C.Emotion and Body Movement in Game Aesthetics

C.Emotion and Body Movement in Game Aesthetics

In a pragmatist view of aesthetics, emotions play a central role in characterizing the underlying, unifying, satisfying quality of an aesthetic experience. As this aspect has been neglected in aesthetics of interaction in HCI inspired by pragmatist aesthetics, we will introduce it in more philosophical terms. We will then turn to the way it can be referred to research on body movement and player engagement in whole-body games.

According to John Dewey [17], an aesthetic experience is possible in a situation where suspense and crisis, disturbance and disorder offer the opportunity for their resolution, and hence the moment of passage from disturbance to harmony, which is the moment of intense life. Living the enjoyment of a period of harmony, and hence an aesthetic experience, is a temporary savoured sense of culmination – a feeling of wholeness. This is because it is the beginning of a new relation to the environment, that implies the disruption of the achieved equilibrium, and hence a new tension between disorder or disruption and the search for a new harmony, which is the rhythm of organic life [17]. In other words, as the material of an experience has variety and complexity, in the course of an experience it evokes a variety of different emotions and hence a variety of emotional responses. But the variety of emotions does not belong to the overall aesthetic experience. What is enjoyed as aesthetic is not the variety which an experience is made of. Rather, it is the complete experience and hence the qualitative unity of an experience that is aesthetically enjoyed [16]. In philosophical terms, this underlying unifying quality of an experience considered to be aesthetic is immediately felt in perception and identified as aesthetic emotion [16:226]. What is enjoyed as aesthetic is a sense of intrinsic completion of an experience, which is its closure. An aesthetic experience comes into being only because of its closure. The term “closure” does not refer to the ending of an experience. Rather, it refers to a closing together which holds within it a new perspective on the world [16:204]. It is the sense that the parts of an experience fit and are appropriated. This corresponds to aesthetic enjoyment in pragmatist terms.

Research on the role of body movement in a player’s engagement by Bianchi-Berthouze [6] makes it possible to apply Dewey’s rhythmic dynamics of aesthetic experience to the study of aesthetic experience in full-body games. In our paper, we argue that, in order to be able to detect and measure the player’s aesthetic experience in full-body games, the quality of movements need to be analysed, as they can shed light on the motivations (e.g., role-play vs. hard-fun) that drive the player, the type of embodied experience the player is going through, the embodied meanings the player is constructing and the transitions in experience that may be taking place. In this paper, we focus on the affective qualities of body movement because of their constitutive role in aesthetic emotion, and in the development of an aesthetic experience, as we have defined it, according to pragmatist aesthetics and aesthetics of interaction in HCI.

In HCI, the question of how aesthetic emotion can be recognized through body movement has not been tackled so far. Our aim in this study is not to provide such an answer. Rather, we aim to open a perspective on the use of automatic recognition of affective body movement for the measurement of aesthetic experience in a video-game scenario.

In other words, in this study we propose and investigate the possibility of automatically recognizing the variety of affective body expressions as this is the element of Dewey’s rhythmic dynamics of aesthetic experience, which contributes to the constitution of its closure and hence to its emotional unifying felt quality called aesthetic emotion. In the next section we review studies on automatic affective body expression recognition that found the basis for our approach.

D.Automatic Recognition of Affective Body Expressions

Recent literature in psychology and neuroscience (e.g., [46]-[51]) have shown that body expressions are an important affective modality which enables us to discriminate not only the intensity of the emotion expressed but also to discriminate between different affective states, as well as to communicate the action tendency in response to the felt emotion. These findings have led the affective computing community to investigate the possibility of automatically recognizing the affective message conveyed by the body movement and posture of an expresser to an external observer.

The initial works done in this area have been carried out on acted body expressions [52]-[57]. Just to provide an example, an interesting work presented in [52] aimed at detecting emotions from acted body motions. The movement modelled in this study is cyclic knocking arm movement, expressing either basic emotions (i.e., angry, happy, sad) or a neutral state. The performance of this automatic recognition system reached 50% and improved to 81% when taking individual idiosyncrasies into account in the description of the movement. These performances were comparable to human observers’ performances (varying between 59% and 71%) for the same set of stimuli, as discussed in [59]. Unfortunately, the types of expressions that are modelled in these studies are very stereotypical and strongly expressed and they do not reflect the variety and subtleness of everyday natural expressions.

More recently, researchers have started to tackle the problem of recognizing naturalistic and often more subtle expressions in order to create systems that can be applied to real-life situations (for a more complete review see [60]). A study that aims at detecting emotional states from non-acted body expressions in full-body games is presented in [61], [62]. The aim of this study was to recognize the player’s emotional state during replay windows, i.e., when the player is observing and re-appraising the game, rather than during the game itself. As the context is quite static, the system was built to recognize the affective message conveyed by the configuration of static postures rather than by the dynamic of the body expressions. The recognition rates for the automatic system were 60% on average for four affective states (concentrating, defeated, frustrated and triumphant). Their results were comparable with human observers’ level of agreement (i.e., 67% recognition rate) reached for the same set of stimuli.

In our paper, we address the recognition of the emotional content expressed by players while playing the game, i.e., a much more dynamic and complex context than the one addressed in [61], [62]. To this purpose, dynamic body features rather than postural features will be used. A benchmark is created from an analysis of the agreement between human observers in order to evaluate the system. The benchmark and the system are built using a dataset collected from players playing Nintendo Wii tennis games. The body movement of the players is collected during matches and represents the affective expressions that occur between the start and the end of a match point (winning or losing the point). An initial version of this study is presented in [66]. Here we extend that work in two directions. First of all, we set the work in the context of evaluating aesthetic experience in full-body movement games and provide a new perspective to evaluate such experience. From a technical perspective, a more thorough description and evaluation of the system are provided. This includes person-independent evaluation and game-session-based evaluation. The next section presents the method used to collect the dataset and build the benchmark on human observers. Section IV presents the automatic recognition system and its evaluation. Finally, we discuss its possible use towards measuring the player’s aesthetic experience and the factors that need to be taken into consideration.

Directory: pub
pub -> Project Scoring Template Pre-Construction Phase New Core/Branch Campus Projects
pub -> Acm word Template for sig site
pub -> The german unification, 1815-1870
pub -> Baltic Olympiads in Informatics: Challenges for Training Together
pub ->  Preparation of Papers for ieee transactions on medical imaging
pub -> Forthcoming meetings and conferences
pub -> Asilomar Conference on Circuits, Systems & Computers, Pacific Grove, ca, November 1978, pp. 55 58
pub -> Forthcoming meetings and conferences
pub -> Harmonised compatibility and sharing conditions for video pmse in the 7 9 ghz frequency band, taking into account radar use

Download 192.99 Kb.

Share with your friends: