The human body shows an impressive variety of sensory mechanisms to detect and evaluate the exterior conditions of its environment. Some of these senses have been exploited for interaction with complex technical interfaces on different stages of development in the course of a multitude of concepts and applications, as the example of a mobile phone depicts: A visual display serves as a primary output device for accessing services, while customized ringtones stress the user's sense of hearing and a vibration alert makes for a means of tactile display. One sensing ability that is of crucial importance for the human body but has long been dragging out a shadowy existence as a quality of human-computer interaction, is our perception of temperature, and more accurately of temperature change. As several research studies have shown, the human body is able to detect temperature changes both for cold as well as heat by means of two different kinds of receptors, both situated in the dermis layer of the skin. Out of the practical reason that receptor density shows maxima in fingertips, elbows and nose, our initial prototype design will focus on an application that can be touched by the user's hands. Alternatively, we designed a prototype of a wearable device that sends thermal impulses to the user's elbow.

Spatial Navigation

One particularly interesting application field for mobile thermal information displays are devices used for navigation purposes. These tools, usually in the form of GPS-enabled mobile phones or PDAs, require the user's immediate visual attention in order to grasp the current geographic situation along with his position as well as the location of his destination. The cognitive limitations of these devices are obvious since they are used in a similar fashion as traditional paper maps. Compared to it, a mobile device equipped with a thermal display might be able to increase the user's orientation performance by guiding him to his destination instead of forcing him to constantly stop, look and make references from the real world to the map representation on his PDA screen. Thinking in concentric „distance rings“ radiating off the destination location, a simple scenario includes a location-aware device that adjusts its temperature gradually according to the user's proximity from the destination; the hotter the output, the closer he gets. Apart from releasing the user from the duty of looking at a map screen every few minutes to make sure he's still on the right track, this scenario employs a completely new concept of wayfinding. Instead of being directed along a specific, prescribed path across the network (as navigational systems usually do), the user merely receives information about the proximity to a desired location. The choice, which exact way (of which in a city are usually more than one) to take is up to him. Particularly for people unfamiliar with a city's geography, such as tourists, this wayfinding concept does not only serve as a safe way of navigating towards a certain location, but also provides the freedom of choice to explore an unknown territory without the danger of getting lost.

Silent Alert

As an add-on feature for mobile phones or PDAs, a thermal information display could serve as a possible extension for silent alert. In addition to ringtones, display illumination and vibration alert, a temperature-changing output element could prove useful in situations where the user does not want to disturb a certain situation (such as a meeting) or even doesn't want to get interrupted herself. Assuming such a device is able to display four different levels of temperature, incoming calls could be classified by the caller's priority level and thus trigger a corresponding thermal effect: While your boss or your girlfriend will let the phone heat up remarkably, the device will stay cool when your mother tries to call during a meeting.

Technical Layout

We created two prototypes to conduct the user tests on thermal information display. The first prototype, intended to test sensing ability of fingertips, consists of a 10-ohm power resistor with a 0.5 W power input as a heating element. The second prototype uses a Peltier element as the thermal display source. This prototype is designed as a wristband containing the element to wear around the elbow. The Peltier's advantage is the possibility to test on heat as well as cold.

User Testing

In a first cycle we tested the ability to distinguish different levels of temperature. In order to stay within the physical limitations mentioned above the setup provided a heat element that could be heated up to three different levels (approx. 32°C, 37°C and 42°C). The test person was supposed to put her finger on the element and tried to tell the correct temperature level (low, middle or high). Of ten persons tested for ten cycles each, an average of 2.7 errors in ten guessing cycles has been made. Minus an error rate based on the test setup, the final error rate was 25 percent. The fact that the majority of errors occurred within the first half of test cycles for each person we concluded a possible customization effect regarding the sensitivity of test persons.

With a second test setup we tried to examine this issue more precisely on a long-term basis. In this test, that took place over the course of ten days, we examined the probability of a learning effect of thermal change recognition. The test person carried a small device with a brass-made touch field that could heat up randomly to one of five different temperature levels. The person then had to guess which temperature level was in effect, and check the guess by pressing a button on the device. An LED display attached to the device would reveal the actual level number. Testing took place inside of buildings at room temperature. As a result, it turned out that the initially high error rate of about 65 % on an average became gradually smaller, converging to a value around approximately 25 % after a cycle of ten days.

