Ambient intelligence: the next generation of user centeredness

XII.4 July + August 2005
Page: 25
Digital Citation

Exploring feedback and persuasive techniques at the sink


Authors:
Leonardo Bonanni, Ernesto Arroyo, Chia-Hsun Lee, Ted Selker

The physical world is full of interactions that at first don’t seem to need or be able to benefit from a computer interface; what scenarios should we expect for computers in physical world? How can we sense what is needed and what are appropriate ways to communicate to a person? The sink, for example, is an important site for interaction in our daily lives. The way we use the sink can directly affect our safety and hygiene. Our behavior at the faucet has an impact on the consumption of clean water and energy.

Sinks pose a challenge to HCI because they are traditionally hostile environments for both people and machines. It is hard to think of a more electronic- and computer-adverse place for computers than a wet sink. Now that context-awareness, sophisticated sensing, and computation can become ubiquitous, computer-human interaction in real-world settings can have positive impact on us and our environment. Having identified the sink as a site for improving safety, hygiene, conservation, and behavior change we prototyped four functional interfaces directed at demonstrating interaction possibilities at the sink: HeatSink, SeeSink, CleanSink, and Waterbot. These prototypes take advantage of the richness of interaction and behaviors associated with water. They are conceived as devices that could be attached to existing fixtures to provide added benefits, especially in homes, hospitals, and restaurants. Our principal design goal was to make these interfaces context-aware, intuitive, and realistic. An iterative design process supported the evaluation and improvement of each device, while a parallel design process produced a board of solutions all informed by each other.

HeatSink. Modern plumbing serves up clean water as though it were effortless to produce and distribute. The colorless, odorless liquid does not make its energetic and natural value evident to the user. One example of this phenomenon is the tendency to leave water running for minutes at a time—something inconceivable in the days when water was carried to its destination. One reason for running the tap is to ensure that we have achieved the right water temperature. HeatSink is a project that seeks to return a sense of value to water exiting the tap; to provide useful information about the temperature of the water without altering the function of the sink. Colored LEDs mounted around the faucet aerator illuminate the stream of water, making it red when it is hot and blue when it is cold (see Figure 1). This simple interface eliminates the need to "test" the water and wet one’s hands or scald oneself. A more subtle consequence is the added perceived value achieved by layering this high-tech interface on an otherwise mundane plumbing fixture. HeatSink can inform subtle behavior change regarding water use while increasing the functionality of the sink.

SeeSink. One impediment to automatic faucets in bathroom and kitchen sinks is the lack of control over temperature and flow. Nevertheless, automation conserves water and promotes hygiene by allowing for "hands-free" operation, including automatically shutting off. The infrared technology used in these systems has limited sensing ability and cannot account for the variety of scenarios possible in a sink. SeeSink is a project that seeks to combine the advantages of faucet automation with context-aware sensing and actuation to be useful in kitchen sinks. A CCD camera mounted to the faucet serves to interpret a variety of tasks and provide the proper flow and temperature of water automatically (See Figure 2). When a user presents his hands, the sink dispenses warm water for washing. When a user presents vegetables, the sink dispenses cold water. A pasta pot calls for filling with cold water, whereas a dish sponge indicates the need for hot water to wash the dishes. A PC interprets the video stream and dispenses the proper temperature and flow of water through an instantaneous heater and pumps. In order to communicate the temperature of the water to a user, a version of HeatSink is installed in the faucet that colors the water according to its temperature. Because of the multitude of possible scenarios for such a system, a provision exists in the software for training new tasks such as setting custom hand-washing temperatures. SeeSink helps to make the sink more functional while improving hygiene and water consumption.

CleanSink. Whereas SeeSink is optimized for situations where water conservation can be effected at the point of use, conditions exist where conservation of water needs to be discouraged in order to promote thorough hand-washing. Hospitals, restaurants, and industrial clean rooms often need to install invasive systems that monitor hand-washing compliance so that non-compliance can be punished.

Unfortunately these systems do not directly prevent non-compliance (which is estimated at 50 percent in hospitals, for example) [12]. Dirty hands are the primary cause for infection, and are certainly very easy to prevent [6].

