The aim of the research presented in this paper is to identify behavior and generic expectations of elderly people (mean=65 years) relevant to Human-Machine Interface (HMI) in the context of health. Our work is based on two industrial cases of health care services. In one case, we conducted user tests on HMI coupled with an eye-tracker to identify one behavior pattern (16 subjects). And in both industrial cases, we realized interviews that reveal generic expectations in terms of information from seniors (16+8 subjects). Then, this double-case approach has enabled us to emphasize the explorer behavior of seniors and expectations “in the periphery of health” with a need for both social and reassuring projection. Based on our results, we propose recommendations for HMI design in our context and which levers (action parameters) for design in other fields.