CogniWin: An Integrated Framework to Support Older Adults at Work David Portugal Marios Belk Sten Hanke, CITARD Services Ltd. University of Cyprus and Markus Müllner-Rieder CITARD Services Ltd. Austrian Institute of Technology Miguel Sales Dias João Quintas Christoph Glauser Microsoft Language Dev. Center Pedro Nunes Institute ArgYou AG Eleni Christodoulou George Samaras Mehdi Snene, CITARD Services Ltd. and University of Cyprus and Dimitri Konstantas University of Geneva CITARD Services Ltd. University of Geneva Figure 1. CogniWin work environment. Figure 2. Mockup design of the intelligent mouse. ABSTRACT adults in nowadays fast-emerging technological working Assisting older adults at work is of critical importance in environments. A number of research works exist that aim to nowadays fast-emerging computerized environments. Therefore, support older adults and motivate them to stay for longer active it is paramount to provide support to mitigate age-related and productive [1, 2]. cognitive degradation and to relieve their fear towards technological changes. In this demo paper, we present CogniWin, In this demo paper, we present an integrated framework, named an integrated framework for providing personalized support to CogniWin, which provides personalized support to overcome older adults at work, which aims to achieve the above goals and to eventual age-related memory degradation and gradual decrease of make them feel more positive in prolonging their stay at work. We other cognitive capabilities, and at the same time assists users to present an overall description of the system components and the increase their learning abilities. Thus, it enables them to cope integration architecture, and highlight the benefits of using the better with software application changes in their organizations. system. The system implements an innovative cognitive-based user model, embracing various cognitive characteristics of the older adults. CCS Concepts Moreover, it provides to older adults personalized tips in order to • Human-centered Computing ➝ Human Computer avoid unwanted age-related health situations at their work via a Interaction (HCI) ➝ Interactive Systems and Tools. well-being advisor that assesses measurements provided by an intelligent computer mouse and an eye tracking device, and Keywords considers adult’s personal health-related characteristics stored in Assisted Living; Adaptive Interactive System; Eye-tracker; the system to infer potential negative trends in well-being at work. Computer Mouse; Older Adults. 2. COGNIWIN – COGNITIVE SUPPORT 1. INTRODUCTION FOR OLDER ADULTS AT WORK Many older adults have bright expectations for an active future CogniWin is an integrated framework that blends different and would like to continue managing their work in an office as a technologies to assist the seamless workflow and learning process paid activity. However, seniors working in highly computerized of older adults in computerized working environments, and at the environments are often required to learn new capabilities and same time provide well-being guidance (cf. Figure 1). In a acquire new knowledge, and to adapt their working way to fast nutshell, CogniWin continuously monitors various user emerging, new or upgraded software systems and methods. This interaction and physiological parameters through an in-house requirement, combined with eventual age-related cognitive developed computer mouse, an off-the-shelf eye tracking device degradations (e.g., limited working memory capacity) makes them and a task analysis recorder for contextualizing the users’ feel mentally stressed or tired to stay longer active at their work, interactions. Accordingly, the data from various sources is fused limits their self-confidence and decreases their productivity. and analyzed in real-time, assisting older adults during unpleasant situations (e.g., when feeling stressed, frustrated, etc.) and when In this realm, both the research community and the industry have facing task completion difficulty [3]. CogniWin entails four come to understand the critical importance of assisting older primary services:  Advanced Monitoring: based on an intelligent mouse (cf. Figure 2) and an eye tracking device, CogniWin measures physiological and visual parameters using sensors that enable the extraction of user states and behaviors;  Learning Assistant: provides personalized tips (audio- visual) based on the users’ cognitive characteristics aiding them to achieve goals and improve performance;  Well-being Advisor: provides personalized well-being advice to prevent unwanted age-related health situations effectively, preserving and improving their well-being status Figure 3. A user testing the CogniWin system. in the work environment;  Working Memory Support: anticipates the next task or system integration, it follows a decoupled architecture. This subtask (e.g., moving the mouse pointer to the concerned allows for components to be implemented using different graphical area) in order to reduce cognitive overload during programming languages, being gracefully integrated via a computerized activities where working memory is highly distributed messaging broker, which enables