Player Performance, Satisfaction, and Video Game Enjoyment

Abstract

An experiment (N = 74) was conducted to investigate the impact of game difficulty and player performance on game enjoyment. Participants played
a First Person Shooter game with systematically varied levels of difficulty. Satisfaction with performance and game enjoyment
were assessed after playing. Results are not fully in line with predictions derived from flow and attribution theory and suggest
players to (1) change their view on their own performance with its implications for enjoyment with increasing game experience
and (2) to switch strategically between different sources of fun, thus maintaining a (somewhat) positive experience even when
performance-based enjoyment is low.

Full text

Player Performance, Satisfaction, and
Video Game Enjoyment
Christoph Klimmt1, Christopher Blake2, Dorothée Hefner2, Peter Vorderer3,
and Christian Roth3
1 Department of Communication, Johannes Gutenberg University of Mainz,
Kleinmann-Weg 2, 55099 Mainz, Germany
2 Department of Journalism and Communication Research, Hanover University of Music
and Drama, EXPO-Plaza 12, 30539 Hannover, Germany
3 Center for Advanced Media Research Amsterdam (CAMeRA), VU University Amsterdam,
De Boelelaan 1105, 1081 HV Amsterdam, The Netherlands
{Christoph Klimmt, Christopher Blake, Dorothée Hefner, Peter Vorderer,
Christian Roth, klimmt@uni-mainz.de
Abstract. An experiment (N = 74) was conducted to investigate the impact of
game difficulty and player performance on game enjoyment. Participants
played a First Person Shooter game with systematically varied levels of
difficulty. Satisfaction with performance and game enjoyment were assessed
after playing. Results are not fully in line with predictions derived from flow
and attribution theory and suggest players to (1) change their view on their own
performance with its implications for enjoyment with increasing game
experience and (2) to switch strategically between different sources of fun, thus
maintaining a (somewhat) positive experience even when performance-based
enjoyment is low.
Keywords: Video games, entertainment, enjoyment, performance, flow,
attribution theory.
Acknowledgment: This research was funded by the European Commission,
project “FUGA: The fun of gaming” (NEST-PATH-IMP 28765). We thankfully
acknowledge the Commission‟s support.
1 Introduction
Many forms of contemporary entertainment computing, most importantly, video
games, apply their interactivity to present tasks and challenges to users. A great
diversity of task types and challenge levels can be found in modern video games, for
instance, tasks that require dexterity and precise timing of control inputs (such as in
first person shooters), tasks that demand logical thinking and problem solving (e.g.,
adventure games, brain trainers), or tasks related to strategic planning and
management of complexity (such as strategy games and business simulations). While
social research on the motivational appeal of video games is still sparse [1], there is

