assumed for a companion. However, there are large demographic
differences between patients who perform software-based therapy. Therefore, the re-
quirements for the characteristics of a companion can vary widely. Likewise, it can vary
how a companion is perceived and which characteristics are attributed to him. To ana-
lyze this, we developed a questionnaire, visualized and integrated different characters
in gender and age, and evaluated them exploratively in a feasibility study. We focused
on the following research questions:
Q1. Which characteristics are desired for a companion?
Q2. Which age of the companion is preferred and why?
Q3. Which gender of the companion is preferred and why?
Q4. Does the reason for the selection match the previously desired characteristics?
Q5. What are the characteristics of the choice of a companion in terms of the age
and gender of the participants?
The main contribution of this paper is to analyze which characteristics are needed
for a companion in cognitive rehabilitation therapy, and what type of companion these
characteristics are attributed to, by which participants. In the cognitive field, analyses
for the representation of companions in software-based therapy, especially with regard
to social relatedness, have so far been neglected. Our results indicate an approach for
the use of NPCs as a companion in software-based cognitive training in rehabilitation
to create lasting social in-game relationships and create possibilities of its use. This may
promote the feeling of social relatedness, motivate intrinsically and support the patient
mentally during the rehabilitation process in the long-term [9].
In the following we first give an overview of social relatedness, avatars and NPCs
used in games and therapy. Then we present the different NPCs we developed in dif-
ferent gender and age groups and the execution of a feasibility study. We sorted the
results, visualized and analyzed them. Finally, we discuss the results and possible fur-
ther uses of NPCs in software based training in cognitive rehabilitation.
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 131
2 Related Work
Social relatedness, along with autonomy and competence, is one of the central factors
of self-determination theory, which describes the psychological needs for intrinsic mo-
tivation [10]. In games, they are able to influence the fun of the game and the future
gameplay independent from each other [11]. We want to use this to support motivation
of the patient in therapy by promoting social relatedness. Social relatedness describes
the human need for integration into a social environment and belongingness [12]. Re-
inforcing this feeling encourages people to become more involved themselves [13].
Common goals and experiences, as well as highlighting cooperation can strengthen the
sense of social relatedness of team members [14]. Social structures can arise on inter-
personal contacts in the real or virtual world, but also on contact with non-realistic
characters. The cooperation with teammates (for example NPCs) successfully affects
the sense of social relatedness [15]. Elements that are assigned to the area of socializa-
tion indicate only neutral to positive influence on players [16].
Gamification elements are elements that are typically used in games, but are inserted
in a different, non-game context [17]. Their effectiveness, is not only based on their
general, but on their conscious, conceptually targeted use [18] and their perceived value
to the user [19]. Therefore, the mechanics of the NPC shall be used as a gamification
element in cognitive rehabilitation and be adapted to the needs of the patient. Serious
games are games that pursue a goal outside the game [17], such as therapy [20] or
education [21]. In the field of motoric rehabilitation, it has been shown that serious
games increase the effectiveness of the corresponding training. However, they combine
different gamification elements, but analyses are required for individual gamification
elements [22].
Avatars are used in current software-based training for self-representation to reflect
the patient's movements or to demonstrate desired movements [23]. This shows that
digital characters in rehabilitation software have already been used with success, which
shows a positive basis for our research. However, these are oriented towards the func-
tionality of the exercise in motoric instead of cognitive rehabilitation and representation
of the user. Avatars are also used in serious games as player-character for cognitive
training of attention for children with attention deficit hyperactivity disorder (ADHD)
[24]. As another possibility of using digital characters, we want to propose an NPC in
our work that guides the patient. The visualization of NPCs is based on the influence
of social norms and should correspond to the expectations of the user. This results in
expectation regarding appearance, behavior and role models, which in turn influences
the behavior of the player [25]. Therefore we analyze whether there is a trend for ex-
pectations. And if there is a trend, we want to use it to influence the patient's behavior
positively in the use of the software-based training.
In computer games, companions are often expected to have the basic attributes
skilled, helpful, nice, attractive and naive [26]. Further, the game identification of the
players is related to the interaction with the companions [27]. However, the focus of
such research is often on a younger healthy target group which plays for fun.
