Fear and anxiety are common emotions that can be triggered by stress. This paper reviews the work examining the role played by specific corticotropin-releasing factor (CRF) receptors in mediating the expression of these emotions. Several lines of evidence taken from CRF(1) transgenic knockout mice, CRF(1) antisense oligonucleotide studies, and CRF(1) receptor antagonist work suggest that the anxiety inducing effects of CRF are mediated by the CRF(1) receptor. Of these three methodological approaches, the work using transgenic CRF(1) knockout mice appears to be the most consistent. In contrast, the work using specific CRF(1) antagonists has produced somewhat varied results that may be explained, in part, by the testing method. When animals are stressed prior to behavioral testing, CRF(1) receptor antagonists appear to have anxiolytic-like effects. In addition, chronic dosing with CRF(1) antagonists may have more potent anxiolytic-like effects, especially in animal models of spontaneous anxiety, than acute dosing procedures. Spontaneous anxiety is defined as behavior that is elicited entirely by the testing situation without current or prior aversive or explicitly induced stress. CRF(1) antisense oligonucleotide work is difficult to interpret because of potential toxicological side effects produced by the antisense oligonucleotide and, in some cases, the absence of verifiable reductions in CRF(1) receptor densities after treatment. Similar methods-CRF(2) knockouts, CRF(2) antisense oligonucleotides, and CRF(2) antagonists-were used to evaluate the function of CRF(2) receptors in emotionality. In comparison to the large number of CRF(1) receptor studies, fewer CRF(2) receptor investigations have been conducted and these studies have yielded mixed results. However, recent work demonstrating a robust reduction in CRF(2) receptors using a CRF(2) antisense oligonucleotide with minimal toxicity, and dose response studies using a peptide CRF(2) antagonist suggest that CRF(2) receptors play a role in stress-induced and spontaneous anxiety. Furthermore, inhibiting the actions of both CRF(1) and CRF(2) receptors produces a greater reduction in stress-induced behavior than inhibition of either receptor alone. Thus, current data suggest that CRF(1) and CRF(2) receptors are involved in the mediation of fear and anxiety behavior.