Previous research on music-evoked autobiographical memories (MEAMs) has focused on cueing MEAMs in laboratory experiments. Such approaches lack ecological validity, and the use of experimenter- selected (typically pop) music may limit the range of MEAMs that are cued. We present two studies designed to capture natural experiences of MEAMs—an online survey and a diary study—to test the feasibility of these methods and compare our results to previous lab- based approaches. In the survey, a representative sample of UK participants (N=800, quota sampled on age, gender, and income) reported the most recent MEAM they could recall, for comparison to the most recent autobiographical memory they could recall as cued by watching TV. In the diary study, participants (N=31) recorded details of MEAMs as they occurred in daily life for 7 days. In both studies, we captured details about the music (e.g., title, familiarity, listening setting) and the autobiographical memories (e.g., content, age of memory, vividness, emotions). In both studies, MEAMs were cued by a range of musical genres (pop, rock, classical, soundtracks, etc.), most typically featured friends or significant others, elicited a predominance of positive or mixed emotions (happiness, nostalgia), and were rated as more involuntarily than deliberately recalled. In comparison to TV-cued memories, MEAMs were rated as more vivid, of greater life significance, and accompanied by greater reliving and stronger emotional responses (in particular, positive emotions such as happiness and love). This was despite the fact that MEAMs and TV-cued memories did not differ significantly in terms of self-reported recency of recall or age of the memory. These studies represent new methodological approaches for capturing naturally occurring MEAMs. Several results have confirmed previous findings, such as the predominance of positive emotions and uniquely vivid nature of MEAMs, suggesting that lab experiments do capture similar aspects to the everyday MEAM experience.