Effective Retrieval Cues Research Articles

Coherence of retrieval cues, rather than additivity, determines prior cuing effectiveness in the rat

In previous studies, we have shown that adding a supplementary cue in a prior-cuing treatment may dramatically reduce its initial effectiveness. This series of experiments was aimed at testing the hypothesis that such a lack of additivity among retrieval cues could be due to the combined reactivation of memory attributes related to incompatible behavioral responses. For that purpose, rats were partially trained to reach the lighted arm (light condition; usual training condition) or the darkened arm (dark condition) of a Y-maze in order to avoid an electric foot-shock. The first experiment demonstrates that animals learned the two versions of the task similarly and that an exposure to the light similarly acts as an effective retrieval cue after a 24-h interval. In the second experiment, the effectiveness of exposure to the grid alone and to the light alone as well as to the combined light + grid cue was investigated after a 21-day training-to-test interval in the light condition, where light was related to safety and grid was associated with the shock, and in the dark condition, where both light and grid were related to the shock. The results indicate that whereas the compound light + grid cue was an ineffective retrieval cue in the light condition, it acted as an effective retrieval cue in the dark condition, supporting the hypothesis that coherence between retrieval cues is a necessary condition for effective cuing.

Contextual Variation and Memory Retrieval at Six Months

In 2 experiments we asked whether training in multiple contexts could eliminate context-dependent memory retrieval at 6 months as it does for adults. We found that 24-hour retention was disrupted when infants were trained in one context and tested in another but not when they were trained in multiple contexts prior to testing in a novel context (Experiment 1). After a long delay, however, training in multiple contexts did not facilitate memory retrieval in a novel context: An otherwise effective retrieval cue (the training mobile) did not alleviate forgetting 3 weeks later when it was presented in a novel context (Experiment 2). These findings demonstrate that multiple learning contexts can override the debilitating effects of an altered context on memory retrieval at 6 months, but only over the short term. The resistance of older memories to retrieval in novel contexts after long periods of disuse may be adaptive, insuring that potentially inappropriate or irrelevant memories will not be expressed.

Marketing with Music

Discusses the use of hit music, parodies of hit music, and originally scored music for products advertised in television commercials. Examines the results of a study undertaken to determine the role of music as a memory factor. Summarizes that music created specifically for a product is a more effective retrieval cue than a parody, which is in turn more effective than the use of original hits.

Contextual gating of memory retrieval.

In two experiments, 3-month-old infants learned to move a crib mobile (the cue) in the presence of a distinctive crib bumper (the context) by operant kicking. In Experiment 1A, infants were trained for 2 days and tested either 1, 3, or 5 days later with one of four same/different cue/context combinations. After all delays, infants tested with the original cue and context exhibited excellent retention, and those tested with a different cue and context exhibited none. Changing the context but not the cue disrupted retention after 3 and 5 days but not after 1 day; in contrast, changing the cue but not the context disrupted retention after all delays. In Experiment 1B, the failure of a contextual change to impair retention after 1 day was replicated. In Experiment 2, three same/different cue/context combinations were used as reminders in a reactivation paradigm, and all infants were tested 1 day later with their original training combination. A change in either the context or the cue significantly impaired the effectiveness of the reminder. These results reveal not only that contextual information is incorporated into the memory representations of very immature infants but also that memory retrieval is highly specific to the context in which an event was originally encoded. This specificity buffers against generalized memory retrieval after long retention intervals. The data are consistent with Reeves and Sperling's 1986 model of attention-gating. The context appears to serve as the initial gate for attention to potentially effective retrieval cues.

Categorization and Memory Retrieval by Three-Month-Olds

84 3-month-olds were tested in 3 studies of the acquisition and long-term retention of category-specific information. Infants who were trained with a perceptibly different member of an alphanumeric category on each of 3 days generalized responding to a novel instance of the original training category but not a novel member of a novel category during a 24-hour novelty test. 2 weeks later, when infants displayed no evidence of remembering their prior training experience, categorization was reinstated if a novel exemplar from the original training category was used as the retrieval cue in a memory reactivation procedure. A novel exemplar from a novel category was not an effective retrieval cue. The effectiveness of the category-specific retrieval cue was a function of its physical similarity to the individual exemplars encountered during training, not testing. The background against which the alphanumeric exemplars were displayed during training was not an effective retrieval cue in either the 24-hour novelty test or the memory reactivation procedure, indicating that all invariant stimulus attributes do not contribute equally as category cues. These data are the first to document retention of category-specific information after extended intervals. A popular account of categorization holds that infants abstract invariant features from individual exemplars and form a schema or distinctive memory representation of these shared features against which subsequent exemplars are compared. The present data provide support for a more parsimonious account of categorization, based on the retrieval of information about individual exemplars, that does not require an assumption of prototype formation.

