It has been claimed that temporal order information is encoded automatically. This claim has not been uniformly supported by experimentation that has included tests of several criteria suggested for identifying automatic processes. The present experiment extends these tests to a case in which temporal order judgments were obtained in a rehearsal task with massed presentation. In this task, temporal order coding was sensitive to strategy manipulations. In fact, in one condition there was no reliable evidence of any temporal order coding. These results contradict the claim that temporal order coding is automatic.In recent years, there has been a growing interest in the study of automatic processes. One frequently cited example of such processes is the encoding of temporal order information. Zacks (1979, 1984) have argued that people automatically code the order in which events occur. To test whether temporal order information is encoded automatically, several researchers have examined a number of criteria that must be satisfied by a process if it is to be considered automatic. The results of these studies are mixed. Some studies have shown that temporal order information is stored under incidentalleaming conditions (Hintzman & Block, 1971;Proctor & Ambler, 1975). Other studies have shown that temporal order encoding is not enhanced by intention to code temporal information (e.g., Toglia & Kimble, 1976;Zimmerman & Underwood, 1968), nor is it influenced by competing task demands (Zimmerman & Underwood, 1968) or by age (Brown, 1973;Perlmutter, Metzger, Nezworski, & Miller, 1981). Other studies, however, have shown temporal order information to be affected by intention (Jackson & Michon, cited in Michon & Jackson, 1984; NavehBenjamin, 1990), by competing task demands (NavehBenjamin, 1990), and by age (McCormack, 1982; NavehBenjamin, 1990). These contradictory results seem to be the result of the operational definitions of some of the criteria in these studies. The varied definitions make it difficult to reach definite conclusions about the automaticity of temporal order coding (see Naveh-Benjamin, 1990, for elaboration).The major assumption underlying all of the criteria tested is that automatic processes should function at a constant level under all circumstances. Thus, if temporal order coding is an automatic process, we would expect that manipulating the coding strategy would not affect the fidelity of temporal order judgments.I would like to thank Sigalit Rozenzweig and Rinat Shemes for their help in data collection and analysis. Requests for reprints may be addressed to Moshe Naveh-Benjamin, Department of Behavioral Sciences, Ben-Gurion University of the Negev, P.O. Box 653, Beer-Sheva 84120, Israel.
181The purpose of the present experiment was to test whether variation in encoding strategy has an effect on temporal order judgments. To render this a powerful test, two encoding strategies were intentionally confounded. The first strategy was the level of processing used by the subjects while encoding the inf...