Control your Game-Self: Effects of Controller Type on Enjoyment, Motivation and Personality in Game

Written by João Costa



Birk, M., Mandryk, R.L. 2013. Control Your Game-Self: Effects of Controller Type on Enjoyment, Motivation, and Personality in Game. In CHI ’13: Proceedings of the 31st international conference on Human factors in computing systems, Paris, France. To appear.


This paper presents a study in which the authors want to understand whether the choice of a game controller has impact on the player’s personality during play. For this, they used different controllers in conjunction with questionnaires (PENS, IMI and PANAS-X) to explain the Player Experience (PX) in terms of enjoyment, motivation and player perceived personality.

Their results show that the controller choice affects player enjoyment and motivation to play, as well as how the player’s needs satisfaction is met within gameplay and how they perceive themselves.


The authors’ Related Work section on Player Experience (PX) evaluation mentions some of the old and new methods of evaluating the user’s experience. Nevertheless, the mention to psychophysiological methods of evaluation of PX lacks referencing of other recent research that are relevant in context to the paper’s direction because this is a growing field with considerable novel research.

Moving forward into related work in the field of psychology, the authors give a thorough description of what inspired them for this study. They begin with the Self Determination Theory (SDT), its sub theory of Cognitive Evaluation Theory (CET), and how it is possible to use the Player Experience of Needs Satisfaction (PENS) to assess the components of SDT and CET on an experience.

The authors also provide insight on another theory called the Self-Discrepancy Theory. This theory is based on the assumption that people compare themselves to so called self-guides. These self-guides represent three domains of self: actual-self (the attributes we actually possess), ideal-self (the attributes we ideally want to possess) and the ought-self (the attributes we believe we should have). A common way to assess these self-guides is to use the five factor model of personality, commonly referred as the Big Five.

Extraversion (E) Tendency to be energetic, outgoing and assertive. People who score high tend to be involved in social activities rather than spending time alone.
Agreeableness (A) Tendency to be friendly, caring and conflict avoiding. People who score high prefer co-operation to competition.
Conscientiousness (C) Tendency to show self discipline, act dutifully, and aim for achievement. People who score high plan activities rather than engaging in spontaneity. Tend to be organized and dependable.
Neuroticism (N) Tendency to be nervous, sensitive and emotionally unstable. People who score high respond poorly to environmental stress and are likely to experience stress and frustration.
Openness to Experience (O) Tendency to be intellectually curious, think abstractly, and to explore. People who score high are likely to try something new rather than sticking to old beliefs and tradition.

These Big Five personality traits have been successfully used to predict the player type, preferred genre and motivation to play, all of which can be used to understand the level of enjoyment of a game. What is particularly useful of the Self-Discrepancy theory is that it considers the difference between self-guides. Past research has shown that game enjoyment affected by the discrepancy between actual-self and ideal-self.

The authors designed a simple 3D game where the user was flying inside of a tornado in a ship and the goal was to collect spinning items. The items would show up in waves, and collecting enough items speeds the player’s ship to the next wave. Three different controller types were used: Traditional – Microsoft XBox GamePad, Positional – Sony PlayStation Move, Gestural – Microsoft Kinect.

The study involved the participation of 78 students, with a within-participants design, who had to fill a questionnaire about their actual-self, ideal-self and current affective state before playing the game with each of the three different controller types. After each controller condition, 5 minutes, players completed questionnaires assessing their experience of the game and of themselves within the game. Game enjoyment was measured with Positive Affect and Negative Affect Schedule-Expanded (PANAS-X), as well as with Intrinsic Motivation Inventory (IMI). Need Satisfaction was evaluated using the Player Experience of Need Satisfaction Scale (PENS), while Personality was assessed with the Big Five Inventory (BFI).

The authors found how the controller choice influences our needs satisfaction differently. They found that the Game-self (the actual self in the context of game play) changed significantly with the use of a different controller, where they say, after a hefty chunk of statistics, that the GamePad increases Neuroticism, whereas the Kinnect enhances Agreeableness. To put it in practical terms, imagine you are designing a game where your character starts frail and gets stronger with the course of time while also having to overcome obstacles (e.g. Deus Ex). The game designer’s intention might be of transmitting the fragility of the main character, and the use of a GamePad might be the ideal choice for this sort of game because Neuroticism makes those obstacles a notch harder. Or, in a game designed as a social icebreaker, the Kinect might be the choice for controller since it improves Agreeableness. The key focal point is to match the player’s actual-self with the ideal-self that the game controller permeates (personality affordances).

In my opinion, the results presented are incredibly thorough and deep in statistics. However, they were perhaps a bit too deep to the point that it makes the understanding of the paper a bit cloudy and convoluted.

Second, the paper is interesting because it tries to tie together several psychology theories under one common purpose (in the related work section). Nevertheless, perhaps because there is so much going on throughout the study, and because so many keywords are the same or similar, it makes it harder to understand the exact purpose of the study, which should have been clearly explained earlier along the paper. I only got there half-way through the experiment description.

Third, in their discussion, the authors only mention the GamePad and the Kinect, but never mention their third controller type, PlayStation Move. In the end, they take some interesting conclusions about the GamePad and Kinect controllers specifically, but what about all the other remaining controllers? It would have been nice to read a result about not the specific controller type in relation to the personality affordances that it allows, but more on these same affordances in relation to the task mapping that the controller has.

Image credits: Side Pose Of Man Playing Videogame –