Sound localization is our ability to determine from where a sound is coming. If a friend calls your name and you decide, based on the sound of the person's voice, that they are behind and slightly to the right of you, then you have localized the sound.
Sound localization primarily depends on two binaural sources of information or cues: the interaural level difference (ILD) (sometimes referred to as the interaural intensity difference [IID] or the interaural amplitude difference [IAD]) and the interaural time difference (ITD). These cues help us determine where in the azimuth plane the sound is located. If you are standing straight up, the azimuth plane is parallel to the ground at approximately ear level. Sound localization also depends on a monaural cue called the head-related transfer function. This cue provides information about the elevation of the sound source -- whether the sound source is above or below the ears.
The ILD is the difference in the intensity of a sound as measured at the ear on the left side of your face and the intensity of the same sound as measured as the ear on the right side of your face. Why should these intensities be different for sounds from most locations? There are two reasons -- first, sound pressure level decreases with the square of distance. That is, if you measure the sound pressure level at a distance of 1 m from the sound source and again at 2 m from the sound source, the pressures will be different. When you double the distance from the source of the sound (e.g. from 1 to 2 m), you quarter the sound pressure level. As the sound wave propagates away from the sound source, it spreads its energy out in two dimensions and thus the pressure at a given point decreases with the square of the distance from the sound source. Because the two ears are not equally distant to most sound sources (that is, one ear will usually be a few cm closer to the sound than the other), the sound pressure level will usually be different at the two ears. Because the two ears are reasonably close together, this difference can be very small.
Question: Where can a sound originate so that it is equally distant to both ears? What would be the ILD for sounds from these locations?
The second reason that the ILD occurs is that the head can create an acoustic shadow of some sounds. With light, a shadow occurs when an object blocks some or all of the light and thus prevents the light from reaching the destination. The shadow is the dark area where the light has been blocked. An acoustic shadow is the same basic thing -- some or all of the sound wave is blocked by an object. But rather than being a dark area as with light, an acoustic shadow is a lessening in the intensity of the sound. Thus, the head can create an interaural intensity difference, but only for sounds that are propagating in a direction that puts one ear closer to the source of the sound than the other ear.
The head is better at creating acoustic shadows for some frequencies than it is for other frequencies. In general, the larger the object (head) is relative to the sound wave, the more capable it is of blocking the sound wave. How big is the sound wave? That depends on its wavelength. Remember that wavelength and frequency are inverses of each other: wavelength = 1 / frequency. That is, a 1,000 Hz tone will have a shorter wavelength (1 / 1000 = 0.001) than will a 100 Hz tone (1 / 100 = 0.01.) For a given size of head, the higher the frequency of the sound wave, the larger the acoustical shadow will tend to be. To understand, think about different sizes of rocks in a fast moving river. The river represents the sound wave and the various sizes of rocks represent different sizes of head. A small pebble (or a small head) has very little effect on the current (or the sound wave). A large bolder (or a large head) has a large effect on the current (or the sound wave) creating an area of relative calm (the acoustic shadow) just downstream of the bolder.
Given the size of most people's heads and the wavelength of the sounds that we can hear, the ILD starts to become an effective localization cue around 1000 Hz. Below 1000 Hz, the head is too small to create an acoustic shadow for the frequencies that humans are sensitive to. The ILD can be as large as 30 dB -- about the maximum protection offered by many over the ear hearing protectors. That is, the difference in the intensity at the left and right ears due to the acoustic shadow can be fairly large.
It should be clear that the ILD occurs, but what does it have to do with localization? The ILD systematically varies based on the location of the sound relative to the ears. A sound that originates directly in front of a person will be equally distant to the two ears (assuming the person is looking directly at the sound source) and any acoustic shadow that arises will be the same for the two ears (because heads are roughly symmetrical.) Thus, sounds coming from directly in front of a person will have an ILD of 0. A sound that is coming from a few cms from the left ear will be five or more times closer to the left ear than to the right ear. Also, if it is a higher frequency, the whole side of the head is there to block it, creating a relatively large acoustic shadow. If you move a sound source in a circle around a person's head, the ILD will be largest when the sound source is directly to the side of either ear and smallest when directly in front of, or behind the head. The ILD decreases somewhat uniformly as the sound source moves away from the ear toward the front and then increases somewhat uniformly as it moves away from the front toward the other ear.
Being good students of perception, you will undoubtedly want to know the physiological basis of the ILD. Park, Klug, Holinstat, and Grothe (2004) demonstrated that the lateral superior olive (LSO) contains neurons that are maximally stimulated by a given ILD.
The interaural time difference (or ITD) is the difference in when a sound reaches one ear compared to reaching the other ear. The ITD also arises because the ears are usually not the same distance to the sound source. Because the sound wave often has to propagate a little farther to reach one ear compared to the other, and sound waves take time to propagate, one ear will often hear the sound a little before the other ear does. Because the ears are relatively close together and sound moves relatively quickly, this difference is often very small.
Question: Where can a sound originate so that its ITD is 0? In conjunction with the previous question, what does this tell us about sounds originating from these locations?
