These are the slides of my Ph.D. thesis oral defence at the University of Waterloo on June 20, 2019.
Gameful design, the process of creating a system with affordances for gameful experiences, can be used to increase user engagement and enjoyment of digital interactive systems. It can also be used to create applications for behaviour change in areas such as health, wellness, education, customer loyalty, and employee management. However, existing research suggests that the qualities of users, such as their personality traits, preferences, or identification with the task, can influence gamification outcomes.
Given how user qualities shape the gameful experience, it is important to understand how to personalize gameful systems. Current evidence suggests that personalized gameful systems can lead to increased user engagement and be more effective in helping users achieve their goals than generic ones. However, to create this kind of system, designers need a specific method to guide them in personalizing the gameful experience to their target audience. To address this need, this thesis proposes a method for personalized gameful design with three steps: (1) classification of user preferences, (2) classification and selection of gameful design elements, and (3) heuristic evaluation of the design.
Furthermore, this thesis describes the design, implementation, and pilot evaluation of a software platform for the study of personalized gameful design. It integrates nine gameful design elements built around a main instrumental task, enabling researchers to observe and study the gameful experience of participants. The platform is flexible so the instrumental task can be changed, game elements can be added or removed, and the level and type of personalization or customization can be controlled. This allows researchers to generate different experimental conditions to study a broad range of research questions.
Our personalized gameful design method provides practical tools and clear guidelines to help designers effectively build personalized gameful systems.
2. Contributions
A. Personalized Gameful Design
1. Classification of user preferences
2. Selection of gameful design elements
3. Evaluation of the design
B. Platform for the Study of
Personalized Gameful Design
2
4. Personalized Gameful
Design
The tailoring
of the gameful design elements
by the providers to the users
based on knowledge about them,
to boost the achievement of the goals
of the gameful system.
4Gustavo F. Tondello. 2019. Dynamic Personalization of Gameful Interactive Systems. PhD Thesis, University of Waterloo.
5. Types of Personalization
5
User-initiated (customization) System-initiated
Access
Lotteries
Boss Battles
Because you recently completed a challenge:
Unlock restricted
areas of the system
You can earn
amazing rewards in
our lottery
Test your skills with
these highly difficult
tasks
6. Why Personalize Gameful Systems?
• Higher engagement, performance, enjoyment
• Boost the achievement of goals
• task completion rate, learning, health, employee engagement…
6
Education Health Fitness Nutrition
Training Customer
relations
Human
resources
Team
management
Image source: www.pexels.com
7. Why Personalize Gameful Systems?
“I feel like it did [influence my
enjoyment] because I felt like I
kind of created my own game
that was perfect for me and so
it felt like I was in control and
added to my enjoyment.”
9
Pilot study with N = 50 MTurk participants.
Gustavo F. Tondello. 2019. Dynamic Personalization of Gameful Interactive Systems. PhD Thesis, University of Waterloo.
9. 1. Classification of User Preferences
11
Classification
of User
Preferences
Selection of
Gameful
Design
Elements
Evaluation of
the Design
1 2 3
10. Gamification User Types Hexad
Do all users equally enjoy all game elements?
12Andrzej Marczewski. 2015. User Types. In Even Ninja Monkeys Like to Play: Gamification, Game Thinking & Motivational Design.
Gamified UK.
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
11. Gamification User Types Hexad
24-item scale
(developed in collaboration with Andrzej Marczewski
and the Austrian Institute of Technology)
13
Image source: screenshots from https://gamified.uk/
G. F. Tondello, R. R. Wehbe, L. Diamond, M. Busch, A. Marczewski, L. E. Nacke. 2016. The Gamification User Types Hexad Scale. In
Proceedings of the 2016 Annual Symposium on Computer-Human Interaction in Play - CHI PLAY ’16, 229–243.
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
12. User Types Hexad: Validation
Four studies
1. Initial validation in English (N = 133)
2. Large-scale validation in English and Spanish (N = 556)
(data gently provided by Alberto Mora)
3. Large-scale validation in English and Spanish (N = 1,328)
(data gently provided by Andrzej Marczewski)
4. Validation of suggested improvements in English (N = 152)
(data collected in collaboration with Andrzej Marczewski)
14
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
[1] G. F. Tondello, R. R. Wehbe, L. Diamond, M. Busch, A. Marczewski, L. E. Nacke. 2016. The Gamification User Types Hexad Scale. In
Proceedings of the 2016 Annual Symposium on Computer-Human Interaction in Play - CHI PLAY ’16, 229–243.
