CREATING AN ARCHETYPAL CHARACTER BASED EXPERIENCE
Mark Chavez, Ina Conradi, Linyi Liu
Nanyang Technological University, School of Art, Design and Media
This paper addresses our current stage of research and our strategy for addressing the challenges of developing a design driven narrative and methodologies.
Cinematics and Narratives (CaN) is a three year research project whose goal is to significantly explore approaches to contemporary animation. CaN is comprised of three integrated objectives; the first is focused on developing and exploiting real-time animation and content within the context of a visual and narrative design based repository of primitives; the second explores the dynamic of context, exposition and expression, e.g. mixing our design primitives into a new form via cinematic narratives, and the third interfaces this system with an audience in such a way as to enable the system to learn from viewer interaction, where the system automatically further refines the design. CaN is focused on integrating computational intelligent agents designed as character archetypes within a dynamically changeable world created to adapt along a possibility of multiple narratives.
In the context of design, this research investigates the viewer experience as played out by their reaction to the work. Using the notion of archetypes, viewer experience and emphasizing character, our work delves into the notion of presence within subjective space and the viewer's immersion therein. As part of our experience our goal is to develop a method to elicit contrasting reactions from the viewer. By designing imagery that emphasizes contrasts, we mingle the 'attractive' and the 'repulsive'1 in a single immersive experience.
Developing a method to capture the viewer emotion through a gesture based system is our primary goal. This is a system that uses proximity and action to assume the viewer's state of emotional being. We incorporate this assumption back into our character system to further agitate reaction from our viewer.
The research attempts to create a bridge of immersion connecting a physical space to a synthetic. To create a space that immerses the viewer in such a way that they are forced to reconsider their impact on the world around them. As referred to by French Renaissance writer Rabelais when writing about medieval carnival, "not a spectacle seen by the people; they live in it, and everyone participates because its very idea embraces"2.
This research's approach plans to leverage upon the mechanics and genre of magic realism. As Gustavo Adolfo Bécquer a Spanish post-romanticist writer of poetry and short stories writes about grotesque mechanisms in literature, "The first of these is the fact that the mechanisms of imagination depend in good measure on distortion and violence. Secondly, this subjective distortion is linked to a larger chaos in the universe, whose tragic force can be glimpsed by poetic fantasy. Third, the manifestations of that universal chaos are found in the grotesque aspects of Nature and art, and particularly architecture. They are also found in a "second" reality, a nether-world of demonic forms. And finally, both realities often fade away in the presence of the commonplace and disillusioning environment that we all live in, which can also be grotesque."3
The current state of our research focuses on quantifying the meaning of images in a series of artistically created works. This research takes shape in two primary approaches. The first approach, proposed by fellow Principle Investigator Chang Yun-Ke, discusses the usage of imagery, generated by student artists under the tutelage of Ina Conradi and Mark Chavez, to analyze and compare the layman interpretation of the image to an artistic one. The second approach uses visual structure methodologies with beat boards set to different components of visual exposition in a time based medium.
The approach of Perception Study on Emotion-Provoking Images
This study as a primary research as both theoretical and empirical foundation to its subsequent application, aims to explore how abstract images affect and provoke the perceived emotion in viewers.
Various studies in information retrieval realm4 have identified that images can carry certain emotional meaning, which could effectively provoke viewers’ emotion. Hence, the intrinsic relationship between image and emotional meaning can be utilized in the study of emotional information retrieval to give the evidence of cognitive process. For these studies, the perception of emotional meaning is largely based on concrete objects in pictures conveying explicit meaning, such as people’s body parts (feet, hand, face, etc.), indicatively emotional gestures like hug, kiss, etc. And only the viewers’ understanding of images has been observed. Limited research in either information science or cognitive psychology done before is focus on the study of relationship between perceived emotion meanings by artists and viewers’ perceptions upon emotion-provoking images. This study fulfills the gap in current theories of how viewers perceive the emotional meaning of abstract imagery which have been embedded into images through creative intuition and observation by artists.
