Precedence Effect

The psychoacoustic phenomenon whereby an acoustic signal arriving first at the ears suppresses the ability to hear any other signals, including echoes and reverberation, that arrive up to about 40 ms after the initial signal, provided that the delayed signals are not significantly louder than the initial signal. A signal arriving after a delay of 40-50 ms is heard as an echo, provided it is not masked. Also called echo suppression and the Haas effect.

As verified by the importance of time delays in binaural hearing, a slightly delayed signal is not entirely ignored, but may influence the exact localization of the sound source. However, because of the precedence effect, echoes and reverberation are minimized for a short period after the original sound, as may be verified by playing a tape of the event backwards, whereby the reverberation is heard first. As a result, the use of echolocation, as practised by the blind and others, requires a sensitization to this type of spatial information.

Compare: Cocktail Party Effect, Phasing.

The precedence effect also plays a role in PA and other electroacoustic systems. Where direct sound is transmitted both acoustically (with a delay because of the relatively slow speed of sound) and electroacoustically, as from a loudspeaker (with no perceptible delay), confusion in localization of the sound source may result unless a suitable time delay is introduced into the electroacoustic system as well. However, in practice, the visual bias, when the sound source can be seen, easily overrides the acoustic precedence effect, and thus the schizophonic ambiguity of direction is generally ignored by most people.

In the diffusion of electroacoustic music, the performer can use the precedence effect to give the illusion that a stereophonic source appears to emanate from a particular speaker by raising the signal level in that speaker, even though the sound is also present in other speakers at a lower level. See: pan.

Ref.: H. Haas, "The influence of a single echo on the audibility of speech," Journal of the Audio Engineering Society, 20, pp. 146-159.