Words are learned in various contexts and over various timescales throughout our lives. The current study explored the role of context in the form of negation in artificial language learning. It was predicted that words trained in an entirely negated context would show lower average correctness in the testing phase than those trained entirely in the affirmative or in the combined contexts. Eighteen artificial nouns were trained using the prefixes “an-” meaning “not the” or “o-” meaning “the” to mark negation. In the testing phase, participants were tested without the prefix on word stems only. Findings indicated words learned solely in the affirmative context led to a higher average correctness while those learned solely in the presence of negation showed the least average correctness in the testing phase.