An aptitude-strategy interaction in linear syllogistic reasoning was tested on 144 undergraduate and graduate students of both sexes. It was hypothesized that the efficiency of each of four alternative strategies--control, visual, algorithmic, and mixed--would depend upon the subjects' pattern of verbal and spatial abilities. Two tests of verbal ability and two tests of spatial ability were administered. Results indicated: (1) that the hypothesis was confirmed; (2) that subjects could be trained to use either a visual presentation or an algorithmic strategy; (3) that algorithmic strategy resulted in lower response times than the other strategies; and (4) that the validity of the mixed model was upheld for the control group. Correlations between parameter estimates for individual subjects and ability factor scores for verbal and spatial abilities are appended. (MH)