Advances in digital capture and storage technologies mean that it is now possible to capture and store one’s entire life experiences in a Human Digital Memory (HDM). However, these vast personal archives are of little benefit if an individual cannot locate and retrieve significant items from them. While potentially offering exciting opportunities to support a user in their activities by providing access to information stored from previous experiences, we believe that the features of HDM datasets present new research challenges for information retrieval which must be addressed if these possibilities are to be realised. Specifically we postulate that effective retrieval from HDMs must exploit the rich sources of context data which can be captured and associated with items stored within them. User’s memories of experiences stored within their memory archive will often be linked to these context features. We suggest how such contextual metadata can be exploited within the retrieval process.