[2-4] G. F. Tondello, A. Mora, A. Marczewski, L. E. Nacke. 2019. Empirical Validation of the Gamification User Types Hexad Scale in
English and Spanish. International Journal of Human-Computer Studies 127, 95–111.
13. User Types Hexad: Validation
Analysis methods
• Exploratory Factor Analysis (EFA)
• Confirmatory Factor Analysis (CFA) with structural equation
modeling (SEM)
Results
The scale is generally consistent and reliable
[1] has already been cited 112 times
15
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
[1] G. F. Tondello, R. R. Wehbe, L. Diamond, M. Busch, A. Marczewski, L. E. Nacke. 2016. The Gamification User Types Hexad Scale. In
Proceedings of the 2016 Annual Symposium on Computer-Human Interaction in Play - CHI PLAY ’16, 229–243.
[2-4] G. F. Tondello, A. Mora, A. Marczewski, L. E. Nacke. 2018. Empirical Validation of the Gamification User Types Hexad Scale in
English and Spanish. International Journal of Human-Computer Studies 127, 95–111.
14. 2. Selection of Gameful Design Elements
18
Classification
of User
Preferences
Selection of
Gameful
Design
Elements
Evaluation of
the Design
1 2 3
15. Groups of Gameful Design Elements
19
Individual
Motivations
Immersion
Progression
External
Motivations
Risk/Reward
Customization
Incentive
Social
Motivations
Socialization
Assistance
Altruism
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
Results of Principal Components Analysis (N = 196)
Image sources: www.pexels.com and Game-icons.net
Gustavo F. Tondello, Alberto Mora, and Lennart E. Nacke. 2017. Elements of Gameful Design Emerging from User Preferences. In Proceedings of the
2017 Annual Symposium on Computer-Human Interaction in Play - CHI PLAY ’17, 129–142.
16. Groups of Gameful Design Elements
28
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
Immersion
Progression
Customization
Incentive
Risk/Reward
Altruism
Socialization
Assistance
Note: Mean Likert scores (1–5) per group (N = 196)
Gustavo F. Tondello, Alberto Mora, and Lennart E. Nacke. 2017. Elements of Gameful Design Emerging from User Preferences. In
Proceedings of the 2017 Annual Symposium on Computer-Human Interaction in Play - CHI PLAY ’17, 129–142.
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
17. Preferences by Gender
29
Note: Mean differences per gender and group (N = 124 men, 53 women)
Gustavo F. Tondello, Alberto Mora, and Lennart E. Nacke. 2017. Elements of Gameful Design Emerging from User Preferences. In
Proceedings of the 2017 Annual Symposium on Computer-Human Interaction in Play - CHI PLAY ’17, 129–142.
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
0.2
-0.29
-0.08
-0.14
0.13
-0.11
-0.37
0.63
0.37
0.31
-0.25
0.23
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0.8
Socialization Assistance Immersion Customization Altruism Incentives
Men Women
18. Preferences by Hexad User Type
30
Note: Based on results from correlation analysis (N = 196)
Gustavo F. Tondello, Alberto Mora, and Lennart E. Nacke. 2017. Elements of Gameful Design Emerging from User Preferences. In
Proceedings of the 2017 Annual Symposium on Computer-Human Interaction in Play - CHI PLAY ’17, 129–142.
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
Participants who scored higher as… …tended to prefer these groups
Free Spirit Immersion
Philanthropist Immersion, Progression, Altruism
Achiever Socialization, Immersion, Risk/Reward,
Progression, Altruism
Socialiser Socialization, Assistance, Altruism
Player Socialization, Risk/Reward, Incentive
Disruptor Immersion, Risk/Reward
20. 3. Evaluation of the Design
32
Classification
of User
Preferences
Selection of
Gameful
Design
Elements
Evaluation of
the Design
1 2 3
21. Gameful
Design
Heuristics
33
G. F. Tondello, D. L. Kappen, E. D. Mekler, M. Ganaba, L. E. Nacke. 2016. Heuristic
Evaluation for Gameful Design. In Proceedings of CHI PLAY ’16 Extended Abstracts.
G. F. Tondello, D. L. Kappen, M. Ganaba, L. E. Nacke. 2019. Gameful Design
Heuristics: A Gamification Inspection Tool. In Proceedings of HCI International 2019.
