The role of attention in perception and memory is one of the longest standing debates in cognitive psychology. Extensive evidence suggests that perception and memory

The role of attention in perception and memory is one of the longest standing debates in cognitive psychology. Extensive evidence suggests that perception and memory are attention dependent. For example, Rock and Gutman (1981) asked subjects to study a red outline shape overlaying a green outline shape and to rate how much they liked it. After many trials of aesthetic judgment, the subjects were given a surprise memory test in which they had to sort out the exposed shapes from novel ones. The subjects were able to recognize the shapes they had attended to, but not the ones they had ignored. In another study, Mack and Rock (1998) presented subjects with a cross shape, followed by a mask. The subjects were asked to judge whether the horizontal or the vertical segment of the cross was longer. After they had done this three times, something else, such as a dot, a word, or a shape, was presented near the cross. A substantial proportion of the subjects, when queried later, denied ever seeing the additional stimulus. These data suggest that attention is the gateway to perception and memory. However, recent studies have shown that unattended objects often leave implicit traces that can be revealed indirectly. For example, using Rock and Gutman’s (1998) stimuli, DeSchepper and Treisman (1996) found that ignored shapes produced a negative priming effect: Responses to the ignored shapes were slower when they later became targets. Using Mack and Rock’s (1998) procedure, Moore and Egeth (1997) found that an unnoticed additional stimulus could affect perception of an attended stimulus. For example, when unexpected stimuli were arranged into arrowheads, they could produce the MüllerLyer illusion. Consistent with such findings, Mack and Rock found that one’s own name is often detected in the inattentional blindness procedure. They suggested that attention is not a gateway to perception and memory in general; it is only a gateway to conscious perception and explicit memory. Implicit processes are exempted from attentional limitations. This study is concerned with the relationship between attention and implicit learning, an issue that has not been fully resolved. The role of attention has previously been studied primarily in the serial reaction time (SRT) task (Nissen & Bullemer, 1987). In this task, four positions are shown on the computer screen, each with a corresponding response key. An asterisk appears randomly in one location, and subjects press the corresponding key. Once they have pressed the key, the asterisk jumps to another location, and the subjects must press that key, and so on. Unknown to the subjects, the sequence of locations in which the asterisk appears follows a fixed 10item sequence, such as BCADBCACBD. Many subjects do not notice the repetition, but their reaction times (RTs) are shorter for repeated than for novel sequences as the experiment progresses. When the SRT task is paired with a secondary tone-counting task, presumed to take away attentional resources, some studies have shown a reduction in learning (e.g., Nissen & Bullemer, 1987; Shanks & Channon, 2002), whereas others have not (e.g., Frensch, Wenke, & Rünger, 1999). A number of factors, individually or jointly, determine whether the SRT effect is affected by a secondary task. Sequences in which one locaThis study was supported by the Milton Fund from Harvard University and by NIH MH071788. We thank Sidney Burks for data collection and Anne Hillstrom, Arien Mack, and Dan Schacter for helpful comments and suggestions. Correspondence should be directed to Y. Jiang, Department of Psychology, Harvard University, 33 Kirkland Street, Cambridge, MA 02138 (e-mail: [email protected]). Implicit learning of ignored visual context