Given the diameter of most people's head (mine's around 18 cm) and the speed of sound (approximately 340 m/s at sea level), the maximum difference in arrival time at the two ears is approximately 0.5 to 0.6 ms.
It is fairly simple to calculate the ITD for a sound in any given location relative to a head. The following figure shows the ITD for sounds located at various angles relative to looking straight ahead. In this figure, -90° is directly to the left of the head, straight out from the left ear. 90° is directly to the right of the head, straight out from the right ear. The Y axis shows the difference in the arrival times to the left ear relative to the right ear. Positive values indicate that the sounds arrives at the left ear first while negative values indicate that the sound arrives at the right ear first. The figure clearly indicates that there is a systematic relation between the location of a sound source at the difference in the times that the sound arrives at each ear.
Question: The figure shows that for any given ITD, there are two angles that will produce that ITD. For example, an ITD of 0.37 ms is produced by angles of -45° and 225° What is the relation between the pairs of angles that produce the same ITD? What does this tell us about auditory localization using only ITD as a localization cue? If we generalize from the two dimensional shown in the figure to the three dimensional case, there is an infinite number of points that will produce a given ITD. What does this tell us about auditory localization? (Hint, think about the cone of confusion.) How are these ambiguities disambiguated?
In addition to having neurons sensitive to the ILD, the Superior Olivary Complex (SOC) also has neurons that are sensitive to ITDs. In particular Yin and Chan (1990) demonstrated that the medial superior olive (MSO) contains a spatial map of ITDs across the anterior - posterior axis of the MSO. Neurons near the anterior part of the MSO tend to respond best to sounds whose ITD is close to 0 while those near the posterior part of the MSO tend to respond best to sounds whose ITD had relatively long (~0.6 ms) delays to the ipsilateral ear.
Head-Related Transfer Function
The final set of auditory localization cues are collectively know as the head-related transfer function (HRTF). The various ridges and troughs of the pinna, and to a lesser extent the head, the shoulders and the upper torso all change, or filter, the sound waves. Furthermore the way the sound waves are filtered or changed is systematically dependent on the location of the sound source. The HTRF provides minimal information about the location of the sound source in the azimuth. However, it does act as a good cue for the elevation of the sound source.
Localization vs. Lateralization
Everything that has been said so far is about localization. Yet the title of the page is lateralization. Do they mean the same thing? The answer is, almost, but not quite. To be able to separately control the IID and ITD, the sounds have to be presented by headphones. When sounds are presented by headphones, the sounds sound as if they originate within the head. Localizing sounds within the head is called lateralization; localizing sounds that appear to come from outside the head is called localization. Lateralization and localization rely on the same binaural cues and mechanisms. However, the simple lateralization study that we will do does not include the monaural head-related transfer cues that are found in localization.
The sounds that you will hear consist of one of two frequencies. The low frequency sounds will be 600 Hz. The high frequency sounds will be 2000 Hz. The ITD, ILD or both of the cues will be systematically manipulated so that the sound will appear to slide from either near the left ear to near the right ear or vice versa. Your task is to indicate whether the sound seems to move from left to right or right to left. You will experience five left to right and five right to left trials for each frequency for each type of localization cue (ITD, ILD or both.) Thus, you will experience (5 + 5) X 2 X 3 = 60 trials. The number of correct lateralization judgments will be recorded and displayed for each frequency and localization cue.
You must do the experiment with headphones or ear buds. You should not try to do it with speakers as you will virtually never keep the speakers exactly the same distance from each ear and any such deviation will severely affect the ITD and ILD.
Always practice safe listening. Always start with a very low volume level and gradually increase it to a reasonable level. If a person standing next to you can hear the sounds while you have the ear buds in your ears, then you have the volume level too high and you must decrease it. The hair cells in the inner ear can be easily, quickly and permanently damaged by loud sounds. The effect of loud sounds is cumulative -- you may not notice the damage now, but each time you expose yourself to loud sounds, you move closer to permanently damaging the hair cells and the resulting loss of hearing.
If you can easily hear both tones and neither is too loud, leave the volume settings alone for the rest of the study unless something becomes too loud.
Question: Before proceeding, try to predict what should happen in the study. For the low frequency sound, will lateralization be better when ILD is present or when ITD is present? For the high frequency sound, will lateralization be better when ILD is present or when ITD is present?
When you are ready to begin the study, click the Start button below.
Remember, you will listen to a sound and then click on one of
two buttons, ">> Left to Right >>" if you think the sound starts near your
left ear and travels toward your right ear, or "<< Right to Left <<" if you
think the sounds tarts near your right ear andt travels toward your left ear. Guess if
Park, T. J., Klug, A., Holinstat, M. & Grothe, B. (2004). Interaural level difference processing in the Lateral Superior Olive and the Inferior Colliculus. Journal of Neurophysiology, 92, 289-301. http://dx.doi.org/10.1152/jn.00961.2003
Yin, T. C. T., & Chan, J. C. K. (1990). Interaural time sensitivity in medial superior olive of cat. Journal of Nuerophysiology, 64, 465-488.