Empirical data from 2 experiments confirmed the format dependence predicted by the combined error model (P. Juslin, H. Olsson, & M. Björkman, 1997). Format dependence refers to the simultaneous observation of over/underconfidence in judgment for the same tasks depending on the choice of response format. The ordering of the over/underconfidence effects with the half-range, full-range, and interval estimation formats was correctly predicted by the model, but the assumption of unbiased cognitive processing perturbed by random error underpredicted overconfidence with interval estimation. The estimation and removal of the effect of anchoring-and-adjustment in Experiment 2 suggested that this heuristic alone is unable to account for the overconfidence with interval estimation, whereas the joint effect of the combined error model and anchoring-and-adjustment can explain the data.