Retrieval Cues at the End of Life: A Two‐Factor Model of Terminal Lucidity in Advanced Dementia

Late-stage dementia, chiefly Alzheimer’s disease, is usually regarded as a permanent decline caused by the gradual destruction of brain tissue. Long-standing clinical and family accounts nevertheless describe people with severe dementia who briefly speak clearly, know where they are and recognise their relatives, typically close to the end of life. Known as terminal lucidity when it falls close to death, and as paradoxical lucidity in the broader sense, the phenomenon has seldom been studied in a systematic way. This narrative review, which did not use a systematic search, synthesises the existing case reports, retrospective surveys and related neurophysiological evidence, and evaluates candidate mechanisms, including terminal neurophysiological surges, retrieval-based accounts of memory loss, and inflammatory or neuromodulatory change. It argues that, if confirmed in prospective studies, terminal lucidity would indicate that some late-stage deficits reflect reversible, network-level dysfunction rather than permanent loss of stored information. It proposes, as a hypothesis, a two-factor model in which lucidity arises when a transient permissive brain state coincides with a salient relational retrieval cue, derives testable predictions from that model, and concludes with ethical considerations and a research agenda for standardised, prospective investigation.