asynchronous solicited. communication between the different components. At its current stage, CogniWin runs on any personal computer (cf. 3. DESCRIPTION OF THE SYSTEM Figure 3) endowed with Microsoft Windows 7, 8 or 10, and the The system architecture is composed at the lower level by: i) an system is capable of identifying and reacting to the following user Intelligent Computer Mouse (CogniMouse [4]); ii) an Eye behaviors when performing a task: normal state, hesitation, Tracking system; and iii) a Knowledge Repository. At this level, drowsiness, vigilance, fatigue, cognitive overload, stress or the system collects anonymously user data from the human- anxiety, and frustration. interface devices while working with the system, and retrieves relevant data stored in the Knowledge Repository such as a priori 4. CONCLUSION AND FUTURE WORK health profiles and specific user capabilities. The intelligent The behaviors recognized by the system are continuously under computer mouse embeds sensors to measure skin conductivity, validation, e.g. as seen in [4], where results revealed links grip force, heart rate, temperature, and inertial measurements between mouse triggering states of user hesitation and user task which are further analyzed aiming to detect user hesitation, completion difficulty. Moreover, two pilot trials at two different frustration and stressful events. The Eye Tracker provides eye end-user institutions have been performed during the project’s gaze point, blinking rate, fixation and saccades rate, and velocity lifecycle. In general, feedback from employees via user as first level parameters which are further processed to get questionnaires and think-aloud protocols has been very positive information about vigilance, hesitation, drowsiness and other regarding the system functionality. They appreciate CogniWin, health and cognitive-related parameters. found it useful (System Usability Scale of 68.3) and felt confident The above information is then fed to the components in the working with the devices, as they did not feel any embarrassment middle-layer level, namely: i) a Contextual Recorder; ii) a Data due to sensors’ usage. Nevertheless, the framework is still not Fusion component; and iii) a Behavior Analysis component. The finalized and we foresee additional upcoming work. In particular, Contextual Recorder is responsible to log the user's keyboard and we intend to improve the timing of advices, enhance the interfaces mouse events, and identify which task, process or services the according to user’s feedback, display the user’s well-being status user is running so as to determine the context according to the so they can monitor their own health parameters, and integrate actions performed. The Data Fusion component combines, filters more precise assistance and training to the user by displaying and synchronizes the outcomes of the lower level modules suitable videos, pictures and text. (CogniMouse and Eye Tracker), and delivers it to the Behavior Analysis module. Also leveraging prior health, personal and 5. ACKNOWLEDGMENTS cognitive characteristics of the specific user and contextual data, This work was partially carried out in the frame of the CogniWin different user behaviors are recognized in real-time in the project (http://www.cogniwin.eu), funded by the EU AAL Joint Behavior Analysis component by means of advanced probabilistic Program (AAL 2013-6-114). reasoning algorithms. 6. REFERENCES Finally, at a higher level stand the user interface components, [1] R. Leung, L. Findlater, J. McGrenere, P. Graf, J. Yang. 2010. which include: i) a Personal Learning Assistant; and ii) a Well- Multi-layered interfaces to improve older adults’ initial being Advisor. In one hand, the role of the Personal Learning learnability of mobile applications. ACM Trans. Access. Assistant is mainly to assist the user in computerized tasks when Comput. 3, 1, Article 1, 30 p. facing difficulties, or at user request. It also provides useful [2] S. Lindsay, D. Jackson, G. Schofield, P. Olivier. 2012. suggestions and helpful tips to provide adaptive support according Engaging older people using participatory design. In Proc. of to the user preferences in order to reduce anxiety or stress. On the ACM CHI '12, 1199-1208. other hand, the Well-being Advisor is triggered when unexpected behavior is detected and provides intervention to prevent [3] S. Hanke, H. Meinedo, D. Portugal, M. Belk, J. Quintas, E. unwanted age-related health situations resulting from user’s Christodoulou, M. Sili, M. S. Dias, G. Samaras. 2015. uncomfortable symptoms. Examples of interventions include CogniWin - A virtual assistance system for older adults at promoting work breaks or stress reducing exercises at specific work. In Proc. of HCII '15, 257-268. times to recreate the user’s productivity. [4] M. Belk, D. Portugal, E. Christodoulou, and G. Samaras. An integrated data model considers relevant historical, contextual, 2015. Cognimouse: on detecting users' task completion sensorial and fused data and incorporates the knowledge difficulty through computer mouse interaction. In Proc. of repository, and the user’s preferences. Furthermore, in terms of ACM CHI EA '15, 1019-102.