why do we need to locate the sources of sounds?
- its important info
- it may orient visual attent
- sound location can separate sequences of sounds arising. from diff locations
the coordinates of spatial hearing: azimuth and elevation
what are the two planes
median plane –> where elevation is calculated
horizontal plane
image in wb - P+C
the coordinates of spatial hearing
horizontal plane
minimum audible angle (MAA)
- first present a first tone at 0 degrees (reference tone)
- second tone presented on the left or right of first tone
- P: on which side of the first tone (L or R) was the second tone presented? –> the dist between the tones will be manipulated until you can just no longer tell if the second tone was coming from the left or right. (this is your threshold, called MAA)
- R: MAA is 1 degree for tones coming from straight ahead at low frequencies. MAA is larger (worse) around 1,500Hz, improves at higher frequencies, but is not as small as it is at low frequencies. MAA is larger (worse) for sounds coming from non-frontal locations
how do we know where the sound is coming from?
ITD and ILD
what is the ITD
when the source of a sound is on the left, the sound reaches the left ear before the right ear
(interaural timing difference)
what is the ILD
when the source of sound is on the left, the head shields the right ear from the sound.
this is the ‘head shadow’
(interaural level difference)
how do ITDs and ILDs work?
- by how much do ITDs and ILDs vary (what is the diff between the L/R ears in terms of time and intensity?
- is the auditory system capable of perceiving such differences? (whats the resolution of the auditory system in terms of timing and level differences?)
- are ITDs and ILDs affected by freq? –> yes
how does ITD vary w/ azimuth?
when a sound comes from one side of an adult head (azimuth +90degrees), the ITD is 0.65 ms –> the duration will vary w/ the width of the head (cf. infants, small/ large animals) –> small = less time, wide = more time (increased dist between ears = increase ITDs)
how does ILD vary w/ azimuth?
when a sound comes from one side of an adult head (azimuth +90), the ILD us abt 20dB at 6,000Hz diminishing to 0dB at 200Hz –> if you have a high freq tone (so 6,000Hz), you will have a very healthy LDA of about 20dB between the two ears.
but, this ILD really decreases to almost nothing w/ low freq’s –> so, ILDs exist for high, not low frew sounds (there isnt a usable ILD que when the sound is low freq)
what is the smallest ITD that listeners can distinguish from an ITD of 0 microsecs?
10us
P. first tone presented w/ ITD of 0. second tone has a non-0 (meaning second tone comes into one ear earlier than the other) ITD, w/ left or right ear leading (manipulate the ITD until you can no longer tell if the second tone is on the right or left of the first tone = establishing a threshold of ITD)
R. corresponds to a diff of 1 degree from straight ahead
. the auditory system has exquisite sensitivity to inter-aural timing differences
. 10 microsecs
how does ILD vary w/ azimuth?
- when a sound comes from one side of an adult head (azimuth +90 degrees), the ILD is about 20dB at 6000Hz (if you have a high freq tone you will have a very healthy LDA of about 20dB between the 2 ears - but this ILD really decreases to almost nothing w/ low freqs, so ILDs exist for high, not low freq sounds) diminishing to 0dB at 200Hz
image in wb
why is there such a freq specific response?
= bc of the head
high freqs:
- the head interrupts the propagation of the sound wave –> high ILDs
- tf meaning it will yield highly exploitable level differences
low frequencies:
- the head does not interrupt the propagation of the sound waves as much –> low ILDs
- this is bc there arent as many ILDs to work w/ –> thats why ILDs are virtually non existant at low freqs but highly exploitable at high freqs
what is the smallest ILD that listeners can distinguish from an ILD of 0dB?
= 1dB (lvl of resol of auditory system)
P. first tone presented w/ ILD of 0dB
. second tone has a non-0 (sound louder in one ear compared to the other), w/ left-ear or right-ear dominating
R.sensitivity is good across a wide range of frequencies
.despite being materially irrelevant for low frequencies –> as we can detect a level diff that doesn’t even exist
C. so… ITDs and ILDs are there for us to use and we use them as cues to localisation, although ILDs are only usable at high frequencies
are ITDs also restricted to a specific freq range?
