Distortions of long-term memory (LTM) in the converging associates task are thought to arise from semantic associative processes and monitoring failures due to degraded verbatim and/or contextual memory. Sensory-based coding is traditionally considered more prevalent than meaning-based coding in short-term memory (STM), whereas the converse is true of LTM, leading to the expectation that false memory phenomena should be less robust in a canonical STM task. These expectations were violated in 2 experiments in which participants were shown lists of 4 semantically related words and were probed immediately following a filled 3- to 4-s retention interval or approximately 20 min later in a surprise recognition test. Corrected false recognition rates, confidence ratings, and Remember/Know judgments reveal similar false memory effects across STM and LTM conditions. These results indicate that compelling false memory illusions can be rapidly instantiated and that, consistent with unitary models of memory, they originate from processes that are not specific to LTM tasks. (Contains 2 tables, 2 figures, and 2 footnotes.)