CleanSink seeks to motivate critical behavior change by augmenting the role of the sink as part of the larger context in which hand-washing compliance is necessary. Several versions of the system have been prototyped. In its most basic form, the CCD camera used for SeeSink confirms the presence of hands under the stream of water and HeatSink provides a subtle prompt by flashing illumination in the water stream when sufficient time has passed. In a more typical setting, the same system is combined with an RFID reader that logs the identity and compliance to a central database (see Figure 3). More persuasive techniques allow the sink to connect with automation in the space around the sink. In a medical examination room scenario, CleanSink was connected to a relay that controls the room lights so that they only brighten once the staff washes their hands. For an industrial clean room, on the other hand, we have prototyped an electric door lock that impedes access until hands are clean. Used in combination, these techniques can directly impact hand-washing compliance at the point of use with a broad impact on health and safety.

WaterBot. Many of the behaviors that lead to wasting water occur at sinks in the home: people leaving the water running while brushing their teeth, washing dishes, etc. WaterBot is installed on household faucets to motivate people to turn off the tap when the water is not being used. It motivates water conservation by providing "just-in-time" feedback to users (see Figure 4). Positive reinforcement is used to persuade people to save water by giving them positive feedback and reminders while using the sink and when closing the tap. Social Validation is employed to allow sink users to compare their performance to other members of the household. An adaptive interface varies feedback modality so that persuasion is more effective and less annoying. Finally, value-added design is used to increase the perceived value of the water and the device through artful colored illumination.

WaterBot presents feedback using non-obtrusive interaction modalities in the form of visual and auditory reminders. Continuous visual feedback helps users track their water usage while appealing color patterns in the water entertains them through the lifecycle of the system. Positive auditory messages and chimes sound every time the tap is closed to act as positive reinforces for having closed the tap. WaterBot chooses feedback modalities depending on how long water has been running and on the type of interaction with the sink. A water flow sensor allows the system to track water usage, water savings, and open tap duration. Finally, WaterBot allows researchers to evaluate different feedback types, persuasive techniques, and how they interact when placed together. WaterBot relies on sound and light to motivate water savings. Since the bathroom sink is not a common place for multimedia interaction, we took precautions to ensure that the interaction is not perceived as a game that would otherwise reduce water savings. The faucet should be treated as a utilitarian object while the water should be seen as a valuable resource.

Figure 5

Conclusion. The designs presented reveal that digital augmentation of the sink can have tangible results and deserve further exploration. Automation can be replaced by persuasive techniques that operate at increasing cognitive levels instead of taking control away from users. Users can learn to improve their own behavior rather than being left out of the process by technology.

In order to design ubiquitous computers that improve daily life in cooperation with their users, the design of their interaction must be considered at multiple cognitive levels. Any device that seeks to promote behavior change must offer pleasing interaction modalities and aesthetic design. The concern for appearance in this design goes beyond marketing and has direct consequences on the success of the persuasive interface. As computers pervade our built environments, it will become crucial to consider the details of interaction to make them pleasing and useful to improve our daily life.

There are opportunities for using alternative feedback approaches in a natural environment. Even in a constrained environment, such as a sink, a hierarchy of helpful goals for feedback can be considered ranging from safety to captology.

References

1. Aitken, C.K., McMahon, T. A., Wearing, A. J. & Finlayson, B. L. Residential water use: Predicting and reducing consumption. Journal of Applied Social Psychology, 24 (2).1994, 136-158.

2. Arroyo, E., Bonanni, L., Selker, T. "Waterbot: Exploring Feedback and Persuasive Techniques at the Sink," in Proc. CHI `05

3. Aronson, E.O.L., M. The relative effectiveness of models and prompts on energy conservation: A field experiment in a shower room. Journal of Environmental Systems, 12 (3).1982, 219-224.

4. Bonanni, L., Lee, C.H., Selker, T. "Smart Sinks: Real-World Opportunities for Context-Aware Interaction." In Proc. CHI `05.

5. Cialdini, R. The science of persuasion. Scientific American, 2001, 76-81.

6. Centers for Disease Control: www.cdc.gov

7. Energy, U.S.D.o. Residential conservation service evaluation report. Office, U.S.G.P. ed. Hearings before the Committee on Energy and Natural Resources of the United States Senate, Ninety-Eighth Congress, World Resources Institute, Washington, D.C, 1984, Chapter 12.

8. Fogg, B.J. Persuasive Technology: Using Computers to Change what we Think and Do. Morgan Kaufmann, 2002.

9. Grigorovici, D., Russill, C., Pearce, J., & Constantin, C. Persuasive Environments: Consequences of Virtual Environments on Effectiveness of Persuasive Messages. PRESENCE 2003, 6th Annual International Workshop on Presence.2003.

10. Midden, C.J., Meter, J. E., Weenig, M. H., & Zieverink, H. J. Using feedback, reinforcement and information to reduce energy consumption in households: A field-experiment. Journal of Economic Psychology, 3.1983, 65-86.

