This research provides an integrated overview of three studies (published previously in English journals) that investigated the objective and perceived reliability of person identifications of older compared to young eyewitnesses. The studies were embedded in an integrative model of eyewitness testimony and its evaluation, which distinguishes an information processing, a metamemory and a judgmental level. In study 1 by Martschuk and Sporer (2018), a meta-analysis integrated 19 studies on the influence of age on face recognition at the information processing level, showing a robust age effect (better performance of younger participants) as well as the presence of an own age bias. Study 2 by Martschuk et al. (2019) shed light on metamemory processes across the lifespan, demonstrating a progressive dissociation between identification performance and confidence as well as decision time across age groups. At the judgmental level, study 3 (Martschuk and Sporer 2020) investigated how mock jurors as fact finders perceive older compared to younger eyewitnesses: Mock jurors were aware of age-related memory changes only to some extent, and did not consider them sufficiently when evaluating the evidence of older eyewitnesses. In sum, the studies demonstrate that memory and metamemory are more fallible in older age than in younger age. Therefore, fact finders, i.e. police investigators, prosecutors, jurors, and judges need to consider an eyewitness’s age, together with decision time and expressed confidence when evaluating identification judgments.