A study explored the degree to which people can capture the meaning of logical terms, using mental imagery. It was then hypothesized that: (1) subjects generally would fail to express logical relationships accurately, and (2) abstract images would be more difficult to capture than would concrete ones. Subjects, 25 college students, were asked to form mental images of two abstract and two concrete sentences, each of which contained a logical term ("either/or,""if/then,""because," or "not") and to write a brief description of the images. These image reports were then coded and analyzed. Results tended to support the first hypothesis, with only 8% of the images reported deemed adequate in a logical sense. Data concerning the second hypothesis were more ambiguous. Concrete stimuli produced more logically appropriate images for "either/or" and "not" sentences, but abstract stimuli worked better for "if/then" and causal reasoning. (FL)