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 132
3 Explorative Feasibility Study
3.1 Development of Elements of the Study
In this study, we analyzed the characteristics required for a companion of patients in
cognitive rehabilitation. In addition, we analyzed preferences for different types of
companions in the representation. For this, we created visual representations of three
female and three male companions (Fig. 1). For each group one companion was created
in the range under 35 years (young), 35 - 49 years (middle aged) and over 50 years
(older). These were developed iteratively on the basis of feedback from three experts
in the field of rehabilitation (psychology, management, design). A consistent graphic
style and consistent clothing is used to minimize further influences by personal prefer-
ences e.g. in the color or clothing style and a thereby possible bias. The choice of the
skin color of the NPC is based on the reason that the study was carried out in Europe
and that the participants are most accustomed to it in their everyday life. Primarily the
face is modified by the change of the age and age-appropriate representation of the hair,
secondarily the age- and gender-appropriate body form.
Fig. 1. Visualization of the six different companions used in the study, A-C Female (age young
to old), D - F Male (age young to old).
3.2 Methods and Material
An explorative feasibility study was conducted based on a structured interview. First,
the participants were informed about the voluntary participation, goals and background
of the study. It was also brought to mind that personal perception and subjective opin-
ions are at the center of attention. The time of rehabilitation was compared to a long
hike. We asked closed questions with predefined answers to focus a decision between
the companions. In addition, we asked open questions with free responses on charac-
teristics and the reasons for the selection of the chosen companion. In this way, we
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 133
wanted to identify the aspects that are personally relevant to the participant without
influencing him or her beforehand. With regard to our questions (Q1 - Q5) we pro-
ceeded as follows:
a) Participants were asked what characteristics a companion needs to have in or-
der to take the companion on the long rehabilitation trail (open) (Q1)
b) Selection of one of three possible female avatars as companions (A, B, C)
based on pictures (closed) and the reason for the selection (open) (Q2)
c) Selection of one of three possible male avatars as companions (D, E, F) based
on pictures (closed) and the reasons for the selection (open) (Q2)
d) Selection between the two avatars previously selected in b) and c) as final
selection based on pictures (closed) and the reasons for the selection (open)
(Q3 and Q4)
e) Collection of demographic data: age group (<20, 20-29, 30-39, 40-49, 50-59,
60-69, 70-79, >80), gender and existence of acquired brain damage (Q5)
With regard to the open questions, we summarized the same answers, counted the
frequency of naming and sorted the answers based on this. Afterwards, for question a),
similar characteristics were clustered. Concerning b), c) and d), we analyzed the occur-
rence for conspicuities with regard to age and gender of the participants.
3.3 Participants
The study was conducted at the open day of a german University Hospital with attached
outpatients clinic for cognitive neurology in September 2018. 40 participants took part
in the study (female: n=24; male: n=16). They are composed of patients in cognitive
rehabilitation (n=5) and persons with professional or personal knowledge and connec-
tion to the topic and / or interest in (n=35). The age groups ranged in steps of ten from
under 20 to over 70. According to experts, the age distribution curve reflects the ap-
proximate distribution of patients in cognitive rehabilitation and is shown in Fig. 2.
Fig. 2. Age distribution of participants in the study.
The participants were personally asked about their interest in participating in the study.
They were selected by us on the basis of their age, which fits approximately to this
distribution, and otherwise without further selection criteria. All participants come from
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 134
the European cultural area. All participants took part voluntarily in the study and were
not rewarded. The study was conducted in Germany.
4 Results, Clustering and Analytical Procedure
In the following, the answers, frequencies and / or percentwise distribution of answers
for a) - d) are presented. The results for step a) are shown in Table 1. It shows the
characteristics that participants list a companion should have. After sorting and cluster-
ing the answers, a focus on character, knowledge and interpersonal behavior is appar-
ent. The results for step b) are shown in Fig. 3.1. In this selection of female companions
in different ages, the middle-aged female companion B is preferred. The results for step
c) are shown in Fig. 3.2. In this selection of male companions in different ages, the
younger male companion D is preferred. The results for step d) are shown in Fig. 3.3.
A final selection was made between the previously selected female and male compan-
ion. 50% of the participants preferred the middle-aged female companion B. Addition-
ally, for b), c) and d), Fig. 3 lists the reasons for the selection on the right axis, or below.
