New Conceptual Associative Learning in Amnesia: a Case Study
Total Page:16
File Type:pdf, Size:1020Kb
Journal of Memory and Language 43, 291–315 (2000) doi:10.1006/jmla.2000.2733, available online at http://www.idealibrary.com on New Conceptual Associative Learning in Amnesia: A Case Study Suparna Rajaram State University of New York at Stony Brook and H. Branch Coslett Temple University School of Medicine We report two experiments in this article that were designed to investigate the role of retrieval constraints and interference in implicit learning of new verbal associations in a densely amnesic participant, C.V., who had presumably sustained medial temporal lobe damage secondary to an anoxic episode. In Experiment 1, repeatedly studied novel sentences produced significant priming with SentenceϩFragment retrieval cues that provided maximal perceptual support as well as per- ceptual priming for the single-word targets. However, little learning was observed when no percep- tual cues were provided for the target itself with the Sentenceϩ??? retrieval cues. In Experiment 2, the effects of intraexperimental interference were measured by examining new verbal learning under the Study-Only, Study-Immediate test, Test-Study training conditions. Unlike in the findings reported in prior studies, C.V. showed little learning with the Sentenceϩ??? retrieval cues even under the minimal interference, Study-Only, condition. Together, these results demonstrate that implicit access to novel verbal associations at a level more abstract than their perceptual configurations is not ubiquitously observed in dense amnesia even when the learning conditions are optimized. These results provide a window into the processes that mediate implicit learning of novel verbal associations when the explicit memory contribution is minimized. © 2000 Academic Press In recent years, empirical studies involving examine the role of these variables on the ac- memory-intact as well as memory-impaired (or quisition of new information in a densely am- amnesic) participants have informed theories of nesic individual. The purpose of our investiga- memory functions and processes. Investigations tion is to isolate the role of implicit processes in involving memory-intact participants have the acquisition and retrieval of new verbal in- demonstrated that repetition, interference, and formation. retrieval cues play a large role in modulating The amnesic syndrome is characterized by memory performance. In the present article, we the selective loss of memory for events subse- quent to the onset of amnesia (also known as We are grateful to C.V. and his family for their enthusi- astic and sustained participation in this project. The assis- anterograde amnesia) while other cognitive tance of Kimberly Feldman and Yoko Yahata in stimulus functions such as language, perception, atten- preparation, data collection, and scoring is gratefully ac- tion, reasoning, problem solving, intelligence, knowledged. Special thanks are due to Teresa Blaxton and and short-term memory are relatively intact. A Endel Tulving for sharing their stimuli. This article greatly loss of memory for events immediately preced- benefited from the constructive comments provided by Douglas Nelson, Endel Tulving, and three anonymous re- ing the onset of amnesia, or retrograde amnesia, viewers. Research reported in this manuscript and the prep- is also commonly reported. aration of this manuscript was supported by an NIH grant, The classic studies by Warrington and R29MH57345-01, to Suparna Rajaram and grant Weiskrantz (1968, 1970) demonstrated that de- RO1AG08870 and RO1DC02754 to H. Branch Coslett. spite grossly impaired explicit memory func- Correspondence may be addressed to Suparna Rajaram, Department of Psychology, State University of New York at tions, as those involved in recall and recogni- Stony Brook, Stony Brook, NY 11794-2500. E-mail: tion, implicit memory functions, as those [email protected]. involved in word fragment completion, may be 291 0749-596X/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved. 292 RAJARAM AND COSLETT largely intact in amnesia (Diamond & Rozin, Vaidya et al., 1995; Warrington & Weiskrantz, 1984; Graf & Schacter, 1985). In the word 1982; Winocur & Weiskrantz, 1976). Yet, oth- fragment completion task, participants are ers have reported impairment on conceptual shown a list of words at study (e.g., elephant) priming in amnesia on a similar group of tasks and are later given fragmented versions of the (Blaxton, 1992; Cermak et al., 1998; Keane et studied(_l_p_an_;solution: elephant) and al., 1997). The reasons for these discrepancies nonstudied words, and they are asked to com- are not clear, but one possibility may be the plete the fragments with the first word that differences in procedure