Page 92 - Master Handbook of Acoustics

insofar as direction is concerned. This is the law of the first wavefront. This identification of the
direction to the source of sound is accomplished within a small fraction of a millisecond.

The Franssen Effect

The ear is relatively adept at identifying the locations of sound sources. However, it also employs an

auditory memory that can sometimes confuse direction. The Franssen effect demonstrates this. Two
loudspeakers are placed to the left and right of a listener in a live room. The loudspeakers are about 3
ft from the listener at about 45° angles. A sine wave is played through the left loudspeaker, and the
signal is immediately faded out and simultaneously faded in at the right loudspeaker, so there is no
appreciable change in overall level. Most listeners will continue to locate the signal in the left
loudspeaker, even though it is silent and the sound location has changed to the right loudspeaker. They
are often surprised when the cable to the left loudspeaker is disconnected, and they continue to “hear”

the signal coming from the left loudspeaker. This demonstrates the role of auditory memory in sound
localization.

The Precedence Effect

Our hearing mechanism integrates spatially separated sounds over short intervals, and under certain
conditions tends to perceive them as coming from one location. For example, in an auditorium, the ear
and brain have the ability to gather all reflections arriving within about 35 msec (millisecond) after

the direct sound, and combine (integrate) them to give the impression that the entire sound field is
from the direction of the original source, even though reflections from other directions are involved.
The sound that arrives first establishes the perceptual source location of later sounds. This is
variously called the precedence effect, Haas effect, or law of the first wavefront. The sound energy
integrated over this period also gives an impression of added loudness.
It is not too surprising that the human ear fuses sounds arriving during a certain time window. After

all, at the cinema, our eyes fuse a series of still pictures, giving the impression of continuous
movement. The rate of presentation of the still pictures is important; there must be at least 16 pictures
per second (62-msec interval) to avoid seeing a series of still pictures or a flicker. Auditory fusion
similarly is a process of temporal fusion. Auditory fusion works best during the first 35 msec after the
onset of sound; beyond 50 to 80 msec the integration breaks down, and with long delays, discrete

echoes are heard.
Haas placed his subjects 3 m from two loudspeakers arranged so that they subtended an angle of
45°, the listener’s line of symmetry splitting this angle (there is some ambiguity in the literature about
the angle). The rooftop conditions were approximately anechoic. Both speakers played the same
speech content and at the same level, but one speaker was delayed relative to the other. Clearly,
sound from the undelayed speaker arrived at the listening position slightly before the sound from the

delayed speaker. Haas studied the effects of varying the delay on speech signals. As shown in Fig. 4-
18, he found that in the 5- to 35-msec delay range, the sound from the delayed loudspeaker was
perceived as coming from the undelayed speaker. In other words, listeners localized both sources to
the location of the undelayed source.