While assistive robotics (AR) have shown promise in supporting seniors with daily life activities and psycho-social development, evaluation of AR systems present novel challenges. From a technical point of view, reproducing HRI experiments has been problematic due to the lack of protocols, standardization, and benchmarking tools, which ultimately impairs the evaluation of previous experiments. On the other hand, working with seniors with cognitive decline presents a major design challenge for researchers, since communication skills, state of mind and attention of participants is compromised. To address these challenges, this paper presents practical recommendations and a protocol for conducting HRI experiments with seniors with mid cognitive decline (MCI).