Memory for spatial location in two-dimensional arrays

Memory for spatial location in a two-dimensional display was studied in two experiments by means of a recognition task. The first experiment investigated the effects of presenting multiple reference points at the same time as sequentially shown to-be-remembered (TBR) locations. In the second experiment, the TBR locations were presented either one by one or four at a time with and without multiple reference points. The reference points tended to decrease the false alarm rate during acquisition, rather than to increase the hit rate, which suggests that they affected the precision rather than the amount of systematic bias in memory for the locations. Similar effects were obtained when the TBR locations were presented four at a time. The reference points were also found to serve as effective retrieval cues in a final recognition test. Reference points may thus facilitate both acquisition and retrieval of information about spatial location.

Optimizing cue effectiveness: Recall of 500 and 600 incidentally learned words.

In three cued recall experiments, prerequisites to optimal memory performance of large amounts of verbal materials were examined. Practically perfect recall of 500 and 600 words was obtained when effective retrieval cues were provided at test. The method used to demonstrate this was to instruct the subjects to define their own retrieval cues by generating properties or features to each word presented. At an unexpected recall test, these self-generated properties were presented as cues, and the subjects were instructed to recall the previously presented items. Cue effectiveness was manipulated by varying amount of retrieval information, type of cues, and retention interval. Distinctiveness and compatibility of retrieval cues are proposed as two necessary prerequisites to perfect recall performance. The well-known studies of Shepard (1967), Standing (1973), and Standing, Conezio, and Haber (1970) have shown extremely high recognition-memory performances of large amounts of pictorial information. Shepard (1967) presented subjects with 600 pictures for a few seconds each and then asked them which member of the various test pairs had been among the 600 items. More than 95% of the pictures presented previously were correctly recognized. At the time of publication, this result was found surprising by memory researchers, who were attuned to the rather meager ability of humans to remember nonmeaningful material such as nonsense syllables and isolated digits (Loftus, 1982). Although these studies are spectacular, it should, however, be pointed out that good memory performance per se is not an exceptional phenomenon in the memory literature. Mnemonic devices have been known to western culture for thousands of years (Yates, 1966), and several experiments have demonstrated that the capacity of human memory can be increased by using specific mnemonic systems (see, e.g., Bellezza, 1981) or some

Personality differences in spontaneous personality inferences at encoding.

In this article, a method is described for studying personality differences in the spontaneous encoding of complex information about others, and is illustrated in a study of authoritarianism (F). Two sets of sentences were developed in pretesting. In the first set, events suggested a consensual disposition for each sentence actor to either high F (HF) or low F (LF) subjects. In the second set, events suggested a different consensual disposition to HF and LF subjects. Then 77 HF and LF subjects read both sets of sentences for a memory study. Consensual dispositions and semantic associates to the actors were used to cue recall. For the first set, there was a significant interaction between subject type, sentence type, and cue type. LF dispositions were more effective retrieval cues for LF than for HF subjects. Subjects had very little accurate awareness of having made trait inferences. No significant effects were found for HF cues alone or for the second sentence set. Results indicated that HF and LF subjects differ in their spontaneous social inferences about others and have little awareness of making these inferences. Implications for integrating trait and cognitive approaches to personality are discussed. The interactionist approach to personality emerging from recent debates over personality traits (Magnusson & Endler, 1977) has taken several forms. Most research has been at a macrolevel, and demonstrates that the expression of traits is conditional upon situational variables, producing statistical interactions between traits and situations. Crowne and Marlowe's (1964) work on the need for social approval, and Christie and Geis' (1970) work on Machiavellianism provide numerous examples. More recent macrolevel research indicates that traits predict the individual's active selection and even creation of different situations (dynamic interactions). Snyder's (1979) work on selfmonitoring, Swann's (1983) research on self-esteem and selfverification, and research on self-consciousness (Turner, Scheier, Carver, & Ickes, 1978) describe some of the macrolevel processes involved. At this macrolevel, the smallest unit of analysis is the individual's response or action. More microlevel analyses break the production of behavior itself into multistage processes (e.g.. attention, encoding, storage, retrieval), and examine effects of individual differences at particular stages. These may be measured with personality inventories more typically used to measure traits at the macrolevel, such as anxiety (Sarason, 1980), locus of control (Lef

The effects of distributed training on retention of operant conditioning in human infants

Long-term retention of operant footkicking acquired in the mobile conjugate reinforcement paradigm was assessed as a function of the distribution of training time. In the first study, 3-month-old infants were trained for either one 18-min session or for two 9-min or three 6-min sessions separated by 24-hr intervals. All infants exhibited retention during a test administered immediately after training, but only those trained in a single session continued to perform the conditioned response during cued-recall tests 7 or 14 days later. Infants trained in three sessions showed no evidence of remembering the contingency even after a week. A warm-up decrement, seen in the day-to-day performance of infants in the distributed conditions, was eliminated in the second study by the interpolation of a nonreinforcement period at the outset of daily sessions. This procedure also enhanced long-term retention such that infants trained in three 6-min session now remembered the contingency for 14 days and did not differ from infants who had received a single 18-min session. Whether distributed training facilitates or impairs long-term retention appears to depend on the opportunity for infants to acquire a sufficient number (or kind) of effective retrieval cues during original learning.

