The reports of many creative individuals suggest the use of mental imagery in scientific and artistic production. A variety of protocols have tested the association between mental imagery and creativity, but the individual differences approach has been most frequently employed. This approach is assessed here through a range of meta-analytic tests. Database searches revealed 18 papers employing the individual differences approach that were subjected to a conservative set of selection criteria. Nine studies (1494 participants) were included in the final analyses. A marginal, but statistically significant, Fisher's Z-transformed correlation coefficient was revealed. Further analyses showed little difference between form and type of self-reported imagery and divergent thinking. Explanations for the failure to account for more than 3% of the variance in the data sets are discussed in the context of anecdotal reports, task validity, and design problems.