some indication that resolving game tasks and mastering game challenges is closely
connected to game enjoyment. Ryan, Rigby and Pryzybylski [2] demonstrated
experimentally that feelings of competence are an important dimension of the
motivational appeal of digital games. Two surveys identified competition as driver of
playing motivation [3] [4]. Both competence and competition are inherently
connected to tasks and challenges presented by games – mastering challenges thus is
probably linked to game enjoyment [5].
The present study addresses the issue of task resolution, mastery of game
challenges, and game enjoyment in more detail. Specifically, the paper attempts to
shed more light on the complexity that is involved in the connection between player
performance and game enjoyment. This complexity stems primarily from the common
understanding that good performance can only occur when the task resolved is not too
easy. Accomplishing an easy task does not hold much merit, and performance is only
valued if it is perceived as mastery of a significant challenge. Building on concepts
from motivational psychology, especially attribution theory [6], this paper takes an
explanatory stance on player performance and video game enjoyment by examining
player responses to and enjoyment of systematically varied challenge levels.
2 Performance, Attribution, Satisfaction, and Game Enjoyment
Research in the psychology of motivation has found consistently that human
individuals respond to good own performance (success) with positive emotions such
as pride and joy. Weiner [6] has emphasized that such positive emotions occur if the
individual identifies her-/himself as origin of the event interpreted as success. Only if
the positive event can be attributed to oneself (e.g., to one‟s talent or one‟s hard
effort), strong positive emotions will arise; if the individual perceives external factors
(e.g., somebody else‟s effort or simple luck) responsible for the event, the resulting
emotions may still be positive, but will not reach the same level of intensity.
Conceptually, the emotion of “pride” will only occur in the case of self-attribution of
the success event, whereas the emotion of “joy” may also occur in the case of external
attribution of the (positive, appreciated) event. Similarly, self-determination theory
[7] argues for the importance of feelings of one‟s own competence for positive
emotions: It is thus the perception that oneself has done something „good‟ or has
achieved a great success that makes the difference in emotional experience.
An important underlying mechanism that connects satisfaction with one‟s
performance to game enjoyment is self-esteem [8]. In general, increases in self-
esteem go along with positive emotions such as pride and joy, and positive
performance feedback or direct experience of competence rise the individual‟s self-
esteem level [9]. Therefore, successful task resolution is theorized to level up self-
esteem, and the increase of self-esteem is experienced as highly enjoyable.
Thus, in the context of video games, the challenges that players are confronted
with represent opportunities to experience own competence by attributing the success
events in the game to one‟s own skill and efforts [10]. Resolving the game tasks
would then be a key to game enjoyment: As most games introduce new tasks and
challenges at high frequencies, players receive ample opportunities to feel competent

and successful, lift their state self-esteem, and thus generate positive self-emotions
continuously throughout game play. The permanent pride of mastering ever-new
game challenges would then constitute an important part of game enjoyment.
However, the link between one‟s own performance, self-esteem and positive
emotions is more complicated, for two important moderators affect the performance-
enjoyment process. One is the difficulty of the task(s) mastered, and the other is the
performance expectation that the individual holds before and during task resolution.
Concerning task difficulty, attribution theory [6] suggests that players cannot derive
pride from the mastery of (very) easy tasks, because in this case, there is no chance to
demonstrate skill or invest serious effort. Without skill demonstration and/or efforts
invested, there is no reason to be proud of: The challenge is simply no challenge. In
the context of video gaming, easy tasks (e.g., enemies that are easy to kill and do not
cause serious damage to the player character) would rather evoke boredom than
enjoyment. In contrast, (very) difficult tasks do not facilitate positive feelings either
[6]. One reason is that difficult tasks are not resolved with high probability, so
experiences of failure and insufficient performance arise more frequently under high
difficulty conditions. Such experiences undermine self-esteem and lead to frustration
and sadness – the opposite of pride, and, when applied to game situations, also the
opposite of game enjoyment. A second important reason for hard tasks interrupting
the effect of success on enjoyment is that if players manage to resolve very heavy
game tasks, there is a often a reasonable chance that their skill and effort alone did not
cause the success, but that additional external factors (e.g., luck) co-occurred, which
would question the self-attribution of the success. The enjoyment value of mastering
very difficult tasks is thus not as „secure‟ as the fun that players can generate from
mastering moderately difficult tasks. For the solution of such tasks, they can claim
full responsibility, with a secure impact on positive emotions.
This consideration converges with flow theory that postulates most positive
experiences resulting from mastering tasks that are not too easy and not too difficult
[11]. Flow theory has also been applied to video game enjoyment [12]. Consequently,
the contribution of success experiences to video game enjoyment is argued to depend
on the difficulty of the game played. Neither very easy nor very hard games should
elicit success-based game enjoyment to a substantial extent; rather, moderately
difficult games should facilitate the highest level of enjoyment.
The link between player performance and game enjoyment is likely to be affected
also by the performance expectations players hold [13]. Expert players will be
convinced of their capacity to master highly difficult challenges and will thus expect
themselves to be quite successful with any given new game. Novice players, in
contrast, will accept the possibility of failure and underperformance in many new
games. Because they are „prepared‟ for failures, they should not suffer from severe
reductions in game enjoyment when they face difficult tasks, as they can still meet
their low performance expectations and need not be disappointed about their
achievements in the game. Expert players, however, may feel frustration more
frequently, as failure to accomplish (seemingly) easy or moderate tasks would imply a
violation of their own high performance expectation and thus reduce their (high) self-
esteem level significantly [13]. On the other hand, expert players hold an objectively
better chance to master any given game task, while novice players will fail with much
greater probability at any given game task. So expert players will succeed frequently

