The influence of closeness centrality, a network science metric that measures a node’s “importance” in the network (Borgatti, 2005), on visual word recognition was investigated. Closeness centrality represents the inverse of the average number of links that must be traversed from one word to all other words in the orthographic language network—a web-like representation of orthographic word-forms in the mental lexicon where links are placed between orthographic neighbors. Words with high closeness centralities are “close” to other words in the lexicon. Regression analyses of behavioral data from the English Lexicon Project (Balota et al., 2007) revealed that closeness centrality was a significant predictor of speeded naming and lexical decision performance. High closeness centrality words were processed more slowly and less accurately than low closeness centrality words in the naming task, whereas high closeness centrality words were processed more quickly than low closeness centrality words in lexical decision. The present findings demonstrate that the network structure of the orthographic mental lexicon influences visual word recognition performance and have important implications for models of visual word recognition.