We discuss the hypothesis that structural factors in the fixed-set memory-search task may contribute to the standard linear set-size effect typically observed in this task. We constructed a lexical decision task that contained the same structural factors as the memory-search task. Two structural factors are discussed: (1) the priming of the probe by the memory set, and (2) the confounding of repetitions per item with set size. Our experiment demonstrated that these structural factors playa role in the set-size effect. The implications of these effects for the memorysearch task and for models proposed to account for memory-search data are discussed.One of the most well-known tasks in modem psychology is the memory-search task popularized by Sternberg (1966Sternberg ( , 1967Sternberg ( , 1969Sternberg ( , 1975. In this task, a subject is asked to comparean item, or probe, againsta previously memorized list of items, called the memory set, and to decide whether or not the probe item is a memberof the memory set. A common finding is that reaction time for the decision on the probe item is a linear function of the number of items in the memory set, both for items that are members of the set (positive probes) and for items that are not (negative probes). Sternberg interpreted these data to indicate a serial and exhaustive comparison of the probe item to the items in the memory set: serial because the functions are linear, suggesting a one-by-one comparison process, and exhaustive because the positive and negative functions are parallel, suggestingthat, in both cases, the subject searches the entire memory set.Sincethe introduction of the memory-search task, it has been used widely and in many variations, and many alternatives to Sternberg's interpretations have followed (see Townsend & Ashby, 1983, for a thorough discussion of these). The central theme in this literature has been the idea that the increases seen in reaction time with the additionof itemsto the memoryset (theset-size effect) reflect the operationof some sort of comparison processin which the probe is compared to the items in the memory set.However, there are structural factors in the memorysearch procedure that may account, at least in part, for the set-size effect without assuming a comparison process. In the fixed-set version of the memory-search task, memory set size is manipulated between blocks; subjects This research was supported by Grant BNS 8510365 from the National Science Foundation to Gordon Logan. The authors would like to thank Michael Corballis, Alice F. Healy, James H. Neely, Henry L. Roediger ill, and anonymous reviewers for their comments on earlier drafts of this paper. Requests for reprints may beaddressed to Michael Stadler, who is now at the Department of Psychology, Louisiana State University, Baton Rouge, LA 70803-5501. use one memory set for several trials in succession. This is in contrastto the varied-set procedure, in which subjects are given a new memory set on each trial before presentation of the probe. In the fixed-set p...