but be disappointed about their few failures, whereas novice players will fail
frequently but will not feel much frustration about it. It is theoretically plausible, then,
to argue that good performance and achievement is virtually irrelevant for the game
enjoyment of novice players, whereas for expert players, performance would be
extremely important, because their self-expectations are high, and they seem to
perceive video games in general as a domain to demonstrate skill and superior
performance [4]. For the present study, we thus focused on expert players to examine
the interplay of performance, satisfaction and game enjoyment.
From these considerations, the actual enjoyment experience during game play can
be modelled. Players begin a gaming session with a self-expectation concerning their
skills and performance capabilities (i.e., they define themselves as rather novice or
rather expert players). The tasks that the game offers – enemies, puzzles, etc. – will
lead to performance-based fun to the extent that (A) players find the tasks challenging
(not too easy, not too difficult) and (B) players find their accomplishments matching
their self-expectations. Because task difficulty and self-expectations are interrelated –
heavy gamers find other game tasks „hard‟ than novice players –, it is difficult to
predict the specific level of optimal performance satisfaction that leads to increase in
self-esteem, pride emotions, and thus achievement-based game enjoyment.
Moreover, recent contributions to entertainment theory [5] suggest that media users
actively „work‟ on their enjoyment experience, for instance, by suspending disbelief
in an implausible drama plot for the purpose of maintaining a suspenseful movie
experience. Because the link between player performance and game enjoyment is
established through players‟ own evaluation of their performance (which uses
perceived task difficulty and self-expectations as indicators), it is possible that players
use the inherent complexity of the task-performance-satisfaction-fun connex
instrumentally to preserve a maximum enjoyment even if they do not perform
optimally in the game. For instance, a player who fails to kill a monster in a first-
person-shooter may attribute his failure to an „unfair‟ game setting that rendered the
task extremely difficult or „impossible to do‟. By justifying one‟s own failure through
external conditions (i.e., the game was unfair), the negative impact of the failure on
self-esteem and the accompanying frustration and loss of fun can be buffered. So
blaming the game for being unfairly difficult may be a strategy to maintain game
enjoyment in spite of underperformance. Vice versa, a player who surprisingly
manages to accomplish a really difficult game task may not acknowledge that luck
was responsible for this success, but rather assign the great victory to his own skill,
thus „creating‟ a reason to be proud, with accompanying positive emotions and
performance-based game enjoyment. In other words, players‟ evaluation of their own
performance, may be biased instrumentally by players in order to maximize fun given
successful game events or to preserve as much enjoyment as possible in the case of
failures in gameplay. More failure may thus not necessarily lead to less enjoyment,
and more success will not automatically facilitate more enjoyment.
The theoretical elaboration so far suggests that there is considerable variability in
player responses to a specified game difficulty level. In order to explore the complex
relationship between game difficulty, player performance and satisfaction as well as
game enjoyment in more detail and to gain empirical evidence for a more accurate
model of performance-based game enjoyment, the following research questions were
derived.

