No abstract

Many people regularly multitask while cooking at home. Juggling household chores, reusing limited kitchen utensils, and coordinating overlapping cooking times for multiple recipes can cause frequent task switching and simultaneous task monitoring while cooking. As a result, the cook occasionally loses track of his cooking progress especially when determining which ingredients have already been added, counting multiple scoops of an ingredient, and keeping watch of cooking times. People compensate for these memory slips by devising memory strategies or deferring to memory aids with varying degrees of success. In this paper, we present a novel memory aid for cooks called Cook's Collage. We describe how the system constructs a visual summary of ongoing cooking activity. Then, we report a task simulation study evaluating the effectiveness of Cook's Collage as a memory aid. We argue that a memory aid is helpful only if it is balanced correctly with a complementary memory strategy and only if the accuracy of the memory aid is trusted. Lastly, we discuss how the six design features of the Cook's Collage suggest a general framework for memory aids in the home, which we term dija vu displays.

No abstract

To achieve an appropriate level of reliance on an automated system, the operator must have an accurate system representation such that he/she is aware of the capabilities and limitations of the system. The appropriate use of an automated system can lead to optimal performance by the human-machine dyad. This study investigates the relationship between an accurate system representation and behaviors associated with human-automation interaction (e.g., reliance). Furthermore, age-related effects are also included in the investigation. A cooking memory aid (Cook's Collage) is used as an automated aid that keeps track of the ingredients used in a specific recipe. Participants are asked to interact with the automated device for 5 sessions (each on different days). Tasks are structured to simulate those in a real kitchen. Preliminary results suggest that that there is a different pattern of interaction with the aid as a function of age. Older adults tend to rely on the aid for real-time feedback while younger adults use the aid as a verification tool that they have executed the recipe as desired.

No abstract