Investigating behavioural effects of alarm systems with adaptable thresholds on younger and older adults using an alarm based pedestrian assistance system

• Applicant: Dr. Rebecca Wiczorek
• Subject Area: Human Factors, Ergonomics, Human-Machine Systems
• Term: from 2017 to 2019
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 387670237

Alarm systems (AS) for pedestrians could help older people in traffic situations. AS can compensate for age-related deficits with regard to perception and attention by notifying pedestrians of approaching cars. However, people often experience problems in adjusting their behaviour to AS. These problems are especially prominent in older people. Moreover, dynamic changes in the amount of traffic imply changes in the base rate of critical events. These fluctuations cause changes in the behaviour of the AS and therefore complicate the adjustment of peoples behaviour even more. The innovative concept of AS with adaptable threshold represents an alternative to conventional AS with fixed threshold. Instead of adjusting their own behaviour to the system, users can customize the behaviour of the AS and adapt it at any time when the situation changes.The aim of the proposed project is to broaden the knowledge about AS with adaptable threshold with regard to user groups and characteristics of the situation. It will be examined (1) whether AS with adaptable threshold generally lead to more appropriate use and better performance compared to AS with fixed threshold, (2) whether older users benefit from AS with adaptable threshold and (3) whether AS with adaptable thresholds are of greater utility than AS with fixed threshold especially in dynamic situations. To answer these questions, four two studies will be conducted. An AS with adaptable threshold and an AS with fixed threshold are each examined within a dual-task paradigm with younger and older users. A newly developed pedestrian simulation serves as test environment. In each of the experiments the base rate will be varied through changes in traffic. Dependent variables are trust, behaviour and performance.Findings from the studies can generate new knowledge in the field of AS basic research. Therefore they can provide solutions for domain independent problems like the cry wolf effect. It is expected that results can be transferred to other use cases as well. This holds true for AS used in the privet sector as well as in different working domains, as long as the AS are used in situations with changing base rates of critical events.

• DFG Programme: Research Grants