RQ1. How does video game difficulty affect satisfaction with one‟s own performance
in expert players?
RQ2. How does video game difficulty affect game enjoyment in expert players?
RQ3. Is the effect of game difficulty on satisfaction stronger than its effect on game
enjoyment?
RQ3 explicitly addresses players‟ (possible) instrumental interpretation of their
own performance: If players actively shape their entertainment experience, they will
protect it against threats from underperformance and according frustration, and will
also derive more enjoyment from mastering easy tasks than it would be appropriate
from an „objective‟ viewpoint (i.e., as a fair estimate of task difficulty within the
attribution process would suggest). Thus, game difficulty may affect game enjoyment
to a smaller degree than it affects player satisfaction: Players may not want
satisfaction to dominate their fun and thus actively work against such an influence.
3 Method
To answer the research questions, an experiment with the first person shooter
(FPS) “Unreal Tournament 2” © was conducted. Overall, 74 voluntary male
university students aged between 18 and 32 years (M = 21.84, SD = 2.73) participated
in the study. All participants said that they played at least “sometimes” computer
games, and they all had at least “some” experience with FPS. Before the students
were invited to the laboratory, they rated their FPS expertise on a 10-point-scale (with
“1” meaning being a novice with almost no experience, “10” meaning being an
absolute expert). Only individuals who rated themselves at “5” or higher were asked
to participate in the study. The reason for this limitation of access was to focus on
game experts (see previous section). Individuals with sufficient FPS experience were
then randomly assigned to play a “duel mode” map of “Unreal Tournament 2” with
either “easy”, “medium” or “very hard” difficulty settings. In the “easy” condition, it
was almost impossible that the player character would get hurt or died, and enemies
were very easy to kill. The version with medium difficulty was supposed to provide
the players with some success and the feeling of competence while a significant level
of challenge was present. At last, the very difficult level was virtually impossible to
win. Players necessarily got killed several times in this condition – independent from
their skills. Everything aside of the difficulty level – appearance of the enemies to be
dueled, map and geographical structure of the game environment etc. – was held
constant across difficulty conditions. Consequently, experimental groups were
confronted with systematically varying levels of game difficulty.
The participants were individually invited to a quiet room with controlled lighting
conditions and were asked to sign a letter of consent to participate in the subsequent
procedure. Before playing, some reaction time data were collected that are not
relevant to the present analysis. Consequently, participants played their FPS level for
10 minutes. They were then requested to complete another reaction-time task and

were handed a questionnaire afterwards. Players‟ objective performance was recorded
from game statistics; for this purpose, the number of enemies killed within the 10
minute play time (“kills”) and the number of times the player character was killed
(“deaths”) were noted by the experimenter.
Analysis of these statistics revealed that the manipulation of difficulty was highly
effective (see table 1). With increasing difficulty, the average number of enemies that
players managed to kill went down sharply, whereas the number of the player
character‟s “deaths” increased substantially. These group differences were highly
significant both for “kills” and “deaths”.
Table 1. Average number of “kills” and own “deaths” across experimental groups of different
game difficulty (n=71).
Enemies killed
Deaths of player character
Game difficulty
Mean
Standard
Deviation
Mean
Standard
Deviation
Easy (n = 25)
24
7.58
1.72
4.52
Moderate (n = 23)
8.96
5.77
15.09
5.59
very difficult (n=23)
2.65
2.81
25.74
4.85
Main effect of game difficulty on enemies killed: F(2,68) = 86.63, p < .0001; η2 = .72.
Main effect of game difficulty on own deaths: F(2,68) = 139.52, p < .0001; η2 = .80.
The post-play questionnaire assessed game enjoyment (with 4 items like “the game
was entertaining”, scaled from “1” meaning “I do not agree at all” to “5” meaning “I
fully agree”, Cronbach‟s α = .93), satisfaction with one‟s own performance (4 items
like “I am proud of my performance in the game”, scaled again from “1” to “5”,
Cronbach‟s α = .80), and perceived difficulty of the game (two items on a 5 p oint
semantic differential such as “the game was… not manageable vs. no challenge”,
Cronbach‟s α = .91). Finally, some additional information (including demographics)
was requested from participants. After responding to the questionnaire, participants
were debriefed and dismissed. Each person received 5 EUR as compensation.
The postplay questions on perceived game difficulty again demonstrated the
effectiveness of the experimental variation in challenge level (table 2). Players rated
the difficulty level of the game level in the way the experimental manipulation had
been designed; this finding also indicates that players were aware of specific objective
difficulty of their game task when evaluating their performance.
4 Results
Results indicate that in general, players evaluate their performance worse if they
have been confronted with harder game difficulty. Obviously, self-assessment of

