Recent progress in neural imaging technologies such as positron emission tomography and dense-array recording of event-related potentials has greatly increased the capacity for in vivo measurement of cognitively relevant processing activity within the human brain. Data from these neural imaging technologies can be combined with behavioral data from standard chronometric techniques to enhance computational modeling of human cognition. Applications of neural data to issues of attention and automaticity are illustrated in the domain of visual word recognition, addressing the question "Is word recognition automatic?"