The purpose of this paper was to investigate finger response time to visual, auditory and tactile modality stimuli in the context of man-machine-interface. A total 94 right-handed Chinese participants (11 to 60 years old) took part in the visual, auditory, and tactile stimuli tests. In the visual and auditory stimuli tests, once the visual or auditory mode of a number was shown, the participants pressed the corresponding number on the keypad with their right or left middle finger as soon as possible. In the tactile stimuli test, the stimuli were produced through a vibrator that was worn on the participants' right wrist or right leg near ankle. Once the participants detected a vibration, they pressed the number key '8' with their right middle finger as quickly as possible. The result showed that the time in response to the tactile stimuli was significantly shorter, followed by the auditory stimuli and then the visual stimuli. The response time of tactile stimuli was 28% and 34% shorter than that of auditory and visual stimuli, and the response time of auditory was 5% shorter than that of visual stimuli. The location of tactile vibrator i.e. wrist and leg did not have any significant influence on response time. Factors like age, gender, education level, time spent on computer, left/right finger, and choice alternative, however, had significant effects on the response time of an individual to visual and auditory stimuli. The response time decreased with an increase of age up to the 21-30 years, and thereafter it increased gradually with an increase of age. Females were found to respond faster than males. The response of tertiary and secondary education groups was faster than that of primary education group. Besides, the longer the time spent on computer in daily life, the shorter was the response time. In addition, the right finger response time was shorter than the left finger response time. The response on single-choice task was the fastest, followed by two-choice task and then four- and eight-choice tasks. The findings of this study provide a useful reference for engineers and designers to realize how the different modality channels could interfere the operators, so as to design a more user-friendly human-machine-interface.