The relations resulting from inclusion of e), the demographic data, between the chosen
companion and the age and gender of the participant are described in Table 2 and at the
end of this section. The focus here is on conspicuities that emerged in the data analysis.
Table 1. Sorted and clustered answers for a). The number in brackets shows the frequency of
naming by all participants. Characteristics without brackets were mentioned once.
Character (44): friendly (4), authoritative (4), empathetic (4), relaxed (3), patient (3),
clear (3), charismatic (smile / be positive) (2), open-natured (2), demanding (2), seri-
ous (2), nice, considerate, understanding, uncomplicated, not too soft, energetic, as-
sertive, distinct, strict, resolute, not so serious but funny sometimes as well, also some-
times sarcastic / black humor, objective, helpful, honest
Type / Optical (4): beeing able to walk well, sporty, more dynamic than oneself, both
feet on the ground
Behavior / Knowledge (23): competent (7), knows what the best way is especially
for me (giving feedback) (5), knows the situation and knows what it is like and how
to deal with the situation (2), someone who has gone through this himself, able to
give expert advice, giving feedback even if it is negative, explain what makes sense /
what doesn't, giving background information, open to questions, interested in the
field, bringing the goals I have in line with the therapy, dealing with it individually
Interpersonal (27): motivating (6), when I don't feel like it / have a low point, moti-
vate me to do it (3), must be able to talk to him (2), must be able and willing to lis-
ten (to one's own problems) (2), friend (2), relaxed atmosphere, person in a position
of trust, right chemistry, says what to do, takes the lead, pays attention if one does
the tasks, supporting, doesn't force you, calming, provide security, able to catch me,
get to know each other on a neutral basis
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 135
Fig. 3. 3.1) percentwise selection of female companion and reasons for selection, 3.2) per-
centwise selection of male companion and reasons for selection, 3.3) percentwise selection of the
final companion.
With regard to Fig. 3.3, the reasons for selection after clustering by frequency of nam-
ing are as follows: sex (11), animating / supporting (6), competence (6), sympathy (6),
empathy (5), sense of security (5), trust (4), assertiveness (4).
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 136
Table 2. 2.1) percentwise selection of companions in b) and c) based on age group of the partic-
ipants, 2.2) percentwise selection of companions in b) and c) based on gender of participants,
2.3) percentwise final selection of companions in d) based on age group of the participants,
2.4) percentwise final selection of companions in d) based on gender of participants
Chosen companion according to b) and c)
female female female male male male
Participants A B C D E F
2.1) age
Young (<30) 5% 7,5% 2,5% 12,5% 2,5% 0
middle aged
(30-59) 7,5% 42,5% 5% 32,5% 17,5% 5%
old
(>60) 7,5% 15% 7,5% 27,5% 2,5% 0
Overall 20% 65% 15% 72,5% 22,5% 5%
2.2) gender
women 20,83% 54,17% 25% 83,33% 8,33% 8,33%
men 18,75% 81,25% 0% 56,25% 43,75% 0%
Chosen final companion according to d)
female female female female male male male male
Participants A B C overall D E F overall
2.3) age
Young (<30) 0 2,5% 0 0 2,5% 0
middle aged
(30-59) 5,00% 32,5% 2,5% 15% 7,5% 2,5%
Old (>60) 2,5% 15% 2,5% 10% 0 0
Overall 7,5% 50% 5% 25% 10% 2,5%
2.4) gender
women 8,33% 45,83% 8,33% 62,5% 33,33% 0 4,17% 37,5%
men 6,25% 56,25% 0 62,5% 12,5% 25% 0 37,5%
Taking account of e), the demographic data, only 7.5% prefer the oldest companion
(female C and male F) in the final selection (Table 2.3). Of these, all were female in the
age range between 40 and 69. The young female companion (A) was chosen by 7.5%,
of whom 5% were female in the age range between 50 and 69 and 2.5% male in the age
range between 20 and 29. The middle-aged male companion (E) was chosen in the final
selection exclusively by men (Table 2.4) who were equally distributed between the age
ranges from under 20 to the ages of 50 - 59. The older man (F) was chosen exclusively
by women between the ages of 50 and 59.
Regarding only the participants with brain damage and cognitive rehabilitation, 60%
chose the middle-aged female companion (B), 20% the younger female companion (A)
and 20% the younger male companion (D). This results in 80% of the participants with
brain damage who chose a female companion.
