Binaural hearing and Sound Localization

Question 1

What are some binaural Summations/Benefits for localization? (4)

Answer 1

1. Increase loudness
2. Improvement in differential limen
3. Better perception in noise: spatial filtering
4. **Binaural fusion and beats

Question 2

How does Binaural summation benefit loudness perception?

Answer 2

Increase loudness perception near threshold: 3 dB perfect
2:1 or 6 dB above threshold

Question 3

What is binaural fusion?

Answer 3

It is one example of binaural interaction on hearing, more examples exists, such as central masking, contralateral suppression, co-modulation release, etc.

Question 4

What does this graph show related to binaural summation for loudness?

Answer 4

Binaural Hearing increases loudness by ~6 dB

Question 5

What is a differential limen?

Answer 5

The smallest change in stimulation that a person can detect

Question 6

What is the difference between unilateral and binaural hearing in discrimination?

Answer 6

Binaural hearing is better than unilateral in discrimination, especially at low sensation levels (SL) due to binaural summation in loudness

Question 7

What is the effect of binaural hearing on frequency and intensity discrimination?

Answer 7

Binaural advantage on frequency and intensity discrimination.

(The binaural benefit can’t be attributed to binaural summation on loudness because it would require more than a 30 dB diff in loudness to produce such a difference in discrimination.)

Question 8

According to Pickle, and Harris (1955), what is the difference in the differential limen between unilateral and binaural hearing?

Answer 8

adjust level to account for loudness advantage—a difference in discrimination threshold between binaural and unilateral hearing (disappeared at low SL)

Question 9

According to Jesteadt (1977), what is the effect of binaural hearing on differential limen of intensity and frequency?

Answer 9

Binaural hearing causes better intensity and frequency discrimination (at 70 dB), not due to loudness advantage.

Question 10

It is understandable why the improved discrimination is related to ________________________________: discrimination is NOT level dependent if signals are _________________________; better discrimination is seen in ________________________.

Answer 10

It is understandable why the improved discrimination is related to binaural summation on loudness: discrimination is NOT level dependent if signals are well above the threshold in SL; better discrimination is seen in binaural hearing.

Question 11

What are the benefits of the potential mechanisms for binaural hearing in noise? (4)

Answer 11

Separates target sound from noise (spatial filtering)
Improves discrimination
Improves stream tracking of target sound
Unmasking (via efferent control and other

Question 12

Binaural fusion is related to: (2)

Answer 12

Binaural cues for the acoustic image in space: Binaural differences in intensity, spectrum, and timing

Fused image

Question 13

Fused image from binaural fusion is related to:

Answer 13

The fact that we do not feel that two ears work separately, but dichotic signals can be different or similar and be connected in certain ways

Question 14

_________________ is required for binaural fusion

Answer 14

The commonality
Model by Cherry and Sayers: binaural fusion from two ears receiving similar signals: commonalities

Question 15

What are three examples of commonality for binaural fusion?

Answer 15

Co-modulation of harmonics presented dichotically (different components go to each ear).
Different speech components to each ear: complimentary for speech.
Residual pitch harmonics are presented dichotically.

Question 16

Binaural beats (BB) occur in the ________ while monaural beats (MB) in the ________.

BB are in the ________________ frequency range than MB.

BB can occur at a ________________ and ____________________between the two tones; one tone can be______________________.

Answer 16

1. BB occurs in CAS, while MB in the cochlea.
2. BB occurs in a lower frequency range than MB.
3. BB can occur at a larger level difference and at a larger frequency difference between the two tones; one tone can be below the audible level.

Question 17

Gestalt principle states that:

Answer 17

When you’re presented with a set of ambiguous or complex objects, your brain will make them appear as simple as possible.

1. Grouping units (sounds) together
2. More than simple addition
3. The whole is larger than the simple sum of all parts

*Principle can be seen in vision

Question 18

In a c complex auditory task like tracking a target talker at a cocktail party, what are the cues that may be used in this situation? (5)

Answer 18

1. Spectrum profile of the talker’s speech
2. Temporal stream of the speech
3. Spatial separation/identification
4. Many more (such as familiarity, dynamic cues etc)
5. Bottom-up and top-down processes involved.

Question 19

What does this graph show related to the segregation of sounds in binaural hearing?

Answer 19

Segregation of the stream by the speed

Question 20

What can we see from the Stream 1 and 2 effect of both speed and frequency difference?

