Most previous studies of semantic processing have examined group-level data. We investigated the possibility that there might be individual differences in semantic decision performance even among the standard undergraduate population and that such differences might provide insights into semantic processing. We analyzed the Calgary Semantic Decision Project dataset, which includes concrete/abstract semantic decision responses to thousands of words and also a vocabulary measure for each of 312 participants. Results of our analyses showed that semantic decision responses had good reliability, and that the speed of those responses was related to individual differences as assessed by vocabulary scores and also by diffusion model parameters. That is, semantic decisions were faster for participants with higher vocabulary scores and for participants with steeper drift rates. Further, in their semantic decision responses high vocabulary participants showed more sensitivity to some lexical/semantic predictors and less sensitivity to others. For responses to both concrete and abstract words, high vocabulary participants were more sensitive to word concreteness and less sensitive to word frequency and age of acquisition. For concrete words, high vocabulary participants were also more sensitive to semantic neighborhood similarity. The results suggest that high vocabulary participants are able to more readily access semantic information and are better able to emphasize task-relevant dimensions. In sum, the results are consistent with a dynamic, multidimensional account of semantic processing.