YES –> they are restricted to low freqs
- ITDs are largest and most exploitable at low freqs –> hence why they are complementary w/ ILDs at telling where a sound is coming from, as ILDs = high but ITDs = low
ITDs are unambiguous at low freqs (they work well a low freqs but become ambiguous at high freqs)
low freqs (<700Hz)
- sound waves have long wavelengths
- peaks dont repeat quickly
- brain can match the same peal at each ear
–> clear time difference –> accurate sound localisation
(long wavelength so clear peak matching and tf accurate ITD)
high freqs (>700Hz)
- waves repeat very quickly
- multiple peaks look the same
- brain cant tell which peaks match between ears
–> ambiguity in sound location
(many repeating peaks so matching becomes ambiguous)
key idea: brain localises sound by pairing similar co-occurring patterns between the two ears
see image in wb
why are ITDs less ambiguous for modulated tones (complex tones) than pure tones?
pure tones
- waves repeat identically
- brain cant tell which peaks match between ears
- –> ambitguous ITD
modulated/ complex tones
- peaks have diff amplitudes (some bigger than others)
- brain can match the distinctive peaks between ears
- –> ITD becomes clearer
key idea: the brain pairs similar co occurring patterns and modulation creayes unique patterns to match
where are ITDs coded?
the superior olive
= analyses the location of sources of sound. this happens early in the ascending auditpry system because it relies on very precise timing - of the order of millionths of a second
measuring ITDs: delay lines and coincidence detectors
see model in wb
this model exploits the speed of neural conduction.
probably how localisation in azimuth is implemented in barn owls.
terrestrial mammals, incl humans, are likely to use an ‘‘opponent-process’’ system
what is opponent-process analysis in sound localisation (ITDs)?
there are two neural channels:
- one tuned to left auditory space (negative azimuth)
- one tuned to right auditory space (positive azimuth)
- each channel fires more strongly when sound comes from its preferred side
- the brain compares the diff in activity between the two channels
–> the relative firing rates tell the brain where the sound is located
front-back ambiguity
sounds coming from directly in front or directly behind generate an ITD of zero –> ITDs and ILDs = useless for telling you the diff between front and back as yhe sound will arrive at the same ITDs whether it arrives at the front or back
front-back ambiguity
how can listeners disambiguate the locaton of such sounds?
ans 1: pinna (if familiar sound). attenuation of high freqs coming from the back (sounds from back will sound more muffled and decrease freq as pinna acts as a sound shadow)
ans 2: by rotating the head. if you turn the head clockwise and the ITD favours the left ear then the sound is coming from the front
elevation
ITDs and ILDs arre unhelpful for elevation differences because a sound reaches both ears at the same time ITDs (or ILDs) regardless of the elevation.
but, were still good at telling where a sound is coming from on a medium plane –> how?
–> because reflections of sound within the pinna boost energy at some frequencies and reduce energy at other freqs, depending on the elevation of the sound source. the shape of the pinna also produces unique resonance patterns that help listeners interpret elevation, especially if theyre familiar w/ the sound
judgements of elevation w/ modified pinnae
image in wb
H. if listeners use those cues, then it should be possible to disrupt their judgements of elevation by modifying their pinnae
P. 4 ptps tested on their perception of azimuth (x axis) and elevation (y axis) over several weeks, w/ their own pinnae, artificial pinnae and their own pinnae again
R. judgements of elevation are severely disrupted by modifying a listener’s phone BUT.. accuracy improves w/ experience –> they learnt their new pinna.
there’s no ‘after effect’: accuracy returns to normal as soon as the moulds are removed. –> they immediately switch back on their ability to use their own pinna as soon as the moulds are removed
C. pinna = v. important
how can we simulate the perception of spatial hearing?
virtual acoustic space (VAS)