Answer 20

When the frequency segregation is larger, you hear two streams at higher speeds.
However, you always hear one stream when the frequency segregation is small (galloping).

Question 21

Give other phenomena and terms in signal processing.

Answer 21

Proximity (similarity): e.g.: similar signals for easy dichotic fusion

Common fate (e.g., on and off together are
the same and likely attribute them together)

Good continuation

Primitive process- Bottom up

Schema-based process- top down

Question 22

What is Schema based learning?

Answer 22

Schema-basedlearning is an active process in which learners construct new ideas or concepts based on their existing knowledge

Question 23

Importance of common onset: example of common fate

Answer 23

A: simple masking: on band of masker upon the signal
B: co-modulation masking release (CMR): reduced masking effect when the noise in the signal band and side bands are co-modulated.
C: CMR disappears due to the mismatched onset of noise between the signal band and the side bands.

Question 24

Give an example of a primitive process:

Answer 24

Bottom-up
Frequency Modulation leads to vowel sensation

Question 25

Example of good continuation in vision

Answer 25

Gliding tone: example of continuation in acoustics from blocking by fence (noise), not a blank in picket fence effect

Question 26

What is the picket fence effect?

Answer 26

It is a bottom-up and top-down process combination causing a continuous signal when blocked by a fence (noise), not a blank gap. The top-down process is involved, especially in the shape perception in the most right graph. (example with gliding tone)

Question 27

What are the two planes for sound localization?

Answer 27

Azimuth (more critical) vs elevation (vertical plan) (our system combines the information from these planes to form the location of the sound source)

Question 28

Azimuth vs Elevation

Answer 28

Azimuth = Horizontal plane Elevation = Medial Plane

Question 29

We investigate the sound source localization ability in two measurements: _________________ and _________________

Answer 29

localization and discrimination

Question 30

What is localization error and what is the task associated with this measure?

Answer 30

The difference between apparent location and physical location The subject must point to the sound source with no reference just hearing the sound

Question 31

What is spatial discrimination and its task?

Answer 31

It is measured as the minimal audible angle (Direction major task, distance, and moving)

Question 32

What are the two listening fields we use to measure sound localization ability?

Answer 32

Open field (closer to reality) Close field

Question 33

What is close field testing in sound localization?

Answer 33

We use headphones which lead to intracranial lateralization (Sound source trapped inside our head likely due to the loss of resonance in the ear)

Question 34

What is open field testing in sound localization?

Answer 34

We use stereophony which leads to extracranial, feeling the sound source (outside our head) localization to both ears (closer to reality)

Question 35

What are the two general issues (questions we ask ourselves for sound localization?

Answer 35

What are the cues for SL and how are they used?

Question 36

What are the approaches to answer these considerations for sound localization? (2)

Answer 36

Behavior studies and neurological mechanisms

Question 37

What are the cues for sound localization?

Answer 37

ITD/IPD (phase difference) IID/ILD

Question 38

What is the duplex theory for localization in the azimuth plane?

Answer 38

ITD/IPD Determined by the size of the head, larger the head which is like a bowl (diameter is like 22-23 cm) IID or ILD From shadow effect of head

Question 39

The ITD higher when sound at____________________________ with largest time difference between both ears of _______ μs and a time difference sensitivity: ____μs difference across frequency.

Answer 39

ITD is higher when a sound at a 90-degree angle (fully lateralized to the side of the head) 660 μs Time difference sensitivity: 10 μs difference across frequency Time converts to angle and phase

Question 40

Shadow effect is larger at_____________________________________________________

Answer 40

High frequency, by shorter wavelength near ear (When the frequency is low longer wavelengths, can go around obstacles, high-frequency sounds are attenuated by obstacles)

Question 41

How does the shadow effect change with frequency?

Answer 41

For higher frequencies, the ITD becomes larger and larger depending on the location of the sound source (90-degree), and for low frequencies almost no shadow effect and no level difference. (Must consider the combined effect of distance and shadow)

Question 42

What is the effect of frequency on ILD?

Answer 42

ILD is higher with higher frequencies and shorter wavelength ILD can go at 20 dB

Question 43

ITD summary: ITD comes from __________________ and is determined by ________________ and is influenced by ____________________________.

