Psychological Review Mental Imagery in Associative Learning and Memory

items. Paivio and Yuille (1967) compared the two mediation set conditions with a control condition in which 5s were instructed to use rote repetition and found that the imaginal and verbal sets produced much better learning than did the repetition set, but the predicted interaction with concreteness did not occur. That is, the imagery set was no less effective than the verbal even with abstract pairs. Yuille and Paivio (1968) attempted to induce stronger mediation sets by having 5s first practice associating pairs of nouns using images or verbal links, after which they learned another PA list using the technique they had practiced. Both imaginal and verbal sets again facilitated learning relative to a repetition control condition (more so when at least one member of the pair was concrete), but the differential effects predicted for imaginal and verbal sets again failed to appear. Moreover, the effects of the set variable had disappeared by Trial 3 whereas the effects of noun concreteness persisted throughout learning. A possible explanation suggested by mediation report data (Paivio & Yuille, 1967) was that associative strategies are only partly controlled by the experimental sets and that, over trials, 5s increasingly revert to associative habits aroused by the semantic characteristics of the to-be-learned items. They may abandon the use of rote repetition generally, resorting to imagery in the case of concrete pairs and verbal mediators in the case of abstract pairs, and so on. Paivio and Yuille (1969) tested the hypothesis using a trial-by-trial probe of associative strategies as a function of instructional set and noun concreteness. Different groups were given one of four instructional sets (Imagery, Verbal, Repetition, and No Set) and subgroups within each were questioned after 1, 2, or 3 trials concerning the type of strategy they had used to learn each pair. The results confirmed and extended the previous findings in that the verbal and imaginal mediation sets facilitated learning in comparison with the No Set as well as the Repetition condition, and the effects of set had essentially disappeared by Trial 3 whereas concreteness remained highly effective throughout. Some of the expected changes in the pattern of reported mediators over trials also occurred. The 5s given the imagery and verbal mediation sets most often reported having used the corresponding type of mediator to learn the pairs, and this was true for all trials. In the case of the RepetiIMAGERY IN LEARNING AND MEMORY 251 tion set, on the other hand, rote repetition was frequently reported after one learning trial and decreased thereafter despite the fact that the twoand three-trial groups had been reminded before each trial to use repetition. The pattern of reports suggested that verbal mediators to some extent replaced repetition on the second trial whereas imagery was clearly favored at Trial 3. The effects were also qualified by the item attribute. Repetition and verbal mediation reports were relatively little affected by concreteness, although verbal mediators tended to show a greater increase over trials for pairs in which both members are abstract than for other pair types. By contrast, imagery reports were strongly affected by concreteness in a manner closely resembling its effect on learning. The parallel effects on reported imagery and recall scores as a function of concreteness and trials are shown in Figure 3. In addition to the identical ranking of the four types of pairs on both response measures, it can be seen that concreteabstract pairs show the most striking inCease over trials on each dependent variable. The findings consistently point to imagery ,TM a preferred and an effective mediation j strategy in the case of pairs in which at least ;' the stimulus is concrete and high /, whereas [ verbal mediation is less affected by concrete\ ness. In these respects the data accord with the two-process theory of associative meaning and mediation, but the results provide only weak support for the suggestion that imagery set ,Ss resort to verbal mediators in the case of abstract pairs. Such changes could actually have occurred but were not detected by the mediation questionnaire. Alternatively, it may be that some 5"s can make effective use of imagery even in the case of abstract pairs. Support for such an interpretation was obtained by MacDonald (1967) in a study of the retention of imaginal and verbal mediators and recall of PAs. The 5s instructed to use images frequently reported the use of a concrete image to symbolize one or both members of abstract pairs (e.g., "boy scout" for the pair, chance-deed) and the mnemonic effectiveness of such mediators was evidenced by high recall of the pair only when the mediator was correctly recalled 003 ec i •5