Player Engagement caused by Graphical Art Styles


Choosing a suitable graphical style for a game can be difficult and might have unpredictable effects on the success of a game. To determine how different graphical styles influence the player’s desire to continue playing, the following problem is stated:

“How do the different art styles cartoon, realistic, and cubic graphics relate,
in regards to causes of engagement that contribute to the player’s desire to continue,
in a first-person video game?”

Research into engagement reveals that players are engaged by different factors. In the article “Sure, I’d like to continue.” (2012) Schønau-Fog & Bjørner presents the factors, on how players are engaged. They split the causes of engagement into six different categories: Emotional, Intellect, Narrative, Physical, Sensory and Social.
The prototype will be developed in four iterations of designing and implementing.
Through the iterations, different issues such as movement speed of the playable avatar, objective design, interactions within the game, and sound design were discovered and accounted for before the final testing.
The final prototype is a game, constructed in three different graphical styles: Realistic, cubic, and cartoon. The test participants will play through one of the graphical styles, after which they will be asked if they want to continue playing. Reaction cards will then be presented to the participants after they have given an answer to the question: Do you want to continue? They will be asked to pick five cards to explain their answer.
Due to a small sample size (n=50), it is difficult to conclude anything, but tendencies can still be found. The results provide recommendations for choice of graphical style.
The results show firstly, that the causes of engagement for the realistic graphical style is more emotional than the other versions of the prototype. Secondly, that the engagement causes of the cubic graphical style is more physical, and lastly, that the causes of engagement for cartoon graphical style is more evenly distributed that the other styles.
More testing will have to be conducted to fully answer the problem statement, but these results function as guidelines and can be further explored in future iterations.


Link to report