Hearing Flashcards
(88 cards)
1
Q
What is the relationship between the local motion of molecules and the direction of travel of the sound wave?
A
Move in parallel
2
Q
Amplitude of the sound definition
A
The pressure variation about
the mean pressure.
3
Q
What is a simple tone
A
A sinusoid (a pure tone)
4
Q
Period definition
A
Time taken for 1 complete cycle
5
Q
Frequency definition
A
Number of cycles per
second (Hz).
6
Q
Phase
A
A portion of a cycle (one
complete cycle corresponds to 360°or 2π radians).
7
Q
Amplitude equation
A
amplitude = maximum amplitude × sin(2π × frequency × time + phase)
a = A sin( 2π ft + p)
8
Q
3 properties of linear systems
A
- The output of a linear system to the sum of two inputs = the sum of its outputs to the two inputs
separately. - If you double the input to a linear system, then you double the output.
- A linear system can only output frequencies that are present in the input; a non-linear system always add extra frequency components.
9
Q
Definition of a period sound
A
A sound that repeats the same pattern over time regularly.
10
Q
What is subjective pitch related to?
A
The repetition rate (the frequency).
11
Q
Fourier’s Theorem in sound
A
Any periodic sound can be described as the sum of a series of sinusoidal
components with specific frequencies, amplitudes and relative phases.
12
Q
What is a component of a periodic sound called?
A
A harmonic (one of the sine waves it is made up of).
13
Q
What is the frequency of the first harmonic?
A
It is equal to the repetition rate of the sound.
14
Q
What is the fundamental component?
A
The first harmonic.
15
Q
What is the relationship between other harmonics and the fundamental component?
A
Other harmonics have frequencies that are integer multiples of that of the fundamental. (The nth harmonic has a frequency which is n times that of the fundamental).
16
Q
What is a sound’s amplitude spectrum?
A
A plot of the maximum amplitude at every frequency.
17
Q
What is the nature of the amplitude spectrum for periodic sounds?
A
The spectrum shows discrete harmonics (a ‘line spectrum’).
18
Q
What is the nature of the amplitude spectrum for non-periodic sounds?
A
The spectrum is continuous (amplitude as a function of frequency).
19
Q
3 components of the peripheral auditory system
A
- Outer ear (pinna).
- Middle ear.
- Inner ear (cochlea).
20
Q
What is the function of the pinna and external auditory canal?
A
Introduce a broad resonance that increases the sound level at the eardrum by about 15 dB.
21
Q
What is the function of the middle ear?
A
Increases efficiency of sound transmission from air into (fluid-filled) cochlea.
Acts as an impedance (resistance of flow of sound energy) matching device.
22
Q
What frequencies do the middle ear work best for?
A
500 - 5000Hz
23
Q
What influences absolute threshold (amplitude) for pure tones (sine waves) depend on?
A
The frequency of the tone.
24
Q
What are the 2 ways to measure the threshold for pure tones (min. amplitude for different frequencies)?
A
- Presenting sounds via earphone. Measure the sound at the ear drum with a small microphone (minimum audible pressure, MAP; monaural).
- Presenting sounds over loud speakers and then measuring the point where the listener’s head was (minimum audible field, MAP; binaural).
25
Which frequencies produce the biggest difference in threshold for monaural vs. binaural hearing?
2 - 5Hz
26
Does monaural or binaural hearing have a lower threshold?
Binaural (particularly for 2 - 5Hz).
27
Where is the basilar membrane?
Inside the cochlea (inner ear).
28
What happens to the basilar membrane when sound is transmitted to the cochlea?
The sound leads to a pressure difference across the BM and this sets up a wave that travels along it.
29
Properties of the BM
1. Stiff and narrow at one end (towards
the stapes; middle ear) -> increasingly flexible and broad towards the other end.
2. The mechanical properties of the BM mean that each place along the BM is tuned to a specific frequency.
3. The peak (max. height) of the travelling wave occurs at the place on the BM that corresponds to the sound's frequency (high frequencies = closer to stapes).
30
Where is the organ of Corti and what does it do?
1. Lies above the BM.
2. Contains inner hair cells and outer hair cells.
31
How do inner hair cells function?
They respond to the relative movement between the BM (vibrates in respond to sound) and the tectorial membrane (stiff structure above the hair cells).
They convert mechanical vibration into nerve spikes (sending the sound info to the brain).
32
How do outer hair cells function?
They act more like amplifiers/tuners and do not send sound info to the brain directly. They influence the mechanical vibrations on the BM.
33
2 properties of neurons in the auditory nerve (IHC -> brain)
1. Each neuron responds best to a limited range of frequencies (derives activity from a single IHC).
2. Each neuron responds at a specific phase of the stimulating waveform (phase-locked responses). They fire action potentials at a particular phase of a sound wave.
34
What are tuning curves?
