Player typology models classify different player motivations and behaviours. These models are necessary to design personalized games or to target specific audiences. However, many models lack validation and standard measurement instruments. Additionally, they rely on type theories, which split players into separate categories. Yet, personality research has lately favoured trait theories, which recognize that people's preferences are composed of a sum of different characteristics. Given these shortcomings of existing models, we developed a player traits model built on a detailed review and synthesis of the extant literature, which introduces five player traits: aesthetic orientation, narrative orientation, goal orientation, social orientation, and challenge orientation. Furthermore, we created and validated a 25-item measurement scale for the five player traits. This scale outputs a player profile, which describes participants' preferences for different game elements and game playing styles. Finally, we demonstrate that this is the first validated player preferences model and how it serves as an actionable tool for personalized game design.

1. “I don’t fit into a single type”:
A Trait Model and Scale of
Game Playing Preferences
Gustavo F. Tondello, Karina Arrambide,
Giovanni Ribeiro, Andrew Cen, Lennart E. Nacke
INTERACT 2019, 6 September 2019

2. Goals
Describe the five player traits
Social Orientation
Aesthetic Orientation
Narrative Orientation
Challenge Orientation
Goal Orientation
Present the measurement scale
Describe the relationship with other models
Describe applications of the player traits
2

3. Bartle’s Player Types (1996)
3
Bartle, R.: Hearts, Clubs, Diamonds, Spades: Players who suit MUDs. Journal of MUD Research 1(1) (1996)
Image source: https://www.interaction-design.org/literature/article/bartle-s-player-types-for-gamification

4. Seeker
Survivor Socialiser
Daredevil Conqueror
Mastermind Achiever
BrainHex (2014)
4
Nacke, L.E., Bateman, C., Mandryk, R.L.: BrainHex: A Neurobiological Gamer Typology Survey. Entertainment Computing 5(1) (2014)
Image source: https://blog.brainhex.com/

5. Issues with Player Types
Players do not enjoy only one type of
experience
Lack of validated scale or unreliable scale
Solution
Player Traits recognize that people's
preferences are composed of a sum of
different characteristics
5

6. Player Traits Model
6
Image source: http://hcigames.com/player-traits/
Copyright 2019 by the HCI Games Group (CC BY-NC-ND 4.0) using icons from game-icons.net (CC BY 3.0).

7. Social Orientation
Players who score high prefer to
play together with others, enjoy
multiplayer games and
competitive gaming communities
Players who score low prefer to
play alone
7

8. Aesthetic Orientation
Players who score high enjoy
aesthetic experiences in games:
exploring the world, enjoying the
scenery, appreciating the
graphics, sound, and art style, etc.
Players who score low focus
more on gameplay than the
aesthetics of the game
8

9. Narrative Orientation
Players who score high enjoy
complex narratives and stories
within games
Players who score low prefer
games with less story and
might skip the story or
cutscenes when those get in
the way of gameplay
9

10. Challenge Orientation
Players who score high prefer
difficult games and hard
challenges
Players who score low prefer
easier or casual games
10

11. Goal Orientation
Players who score high enjoy
completing game goals and like
to complete games 100%, explore
all the options, and complete all
the collections
Players who score low might
leave optional quests or
achievements unfinished
11

12. Measurement Scale
We created a measurement scale with 25 Likert items
(5 items per trait)
Validated with exploratory (N = 175) and confirmatory (N = 157)
factor analysis, and test-retest reliability (N = 70)
Model Fit Indices (from CFA)
CFI = .927 (good if ≥ .90)
RMSEA = .058 (good if < .08)
SRMR = .067 (good if < .08)
12

13. Mean Scores and Reliability
Player Traits Mean
Std.
Deviation
Consistency
(α)
Test-retest
reliability (r)
Social orientation 51.4% 24.7 .914 .906
Aesthetic orientation 80.1% 14.8 .753 .763
Narrative orientation 77.7% 18.6 .843 .768
Challenge orientation 64.8% 18.6 .854 .813
Goal orientation 58.2% 19.9 .819 .844
13
(with N = 332) (from EFA; N = 175) (with N = 70)

14. Correlation with Personality Traits
Player Traits Extrav. Agree. Consc. Neurot. Open.
Social orientation .254 .149 - -.129 -
Aesthetic orientation - - - - .248
Narrative orientation -.169 - - - .127
Challenge orientation - - - -.175 -
Goal orientation - - .141 .118 -
14
Included coefficients (Pearson’s r) are significant at p < .05.

15. Correlation with Game Elements
Player Traits
Strat. Res.
Manag.
Puzzle
Artistic
Movement
Sports &
Cards
Social orientation .205 - .154 .199
Aesthetic orientation - .163 - -.130
Narrative orientation - - -.113 -.224
Challenge orientation .202 .234 - .130
Goal orientation .131 .180 - -
15
Included coefficients (Pearson’s r) are significant at p < .05.

16. Correlation with Game Elements
Player Traits
Role-
playing
Virtual
Goods
Simulation Action
Social orientation - .229 - .241
Aesthetic orientation .479 .305 .521 .311
Narrative orientation .492 - .396 -
Challenge orientation .111 - - .403
Goal orientation .210 .248 .133 -
16
Included coefficients (Pearson’s r) are significant at p < .05.

17. Correlation with Playing Styles
Player Traits Multipl.
Abstract
Interac.
Solo
Play
Comp.
Comm.
Casual
Play
Social orientation .818 - -.115 .460 .115
Aesthetic orientation - - .363 - -
Narrative orientation -.166 - .256 -.133 -
Challenge orientation .263 .145 .238 .271 -.173
Goal orientation - - - - -
17
Included coefficients (Pearson’s r) are significant at p < .05.

18. Takeaways
We introduced a new player traits model that
solves the issues identified in previous work
We created and validated a 25-item
measurement scale
We showed that player traits are somewhat
correlated, but different than personality
traits
We showed that player traits are correlated to
preferred game elements and playing styles
18

19. Applications
To select participants for game tests
To better understand game tests
according to participant’s gaming
preferences
To give designers and game studios
more accurate insights about their
audience
To target market campaigns to the
right audience
19

20. Future Work
Continue validating the
scale with larger samples
Continue studying
correlations with other
models
Compare participants’ self-
reported preferences with
their actual behaviour in
games
20

21. Thank you!
A Trait Model and Scale of Game Playing Preferences
http://hcigames.com/player-traits
CONTACT
Gustavo F. Tondello
gustavo@tondello.com
@GustavoTondello
Acknowledgments: This work was supported by the CNPq Brazil, SSHRC (IMMERSe),
NSERC Discovery, NSERC CREATE SWaGUR, and CFI, and presented at INTERACT 2019.
21
Take the test!