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 137
5 Discussion
In this work, requirements for the characteristics of a virtual companion in software-
based cognitive therapy in rehabilitation and selection criteria were analyzed. These
were based on age and gender of companions and participants.
To support the patient's motivation, the companion can guide him starting from the
initial training. He can show the patient on a map his progress in rehabilitation, remind
him of good experiences, note frequent mistakes made during training and can remind
for the next training. Comparable use cases, such as informing about a task or helping
the player, were included in classifications within research in games [25]. The com-
panion's skills and clothing improve through regular training or can be selected. This
enables companion to develop further together with the patient. The aim is to support
the regular performance of the training required for rehabilitation.
Software-based training in cognitive therapy often involves several individual train-
ings (e.g. for deficits in attention, memory or executive functions). These differ in rules
and tasks. In the therapy for adults the trainings are carried out one after the other. There
are rarely integrated characters or social relatedness. Through our suggested use of
NPCs these can be modified. They can be used in individual trainings as well as in an
overarching role to connect the trainings. Previous research has shown that due to social
norms there are expectations of the user to the NPC [25]. Based on this, the patient's
behavior could be positively influenced by a companion considered as positive as pos-
sible, which may support regular training. This will be considered in further studies.
In the presented study the desired companion is predominantly described as an em-
pathetic friend and competent at the same time (Q1). The companion has to take the
leading role and be motivating, but also strict if necessary. This shows the desire for
trust and guidance in a situation that cannot be assessed by the individual because the
necessary skills are lacking. The focus is therefore not on appearance. That is highly
relevant for the development of a companion for use in software-based therapy for
which the characteristics analyzed in a) should be in the focus. Developed NPCs should
also be evaluated on this.
The stated needs with regard to characteristics and that being naive is no require-
ment, differs from the expectations of a classic companion in computer games. Rather,
a combination of companion and mentor is the result. The mentor is additionally de-
scribed by Rogers et al. [26] as wise and intelligent, but also old, which does not cor-
respond to the expectations regarding a companion in rehabilitation. This implies that
the requirements for the use of a companion in rehabilitation differ from those of normal
players. Existing classifications [25, 26] are an important basis for comparing these
systems with the needs of patients and possibly extending them.
The selection of the younger to middle-aged companion (Q2) could be based on the
fact that in long-term strength may be needed for support. If older avatars are chosen,
patients frequently stated that the reason is due to the closeness to one's own age. This
may indicate the need to bring up empathy for one's own situation. On the basis of age
(Q3), the middle-aged (female companion B) is preferred for the female companions
and among others described as pleasant and competent. Among male companions, the
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 138
younger one (male companion D) is preferred and described as sympathetic and dy-
namic. The desired characteristics, which were stated previously by each participant,
were later often linked to the choice of gender when the reasons for the final selection
were stated. The overall desired qualities for the companion are predominantly at-
tributed to women in everyday life, which could explain the final tendency towards the
female companion. Compared to the previous assessment (Q4), however, none of the
participants had expressed gender-specific wishes. The companions selected in the final
selection were otherwise attributed congruent characteristics as desired before. These
were thus confirmed based on the representation of the companions. This may show on
the one hand that in Q1 the gender classification did not seem relevant for the partici-
pants. On the other hand, it may also show that previous characteristics were assigned
to the characters shown. Regarding the choice of age and gender of respondents (Q5),
it was found that women also chose the older characters, while men did not. Women
were more likely to choose the younger male companion, men also the middle-aged.
The younger female companion is preferred by older women and younger men. Here
various aspects, such as the protective instinct of the older participants or the sexuality
of the younger participants may be involved. Due to the low final selection of some of
the companions, this can only be seen as an indication.
The comparison of data between participants with and without brain damage shows
that the distribution of the selection is very similar. That is apparent from the fact that
the trends in the selection of companions, both female and male, and the final selection
correspond. It can be assumed that the presence of brain damage does not change the
needs in the presented situation, because all participants were asked about their personal
preferences. However, the observed trend cannot be generalized due to the small sample
of patients. Persons without acquired brain damage know the situation of patients, but
are in a different emotional situation. The acquired brain damage and the resulting ex-
perience may lead to further effects in the use of NPCs. Further studies should be car-
ried out with patients and the effect on patients during use of NPCs should be examined
more closely. A further influence on the data can be assumed by the unequal distribu-
tion of the participants between men and women. However, if the percentual distribu-
tion of the final selection is compared between female and male participants, it is ex-
actly the same. 62.5% of both men and women chose a female companion and 37.5%
a male companion in the final selection (Table 2.4). From this, it can be assumed that
the gender of the participants does not influence the distribution of the selection of the
gender according to the companion. In this study, one visualization was used per age
and gender. There may be an influence from preferring a particular graphical represen-
tation. This can be further investigated by using different visualizations per category.