Answer 43

1. ITD comes from a distance difference 2. Determine by azimuth (angle), little effect of distance 3. Influenced by the shape of the head, changes with frequency

Question 44

ITD produces IPD for __________________________________

Answer 44

for periodical signals (Interaural time difference can be converted into phase delay)

Question 45

IPD depends on periodical signals and also on:

Answer 45

on frequency: for certain IDT, high frequency will have a larger IPD value

Question 46

Larger IPD ≠

Answer 46

stronger signal

Question 47

For a fixed IPD, ITD ___________________________________

Answer 47

ITD decreases with frequency

Question 48

IPD is a useful cue when it is __________ over that it will ___________

Answer 48

below 360 degrees, over that, it will repeat itself (sine When 360o>IPD>180o, the IPD acts the same as of IPD<180o

Question 49

To make a clear decision based on IPD, ½ period must be:

Answer 49

larger than the maximal time difference (MTD) (660 microseconds), so that IPD <180.

Question 50

How does the frequency affect ITD and IPD?

Answer 50

360 phase difference for different frequency Higher the frequency, the shorter the time difference for IPD<180o

Question 51

MTD and the highest frequency for useful IPD (to generate useful IPD):

Answer 51

To make the ½ period longer than MTD, the Frequency must be smaller than a value. If MTD=0.7(or 660) microseconds, what is the max Frequency to ensure this? To make period/2 larger than 0.7 ms, 1000/1.4 = 714 Hz Therefore, max Fre for IPD < 180o is ~700 Hz

Question 52

What does this graph show related to the Intensity difference/time change with frequency?

Answer 52

a and b show the ITD at 90 and 45 degree: shorter the ITD, higher the frequency for 180 phase difference. Limiting conditions for intensity and phase For phase: the curves tell how much ITD will produce targeted phase difference, as a function of frequency

Question 53

Why are low frequencies better for ITD and IPD processing?

Answer 53

Temporal coding is better for low frequency No ILD available at low frequency

Question 54

Smaller animals use _____, big animals ______ because MOC neurons ____________________________________.

Answer 54

Smaller animals use ILD, big animals ITD MOC neurons are more dominant and they require low frequencies to use ITD to localize a sound source.

Question 55

What are the elements for identifying sound sources at low frequencies?

Answer 55

- Identity circle - Must be < 180 degrees, change with frequency At 90 azimuth, ITD 650 us = ½ cycle of 770 Hz At 45 azimuth, ITD 350 us = ½ cycle of 1400 Hz Close to 0 azimuths, ITD →0, frequency limit increase, but still low f signals make strong IPD cues!

Question 56

Sound localization accuracy in azimuth is always the best at 0 ____________ No matter what types of cues are used and largest ITD/IPD/ILD at ____________ but larger ITD/IPD/ILD does not mean _________________

Answer 56

Always the best at 0 azimuth No matter what types of cues are used Largest ITD/IPD/ILD at 90 degree But larger ITD/IPD/ILD does not mean strong stimulation

Question 57

Why are we better at localization at 0 degrees related to neurons?

Answer 57

More neurons for localization are organized that they work best at 0 azimuth

Question 58

What is the duplex theory?

Answer 58

High frequency, rely on ILD Low frequency, rely on ITD, predictable from sphere model (At high fre, ITD also play roles At low fre, ITD differ from sphere model (< 500 Hz, ITD = 800-820 us))

Question 59

The duplex theory depending on studies using ____________________________________

Answer 59

Depending on studies using pure tones, the real sounds are normally complex

Question 60

What can we say about our localization accuracy across frequency?

Answer 60

The higher the line, the poorer our accuracy: ITD/IPD is better at low frequencies, ILD is better at high frequency If our localization depend on duplex theory (pure-tones) then it is poorer performance at middle frequencies (which is rare for our system to be poor at middle frequencies)

Question 61

What are the limitations of duplex theory? (3)

Answer 61

- We do not rely upon pure tone for localization - High-frequency sound can have time cues, e.g., when modulated by low frequency - Break down front-back confusion and identical circle by pinnae cues

Question 62

Why should we use earphones instead of speakers?