and different groups of comes to mind. The advantage in completing amnesic participants employed across these studied fragments over nonstudied fragments studies. constitutes the measure of implicit memory, or The database on the perceptual and concep- priming (Tulving et al., 1982). Such implicit tual priming effects in amnesia has informed memory tasks are largely data-driven such that and refined our theories of intact memory func- changes in modality across study and test re- tions by providing a window into the operation duce priming (Blaxton, 1989; Rajaram & Roe- of implicit memory when the contribution of diger, 1993; Srinivas & Roediger, 1990), but explicit memory is minimized (Blaxton, 1995; meaningful encoding does not necessarily in- Cohen & Squire, 1980; Gabrieli, 1995; Mosco- crease priming relative to surface-level analysis vitch, 1992, 1994a; Roediger et al., in press; of study material (see review by Roediger & Schacter, 1990; Tulving & Schacter, 1990). The McDermott, 1995). studies reviewed thus far reveal both the pro- Priming is also observed in implicit memory cessing and neural requirements for the reacti- tasks when the task relies on conceptual, and vation of learned information under different not perceptual, processes. For example, in a retrieval conditions. Specifically, the percep- conceptual implicit task such as category pro- tual, lexical, and conceptual representations of duction, participants study exemplar names words such as “elephant” exist in the cognitive (e.g., elephant) of various categories. At test, system prior to the experimental context. only the category names are provided (e.g., an- A key question that has concerned research- imals), and participants are asked to write down ers in recent years is the role of implicit memory the first few exemplars of that category that in mediating the learning of new verbal infor- come to mind. This task is considered to be mation. It is obvious that explicit memory or conceptual in nature because the exemplars and conscious awareness plays a vital role in the category names are conceptually related, and no acquisition of new information, as evidenced by perceptual overlap exists between study and test the performance of memory-intact individuals stimuli. Predictably, modality changes do not in the laboratory and in the real world. Whether impair such priming whereas manipulation of implicit memory also supports the acquisition meaningful encoding enhances such priming of new verbal information, and if it does, what (e.g., Blaxton, 1989; Hamann, 1990; Srinivas & the nature of such learning might be, constitute Roediger, 1990). the questions examined in the present article. Both perceptual and conceptual forms of We investigated these questions in a densely priming, or implicit memory, have been exten- amnesic individual because the minimal opera- sively examined in amnesia and found to be tions of explicit memory may help isolate the generally preserved (see Moscovitch et al., implicit processes that mediate new verbal 1993, for a review), although the evidence for learning. preserved conceptual priming is somewhat Recent studies have investigated the role of mixed. Many studies have shown preserved implicit processes in mediating new verbal conceptual priming in amnesia across a variety learning in the memory-intact as well as the of conceptual implicit tasks (Gardner et al., amnesic populations. The latter set of studies is 1973; Hamann, 1990; Keane et al., 1997; directly relevant here, and these studies fall into Schacter, 1985; Shimamura & Squire, 1984; three general categories. In one set of studies, NEW LEARNING IN AMNESIA 293 the acquisition of hitherto unknown words was 1987), and name–face associations (Thoene & examined in the amnesic population. The results Glisky, 1995), the most extensively studied par- of these studies are mixed. Some studies have adigm in this domain involves the learning of shown little, if any, evidence of acquisition of novel word associations. In a typical experiment new vocabulary in the medial temporal lobe of this sort, participants are presented with un- patient H.M. (Gabrieli et al., 1988) or in Kor- related word pairs (window–reason, march– sakoff’s patients (Grossman, 1987). In contrast, shave) and are later presented with intact pairs positive evidence of new learning has been re- (window–rea_____) or recombined pairs ported in amnesics in some studies. For exam- (march–rea_____) during the stem completion ple, there have been reports of progress in the task. The advantage in stem completion priming academic training of a young amnesic girl for intact pairs over recombined pairs provides (Wood et al., 1989), normal acquisition of new the measure of new learning. The evidence with French vocabulary (Hirst et al., 1988),