JOURNAL OF VERBALLEARNING AND VERBALBEHAVIOR 17, 455--463 (1978) Type i Rehearsal and Recognition ARTHUR GLENBERG AND FREDERICK ADAMS University of Wisconsin-Madison Rote, repetitive Type 1 Rehearsal is defined as the continuous maintenance of information in memory using the minimum cognitive capacity necessary for maintenance. This definition is operationalized in an incidental paradigm where pairs of words are overtly rehearsed 1, 5, or 10 times (maintained for 1.33 to 13.33 seconds). An analysis of the types of errors made on a forced-choice recognition test supported the hypothesis that acoustic-phonemic components of the memory trace, as opposed to semantic and contextual components , are added or strengthened by this rehearsal process. Furthermore, it was observed that co rehearsed words did not effectively cue one another's recall. These results are discussed in relation to the use- fulness of the distinction between Type I and Type II rehearsal. The terms Type I and Type II rehearsal were proposed by Craik and Lockhart (1972) to refer to two theoretically distinct kinds of processing. Type II rehearsal results in deep (semantic) analysis or mnemonic elaboration and is beneficial for long-term performance. Type I rehearsal was characterized as a rote repetition of analyses that have already been completed. This process was supposed to maintain information for immediate perform- ance at a given level of analysis (within a level- of-analysis framework), but not to enhance long-term performance. A similar dichotomy was proposed by Woodward, Bjork, and Jongeward (1973). This article is primarily concerned with the definition of Type I rehearsal, and with the effects of Type I rehearsal on the memory code as measured, primarily, with a recognition procedure. Research on Type I rehearsal has been hampered and roundly criticized (e.g., Nelson, 1977) for lack of an operational definition. Many thanks to Steven Smith for his valuable conver- sations throughout the course of this project. This research was supported in part by Grant MH26643 from N1MH and Grants 180299 and 160076 from the Graduate School, University of Wisconsin-Madison to Arthur Glenberg. Reprint requests should be sent to Arthur Glenberg, Department of Psychology, University of Wisconsin, Madison, Wisconsin 53706. 455 Some investigators (e.g., Dark & Loftus, 1976; Rundus, 1977) have accepted a circular definition: The duration of Type I rehearsal is not related to delayed recall performance. Glenberg, Smith, and Green (1977) proposed a number of distinctions and criteria to begin to clear up the definitional problems. First, the term Type I Rehearsal was proposed as the generic name of many possible low-level, rote, repetitive rehearsal processes. Maintenance rehearsal (Craik & Watkins, 1973) was reserved to refer to a process that maintains information for immediate recall and does not influence delayed memory performance. Primary rehearsal (Woodward et al., 1973) was reserved to refer to a Type 1 process that does affect the memory trace, as, for example, measured with a recognition procedure. Glenberg et al. (1977) also proposed three criteria as a guide toward operationalizing definitions of Type I rehearsal. First, paradigms that elicit Type I rehearsal should control the subjects' processing activity to be reasonably sure that the subjects are not trying to form associations among to-be-remembered (TBR) items, nor are they trying to increase the depth of analysis (Craik & Lockhart, 1972) or elaboration (Craik & Tulving, 1975) of the TBR items. Second, the paradigm 0022-5371/78/0174-0455502.00/0 Copyright~) 1978 by Academic Press, Inc. All rights of reproductionin any form reserved. Printed in Great Britain