This multi-discipline collaboration is involved with artists’ creative work, information retrieval science, and media-effect studies. A cognition-affect model5 grounded in social psychological theories will be included as the approach for examining the process that how perception and cognition influence the emotion.
Existing emotional image retrieval studies are mostly based on tests of viewer’s taggingevaluating behavior and quantitative methodologies.6 But few studies have been devoted to testing the dynamic relationship between viewers’ perception and components of image.7 As a foundation research in both fine art and information science, this study will be focus on how the change of imagery influences the viewers’ perception on its emotional meaning.
Classification of Emotion
Everyone knows what emotions refer to, but barely could point out what a consensual concept of emotion exactly is. Basically, an emotion is considered as a mental state associated with feelings, thoughts and behaviors and so on, although to find out a consensual concept of emotion is never an easy task. An emotion state can be stimulated by either extrinsic conditions or intrinsically physical needs (such as hunger activates anger). Previous arguments in emotion studies also have revealed that human emotion can be either the result of cognitive process, or functioning differently from cognitive system.8 Psychologists and other scholars try to approach human emotion via both subjective description of emotional experience9, including some studies made to dissociate the emotion expression from emotion experience10, and physical data such as facial expression, body reaction and so on assumed as relevantly objective measurement which is able to tangibly reflect the subjective estimation of emotion degree11. Scherer pointed out in his article that “defining emotion is a ‘notorious’ problem”.12 Although he tried to highlight the importance of this definitional issue in emotion research, so far not a single scholar or a study claims that they could give out a consensual definition of emotion. Hence, the more important thing in this study seems to find out a reasonable classification of emotion in lieu of struggling with the definition.
Emotions can be roughly divided into three types, positive, negative and neutral, although neutral emotion is automatically wiped from the classification. According to Parrot13, emotions are categorized into a short tree structure constructed by primary (basic), secondary and tertiary groups. The first-tier emotions, which are supposed to be the primary feelings of human beings, comprise love, joy, surprise, anger, sadness and fear. And Paul Ekman14, who is a pioneer in the study of emotion and their relation to facial expressions, also classified emotions into anger, disgust, fear, happiness, sadness, surprise. Although Scherer 15 argues that the classification of emotions could only show the “prototypically patterned types of significant events in the life of organisms”, there are a lot more different categorization methods depending on the theory the scholar adopts.16 We adopt the method of emotion classification from Ekman developed in 1978, that is, emotions incorporate categories of anger, disgust, fear, happiness, sadness, and surprise.
Creation of Abstract Imagery
During the course of study when trying to depict emotions, we used term abstract as equivalent to the expression of “nonfigurative art” or the broader term “nonrepresentational” art. Artists were not only to distort or exaggerate figurative, emphasize or simplify ‘what things look like’ rather had to eliminate figurative depiction all together to create an image of emotion in itself charged with affective energy of expression. The abstract imagery painted was to symbolize the artist’s subjective emotional experience. It is believed that figurative representation limits artist’s capacity to express the actualities of experience, including spiritual experience, and emotions with the kind of intensity or clarity that would reveal its true nature and hidden relations between things. 17 Therefore, the abstract components of design and structure, line and shape, texture and facture, rhythm and interval, light and shadow, color and tone were used independently from naturalist representation, rather for themselves as representation of sensuous experiences of emotion alone.
Meanings are created as these concrete actualities impinge, through senses, upon the receiving imagination. It is in this discord around art that later words will come in play: spoken or written, language answers to image, articulating personal responses that enable the negotiation of shared aspects of meaning. Abstract art has many ways of touching upon things known, but its reference to events in familiar narratives is never literal and unequivocal: it always demands imaginative extrapolation .
The starting point for this study image generating was research into diversity of stylistic methods of abstract art: abstract expressionism, such as Franz Kline, Philip Guston, Hans Hofmann, Barnett Newman, Ad Reinhardt, Robert Motherwell, Arshile Gorky and Joan Mitchell, painterly automatism with free creation of imaginary forms, action painting like Jackson Pollock and Willem de Kooning, expressive mark-making such as Cy Twombly, techisme, Lyrical Abstraction as the Cobra group artists and Japan's Gutai group, Color field painting with highly articulated and psychological use of color, typically as Morris Louis, Mark Rothko, Clyfford Still, Ad Reinhardt, Helen Frankenthaler, Richard Diebenkorn, Sam Francis and Mark Tobey, and last but not least, calligraphy.