A plot of the threshold amplitude of a neuron responding to a sine wave as a function of the frequency. Record the responses to a frequency sweep.
35
What is the characteristic frequency?
The frequency at which the amplitude threshold is lowest (frequency neuron is most sensitive to).
36
An example of non-linearity in the auditory system
Two-tone suppression. If it were linear, the firing rate should be unchanged or increased by the addition of an extra tone.
37
What is two-tone suppression?
A neuron will fire for a tone at the neuron's characteristic frequencies. However, if a 2nd tone is played at a frequency hear the CF and a level further above the threshold, the firing rate will be reduced. (Characteristic of the normal ear).
38
For which frequencies is phase-locking most precise?
Low frequencies (disappears when frequency > 4 - 5Hz).
39
Phase-locking meaning
The time intervals between the action potentials are roughly integer multiples of the period of the waveform.
40
2 theories for the perception of pitch
1. Place theory of pitch perception.
2. Temporal theory of pitch perception.
41
2 postulates of the place theory of pitch perception
1. Cochlea converts frequency into place in the BM.
2. Perceived pitch is inferred from place of max. response on the BM (or the CFs of the neurons responding most strongly in the auditory nerve).
42
2 postulates of the temporal theory of pitch perception
1. The time pattern of neural impulses reflects the frequency of the stimulus (phase-locking).
2. Perceived pitch is inferred from the time intervals between nerve spikes (so can only apply for frequencies up to about 4-5 kHz).
43
2 findings to support the temporal theory
1. Ability to detect changes in frequency worsens markedly for frequencies above 4-5 kHz
2. The sense of melody and of musical intervals is lost for frequencies above 4-5 kHz.
44
What is masking?
Situations in which the
threshold for one stimulus (the signal) is increased by another stimulus (the masker).
45
How is the selectivity of auditory channels estimated?
Using masking.
46
Fletcher (1940) masking experiment
Measured the threshold for detecting a sine wave in a white noise masker with varying bandwidth (the frequency ranges).
At first as bandwidth increased, it became harder to detect the tone (detection threshold increased). But after a certain point, the threshold flattened off.
47
Fletcher's (1940) explanation for his masking experiment
The signal is detected by monitoring the output of the single auditory channel centred at the signal frequency.
Noise bandwidth < auditory channel bandwidth: increase in noise bandwidth -> increase in noise passing through the channel so threshold increases.
Noise bandwidth > auditory channel bandwidth: increasing the noise bandwidth does not change the amount of noise passing through the channel.
48
What is the critical bandwidth of an auditory channel?
The bandwidth at which the threshold reached its asymptotic value. (Provides an estimate of the selectivity of the auditory channel).
49
How to measure the shape of an auditory channel (how it responds to different frequencies)?
Using psychophysical tuning curves (PTCs). Use a masker and determine the level required to just mask the signal.
Shows the inverted shape of the auditory channel centred at the signal frequency.
50
What 3 things affect the shape of the PTC?
1. Beats.
2. Combination tones from non-linear interactions.
3. Off-frequency listening.
51
What is the bandwidth for CFs above 1kHz?
About 12-13 % of the CF.
The bandwidth of the auditory channel increases with CF.
52
What are beats?
Adding 2 sinewaves of slightly different frequencies produces an amplitude modulation that has a
frequency set by the frequency difference between the components.
It doesn’t add additional frequencies to the stimulus, but modulates amplitude of the resultant waveform (so may detect beat rather than signal).
53
What are combination tones?
If two tones at frequencies f1 and f2 are played to the same ear simultaneously, a 3rd tone is heard. (At frequency 2f1-f2, provided that f1 and f2 are close in frequency and at similar levels.
54
What is off-frequency listening?
Participant may use a channel that is not centred at the signal frequency (so effect of masker is lessened).
55
How to sidestep notched-noise masking?
Notched-noise method (maskers on both sides of the amplitude/frequency curve).
56
What is the relationship between sound intensity and decibels?
Sound intensity is proportional to the square amplitude of the sound.
57
What are filters?
Selective auditory channels - a filter that lets through some frequencies but not others. Each filter corresponds to a specific place on the BM.
58
Passband
The frequency range over which the response of a filter is high. (Width = bandwidth).
59
2 measures of bandwidth
1) The difference between the two frequencies at which the response of the filter is 3 dB below the maximum response (the 3-
dB bandwidth, also called the half-power bandwidth since a change in level of –3 dB corresponds to a halving of power).
2) The equivalent rectangular bandwidth (ERB). The ERB of a filter is the bandwidth of a rectangular filter which has the same
peak transmission as that filter and which passes the same total power for a white noise input.
60
What does a 10dB increase correspond to in sound intensity?
A 10x increase in intensity and √10 change in amplitude.
61
What is the pitch of a complex tone?
