Temporal contiguity in incidentally encoded memories

Abstract

Abstract Thinking of one event often triggers recall of other events experienced nearby in time. This Temporal Contiguity Effect has been extensively documented in laboratory list learning tasks, but its source is debated. Is it due to task-general automatic processes that operate whenever new memories are formed? Or is it due to task-specific encoding strategies that operate only during deliberate rote learning? I test these theories by presenting over 3500 subjects with a surprise free recall test after various incidental encoding tasks. Experiments 1 and 2 show that temporal contiguity is dramatically reduced under incidental encoding. Experiments 3 and 4 show that although the effect is reduced, it is not eliminated—temporal information is encoded incidentally and is used to guide memory search during both free recall and serial recall. These results demonstrate that contiguity is not an artifact of strategy, but the dramatic reduction of the effect also challenges models that posit a strong link between successful memory encoding and contiguity.