Answer 62

Speakers require too much equipment, a testing environment is more strict Earphones: - can avoid these problems - reduce the burden of equipment. - We can change the parameters to change phase, and level independently in each ear

Question 63

With earphones, Halverson studied using:

Answer 63

500 Hz tone, 0-180o change in phase converted to 0-90o azimuth Found that frequency limit for useful IPD Phase change leads position image change when frequency < 1400 Hz

Question 64

When using earphones: The relative effectiveness _______________________________ Sound trapped in the head due to _____________________ ______ more important than______ when using earphones ITD: only works at _______________ IPD: works for ________________

Answer 64

The relative effectiveness of ITD and ILD can be evaluated Sound trapped in the head due to the loss of the pinna effect ITD is more important than IPD when using earphones ITD: only works at onset and offset IPD: works for continuous signals

Question 65

Explain the contributions from the ITD at onset and offset:

Answer 65

In virtual hearing (via earphone): early onset in the near ear leads to sound coming from the nearer ear (the effect of onset discrepancy), whereas early offset in the near ear leads to sounds coming from the farther ear (the effect of offset discrepancy) Overall, the onset ITD is dominant: in real hearing, we hear sound based upon the onset discrepancy. Also, there are studies comparing the effect of onset ITD and ILD: in virtual hearing, the near ear can have early onset but weak sound.

Question 66

In research, the major progress with improvements in technology showed that virtual hearing (earphones) can better mimic real hearing from: (3)

Answer 66

1. From simple sounds to complex signals 2. Use of headphones: lost spectrum cues 3. Digital technology can put back the spectrum cues

Question 67

Explain MMA (Minimum Audible Angle)

Answer 67

Smallest detectable difference between the incident directions of two sound sources Much more accurate than sound localization Largest around 1-3 kHz Smallest at 0o azimuth Concurrent MAA (CMAA): two signals at the same time

Question 68

MAA according to Yost states:

Answer 68

1. IPD shift required for image shift remains constant when f<900 Hz 2. IPD shift required for image shift increases with original IPD 3. Upper freq limit: 1200 Hz

Question 69

What does this graph show related to the IDL threshold and frequency in lateralization?

Answer 69

Y-axis: Sound level required for the subject to tell where the sound is. The IDL threshold does not change with frequency 0 dB: middle line (0 azimuth) Better performance 9 dB: 45o = poor performance 15 dB: 90o = poorest performance

Question 70

What does this graph show related to the IPD in lateralization using earphones?

Answer 70

No effect at 2000 Hz, the performance is so poor it's not useful For lower frequencies, we can see the change of phase difference that is required to identify sound source smaller value around 0 azimuth and poorer at 90

Question 71

The Minimal angle for earphones remains _____________________________and is proportional to the original (or baseline) phase difference at _________ a just detectable phase angle change is _________________________________________

Answer 71

The Minimal angle for earphones remains constant for up to 900 Hz and is proportional to the original (or baseline) phase difference at 500 Hz, a just detectable phase angle change is 2 degrees or 11 microseconds

Question 72

In Lateralization: Just noticeable phase diff at 500Hz: ________ or _______ At 1200 Hz: _________ or ___________ In Localization: just noticeable phase difference at ____________ or ____________ MAA is __________________localization error

Answer 72

In Lateralization: Just noticeable phase diff at 500Hz: 2o or 11 microsec At 1200 Hz: 12o or 27 microsec In Localization: just noticeable phase difference at 100 Hz:3o or 5.83 micros MAA is much smaller than localization error

Question 73

What does this graph show related to the MMA change with frequency?

Answer 73

Three different locations When at 60 degrees the MMA is huge at middle frequencies Smallest at 0 azimuths its roughly at 2 degrees

Question 74

What occurs in front-back confusion in sound localization?

Answer 74

Sound sources on the cone of confusion are equidistant from the left ear and right ear and thus provide identical ITD and ILD for a listener and the duplex theory doesn't work. At any point on the cone surface, binaural cues are the same.

Question 75

To get rid of the confusion cone effect, we need to use:

Answer 75

Spectral cues

Question 76

What are the spectral cues and their use?

Answer 76

We use them for location and to avoid errors in azimuth (confusion cone effect) in the vertical plane from the pinna effect ear canal resonance.

Question 77

What are dynamic cues?

Answer 77

Dynamic versus stationary sound source (referred to location) Stationary stimuli can be dynamic when the head is moving We can break front-back confusion by moving head The head movement helps monaural localization (Small effects reported)

Question 78

What are the cues for estimating the distance of a sound source? (5)

Answer 78

Sound Level The ratio of direct-to-reverberant energy Spectral Shape Binaural Cues (ILD) Familiarity or Experience

Question 79

What is the precedence effect?

Answer 79

The sound that is heard first takes the dominant role in the localization and is the most contributor (onset time difference is more important than offset).

