The study was designed to assess the impact of arousal, induced by white noise, upon priming in lexical decisions. The subjects classified letter strings as words or nonwords. “Priming effect” refers to the fact that it takes less time to classify “butter” as a word after making a decision about a related word, like “bread,” than after classifying an unrelated one, like “nurse.” Two hypotheses were compared. (1) The debilitation hypothesis stated that noise would impair all lexical decisions, raising classification times. (2) The enhancement hypothesis stated that arousal (noise) biases the sampling of information toward “dominant” sources, increasing the priming effect. Neither hypothesis received support from the data. Instead, the impact of noise was as follows: Targets were slower than their preceding primes with noise, but there was no difference without it. This interaction suggests that noise makes it difficult to traverse lexical memory from one node to another, regardless of the relatedness of target to prime.