Due to the exploratory approach to the feasibility study, the results of the individual
categories should be considered as trends. Slight to strong deviations from the trend can
arise individually due to the unique personalities and resulting needs of patients. The
results generally represent the field of cognitive rehabilitation. Whether there are dif-
ferences between various neurological diseases needs further analysis. The participants
come exclusively from the European cultural area, so the results can only be related to
this particular demographic. Therefore, the results provide a basis for further studies to
consider the conspicuous results found here separately.
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 139
6 Conclusion
This work shows an analysis of possible representations of NPCs as a companion in
cognitive therapy during rehabilitation based on characteristics, age and gender of com-
panion and participants. This was done using three visualized female and male charac-
ters in each case as selections in combination with a questionnaire on preferred charac-
teristics and reasons for selection. Results show a trend to use a middle-aged female or
a younger male companion for software-based cognitive training in rehabilitation. The
relevant characteristics are to be competent and give feedback, motivating but at the
same time determinant, friendly and empathetic. It worth further investigation because
a trend has been shown, which should be considered further if companions are inte-
grated in cognitive training software. Furthermore, various possible uses for the com-
panion should be evaluated. In the future, it could be possible to present the progress of
the patient, to remember training times or to give hints for improvements in training
and everyday life using the optimized companion as an application purpose.
Acknowledgments. We would like to thank Juliane Weicker, Angelika Thöne-Otto
and Michael Preier for support with professional questions and conducting the study.
This work was funded by the European Regional Development Fund under the opera-
tion numbers ZS/2016/04/78123 and ZS/2017/01/83843 as part of the initiative
”Sachsen-Anhalt WISSENSCHAFT Schwerpunkte” and FEM-POWER under the
operation number ZS/2016/09/81572.
References
1. Yoo C., Yong M., Chung J., Yang, J.: Effect of computerized cognitive rehabilitation pro-
gram on cognitive function and activities of living in stroke patients. Journal of Physical
Therapy Science 27(8), pp. 2487–2489 (2015).
2. Mendis, S.: Stroke disability and rehabilitation of stroke: World Health Organization per-
spective. International Journal of Stroke 8, 3-4 (2012).
3. Kleim, J., Jones, T.: Principles of experience-dependent neural plasticity: implications for
rehabilitation after brain damage. Journal of speech, language, and hearing research 51(1),
225-239, (2008).
4. Glass, T., Matchar, D., Belyea, M., Feussner, J.: Impact of Social Support on Outcome in
First Stroke. Journal of the American Heart Association. 24, 64-70 (1993).
5. Ruberman W., Weinblatt E., Goldberg J., Chaudhary B.: Psychosocial influences on mor-
tality after myocardial infarction. The New England Journal of medicine 311(9), 552-559
(1984).
6. Eysenbach, G.: The law of attrition. Journal of Medical Internet Research 7(1) (2005).
7. Pfammatter, M., Junghan, U., Tschacher, W.: Common factors of psychotherapy: concepts,
contradictions and a synthesis. Psychotherapie in Psychiatrie, Psychotherapeutischer
Medizin und Klinischer Psychologie 17 (1), 17-31 (2012).
8. Rogers, C.: Client-centered Therapy. Journal of Clinical Psychology 7(3), pp. 560 (1951).
9. Ryan, R., Deci, E.: Self-determination theory and the facilitation of intrinsic motivation,
social development, and well-being. American Psychologist 55, 68-76 (2000).
10. Deci, E., Ryan, R.: Intrinsic motivation and self-determination in human behavior. Springer
US, New York (1985)
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 140
11. Ryan, R., Rigby, C., Przybylski, A.: The motivational pull of video games: A self-determi-
nation theory approach. Motivation and Emotion 30(4), 347-363 (2006).