playing performance was made under the impression of the number of kills (and
deaths) without taking the objective difficulty of the game into account. The many
kills that virtually all players achieved in the easy game version seem to have caused
players to rate their performance as very good in spite of the low challenge that this
game version imposed. Vice versa, the many own deaths in the hardest condition
obviously caused players to evaluate their performance negatively in spite of the
objective difficulty level. However, overall performance ratings did not differ as
strongly between players of the moderate and the hard game version, which suggests
that players begin to „defend‟ or „justify‟ their performance when confronted with
extremely difficult tasks and are not ready to accept any negative gaming outcome as
consequence of their insufficient performance (table 3).
Table 2. Average ratings for game difficulty across experimental groups (n=71).
Perceived difficulty of game
Game difficulty (experimental factor)
Mean
Standard Deviation
Easy (n = 25)
1.58
0.61
Moderate (n = 23)
2.98
0.71
very difficult (n = 23)
3.85
0.75
Main effect of experimental variation in difficulty: F(2,68) = 66.45, p < .0001; η² =
.66
A similar analysis was conducted to examine the impact of game difficulty on
game enjoyment (see table 4). Overall, the easy game version that facilitated the
highest number of success experiences (enemies killed) together with the lowest
number of failures (deaths of the player character) generated the most intensive game
enjoyment. Enjoyment of the harder game versions was lower, with the mean
difference between the moderate and the highly difficult version being smaller than
the difference between the easy and the moderate condition. The greater satisfaction
with one‟s own performance in the easier difficulty conditions thus comes along with
greater enjoyment, which contradicts the assumption elaborated earlier that the
moderate difficulty condition would generate more fun than the easy and the hard
condition. The difference in effect sizes should be noted, as they are important for the
interpretation of findings. While the effect of the manipulated game difficulty was
very strong on actual performance (kills/deaths), it was a little lower but still strong
(² = .66) for the game difficulty rating, again substantially lower for players
satisfaction (² = .46), and went down to a moderate effect size for game enjoyment
(² = .17)
Table 3. Satisfaction with playing performance across experimental groups of
different game difficulty (n=71)

Game difficulty
Mean
Standard
Deviation
Easy (n = 25)
4.06
0.63
Moderate (n = 23)
2.89
0.69
very difficult (n = 23)
2.51
0.81
Main effect of game difficulty level: F(5,65) = 28.17, p < .0001; η² = .46
Table 4. Game enjoyment across experimental groups of different game difficulty
(n=71)
Game difficulty
Mean
Standard
Deviation
Easy (n = 25)
4.29
0.59
Moderate (n = 23)
3.84
0.83
very difficult (n = 23)
3.53
0.86
Main effect of game difficulty level: F(5,65) = 6.49, p < .01; η² = .17
5 Discussion
The experimental variation of game difficulty produced patterns of game
enjoyment (RQ1 and RQ2) that are not fully in line with flow theory [12] and
attribution theory of motivation [6]. From these frameworks, maximum enjoyment
would be predicted for moderate game difficulty, because under such conditions,
players can perceive their own skills and efforts, and attribute occurring success
events to themselves. Too easy and too hard difficulty levels would either lead to
boredom (or no reason to be proud on one‟s performance) or frustration (or the
suspicion that luck helped to overcome the extreme challenges). Interestingly, our
experiment found that players enjoyed the FPS the most when they were given a very
easy condition with many success events (enemies killed) and very few (if any)
failures (own deaths). With increasing difficulty, the number of success events went
down and the number of failures events went up, and both satisfaction with the own
performance and overall game enjoyment were lower than in the easy condition.
Some methodological issues need to be addressed before a conceptual discussion.
First, the results may be caused by a misinterpretation of objective difficulty le vels.
Maybe players found the condition that the investigators labeled “easy” actually
challenging, which would suggest to reconsider if the findings do in fact match with
attribution and flow theory. However, players‟ own ratings of game difficulty were in

