Language production depends upon the context in which words are named. Renaming previous items results in facilitation while naming pictures semantically related to previous items causes interference. A computational model (Oppenheim, Dell, & Schwartz, 2010) proposes that both facilitation and interference are the result of using naming events as “learning experiences” to ensure future accuracy. The model successfully simulates naming data from different semantic interference paradigms by implementing a learning mechanism that creates interference and a boosting mechanism that resolves interference. This study tested this model’s assumptions that semantic interference effects in naming are created by learning and resolved by boosting. Findings revealed no relationship between individual performance across semantic interference tasks, and measured learning and boosting abilities did not predict performance. These results suggest that learning and boosting mechanisms do not fully characterize the processes underlying semantic interference when naming.