Infographic by Dennis Kappen and Marim Ganaba
Heuristics
General design principle or
guidelines
Heuristic Evaluation
Use of said principles to identify
design problems
Gameful Design Heuristics
Set of guidelines for heuristic
evaluation of gameful
applications
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
22. Gameful Design Heuristics
How to conduct a heuristic evaluation of a gameful system
1. Familiarize yourself with the application
2. Use the heuristics checklist
3. For each heuristic:
a. Familiarize yourself with the heuristic
b. Think about the supporting questions in relation to the app
c. If you identify any issue in the app related to the heuristic, write it down
4. Finally, count the number of issues identified for each category to
identify those with more issues
34
Gustavo F. Tondello, Dennis L. Kappen, Elisa D. Mekler, Marim Ganaba, and Lennart E. Nacke. 2016. Heuristic Evaluation for Gameful Design.
In Proceedings of CHI PLAY ’16 Extended Abstracts, 315–323.
G. F. Tondello, D. L. Kappen, M. Ganaba, L. E. Nacke. 2019. Gameful Design Heuristics: A Gamification Inspection Tool. In Proceedings of HCI
International 2019.
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
23. Usefulness of this evaluation method
35
1. Classification of User Preferences 2. Selection of Gameful Design Elements 3. Evaluation of the Design
Note: N = 2 without heuristics. N = 3 with heuristics for Habitica, 1 for Termling.
G. F. Tondello, D. L. Kappen, M. Ganaba, L. E. Nacke. 2019. Gameful Design Heuristics: A Gamification Inspection Tool. In Proceedings of HCI
International 2019.
Results
• Gameful design heuristics help evaluators who are not familiar with gamification to
evaluate a system as well as a gamification expert who does not use the heuristics
• Gameful design heuristics improves the ability of gamification experts to perform an
heuristic evaluation
7 8
13
24
0
5
10
15
20
25
30
Habitica Termling
Number of motivational issues found
Without Heuristics
With Heuristics
28. Frequency of Element Selection
36
30 30
23 23
20
16 16
6
0
5
10
15
20
25
30
35
40
Progress
Feedback
Levels Power-ups Leaderboards Chance Badges Unlockable
Content
Challenges Moderating
Role
40
Pilot study with N = 50 MTurk participants.
Gustavo F. Tondello. 2019. Dynamic Personalization of Gameful Interactive Systems. PhD Thesis, University of Waterloo.
29. Rating of the Experience
34
36
42
39 39
9 8
6 5
11
7
1 2 3
0
0
5
10
15
20
25
30
35
40
45
Overall experience Element selection Satisfaction Preference matching Enjoyment
Positive Neutral Negative
41
Pilot study with N = 50 MTurk participants.
Gustavo F. Tondello. 2019. Dynamic Personalization of Gameful Interactive Systems. PhD Thesis, University of Waterloo.
31. Personalized Gameful
Design
43Gustavo F. Tondello. 2019. Dynamic Personalization of Gameful Interactive Systems. PhD Thesis, University of Waterloo.
The tailoring
of the gameful design elements
by the providers to the users
based on knowledge about them,
to boost the achievement of the goals
of the gameful system.
35. Open Research Questions
Do user preferences vary by context or country?
Do users’ behaviours correspond to their self-reported
preferences?
Do different designs of gameful elements influence user
preferences?
What other evaluation methods can be devised?
47
36. Empirical Studies of Personalized
Gameful Systems
Personalized vs Generic systems
Partial vs Full Personalization
Personalization vs Customization
48
37. Friend invite
Social discovery
Trading
Access
Lotteries
Boss Battles
Because you recently completed a challenge: Because you recently joined a team:
Unlock restricted
areas of the system
You can earn
amazing rewards in
our lottery
Test your skills with
this highly difficult
tasks
Invite your friends
to work with you
Find other users
with similar
interests
Exchange your
spare items with
other users
Recommender Systems for
Personalized Gamification
49
Models of user types
and design elements
Recommendation
algorithms
Machine learning
38. What else can we personalize?
Activities
Game
Elements
Persuasive
Strategies
Difficulty Rewards
50Image source: www.pexels.com
39. Questions and Discussion
51
Dynamic Personalization of
Gameful Interactive Systems
Gustavo Fortes Tondello
Acknowledgments:
This research was supported by:
• The National Council for Scientific and Technological Development – Brazil (CNPq)
• University of Waterloo (Cheriton School of Computer Science, Games Institute, HCI Games Group)
• NSERC (grants RGPIN-418622-2012; RGPIN-2018-06576)
• SSHRC (grant 895-2011-1014, IMMERSe)
• Mitacs (grant IT07255)
• CFI (grant 35819)
• NSERC CREATE SWaGUR