In order to study the cognitive function rhythm related to the auditory frequency system for people who prefer to be active in the morning and at night, we conducted an experiment during morning (09:00), evening (17:00) and late-night (01:00) periods. On the basis of a morningness/eveningness questionnaire, six moderately morning-type subjects (M-types) and seven evening-type subjects (E-types) were selected. Diurnal variation of event-related potential (ERP) were assessed under low-frequency (250/500 Hz) and high-frequency (1000/2000 Hz) condition using an oddball task. M-types were tested during the morning (09:00) and evening (17:00) periods, and E-types were tested during the evening (17:00) and midnight (01:00) periods. Subjects were asked to press a button when the target stimulus was detected. We found that the P300 amplitude at 09:00 was significantly greater than that at 17:00 for M-types, was significantly greater at 17:00 than that at 01:00 for E-types. A significant difference of P300 latency and P300 amplitude was observed at 17:00 between M-types and E-types. The P300 amplitude obtained after a low-frequency stimulus was significantly greater than that after a high-frequency stimulus at 09:00 for M-types, and at 01:00 for E-types. These results revealed that stimulus frequency had effects on the diurnal changes of human cognitive function, and circadian typology had a direct effect on the diurnal change of human cognitive function. This study has extended the previous findings of auditory P300 studies on diurnal variations in terms of circadian typology and stimulus parameter.