The direct influences were investigated of test-based measures of intelligence and personality on self-estimates of intelligence, as well as the potential moderating role of personality on the accuracy of self-estimates. Participants (N=189) completed a Big-Five personality traits measure, self-estimated their Fluid reasoning (Gf), Comprehension-knowledge (Cc), and Visual processing (Gv), and completed related subtests from the Woodcock-Johnson Tests of Cognitive Abilities. Regression analysis demonstrated that test-based measures of Gf. Cc, and Gv significantly predicted parallel self-estimates, and while Neuroticism, Openness and Agreeableness accounted for a significant amount of additional variance in self-estimated Gf, personality did not explain significant variance in self-estimated Cc or Gv. Personality was also found to moderate the relationship between test-based and self-estimated Gf and Gv with low extraversion resulting in more accurate self-estimates of Gf (b = 0.65, p < 0.001) and Gv (b = 0.42, p < 0.01). and high agreeableness in more accurate self-estimates of Gf (b = 0.65, p < 0.001). Understanding predispositions that distort self-perceptions of intellectual capacities has important clinical ramifications for improving inappropriate expectations and self image. (C) 2012 Elsevier Ltd. All rights reserved.