It is widely hypothesized that separate recollection and familiarity processes contribute to recognition memory, The present research measured event-related brain potentials (ERPs) from 128 head locations to identify patterns of brain activity related to recollection and familiarity, In two experiments, subjects performed a recognition memory task requiring discrimination between previously studied words, similar words that changed plurality between study and test, and new words (following Hintzman & Curran, 1994), The FN400 ERP component (300-500 msec) varied with the familiarity of words (new> studied = similar), The parietal component (400-800 msec) was associated with the recollection of plurality (studied> similar = new). Differences in the timing and spatial topography of the FN400 and parietal effects support the view that familiarity and recollection arise from distinct neurocognitive processes.Dual-process theories of memory posit that recognitionjudgments can be based on two different types of information: familiarity and recollection (Brainerd, Reyna, & Kneer, 1995;Hintzman & Curran, 1994;Jacoby, 1991;Mandler, 1980;Yonelinas, 1994). Familiarity is generally thought to reflect an assessment of the global similarity between studied and tested items (e.g., Clark & Gronlund, 1996;Gillund & Shiffrin, 1984; Hintzrnan, 1988;Humphreys, Bain, & Pike, 1989;Murdock, 1982). Recollection entails the retrieval of specific information about studied items, such as physical attributes (Chalfonte & Johnson, 1996;Hintzman & Caulton, 1997;Hintzman & Curran, 1994), associative/contextual information (Clark, 1992;Clark, Hori, & Callan, 1993;Gronlund & Ratcliff, 1989;Humphreys, 1978;Mandler, 1980;Yonelinas, 1997), or other source-specifying information (Hintzman, Caulton, & Levitin, 1998;Jacoby, 1991;M. K. Johnson, Hashtroudi, & Lindsay, 1993).Recognition memory research has addressed two major questions regarding dual-process theories. First, does recognition memory truly depend on more than one process? In some cases, single-process theories can' account for results that are commonly interpreted as requiring two processes (w. Donaldson, 1996;Hirshman & Master, 1997;McClelland & Chappell, 1998;Ratcliff, Van Zandt, & McKoon, 1995;Shiffrin & Steyvers, 1997 each process? Separating the distinct contributions ofrecollection and familiarity to recognition performance is difficult. Jacoby (1991) developed the process dissociation procedure for this purpose. Others have attempted to differentiate separate phenomenological states by asking subjects to introspectively differentiate "remembering" from "knowing" (reviewed by Gardiner & Java, 1993;Rajaram & Roediger, 1997). Others have used responsesignal, speed-accuracy tradeoff (SAT) procedures to separate fast familiarity from slower recollection processes (e.g., Hintzman & Caulton, 1997;Hintzman et al., 1998;Hintzman & Curran, 1994;McElree, Dolan, & Jacoby, 1999). The utility of these approaches has been debated elsewhere: process dissociation Graf, 1995;Graf & Komatsu, 1994;Jacob...