Some Tests of the Encoding Specificity and Semantic Integration Hypotheses

The encoding specificity hypothesis of Thomson and Tulving, as stated in 1970, holds that only those cues that have been present at encoding will be effective retrieval cues. In each of eight experiments, recall was generally best when the retrieval context matched the encoding context, thus supporting the hypothesis. The hypothesis was not supported, however, in several experiments in which, under a wide range of conditions, recall was better when strong extralist retrieval cues were present than when no retrieval cues (Experiments 2-7) or when weak extralist cues were present at retrieval (Experiment 8). These results were, however, accommodated by Tulving and Thomson's encoding specificity principle of 1973 combined with the assumption that the strong cues present at retrieval were also present at encoding. Finally, results from several of these experiments indicated, contrary to Baker and Santa's semantic integration hypothesis, that the facilitation of recall deriving from the presence at retrieval of strong extralist cues was independent of the degree to which a target word and its cue word were integrated.

Interference in pigeons’ long-term memory viewed as a retrieval problem

In Experiment 1, three (experimental) groups of pigeons (n=8) acquired a successive wavelength discrimination in Phase 1 and a reversal in Phase 2; then, after a 24-h delay, they received a wavelength generalization test in extinction (Phase 3). For one group (“Context Same”), the same context was present throughout; for both others, a different context was used for Phase 1 and Phase 2. One group received the generalization test in the presence of Context 1, the other in Context 2. The Context Same and Context 2 experimental groups showed “recency,” with all gradients peaking at the reversal S+ value. The Context 1 group yielded several different response patterns but never showed recency, thus revealing context-generated proactive interference. In Experiment 2, eight subjects learned the original discrimination and its reversal in different contexts, and each bird was tested alternately (within a session) in both contexts. Under this condition, the test contexts were effective retrieval cues. In every case, the gradients obtained in each context peaked sharply at the appropriate S+ value. These experiments indicate that conflicting memories may be stored along with their associated contexts such that they can be retrieved by an appropriate manipulation of contextual cues at the time of retention testing.

The advantage of specific terms over general terms as cues for sentence recall: Instantiation or retrieval?

Anderson et al. (1976) found that specific terms, which had not appeared in but fit the context of sentences, were better sentence recall cues than the general terms appearing in the sentences. They interpreted this as indicating that general terms in context were encoded on the basis of appropriate instantiations. Experiment 1, using part of the same material, demonstrated that the advantage of specific cues increased when the general terms were omitted from the to-be-remembered materials, and that other specific terms were also effective cues for the same material. Experiment 2 demonstrated that terms that could not be possible intantiations of any words in the sentences, but that shared common meaning with the sentences as a whole, were as effective retrieval cues as the specific terms. The advantage of specific cues was interpreted as attributable to feature overlap between the specific terms and the to-be-remembered materials rather than to instantiation of the general terms.

The role of the verb in sentence memory

Investigators have frequently found that nouns are more important than verbs in the memory representation of sentences. This pattern of results is incompatible with recent theories of sentence memory that have emphasized the verb and with the results of sentence comprehension studies that have demonstrated the importance of the verb. The four experiments reported here were designed to reexamine the role of the verb in sentence memory. Linguistic variables other than sentence function were held constant in three studies. Verbs were recalled as well and were as effective retrieval cues as nouns in these experiments. The to-be-remembered sentences were presented in the context of paragraphs in the fourth experiment. This reduced the difference in level of recall between subjects, verbs, and objects. The conclusion was drawn that the verb is no less important than sentence nouns in the memory representation.

Encoding specificity: Relation between recall superiority and recognition failure.

The results of four experiments show that (a) recall superiority over recognition is reversed by the use of unrelated word pairs in the study list, and (b) the reversal of recall superiority leaves intact the phenomenon of recognition failure of recallable words. These results extend the generality of encoding specificity and suggest that although recall superiority is a sufficient condition for recognition failure of recallable words it is not a necessary condition. In memory experiments in which people are required to remember the items of a list, the most effective retrieval cues are almost always nominal copies of to-be-remembe red items. If such cues are ineffective, it is unlikely that any other cues will succeed. Under certain experimental conditions, however, copies of list items are not the most effective cues for retrieval—in the presence of other cues, people do recall learned list words that they fail to recognize. This phenomenon of recognition failure of recallable words has been manifested in a number of experiments in which the same procedure, with only minor variations, has been used (e.g., Tulving, 1974; Tulving & Thomson, 1973; Watkins & Tulving, 1975; Wiseman & Tulving, 1975). In these experiments, learners studied a list of pairs of ,associatively related words. One word of each pair was designated as the to-be-remembered (TBR) or target word; the other word was

Effectiveness of Retrieval Cues and the Issue of Independent vs Dependent Storage

The purpose of this experiment was to show that an understanding of retrieval cue effectiveness is basic to the controversy on dependent versus independent storage. Ss were tested in one of 4 conditions. The stimulus lists contained sets of asscciatively related words or unrelated words, and recall was either free or cued. The resulting interaction of list types and recall conditions indicated that, when experimental conditions result in effective retrieval cues, the hypothesis of dependent storage is supported, otherwise the hypothesis of independent storage is supported.