An experiment designed to test the generality of Paivio's two factor theory of meaning to sentence comprehension was carried out, and two scoring procedures were investigated: a verbatim score in which sentence elements had to be recalled exactly as they appeared in the original, and a substance score in which synonyms were accepted. The subjects in the experiment were 30 graduate students, 15 of whom were assigned to the Imagery group and 15 to the Verbal group. Ten concrete (high imagery ratings) and 10 abstract (low imagery ratings) sentences were selected for the study (see Appendix for the complete list). The design was 3 x 2 repeated measures design, with repeated measures on the second factor--sentence type (concrete vs. abstract). The first factor was instructional set: Imagery vs. Verbal vs. Rote. With one exception, it was demonstrated that subjects employing verbal and/or imaginal processing techniques will recall more sentence elements than a group instructed to repeat a sentence over and over. The instructional sets had somewhat different effects upon the recall of nouns, verbs, and adjectives. Tables and figures present the results of the experiment. (DB)