The effects of perceived control and design characteristics on evaluations of a home control system: Attitudes towards a working prototype versus a paper concept

Two studies were conducted to investigate users' attitudes toward centralized home control. The first study (live version) evaluated users' attitudes and behaviors after interacting with a prototype of a centralized home control system in a laboratory setting. The second study (paper version) evaluated users' imagined attitudes towards the same prototype used in Experiment 1, after reading a written description and studying: 4 picture of the interface. Both studies employed a 2 x 2 x 2 mixed design, with perceived control (high or low) and organization of devices (generalist or specialist) as the between-subjects factors; and device group membership (entertainment or utility) as the within-subjects factor. Perceived control was partly conveyed by the presence (low) or absence (high) of a text-based interface agent. The attitude measures assessed were trust for the system, anxiety of the user, and performance, acceptance and integration of the system. In the first study, a behavioral measure was used to assess compliance with the system. The devices controlled by the system were six common household items: three entertainment devices (VCR, DVD Player, and CD Player) and three utility devices (Temperature, Lights, and a Security System). Both studies found significant main effects for perceived control, with low perceived control (LPC) leading to significantly lower trust and acceptance and increased anxiety. Both studies also found significant crossover interactions for the between-subjects factors for trust, anxiety, performance and acceptance, with the low-perceived control specialist being rated more positively than the low perceived control generalist, while the high-perceived control generalist was rated more positively than the high-perceived control specialist. There was a significant main effect for version on all measures, with the paper version producing stronger effects than the live version for perceived control, anxiety and trustworthiness; and the live version producing stronger effects for perceived performance and acceptance.