Recent advances in neuroscience have given us unprecedented insight into the neural mechanisms of false memory, showing that artificial memories can be inserted into the memory cells of the hippocampus in a way that is indistinguishable from true memories. However, this alone is not enough to explain how false memories can arise naturally in the course of our daily lives. Cognitive psychology has demonstrated that many instances of false memory, both in the laboratory and the real world, can be attributed to semantic interference. Whereas previous studies have found that a diverse set of regions show some involvement in semantic false memory, none have revealed the nature of the semantic representations underpinning the phenomenon. Here we use fMRI with representational similarity analysis to search for a neural code consistent with semantic false memory. We find clear evidence that false memories emerge from a similarity-based neural code in the temporal pole, a region that has been called the "semantic hub" of the brain. We further show that each individual has a partially unique semantic code within the temporal pole, and this unique code can predict idiosyncratic patterns of memory errors. Finally, we show that the same neural code can also predict variation in true-memory performance, consistent with an adaptive perspective on false memory. Taken together, our findings reveal the underlying structure of neural representations of semantic knowledge, and how this semantic structure can both enhance and distort our memories.false memory | semantic | temporal pole | fMRI | pattern similarity E ach of us has a vast store of semantic knowledge that we apply to incoming sensory data to extract meaning from the world around us. Semantic representations are capable of capturing important structural features of the world at many different levels of abstraction, which allows for rapid and flexible responses to a diverse array of environmental challenges. This preexisting knowledge structure guides ongoing cognition, which usually aids performance, but under some circumstances can lead us into error (1-3). A striking example is the widely studied DRM (Deese, Roediger, and McDermott) false-memory illusion (4, 5). In a typical DRM task, subjects are asked to memorize a set of words such as "snow," "winter," "ice," and "warm." After a delay, subjects will typically falsely remember having seen the semantically related word "cold." It is widely agreed that this memory illusion is driven by the semantic relatedness between words contained in the encoding list (e.g., "snow") and falsely remembered words that were not actually presented (e.g., "cold"). As such, it is thought that each list item automatically, but weakly, activates the semantically related concept (Fig. 1A). This activation leads to memory confusion, either through a cumulative priming of the related lure (5, 6) or the encoding of the semantic overlap as a "gist" memory (3), resulting in a false memory unless the error is detected by some internal monitoring p...