In a third test we focussed on a concrete application scenario for a thermal display device in order to assess its feasibility under real-life conditions. The scenario we chose to explore more profoundly was the application of a mobile navigation device that uses a thermal display as the primary output channel. The test person was supposed to find his way across an unfamiliar city from a starting point to a certain destination. The navigation device he carried assessed the person's walking direction and accordingly produced a output to a touchable thermal display; if he was walking straight towards the destination, the display was set to the highest temperature level, in other cases it cooled down accordingly to the degree of deviation form the target location. The thermal display was able to produce five different levels of temperature. As a result, the user was able to find his way to the destination, although some issues for consideration appeared. While the application of the device in a compass-style fashion gives relatively precise hints on the walking direction, it turned out to be desirable to retrieve information on the distance from the target location. Thus, an alternative version of the thermal display that maps proximity levels instead of deviation levels needs be considered and tested. Furthermore, a combination of both variables, direction and proximity resolved in the same thermal display was discussed. Generally, it can be stated that the more sophisticated the display becomes, the more training is required to get familiar with the device and obtain satisfying results.

Comment on Test Results

It is obvious that thermal display is limited to a few clearly distinguishable states of information. However, within these limitations the user tests proved that it is indeed possible to convey information by means of a display that can change its temperature. Particularly the long-term test showed encouraging results regarding the training effect of a regular use of such a device and the user's ability to familiarize with different display states.

In conclusion, it has been shown that within a given range of approximately 10°C it is possible to employ a display with up to five different temperature levels an unfamiliar user can detect and distinguish from each other. As long-term tests have shown, the sensitivity for thermal variations increases with regular exercise. We believe that skilled user who has been using the device over a longer period of time on a regular basis might be able to distinguish up to six or seven temperature levels. This prospect of accordingly increasing the devices information capacity provides a clear motivation for further research on the topic. One important feature a thermal display needs to contain is a calibration functionality that allows the customization of a device's output qualities (such as display level granularity or minimum and maximum temperature) according to a user's individual abilities and limitations of thermal perception.

Outlook

Despite technical hurdles, from a conceptual point of view the prototype device provided valuable insights into the concrete application of the idea of ambient information display. It showed that interaction between technical devices and their human users do not need to be confined to visual and audible signal channels. Instead, the human body's complex sensory system provides numerous possibilities for the perception of and response to ambient information.

Weiser, M. and Brown, J.S. (1996). Designing Calm Technology. PowerGrid Journal, 1:1, 1996.

Pousman, Z. and Stasko, J. (2006). A taxonomy of ambient information systems: four patterns of design. In Proceedings of the Working Conference on Advanced Visual interfaces (Venezia, Italy, May 23 - 26, 2006). AVI'06. ACM Press, New York, NY, 67-74.

Sines, P. and Das, B. (1999). Peltier Haptic Interface (PHI) for Improved Sensation of Touch in Virtual Environments. In: Virtual Reality. 4: 260-264. London: Springer-Verlag, 1999.

Fujita, H. and Nishimoto, K. (2004). Lovelet: a heartwarming communication tool for intimate people by constantly conveying situation data. In CHI '04 Extended Abstracts on Human Factors in Computing Systems (Vienna, Austria, April 24 - 29, 2004). CHI '04. ACM Press, New York, NY, 1553-1553.

Nam, C.S.; Di, J.; Borsodi, L.W. and Mackay., W. (2005). A Haptic Thermal Interface: Towards Effective Multimodal User Interface Systems. In: Human-Computer Interaction. IASTED-HCI 2005 (Phoenix, AZ, USA, November 14-16, 2005). ACTA Press, Calgary, AB, Canada.

Kron, A. and Schmidt, G (2006). „Multi-Fingered Tactile Feedback from Virtual and Remote Environments“ haptics , p. 16, 2003.

Frey, Martin (2006): CabBoots. Shoes with an integrated Navigation System. In: Simplicity. The Art of Complexity. Ars Electronica, p. 178, 2006.

LapLogic (2003). Mechanisms of Heat Perception Applied to Laptop Computer Users.

Ottensmeyer, M. and Salisbury, J.K. (1997). Hot and Cold Running VR: Adding Thermal Stimuli to the Haptic Experience. Proceedings of The Second PHANToM User's Group Workshop, October 1997.

Jones, L. A. and Berris, M. (2002). The Psychophysics of Temperature Perception and Thermal-Interface Design. In Proceedings of the 10th Symposium on Haptic interfaces For Virtual Environment and Teleoperator Systems (March 24 - 25, 2002). HAPTICS. IEEE Computer Society, Washington, DC, 137.

Ambient Devices, Inc (2006). The Ambient Orb. Ambient Devices . Last download December 16, 2006

Ishii, H.; Wisneski, C.; Brave, S.; Dahley, A.; Gorbet, M.; Ullmer, B.; Yarin, P. (1998). ambientROOM: Integrating Ambient Media with Architectural Space (video). In Conf. Summary of CHI '98, ACM, 1998.