12. Deci, E., Vansteenkiste, M.: Self-determination theory and basic need satisfaction: Under-
standing human development in positive psychology. Ricerche di Psicologia 27(1), 23-40
(2004).
13. Ling, K., Beenen, G., Ludford, P., Wang, X., Chang, K., Frankowski, D., Resnick, P., Kraut,
R.: Using social psychology to motivate contributions to online communities. In: Proceed-
ings of the 2004 ACM conference on Computer supported cooperative work, pp. 212–221,
ACM New York, Chicago, Illinois, USA (2005).
14. Sailer, M., Hense, J., Mandl, H., & Klevers, M.: Psychological perspectives on motivation
through gamification. Interaction Design and Architecture(s) Journal 19, 28-37 (2013).
15. Sailer, M. Ulrich Hense, J., Mayr, S., Mandl, H.: How gamification motivates: An experi-
mental study of the effects of specific game design elements on psychological need satisfac-
tion. Computers in Human Behavior 69, 371-380 (2017).
16. Tondello, G., Mora, A., Nacke, L.: Elements of Gameful Design Emerging from User Pref-
erences, In: Proceedings of the Annual Symposium on Computer-Human Interaction in Play,
pp. 129-142, ACM New York, Amsterdam, Netherlands (2017).
17. Deterding, S., Khaled, R., Nacke, L., Dixon, D.: Gamification: Toward a definition. In: CHI
2011 Gamification Workshop Proceedings, pp. 12-15, Vancouver, Canada (2011).
18. Deterding, S.: Eudaimonic design, or: Six invitations to rethink gamification. In: Rethinking
Gamification. pp. 301-331, 1st Edition, Meson Press, Lüneburg (2014).
19. Yang, R., Chen, V.: Gamification: Influencing Value-Perception of Target Behaviors. In:
CEUR Workshop Proceedings. pp. 120-128, Pori, Finland, May 9-10, (2017).
20. Burke, J. W., McNeill, M., Charles, D., Morrow, P. J., Crosbie, J. & McDonough, S.: Serious
Games for upper limb Rehabilitation following Stroke. In: 2009 Conference in Games and
Virtual Worlds for Serious Applications, pp. 103-110, IEEE, Coventry, UK (2009).
21. De Gloria, A., Bellotti, F., Berta, R.: Serious Games for education and training. International
Journal of Serious Games 1(1) (2014).
22. Tăut, D., Pintea, S., Roovers, J., Mañanas, M., Băban, A.: Play seriously: Effectiveness of
serious games and their features in motor rehabilitation. A meta-analysis. NeuroRehabilita-
tion 41(1), 105-118, (2017).
23. Agopyan, H., Griffet, J., Ginon, C., Bruno, M., Bredin, J.: Personalized and parametrized
AVATAR for interacting with post-stroke action-perception loop during rehabilitation. Neu-
rophysiologie Clinique/Clinical Neurophysiology 46(4-5), 240 (2016).
24. Blandón, D., Muñoz, J., Lopez, D., Gallo, O.: Influence of a BCI neurofeedback videogame
in children with ADHD. Quantifying the brain activity through an EEG signal processing
dedicated toolbox. In: 2016 IEEE 11th Colombian Computing Conference (CCC), pp. 1-8,
IEEE, Popayan, Colombia (2016).
25. Warpefelt, H.: The Non-Player Character: Exploring the believability of NPC presentation
and behavior. 1st Edition. Doctoral thesis. Stockholm: Department of Computer and Sys-
tems Sciences, Stockholm University (2016).
26. Rogers, K., Aufheimer, M., Weber, M., Nacke, L.: Towards the Visual Design of Non-
Player Characters for Narrative Roles. In: Proceedings of the 44th Graphics Interface Con-
ference, 145-152. Canadian Human-Computer Communications Society, Mississauga, Can-
ada (2018).
27. Rogers, K., Aufheimer, M., Weber, M., Nacke, L.: Exploring the Role of Non-Player Char-
acters and Gender in Player Identification. In: Proceedings of the 2018 Annual Symposium
on Computer-Human Interaction in Play Companion Extended Abstracts, pp. 271-283,
ACM New York, Melbourne, VIC, Australia (2018).
GamiFIN Conference 2019, Levi, Finland, April 8-10, 2019 141