Towards a Pattern Language for Affective Systems JAVIER GONZALEZSANCHEZ, Arizona State University MARIA ELENA CHAVEZECHEAGARAY, Arizona State University ROBERT ATKINSON, Arizona State University WINSLOW BURLESON, Arizona State University There is growing interest about personalizing a computer’s responses according to a user’s affective states (emotions). This is particularly useful for learning, health care, and entertainment systems. However, systems able to recognize, understand, and respond to human affect are still designed and developed from scratch and the experience from their implementation is rarely documented, resulting in forcing developers to “reinvent the wheel” each time. This paper presents an approach to record the design knowledge related to affective systems in the form of patterns, aiming to develop a pattern language for them. Categories and Subject Descriptors: D.2.11 [Software Engineering]: Software ArchitecturePatterns; D.2.13 [Software Engineering]: Reusable SoftwareReuse models; H.1.2 [Models and principles]: User/Machine Systems  Human information processing General Terms: Design Additional Key Words and Phrases: Affective systems, emotion recognition, pattern languages, patterns ACM Reference Format: GonzalezSanchez, J., ChavezEcheagaray, M.E., Atkinson, R. and Burleson, W. 2012. Towards a Pattern Language for Affective Systems. 19th Conference on Pattern Languages of Programs (PLoP), Tucson, Arizona, USA (October 2012), 9 pages. 1. INTRODUCTION Affective Systems (AS) have the ability to recognize and appropriately react to human affective states (emotions). AS incorporate hardware devices to measure human physiological signals and machine learning algorithms to infer affective states from gathered measurements. Measured signals include brainwaves, heart rate, skin conductance, and facial expressions (GonzalezSanchez, Christopherson, et al. 2011). Examples of research conducted on creating AS include AS that support learning (Arroyo et al. 2009; Cooper et al. 2009; D'Mello et al. 2010), AS that monitor patients’ health (Chao and Zhiyong 2008), and AS that enhance video gaming experience (Gilleade et al. 2005; Bernays et al. 2012). However, most of this research has been focused on creating a proofofconcept AS to collect data and validate technology approaches; therefore, the knowledge gained over several years of trialanderror attempts to design and implement AS has gotten lost. Systematic disciplined approaches must be devised to document AS design knowledge and to facilitate the use of that knowledge to leverage the complexity and assortment of AS. One such approach is the use of patterns. Patterns describe a problem which occurs over and over again and then describe the core of the solution to that problem in such a way that this solution can be used a million times over (Alexander et al. 1977). Patterns are not conceived but rather are discovered or mined after numerous implementations of the same solution for a given problem, usually by different people. Patterns that are related to each other and collaborate inside the boundaries of an application domain form a pattern language (Alexander et al. 1977). Patterns and pattern languages have been used for various disciplines and offer designexpertisereuse to the corresponding communities; for instance, in the objectoriented software community (Gamma et al. 1994; Buschmann et al. 1996), the hypermedia community (Hypermedia Patterns Repository 2012), the HCI community (HCI Patterns 2012), and the learning community (Iba et al. 2011). We are grateful to Filipe Correia for his support during the writing process of this paper. This research was supported by the Office of Naval Research under Grant N000141010143 awarded to Dr. Robert Atkinson. Author's address: Javier GonzalezSanchez, University Drive and Mill Avenue, Tempe AZ 85287; email: javiergs@asu.edu; Author’s address: MariaElena ChavezEcheagaray, University Drive and Mill Avenue, Tempe AZ 85287; email: helenchavez@asu.edu; Author's address: Robert Atkinson, University Drive and Mill Avenue, Tempe AZ 85287; email: robert.atkinson@asu.edu; Author’s address: Winslow Burleson, University Drive and Mill Avenue, Tempe AZ 85287; email: winslow.burleson@asu.edu Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission. A preliminary version of this paper was presented in a writers' workshop at the 19th Conference on Pattern Languages of Programs (PLoP). PLoP'12, October 19 21, Tucson, Arizona, USA. Copyright 2012 is held by the author(s). ACM 9781XXXXXXXXX