A three‐alternative forced‐choice technique was employed to estimate the phase and amplitude of the combination tone f2−f1 and 2f1−f2. The measurement procedure is based on a simple vector description of combination tone interactions. The task of the observer was to discriminate a change in intensity of the combination tone produced by a change of either the phase or amplitude of the cancellation tone. Measurements of f2−f1 essentially constitute a replication of previously reported data obtained with adjustment techniques. Considerable difficulty was encountered, however, in the measurement of the phase and amplitude of the cubic difference tone. Masking of additional low‐frequency combination tones with low‐pass noise did not eliminate the problem. It is concluded that phase induced changes in the envelope of the acoustic waveform as well as the magntiudes of higher‐order combination tones contribute to a monaural phase effect and hamper the forced‐choice measurement of 2f1−f2.