Question 80

Which type of sound was used to test the precedence effect?

Answer 80

Found out from a classical click experiment and fusion occurs when the click interval is less than 5ms Summing localization—fused when interclick interval <1 ms Localization dominance, when click interval 2-5 ms, pair interval between 10-100 ms. Causes Discrimination suppression

Question 81

Explain localization dominance.

Answer 81

τ1 = interval first pair of clicks τ2 = interval second pair of clicks τ3 = interval between the pairs of τ1 and τ2 When τ1, τ2, and τ3 are small enough (τ1, τ2 < 5 ms, 10ms<τ3 <100ms), the listener tells the sound comes from the left ear (the summing localization of the leading ear dominates the result). summary: The first pair is left lead and the second pair is right to lead. Playing the first pair along, the subject feels that the two clicks come from the left. Play the two pairs together, and the subject hears the fused image from the left.

Question 82

Explain how discrimination is suppressed from the interval between clicks.

Answer 82

a) summing localization since the interval between click pairs is too small less than 1 ms (fused) b) summing localization since the interval between click pairs is too small about 3 ms (fused) lateralized to the leading ear c) No summing localization since the interval between click pairs is large enough about 15 ms (not fused) c) Left leading suppress localization at right (poorer)

Question 83

Discrimination suppression occurs when _____________________________________________________.

Answer 83

the leading ear receiving the sound first causes suppression of sound in the lagging ear.

Question 84

What cues do we use for localization? (summary)

Answer 84

Binaural - azimuth - divided into ITD (low frequency, big animals) and ILD (high frequency, small animals) Spectral cues - vertical to avoid cone of confusion using pinna effect and EAC resonance Dynamic cues - moving head or sound source for better localization to break down front-back confusion, or monaural localization

Question 85

What is head-related transfer function?

Answer 85

- How our head changes the transmission of sound depending on its location - directionally related to the sound source (spectrum change with direction) - causes timbre changes

Question 86

What is binaural fusion?

Answer 86

It is one example of binaural interaction on hearing between ears from many sound cues

Question 87

Explain the masking level difference between dichotic and diotic presentations.

Answer 87

S: signal, N: noise, o: in phase, pi: out of phase (180 degree), tau: time delay a) signal and noise in monaural b) signal and noise binaural in phase c) signal in monaural and noise both ears 9 dB masking level goes down (dichotic) due to efferent d) diotic signal both ears noise both ears out of phase 13 dB of masking level difference e) LARGEST signal out of phase both ears (dichotic) noise in phase both ears (diotic) 15 dB masking level threshold f) signal diotic in phase noise time delay. 3-10 dB

Question 88

In which presentation is the masking level higher than the others? Why?

Answer 88

SpiNo LARGEST signal out of phase both ears (dichotic) noise in phase both ears (diotic) 15 dB masking level threshold -Indicating mechanisms related to phase locking -Supported by MLD change with fre (Larger MLD for low fre) - Larger for higher spectrum level of noise

Question 89

What does this graph show us related to the masking level difference depending on the frequency of signals?

Answer 89

There is a larger masking effect for low frequency signals

Question 90

Give the binaural responses of neurons. (ipsilateral stimulation)

Answer 90

Table Below

Question 91

Give the spatial distribution of binaural neurons in the MSO and LSO.

Answer 91

Low Frequency neurons in MSO and IC: EE type dominant - ITD High Frequency neurons in LSO: IE type dominant; in IC: EI type dominant - ILD

Question 92

How do neurons code sound time difference from both ears?

Answer 92

Binaural processing: Period Excitation and inhibition: Curve = sound wave ITD—onset disparity IPD—ongoing disparity Maximal interaural delay: 0.8 ms IPD depends on frequency

Question 93

What does this slide show related to the fact that MSO neurons have characteristic delay ?

Answer 93

This slide shows the time delay of neurons in respone to contralateral stimuli vs ipsilateral stimuli Ipsilateral stimuli neuron in white has shorter latency Contraleral stimuli neuron in black has bigger latency which is due to the length of the traveling wave to the cochlea

Question 94

What would happen if we would make the stimulation to the contralateral ear earlier (near ear)?

Answer 94

Shorter time delay stimulation when sound to travel from the speaker to the neuron if contralateral ear earlier (near ear) Jeffery’s Coincident theory Interaraul delay is countered by neural delay to make coincident