11. Selker, T. Attentive User Interfaces. in Berkshire Encyclopedia of Human-Computer Interaction, Berkshire Publishing Group, 2004, 51-54.

12. University of California at San Francisco-Stanford University Evidence-based Practice Center. "Making Health Care Safer: A Critical Analysis of Patient Safety Practices." Agency for Healthcare Research and Quality, Contract No. 290-97-0013. [http://www.ahrq.gov/clinic/ptsafety/]

Authors

Leonardo Bonanni
MIT Media Lab
amerigo@media.mit.edu

Ernesto Arroyo
MIT Media Lab
arroyo@media.mit.edu

Chia-Hsun Lee
MIT Media Lab
jackylee@media.mit.edu

Ted Selker
MIT Media Lab
selker@media.mit.edu

About the Authors:

Leonardo Bonanni is a master’s candidate in the Media Lab’s Counter Intelligence group (the MIT kitchen of the future). He received a B.A. in architecture and sculpture from Columbia in 1999 and an M. Arch. from MIT in 2003. His work concentrates on augmenting user experience with products and spaces through digital augmentation. He has worked as a consultant, architect, furniture and product designer in Chicago, Los Angeles, New York, Tokyo, and Boston. He has been awarded prizes in two international design competitions and most recently as a finalist in the Microsoft Student Research Competition. He has published a number of papers on his work at the Media Lab at SIGGRAPH `04, IUI `05 and CHI’05; articles featuring his work have appeared most recently in Metropolis and the Guardian and online in the ACM’s MemberNet. www.media.mit.edu/~amerigo

Ernesto Arroyo is a Ph.D. candidate at MIT Media Lab. He investigates disruption and interruption in human-computer interaction. He creates novel interfaces that affect the interruption process directly, allowing for deliberate control over its results, such as costs and benefits. He is also interested in self-adaptive interfaces that track users’ reactions and adjust in order to maximize effectiveness. He worked on instrumentation research projects at the National Astronomical Observatory of Mexico. Arroyo earned a MS from MIT Media Laboratory in 2002 and a bachelor’s degree in computer science and engineering from the University of Baja California in 1999. www.media.mit.edu/~earroyo

Jackie Chia-Hsun Lee is a Ph.D. candidate in the context-ware computing group at MIT Media Lab. He is interested in designing 3D and augmented reality user interfaces. Currently, he is designing and implementing an intelligent spatial information system for the Kitchen of the Future. He received 3rd place in ACM SIGGRAPH’s Student Research Competition in 2004. Before coming to the Media Lab, he built many digitally-augmented tools for architectural design in Taiwan, where he earned his B.Arch and M.Arch degree. www.media.mit.edu/~jackylee

Ted Selker is an associate professor at the MIT Media and Arts Technology Laboratory, director of the Context Aware Computing Lab, and of the Counter Intelligence Special Interest Group. Prior to joining MIT, Ted directed the User Systems Ergonomics Research lab at the IBM Almaden Research Center, where he became IBM Fellow in 1996. He has also served as a consulting professor at Stanford University, taught at Hampshire College, University of Massachusetts at Amherst, and Brown Universities, and worked at Xerox PARC and Atari Research Labs. Ted is known for the design of the "TrackPoint III" in-keyboard pointing device now found in Compaq, Fujitsu, HP, IBM, Sony, TI, and other computers, for creating the "COACH" adaptive agent that improves user performance (Warp Guides in OS/2), and for the design of the 755CV notebook computer that doubles as an LCD projector. Ted is the author of over 50 patents, and 20 papers in refereed journals and conference proceedings. His inventions have received more than 30 awards from publications like Scientific American, PC Magazine, Business Week, and BYTE. www.media.mit.edu/context

Figures

F1Figure 1. HeatSink illuminates the stream of water according to its temperature, becoming red when hot and blue when cold.

F2Figure 2. SeeSink interprets a variety of tasks being performed by the user to provide useful hands-free control of water temperature and flow.

F3Figure 3. Clean Sink’s indicator (left) RFID reader (middle) and sink (right)

F4Figure 4. WaterBot offers continuous water usage feedback. It does not interfere with tasks performed at a typical sink.

F5Figure 5. Random patterns in color illuminated water act as reinforcers to reward consistent water-saving behavior. Reinforcers also help to keep users interested.

©2005 ACM  1072-5220/05/0700  $5.00

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 and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2005 ACM, Inc.

 

Post Comment


No Comments Found