University students participated in five experiments concerning the effects of unmasked, orthographically similar, primes on visual word recognition in the lexical decision task (LDT) and naming tasks. The modal prime-target stimulus onset asynchrony (SOA) was 350 ms. When primes were words that were orthographic neighbors of the targets, and relative to primes sharing no letters in position with the targets, interference with responses to targets was observed consistently in both tasks. Nonword primes did not consistently interfere except with high-frequency targets in the LDT. Phonological similarity and congruity and incongruity in spelling-sound connections did not account for the effects of orthographic similarity. The results are consistent with competition in lexical processing and suggest shared mechanisms of inhibition for masked and unmasked neighbor primes. (Contains 8 tables and 2 figures.)