Subjects were taught two eight-term linear orders of the form" A taller than B taller than C ...." They were then asked to choose the "taller" term in all possible pairwise combinations within each series, and reaction time was measured for each pair. In addition, subjects performed a further task in which they judged whether or not two terms were adjacent in the ordering. In subsequent sessions, subjects were told that the "shortest" term on one list was taller than the "tallest" term on the other, so that the two lists were merged into a single 16-term series. They were then required to choose the "taller" term for both within-groups and between-groups pairs. Subjects did not appear to use the initial groupings in performing this task. even when given training on differential categorical codes ("tall" vs. "short") for the two sublists. Rather, subjects in all tasks appeared to represent the items as ordered positions along an internal array, so that comparison times depended largely on the differential discriminability of the item positions. In each task decisions were made more quickly if the terms being compared were near the ends of the ordering, rather than near the middle.The problem of how information about serial order is stored in memory has concerned psychologists from Ebbinghaus (1885Ebbinghaus ( /1964) to the present (for an excellent historical review, see Crowder, 1976, Chapter 12). This classic problem has recently been investigated in a relatively new reaction time paradigm, in which subjects answer questions about an arbitrary ordering of stimuli along a dimension. A typical procedure is to first train the subjects on adjacent pairs of stimuli (e.g., Henry taller than Pete, Pete taller than Bill, Bill taller than Steve, etc.), and then ask questions about all possible pairs (e.g., Pete taller than Steve?). This paradigm began with the study of "three-term series problems" (Burt, 1919;Clark, 1969; DeSoto, London, & Handel, 1965;Huttenlocher, 1968). More recent investigators have used larger series and emphasized different experimental phenomena; however, the primary emphasis of the earlier work, the process by which people make transitive inferences, remains a major concern (Trabasso, 1975 A major finding in the recent work on linear orderings is that during the testing phase, subjects can compare nonadjacent pairs more quickly than they can compare adjacent pairs (potts, 1972 Trabasso & Riley, 1975; Trabasso, Riley, & Wilson, 1975). The advantage of the nonadjacent or "remote" pairs is particularly interesting in that these pairs are not presented during training. Subjects are actually faster to compare items never previously paired (e.g., B > D) than to compare items that were explicitly learned together (e.g., B > C). It seems that subjects do not simply store the adjacent pairs during training and then use them to draw transitive inferences at the time of testing (as an "associative chaining" model might suggest). Rather, people appear to form a unified representation of the entire orderin...

A model for memory scanning is proposed in which the encoded representation of a probe is compared in parallel with encoded representations of each item in the positive set. The within item matches are serial feature by feature comparisons that terminate when either a positive or negative criterion is reached. This model is shown to predict the results of a probe similarity experiment. The serial location of a similarity within an item affects negative reaction times, but the number of items in the positive set to which the probe is similar has no main effect. The model is also shown to yield predictions consonant with existing data on the relation between reaction times and set size and speed-accuracy trade offs. In a memory scanning recognition task a subject must decide as rapidly as possible whether or not a presented test item was included in a previously presented set of items. Typically, a subject is asked to remember a list, s, of digits, letters or words presented visually or auditorily. He may or may not be allowed to rehearse the list. On a test trial he is presented with a test item, d, and asked to decide whether it was in the positive set, s. Usually, reaction time is the dependent variable of interest, and performance is designed to be nearly error free. Sternberg's (1966, 1969) well known exhaustive scanning model for explaining recognition times in memory scanning was based on three major findings from his early experiments. First, recognition times increase linearly as a function of positive set size. Second, the slope of the positive and negative reaction times plotted against set size are about equal. Third, positive recognition times are independent of the serial position of the