The abstract visual language and its linear complexity, energy, suspense and coloristic brilliance were used to express dynamic relations between different emotion categories. They differ in terms of color palette from overly of saturated pinks reds and greens in disgust category, following with extreme contrast of morbid black and white hues in anger, dissolving hues in sadness to the delirium of contrasting blues reds and yellows in happiness. Linear sweep also changes from ornate swellings and curved spilling in fear, articulate linear dribbles and whorls in anger, arabesques and splashes in joy to stains and splotches in disgust.
Figure 1 Jealousy and Mortification
Figure 2 Happiness and Fear
Image Selection and Test Design
More than fifty images were specifically created for each basic emotion category by faculty members and graduates from School of Art, Design and media, NTU, which comprise of abstract paintings and photographs. Fifteen images were manually selected for each group, which were supposed to be the most representative for the given emotion. A three-round selection procedure is employed, which means (1) Initial setting of image pool, (2) Crossed Agreement on image selection and (3) the final decision carried by the main supervisor. There is also a group of relevantly neutral image is selected as control group. Then, participants are invited to use emotion-describing tags (anger, disgust, fear, happiness, sadness, and surprise) to indicate at least five images of each emotion group. To avoid unnecessary work on determining the semantic level of terms participants may use, viewers can only choose tags from prescribed words from these six categories. Furthermore, they can adjust a scroll bar to suggest their emotion intensities, which will be automatically calculated as numeric value in the system to facilitate the importing of database.
Prior including the large sample size, a pilot study with a number of students and staff in School of Art, Design and Media (max. 30) will be conducted as a pre-test. A website dedicated to the formal test will be launched to collect all necessary data, including personal particulars and tagging-scrolling results. The designed and ideal size of valid respondents should be no less than 500 people. The test will open to all participants for about two months.
All test results will be imported to a database then processed by the statistical analysis software SPSS. Coding will be done by authors. Data analysis will be based on result of descriptive statistics (mean and standard deviation) and correlations (Pearson, two-tailed).
However, abstract art also challenges the viewers in a particular way: they are required to look with fresh eyes at pictures that are different. They have to discard old habits, such as the desire to recognize something. Abstract art does not imitate; it represents in a different way. Viewers find no affirmation of themselves in what they see. They are denied the satisfaction of re-encountering a known reality. One of abstract art’s great discoveries is undoubtedly to have made reality’s energetic side visible again. It helps us to comprehend that Nature is just as invisible, immaterial and dynamic as it is tangible, concrete and static. The importance of the in-between is rediscovered. The abstract representation of reality is founded in the towway flow of visual energies.
- The study is unable to investigate the pre-condition of viewers, that is, whenimages were exposed to viewers, their emotional mood before seeing pictures wasunknown and thus ignored. They are assumed to be neutral and calm to judge thepictures. Apparently, some of the audience may not be as calm and neutral as weassumed.
- Viewers are required to tag images with proper emotion-describing words toreflect their feeling of certain image. The tagging process can be considered as away of confirmation of viewers’ emotion, but on the other hand, this also could beconsidered as an evaluation process, that is, emotion-describing words merelyplay the role of markers instead of emotional meanings. Viewers think that thispicture stands for certain emotional meaning doesn’t mean they really feel in thatway, which means that even viewers can tag certain emotion category on a imagedoesn’t necessarily mean the image is able to provoke such a feeling for viewers, 43 Chavez 9since the emotion itself is thought to be largely unconscious. That’s why we onlyset the objective of test as figuring out the perception of viewers on emotionalimage, instead of inspecting viewers’ emotion state activated by the picture. Soprobably, in the further study, we can include the third-person effect theory andobjective capture device (such as facial expression capture) to approach thedistinctions between their judgment on emotional meanings and their real emotionstates.
