1 Towards a Model of Information Aesthetics in Information Visualization Andrea Lau and Andrew Vande Moere Key Centre of Design Computing & Cognition, University of Sydney, Australia {andrea andrew}@arch.usyd.edu.au Abstract This paper proposes a model of information aesthetics in the context of information visualization. It addresses the need to acknowledge a recently emerging number of visualization projects that combine information visualization techniques with principles of creative design. The proposed model contributes to a better understanding of information aesthetics as a potentially independent research field within visualization that specifically focuses on the experience of aesthetics, dataset interpretation and interaction. The proposed model is based on analysing existing visualization techniques by their interpretative intent and data mapping inspiration. It reveals information aesthetics as the conceptual link between information visualization and visualization art, and includes the fields of social and ambient visualization. This model is unique in its focus on aesthetics as the artistic influence on the technical implementation and intended purpose of a visualization technique, rather than subjective aesthetic judgments of the visualization outcome. This research provides a framework for understanding aesthetics in visualization, and allows for new design guidelines and reviewing criteria. Keywords -- information aesthetics, information visualization, aesthetics, visualization art. 1. Introduction Information visualization has recently emerged as an independent research field which aims to amplify cognition by developing effective visual metaphors for mapping abstract data [1]. The design of such effective data representations are generally supported by insights from visual cognition and perception research [2], as well as taxonomies which match data types to the most effective mapping technique [3, 4]. Some researchers have suggested that information visualization may be further augmented by engaging in an interdisciplinary discourse with design and art communities, or vice versa, and have proposed that artistic expression can be effectively supported by better understanding existing information visualization techniques [5-7]. Driven by a parallel stream of independent designers and artists, an increasing number of such visualization art (or data art) works have emerged that aim to express the subjective experience of our information society by artistically motivated but data- driven visual forms [8]. However, such works have not yet been readily acknowledged in either the art or visualization community. While information visualization predominantly focuses on effectiveness and functional considerations, it may be neglecting the potentially positive influence of aesthetics on task-oriented measures. Conversely, the reflection of artistic intent in visualization art often disregards functionality, making some works unintentionally incomprehensible. We propose that information aesthetics bridges this apparent gap between functional and artistic intent by focusing on aesthetics as an independent medium that augments information value and task functionality. Aesthetics has been identified as one of the key problems yet to be solved in current information visualization research [9]. Accordingly, this paper proposes a conceptual model of information aesthetics in an aim to better understand its core characteristics, as well as its commonalities and differences with the fields of information visualization and visualization art. By better appreciating its intentions and employed techniques, this research aims to describe how data can be represented in insightful and appealing ways. 2. Background 2.1. Aesthetics & Information Aesthetics Aesthetics has already been discussed as a key factor in several subfields of information visualization. This is reflected in ambient visualization ��� informative displays communicating information in the periphery of attention ��� which explicitly recommends aesthetics as a method to ensure displays remain unobtrusive in the physical settings in which they are placed [10]. Metrics for aesthetics have also been defined in the field of graph drawing, in terms of readability, such as minimising the number of edge crossings or maximising symmetry [11]. In the context of industrial design, the scientific discipline of engineering aesthetics proposes more rigorous empirical methods for evaluating aesthetics. It aims to systematically identify how people���s multiple senses work together to form aesthetic judgements to assess the potential success of products in the marketplace [12]. Research in aesthetics is also a focus in the fields of affective computing [13] and user experience research [14, 15], which aim to develop computational interfaces that react to or provoke human emotions.