Ap Cam

Find The Best Tech Web Designs & Digital Insights

Technology and Design

Understanding Sound Localization Cues

The ability to locate sound sources is crucial for safety and survival. This review focuses on the nature and significance of localization cues, primarily dealing with the direction of sound sources.

Localizing sound involves the use of both monaural and binaural cues. To completely localize a sound, we need to know three things: (1) angular location in the horizontal plane (0° is straight ahead, +90° is to the right, −90° is to the left), (2) angular location in the vertical plane (0° is straight ahead, +90° is directly above, −90° is directly below), and (3) distance.

What cues do we have to horizontal location? Since we have two ears, we can compare the sound that each ear detects.

How Your Brain Pinpoints Sound: The Science of Sound Localization

The Duplex Theory: ITD and ILD

The Duplex theory, proposed by John William Strutt (Lord Rayleigh), suggests that sound localization relies on interaural time differences (ITD) at low frequencies and interaural level differences (ILD) at high frequencies.

If a sound comes from an off-center location, it creates two types of binaural cues: interaural level differences and interaural timing differences. Interaural level difference refers to the fact that a sound coming from the right side of your body is more intense at your right ear than at your left ear because of the attenuation of the sound wave as it passes through your head. Interaural timing difference refers to the small difference in the time at which a given sound wave arrives at each ear.

Interaural Time Difference

Interaural Time Difference

See for an illustration of the binaural cues of interaural time difference (ITD) and interaural intensity difference (IID). showed how ITDs arise and demonstrated that they are effective, but how does the brain measure those time differences? In 1948, Lloyd Jeffress proposed a way in which different ITDs could activate different neurons.

gives a simple illustration of his model. Although widely accepted, there was no anatomical evidence for this model until 1988 when Catherine Carr and Mark Konishi reported such a circuit in the brainstem of owls (Carr and Konishi, 1988).

Jeffress model

Jeffress model

Limitations of ITD and ILD

If the head remains stationary neither a given ITD nor an ILD can sufficiently define the position of a sound source in space. On such a theoretical basis cones of confusion which open outward from each ear can be predicted ambiguously projecting any source on the surface of such a cone onto an interaural axis. Our restricted ability at localizing sound sources in the vertical median plane is another example of possible ambiguity.

The Role of the Pinna and Spectral Cues

At the end of the 19th century scientists already realized that occlusion of the pinnae cavities decreases localization competence. Each pinna interacts with incoming sound waves differently, depending on the sound’s source relative to our bodies. This interaction provides a monaural cue that is helpful in locating sounds that occur above or below and in front or behind us.

As a result of later achievements in physics and signal-theory it became more obvious that the pinnae may provide an additional cue for spatial hearing and that the outer ear together with the head and the upper torso form a sophisticated direction-dependent filter. The action of such a filter is mathematically described by the so-called Anatomical Transfer Function (ATF). The spectral patterning of the sound produced by the pinnae and the head is most effective when the source has spectral energy over a wide range and contains frequencies above 6 kHz, that is it contains wavelengths short enough to interact with the anatomical characteristics of the outer ears.

Scientific findings further suggest that spectral patterns like peaks and notches may also be exploited monaurally, albeit an a priori-knowledge at the central-auditive level concerning the corresponding transfer functions and relevant real-world sounds is required. Binaural spectral cues are more likely to play a major role in localization. They are derived from another transfer function, the so-called Interaural Transfer Function (ITF), being the ratio of the ATFs at the two ears.

Head Movements and Ambiguity Resolution

The contributions of all these cues may sometimes not be enough to prevent the listener from opting for the wrong direction. But things can be eased by allowing head-movements: More than 60 years ago science reasoned that small head movements could provide the information necessary to resolve most of the ambiguities.

Using empirically determined “head-related transfer functions,” sounds can be digitally altered to mimic the acoustic effects of the head and pinna to give a realistic impression of location in three-dimensional space.

Of course, ITD and IID do not completely localize a sound. They give no cues to distance or vertical elevation, for example.

What location(s) should give rise to the smallest ITDs?

References

  • Campbell RAA, King AJ (2004). Auditory neuroscience: A time for coincidence?
  • Carr CE, Konishi M (1988). Axonal delay lines for time measurement in the owl’s brainstem. Proc. Natl. Acad. Sci.
  • Jeffress LA (1948). A place theory of sound localization.