Chlorhexidine, a bitter bis-biguanide antiseptic, is the only known blocker of the human salty taste. In order to characterize the effects of chlorhexidine on stimulus identification, taste confusion matrix (TCM) performance was measured for subjects treated with 1.34 mM chlorhexidine gluconate (n = 9) and water controls (n = 9). Ten stimuli [water, 0.1 M NaCl, 0.1 M KCl, 0.1 mM quinine-HCl (QHCl), 0.1 M monosodium glutamate (MSG), 3 mM citric acid, 0.3 M sucrose and mixtures of NaCl, QHCl and citric acid with sucrose] were presented in 10 replicates for identification from a list of 10 stimulus names. T(10), a measure of performance consistency from information theory, was lower for chlorhexidine-treated subjects (2.02 +/- 0.11 bits) than controls (2.73 +/- 0.11 bits) (P < 0.0001). T(2), an indirect measure of pairwise stimulus discrimination, approached chance levels (0.40 bit) in chlorhexidine-treated subjects for all possible pairs of NaCl, KCl, QHCl and water, as well as pairs composed of sucrose and the NaCl-sucrose and quinine-sucrose mixtures. In controls T(2) values approached perfect scores (1.00 bit) for all stimulus pairs except NaCl-KCl and NaCl-MSG. The results demonstrate a decreased ability to identify taste stimuli that is consistent with alterations in the ability of stimuli to elicit salty and bitter taste perceptions. As a selective, effective, persistent and reversible blocker of taste perceptions, chlorhexidine should prove useful in defining taste mechanisms in humans.