line with the experimental manipulation: So players enjoyed the game condition the
most for which they admitted a low difficulty level, which suggests that the according
results are no artifact of unrealistic experimental manipulation. Second, players used
the game in a laboratory setting, which may have reframed participants‟ situation
perception in a way that made performance requirements more salient than in
conventional home use situations. Reports of satisfaction and enjoyment may thus be
biased due to players‟ intentions to impress the experimenter; however, there is no
evidence for the type of bias that may have occurred. Field replications of the study
may illuminate this possible problem in the future.
From a theoretical perspective, there is a need to reconcile the present results with
previous studies who found evidence for flow experiences to be connected to video
game enjoyment [14] and for expert players‟ preference for extremely difficult over
simple tasks [15] – findings that seem to be just the opposite of what the present study
revealed.
Our attempt to integrate these findings with previous research is focused on the
issue of playing time. We suggest that the relationship between game difficulty,
success rate, internal attribution of success, satisfaction with own performance, and
overall game enjoyment changes over the time of using a given game. Our findings
picture the situation of players beginning an unfamiliar game. In this early stage,
enjoyment seems to be driven by „quick success‟, that is, a high number of explicit
positive feedbacks fuels game enjoyment. The fact that this fast stream of success
experiences is caused by low difficulty seems to be „ignored actively‟ by players:
They know that the game „makes it easy‟ for them, but still they have fun with being
successful, although they „objectively‟ did nothing to be really proud of [6]. In turn,
high numbers of failure events reduce game enjoyment in the beginning stage of game
use, although players know that it was an „objectively‟ hard task. So our interpretation
of the found link between game difficulty and game enjoyment is that during the early
stage of getting acquainted to a new game, players heavily depend on visible success
and positive feedback provided by the game. Internal performance evaluations such as
“I was good, but the game was really hard” seem to be less important at this stage.
Without such cognitive rationalization of failure, game enjoyment cannot be
preserved at harder difficulty levels and goes down. Overall, players starting an
unfamiliar game depend on the game‟s feedback of good performance; only if the
game delivers such success feedback (regardless of objective difficulty), the
beginning stage of game use is notably enjoyable.
These patterns of game difficulty and game enjoyment may change with increasing
playing time. After eight more hours of experience with the same game, for instance,
the experimental levels with easy, moderate, and difficult challenge would have
different meanings for players. More importantly, players would hold more
knowledge to judge their own performance independently of the game‟s direct
success feedback. That is, with more knowledge about the game and one‟s own skill
level, players can evaluate their performance without relying on the game‟s
performance feedback alone. Such an experience-based interpretation of performance
may then lead to a shifting pattern of performance and game enjoyment. At a later
playing stage, the fact that there was no challenge involved in achieving many kills
may hold more weight in the generation of game enjoyment. In turn, with more game
experience, players learn to value really good performance in objectively difficult

