Studies generally have shown that even when the IQ scores of the two groups are comparable, good readers perform better than poor readers with paired-associate (PA) learning tasks (Walters & Doan, 1962;Walters & Kosowki, 1963;Giebink & Goodsell, 1967; Otto, 1961, in press). However, when Cooper (1968) used a list of pictures of common objects paired to minimize associational bonds, she found no significant difference in good and poor readers' performance with the PA task. The explanation may be inherent in the fact that Cooper's list included pictures only, while the lists used in the other studies included words or word-like stimuli. The finding of Rohwer, Lynch, Levin, & Suzuki (1967) that third and sixth graders unselected as to reading ability learned more pairs when PA items were pictures rather than words provides some support for this notion: The general advantage derived from responding to pictures might be even greater for poor readers, who have every reason to respond negatively to words.The present study was done to determine whether verbal and pictorial presentations of items would differentially affect good and poor readers' performance with a PA task. The predictions were that (a) in general, good readers, fifth graders and Ss responding to pictorial items would make fewer errors in learning a PA list than poor readers, second graders, and Ss responding to verbal items, and (b) specifically, poor readers would make more errors than good readers with verbal items, but with pictorial items the performance of the two groups would not differ.Subjects and Design. The Ss were good and poor readers from grades two and five in a small city school system. The 10 teachers involved were asked to identify all of their pupils whom they judged to be (a) in the average (90-115) intelligence range, and (b) in the upper and lower thirds of their classes in reading achievement. The teacher judgments were checked against California Test of Mental Maturity IQ scores and reading subtest scores from the Stanford achievement battery, and pupils whose test scores were in accord with the teacher judgments were considered potential Ss. From this pool of Ss, equal numbers of boys and girls from each reading level and each grade were assigned to the two treatment groups. Thus, the design was a 2 by 2 by 2 factorial, the independent variables being grade (second and fifth), reading ability (good and poor), and treatment (verbal and pictorial lists). Six boys and six girls were assigned to each cell, for a total of 96 Ss.Materials and Procedure. The PA list included eight pairs of common objects-mop -cake, tree -hat, clock -house, fish -bed, cat -shoe, soap -fork, comb -glass, and cow -ball-which were .verbally or pictorially represented on 4 in. x 4 in. squares of poster board. The pictures were line drawings from a first grade workbook and the words were hand printed in lower-case letters.The Ss were tested individually. The cards were arranged in pairs and placed face down in a constant order to begin. The E said, "There a...