- Cultural differences are ignored in the study, but psychology scholars havetestified that there are different cognitive styles between Westerners and EastAsians. The prominent distinctions in the attention allocation to visuals refer tothat Westerners tend to pay more attention on focal objects while Easterners caremore about the whole background. So it could be inferred that such distinctionsalso exist in different races of people in our tests.
Mark Chavez is an animator who's work has focused on emergent computer animation including sythetic sculpture with computer toolsets and related forms in popular culture. His research interests are in characterization and storytelling with real-time and prerendered imagery exploring visual and behavioral representation in the animated form; including the creation of intelligent animated forms with a richness in personality and emotive evocative states that are flexible enough to respond to the viewer within a predetermined simulated performance
1 Michael Hollington, Dickens and the Grotesque. Singapore: Taylor & Francis 1984, p. 143.
2 Mikhail Bakhtin, Rabelais and His World (trans. by Helene Iswolsky). Bloomington: Indiana University Press 1984, page 7.
3 Paul Ilie, Bécquer and the Romantic Grotesque. In: PMLA 83, no. 2, 1968, p. 312-331.
4 Rainer Reisenzein, What is a definition of emotion? And are emotions mental-behavioral processes? In: Social Science Information 46. no. 3, 2007, p. 424-428; Klaus R.Scherer, What are emotions? And how can they be measured? In: Social Science Information 44, no.4, 2005, p. 695-729.
5 Yan Jin – Paul Bolls, A Cognition-Emotion Integrated Model of Media Message Processing. Paper presented at the annual meeting of the International Communication Association, Sheraton New York, New York City.
6 James J. Gross – John Oliver, The Dissociation of Emotion Expression from Emotion Experience: A Personality Perspective. In: Personality And Social Psychology Bulletin 26, no.8, 2000, p. 712-726.
7 Paul Ekman, Universals and Cultural Differences in Facial Expression of Emotion. In: J.R. Cole (ed.), Nebraska Symposium on Motivation 19. Lincoln: University of Nebraska Press 1972, p. 207–83; id., An Argument for Basic Emotions. In: Cognition and Emotion 6, 1992, p. 169–200; Alan J. Fridlund - Karen G. Kenworthy, Audience effects in affective imagery: Replication and extension to dysphoric imagery. In: Journal of Nonverbal Behavior 16, no. 3, 1992, p. 191-212; Mel Gooding, Abstraction: An Introduction. In: Mel Gooding, Movements in Modern: Abstract Art. Cambridge: Cambridge University Press 2001, p. 6-7.
8 Richard S. Lazarus, Thoughts on the relations between Emotion and Cognition. In: American Psychologist 37, no.9, 1982, p. 1019-1024.
9 Michael D. Robinson - Gerald L. Clore, Simulation, Scenarios, and Emotional Appraisal: Testing the Convergence of Real and Imagined Reactions to Emotional Stimuli. In: Personality and Social Psychology Bulletin 27, no. 11, 2001, p. 1520-1532.
10 J. Gross, c. d. (note 6).
11 A. Fridlund . K. Kenworthy, c. d. (note 7); Daniel Messinger – Tricia Cassel, Infant Smiling Dynamics and Perceived Positive Emotion. In: Journal of Nonverbal Behavior 32, no. 3, 2008, p. 133-155.
12 K. Scherer, c. d. (note 4).
13 W. Gerrod Parrott, Emotions in Social Psychology. Philadelphia: Psychology Press 2001.
14 Paul Ekman, Facial Signs: Facts, fantasies, and possibillities. In: Thomas A. Sebeok (ed.), Sight, Sound, Sence. Bloomington: Indiana Press 1978, p. 124 -156.
15 K. Scherer, c. d. (note 4).
16 P. Ekman, c. d. (note 7, 1992); Carroll E. Izard, The Face of Emotion. New York: Appleton-Century- Crofts 1971; id., The Psychology of Emotions. New York: Plenum Press 1991; D. Messinger – T. Cassel, c. d. (note 11).
17 M. Gooding, c. d. (note 7).
© Nanyang Technological University, School of Art, Design and Media, Přehlídka animovaného filmu Olomouc, 2009