circumstances. Their experience provides the arguments to defend enjoyment even
when confronted with many failures. Vice versa, their experience also enables the
state of boredom when confronted with too easy tasks – the many wins that are
demonstrably fun at the early playing stage become boring with increased experience.
In sum, we argue that players who are more familiar with a given game would
display the pattern of game difficulty, satisfaction with own performance, and game
enjoyment that is predictable from attribution theory and flow theory (see above):
Game events and knowledge-based interpretation of performance jointly render
moderately difficult tasks more enjoyable and create circumstances of less enjoyment
for too easy and too difficult game tasks. For players who are at the beginning to
using a new game, however, the pattern of game difficulty and enjoyment seems to be
mostly driven by the explicit feedback given by the game, regardless of players‟
internal evaluations of difficulty levels. For players starting a new game, the fun of
gaming thus seems to arise from what the game offers in terms of positive and
negative feedback, whereas for players who are very familiar with a game, their own
interpretation of the game‟s feedback (in terms of “how difficult was that situation?”)
plays a greater role in the generation of game enjoyment. This difference of patterns
between early and later stages of game familiarity is thus the conceptual resolution
that comes out of the present findings that seem to contradict patters reported in
earlier work [6] [12].
In addition to the experimental group mean differences in average satisfaction and
enjoyment, the effect sizes observed in the current experiment deserve conceptual
attention. In the easy game condition, players did not „die‟ very often; in the „very
hard‟ condition, players got killed very frequently. Effect sizes were substantially
lower for players‟ satisfaction with their performance, and still lower for enjoyment
rating, however. This means that strong group differences in objective performance
led to smaller (yet massive) group differences in satisfaction with performance and to
rather small (actually moderate) group differences in game enjoyment.
Concerning RQ3, this pattern of effect sizes indeed suggests that players actively
manage and protect their enjoyment experience. Following arguments from general
entertainment theory [5], we argue that players‟ strategy to maximize game
enjoyment is to switch between different origins of fun instrumentally. Research on
video game enjoyment has identified various sources of fun beyond „performance –
self-esteem – enjoyment‟, for instance, identification with the game character, or
curiosity and surprise (see, for instance, [16] [17]). This means that the game
experience does not fully depend on performance issues but that other factors can also
affect enjoyment.
Our conclusion is thus that players do not allow performance to dominate their
experience, but that they attend also to other „fun factors‟ in order to maintain a
positive play experience. Instrumental use of a game‟s fun factors would then imply
to focus in performance-based enjoyment when performance is good and satisfaction
is high (i.e., pride as dominating mode of game enjoyment) and to focus on other fun
factors when performance is bad and satisfaction is low (i.e., curiosity or suspense as
dominating mode of game enjoyment). Players seem to „intelligently handle‟ the
various types of fun that video games have to offer, and they seem to be able to take
the most fun out of the game even if one important factor (performance) does not
reach optimal values. Thus, the effect size of difficulty on enjoyment was much lower

than the effect size of difficulty on actual performance. However, game difficulty and
performance still have an impact on game enjoyment in spite of players‟ active
management and protection of their fun experience (i.e., players in the high-failure
condition did not report exactly the same enjoyment as players in the medium-failure
condition). We thus argue that players cannot fully override game-based determinants
of fun, but that they can only partly buffer the impact of fun factors on their
experience.
From the perspective of game development, finally, our results provide support to
the common techniques of adapting the way a game operates difficulty management
to the player‟s current stage. For players who have just begun a new game, it is
important to offer a high number of positive events (success experiences) in order to
facilitate enjoyment right from the start. This can be achieved by reducing difficulty
levels to the absolute minimum. Examples from successful games and their „tutorials‟
for beginning players illustrate the viability of this design strategy. With ongoing
game use, the difficulty level should rise stepwise in order to offer more and more
information that players can use for their own individual assessment of their
performance, which will contribute to sustained game enjoyment in later stages of
game use. Another related strategy of game developers is to adapt the difficulty
automatically and dynamically in order to adjust the game to individual performance
and provide „credible‟ success experiences in any stage of game exposure. The
interpretation of the present results implies that such automatic difficulty adaptation
should not attempt to maintain a similar level of success experiences for beginners
and advanced players alike, but rather increase difficulty disproportionately fast with
increasing player performance. This way, players reaching a higher game-related
expertise level would be confronted with a much greater frequency of failure than
early-stage players, and this change of the game‟s requirement profile would better fit
to advanced players‟ expectations (because advanced players expect to run into
extremely heavy game challenges and are resilient against much of the frustration that
comes along with them). The present study thus suggests to compare different logics
of dynamic difficulty adaptation systematically to determine their impact on game
enjoyment over playing time. In general, however, the present results are nicely in
line with what is widely practiced in game design concerning difficulty management
and adaptation.
Finally, the findings on different effect sizes of game difficulty on player
performance versus game enjoyment open interesting theoretical and applied
perspectives on the video game experience. We have suggested an interpretation of
the multi-causality of game enjoyment [13], which has implications for future
research on user experiences in entertainment computing (i.e., to observe multiple
dimensions of enjoyment simultaneously in one study and to compare expert and
novice players) and for game design (i.e., orchestrate different fun factors to support
players‟ instrumental switching between modes of fun). The overall conclusion is thus
that games do facilitate fun because they are task-based environments and allow self-
experiences of competence and pride, but that players are also „smart‟ in construing
their entertainment experience and can handle the different fun factors of video games
instrumentally to maximize their emotional benefits.

