Biased health care decision making has been regarded as responsible for inefficient behaviours (for example, the limited insurance purchase). This paper empirically examines two sets of biases in the perception of old age disability and longevity. Particularly, we test for the existence of a so called cumulative bias and, secondly, a so called optimism bias. Findings are suggestive of a significant overestimation of disability risks but no overestimation of longevity expectations is found. Both disability and longevity perceptions appear to exhibit a "cumulative" pattern when mapped over time. Healthier individuals are less likely to perceive high disability and longevity risks whilst female and younger respondents perceive a higher risk of disability in old age at a population level but not at an individual level.