The concept of attention plays a central role in theories of memory, a view supported by a long history of findings that attentional state during encoding affects later memory performance (e.g., Broadbent, 1958;Cowan, 1995;Norman, 1969). Traditional research on this topic has focused on explicit memory. More recently, researchers have addressed the relationship between attention and implicit memory, spurred by the well-documented dissociations between implicit and explicit memory (Mulligan, 2003b;Roediger & McDermott, 1993), as well as the possibility that attention during encoding may not be as crucial to implicit memory as it is to explicit memory.Several initial studies of attention and implicit memory reported the rather startling result that dividing attention reduced explicit test performance but left implicit tests unaffected (e.g., Bentin, Kutas, & Hillyard, 1995;Mulligan & Hartman, 1996;Parkin, Reid, & Russo, 1990;Russo & Parkin, 1993; SchmitterEdgecombe, 1996). Other early studies demonstrated that retention of very poorly attended stimuli is more likely to be detected with implicit than explicit memory tests (e.g., Bornstein, Leone, & Galley, 1987;Eich, 1984;Jelicic, Bonke, Wolters, & Phaf, 1992;Mandler, Nakamura, & Van Zandt, 1987;Merikle & Reingold, 1991; cf. Berry, Shanks, & Henson, 2006). Such findings imply that attention plays a larger role in encoding for explicit than implicit memory. Other researchers have gone farther in interpreting these results, proposing that implicit memory is chiefly the result of automatic encoding processes (the automaticity hypothesis, e.g., Aloisi, McKone, & Heubeck, 2004;Bentin et al., 1995;Shallice et al., 1994;Szymanski & MacLeod, 1996;Wolters & Prinsen, 1997). Much subsequent research on this topic took place under the rubric of the transfer-appropriateprocessing (TAP) view of implicit and explicit memory (e.g., Roediger, 1990;Roediger & McDermott, 1993), which provides a more differentiated view based on the distinction between conceptual and perceptual priming. This view suggests that implicit tests drawing heavily on conceptual processes should depend heavily on attention during encoding. However, perceptual priming tasks, sensitive to relatively automatic perceptual encoding processes (Roediger, 1990), should be little affected by manipulations of attention (see Mulligan & Hartman, 1996, for a more detailed development of the TAP predictions). With respect to conceptual implicit memory, these expectations are largely borne out: divided attention typically reduces conceptual priming (see Mulligan & Brown, 2003, for a review).Studies of perceptual implicit tests, however, have produced mixed results. As noted above, some of the initial studies on this topic found no effect of divided attention on perceptual priming despite robust effects on explicit memory (e.g., Mulligan & Hartman, 1996;Russo & Parkin, 1993). For example, in Mulligan and Hartman (1996), study words were presented visually under one of two conditions. In the full attention (FA) condition, th...