Interactive recipe instructions: supporting cooks with better designs

In the last decade in HCI there has been a growing interest in using interactive technology to support human-food activities such as meal planning, shopping and cooking. However, despite the principles of UCD, few researchers have made use of empirical research of users’ human-food activities to ground their system designs, few evaluated their systems with users against a control and most have designed systems with the needs of younger people in mind. In the first two studies I investigated the food related activities of older adults using focus groups and a food and shopping diary with post diary interview. I found that older adults’ food related activities were both routinized and flexible. They were knowledgeable about food, how to plan meals, shop and prepare food and did not have need for technological support. The design assumptions of early food-related systems do not generalise to older adults. In the second two studies I investigated how cooks interact with recipe instructions in order to inform the design of interactive recipe systems. In the first cooking study I investigated the effect (1) of shorter recipe steps and (2) of integrating the instructions and information normally located in the ingredient list into the text of recipe steps. I found that cooks do not read or perform the tasks of recipe instructions in sequence. In part this is because of the nature of recipe procedures and in part because they read ahead to understand the narrative of ingredients to help them interpret the current step. Cooks found instructions with ingredient information integrated clearer and easier to read because integrating the information reduces the need for the cook to look up and mentally integrate information from different sources. In the second cooking study I investigated the effects (1) of adding pictures to recipe steps and (2) of presenting an overview of the structure of the recipe procedure. I found that cooks liked pictures with the recipe steps but they did not reduce problems or errors. Cooks preferred recipes with a structural overview because it enabled them to understand the process and to plan and perform tasks in parallel. Results suggest that the overview presentation, by highlighting the sub-goals of recipe instructions, reduced problems due to recipe content experienced by cooks. The implications for design show that future interactive recipe systems should integrate ingredient information into the recipe steps, support cooks to move around the recipe steps non- sequentially, support cooks to perform more than one step in parallel and provide an overview of the procedure with sub-goals.