References
1. Vorderer, P., Bryant, J. (eds.): Playing video games: Motives, responses, consequences.
Lawrence Erlbaum Associates, Mahwah (2006)
2. Ryan, R., Rigby, C.S., Przybylski, A.: The motivational pull of video games: A self-
determination theory approach. Motivation and Emotion, 30, 347-363 (2006)
3. Vorderer, P., Hartmann, T. & Klimmt, C.: Explaining the enjoyment of playing video
games: The role of competition. In: Marinelli, D. (ed.) ICEC conference proceedings 2003:
Essays on the future of interactive entertainment, pp. 107-120. Carnegie Mellon Press,
Pittsburgh (2006)
4. Jansz, J., Tanis, M.: Appeal of playing online first person shooter games. CyberPsychology
and Behavior, 10, 133-136 (2007)
5. Vorderer, P., Klimmt, C., Ritterfeld, U.: Enjoyment: At the heart of media entertainment.
Communication Theory, 14, 388-408 (2004)
6. Weiner, B.: An attribution theory of achievement motivation and emotion. Psychological
Review, 92, 548-573 (1985)
7. Deci, R. M., Ryan, R. L.: Self-determination theory and the facilitation of intrinsic
motivation, social development, and well-being. American Psychologist, 55, 68-78 (2000)
8. Seery, M. D., Blascovich, J., Weisbuch, M., Vick, S. B.: The relationship between self-
esteem level, self-esteem stability, and cardiovascular reactions to performance feedback.
Journal of Personality and Social Psychology, 87, 133-145 (2004)
9. Higgins, E. T.: Self-discrepancy: A theory relating self and affect. Psychological Review,
94, 319-340 (1987)
10.Klimmt, C.: Dimensions and determinants of the enjoyment of playing digital games: A
three-level model. In Copier, M., Raessens, J. (eds.): Level Up: Digital Games Research
Conference, pp. 246-257. Faculty of Arts, Utrecht University, Utrecht (2003)
11.Csikszentmihalyi, M.: Flow: The psychology of optimal experience. Harper Row; New
York (1990)
12.Sherry, J. L.: Flow and media enjoyment. Communication Theory, 14, 328-347 (2004)
13.Klimmt, C., Hartmann, T.: Effectance, self-efficacy, and the motivation to play video
games. In Vorderer, P., Bryant, J. (eds.): Playing video games: Motives, responses, and
consequences, pp. 132-145. Lawrence Erlbaum Associates, Mahwah (2006)
14.Rheinberg, F., Vollmeyer, R.: Flow-Erleben in einem Computerspiel unter experimentell
variierten Bedingungen [Flow experience in a video game under experimentally varied
conditions]. Zeitschrift für Psychologie, 201, 161-170 (2003)
15.Behr, K.-M., Klimmt, C., Vorderer, P.: Der Zusammenhang zwischen Leistungshandeln und
Unterhaltungserleben im Computerspiel [The relationship between performance and
enjoyment in video games]. In Quandt, T.´, Wimmer, J., Wolling, J. (eds.): Die
Computerspieler – Studien zur Nutzung von Computergames, pp. 225-240. VS Verlag,
Wiesbaden (Germany) (2008)
16.Hefner, D., Klimmt, C., Vorderer, P: Identification with the player character as determinant
of video game enjoyment. In Ma, L., Nakatsu, R., Rauterberg, M. (eds.): International
Conference on Entertainment Computing 2007, LNCS vol. 4740, pp. 39–48. Springer,
Berlin
17.Ravaja, N., Turpeinen, M., Saari, T., Puttonen, S., Keltikangas-Järvinen, L.:. The
psychophysiology of James Bond: Phasic emotional responses to violent video game events.
Emotion, 8, 114-120 (2008)