Can technology help us spot a false memory? A linguistic analysis of rich false memories of committing crime

Whether or not false memories can be reliably distinguished from true memories carries tremendous legal and therapeutic implications. At the moment the false memory literature is somewhat divided on this issue, with some researchers finding notable differences between true and false memories, and others claiming that true and false memories look and feel the same. If an objective technique were to be able to reliably distinguish between true and false memories this would be incredibly useful for contexts where the reliability of memory is a constant concern. The present study adds to the debate by being the first to compare true emotional memories to lab-generated rich false memories of committing crime (false confessions) and other emotional events. 60 sets of videos, each set depicting the same individual recalling a rich true and a rich false memory, were examined using the linguistic software LIWC (2007). These videos were generated in a previous study (Shaw & Porter, 2015) which implanted the false memories experimentally through suggestive interviews. Results indicate that there is considerable overlap between the linguistic style of true and false accounts, which may result in false memories being linguistically indistinguishable from true memories in non-experimental contexts. The breakdown of specific linguistic characteristics will be examined, the potential usefulness of LIWC in legal contexts will be considered, and directions for future research will be discussed.