Psychology ETDs


Doris C. Sahd

Publication Date



This research examined imagery effects in the learning and long-term retention of educable mentally retarded subjects. Between-subject factors included the level of stimulus imagery (low imagery versus high imagery), instructional procedure (standard paired-associate instruction versus imagery instruction), and length of retention interval (immediate, one week, and two weeks). Materials used were two lists of paired-associate nouns, selected so that meaningfulness was held constant while imagery was varied. Imagery was low for the stimulus terms of one list and high for the stimulus terms of the other list. Subjects were 120 non-institutionalized educable mentally retarded children (mean MA of 9.19) from public school EMR classes. Ten subjects were assigned to each of the 12 experimental cells, where they proceeded individually through the four stages of this experiment. The four stages included a pretraining phase, a training phase, a task presentation phase, and a retention measuring phase. In the task presentation phase, word pairs were presented in a single trial, experimenter paced (12 seconds per pair) study-test procedure. In the retention measuring phase, each subject was tested after one of the three different retention intervals, first for cued recall and then for recognition memory. Different random orders of the lists were used for training, cued recall retention, and recognition retention. It was found that recall retention dissipates under these conditions to near zero after one and two weeks. In the immediate retention interval cued recall was significantly facilitated when imagery instruction was provided for high-high (H-H) imagery level word pairs . Error analysis of the recall data, based on extra-experimental intrusions, showed that the number of extra-experimental intrusions increased significantly when the retention interval was extended to one or two weeks, but held relatively constant over these two retention intervals. In addition, a significant interaction of instruction by level of stimulus imagery was found, in which fewer extraexperimental errors were made when imagery instruction was given, provided word pairs were high-high (H-H). The retention pattern found in recognition memory was very similar to that shown in the recall intrusion data. Recognition retention was quite high immediately following training. While retention dissipated significantly between the immediate and one week interval, it held relatively stable between one and two weeks and remained well above the chance expectancy level. The significant instruction by level of stimulus imagery interaction shows that recognition retention is facilitated by imagery instruction when level of stimulus imagery is high. but impaired by instruction when level of stimulus imagery is low. Post-experimental inquiry indicated that no subjects who received standard paired-associate training reported using imaginal mediational techniques. Although 60% of subjects who received i magery training reported using imagery techniques, subjects were differentially influenced by the instructions depending on the nature of the materials presented. Specifically, 77% of imagery trained subjects who learned high-high (H-H) imagery level word pairs reported using imagery to help them learn and remember. Only 43? of imagery trained subjects who learned low-high (L-H) imagery level word pairs reported using imagery.

Degree Name


Level of Degree


Department Name


First Committee Member (Chair)

Sidney Rosenblum

Second Committee Member

Henry Carleton Ellis

Third Committee Member

Thomas Patrick Friden

Fourth Committee Member

Samuel Roll

Fifth Committee Member

John M. Rhodes



Document Type


Included in

Psychology Commons