Corresponds to the fundamental frequency (F0).
62
What is the phenomenon of the missing fundamental?
The pitch of a complex tone remains the same even if the fundamental component is removed.
63
Temporal theory of the pitch of complex tones
1. Pitch is derived from a place on the BM where harmonics are interfering (i.e. high harmonics - many frequencies per channel leading to complex but repeating waveform).
2. Nerve spikes are evoked by the largest peaks in the waveform on the BM. The most prominent time interval between spikes is assumed to determine the pitch.
64
Pattern recognition theory of the pitch of complex tones
1. Pitch is determined by a central processor
from info about the frequencies of
resolved harmonics (i.e. lower harmonics). Info could be on place (local peaks in the vibration pattern on the BM) or temporal (phase locking to individual harmonics).
2. The central processor tries to find a
fundamental frequency whose harmonics
would match those actually presented.
65
Evidence for temporal theory vs. pattern recognition theory
1. Support for pattern-recognition: a complex tone containing only low harmonics (but
excluding the fundamental component) evokes a clear pitch, but a complex tone containing only high harmonics has a much weaker pitch.
2. Support for temporal: a (weak) pitch can
be still heard even when only high unresolved components are present.
Both pattern-recognition and temporal processes might operate.
66
What is subjective timbre determined by?
The waveform characteristics of the sound (generally determined by relative phases and relative amplitudes of harmonics, but relative phase does not influence timbre).
67
Does relative phase influence timbre?
No (negligible).
1. Timbre is quite insensitive to our position in a room, even though position affects the phase relations of sound reflections.
2. Plomp & Steeneken (1969) confirmed that the effect of harmonic phase on the perception of timbre is negligible compared with the effect of harmonic amplitude.
68
How do we localise sounds (left/right)?
Comparing the sounds reaching each of our 2 ears.
69
2 types of inter-aural difference
1. Interaural time differences (ITDs): time delay between ears. Mainly for low-frequency sounds.
2. Interaural level differences (ILDs): loudness difference between ears. Mainly for high-frequency sounds.
70
Duplex theory of sound localisation
Using both interaural time differences and interaural level differences to localise sound.
71
How do we localise sound for up/down, infront/behind sounds?
1. Moving our heads and monitoring changes in ITD and ILD.
2. Information provided by pinnae (sounds can either enter directly or after reflection from pinna, which can cause interference effects that are unique for each direction in space).
Interaural differences do not indicate information on up/down or infront/behind.
72
Echolocation definition
The process of emitting calls and using echoes of those calls to navigate and locate nearby objects.
73
How good are humans at echolocating?
1. Distance threshold: the order of 10cm depth at a distance of 60cm.
2. Location threshold: the order of 4 degrees.
3. Size threshold: as small as 8 degrees at 75cm.
Echolocation can be learnt.
74
Milne et al. (2014) size constancy in echolocation
1. Acoustic angle of a sound-reflecting surface decreases as its distance increases.
2. Changes in size lead to changes in the level and spectrum of the echo.
3. Changes in distance lead to changes in the pulse-echo delay.
Nonetheless, there was accurate identification of small and large objects (at near and far distances). Size constancy demonstrated for expert echolocators (judge size regardless of distance).
75
6 Gestalt rules for the grouping of elements (perceived as coming from a single sound source)
1. Similarity.
2. Proximity in space.
3. Good continuation.
4. Common fate.
5. Closure.
6. Simultaneous sinusoids are hard as a single sound if their frequencies form a harmonic series.
76
Ways in which grouping works
1. Grouping across frequency of simultaneous frequency components.
2. Grouping across time of successive components.
77
What is the rule of similarity?
The elements in an auditory or visual scene will be grouped and perceived as one object or one sound source if they are similar.
Hearing: similarity in frequency for successive tones; similar timbre; speech sounds of similar pitch.
Vision: similar elements in shape and orientation.
78
What is the rule of proximity in space?
Elements in an auditory or visual scene tend to be grouped and perceived as one object
or one sound source if they are close together in space.
(Sounds from different locations are harder to group together across time than those from the same location).
79
What is the rule of good continuation?
If the elements in an auditory or visual scene form a simple pattern they tend to be perceived as one object.
Vision: elements in a visual scene that fall on a straight line or a smooth curve tend to be grouped.
Hearing: smooth trajectories in frequency (glides) often lead to grouping (rather than an inverted-U).
80
What is the rule of common fate?
If the elements change in a correlated way they tend to be grouped.
Vision: elements of a visual scene that move together/segregated from a background that is static or moving in a different way.
Hearing: sounds from a common source tend to start and stop at the same time, and common amplitude fluctuation promotes grouping of different (simultaneous) frequency components (e.g. vibrato).
81
What is the rule of closure?
If part of a scene is missing or obscured by other parts, our perceptual systems may “fill in” the missing part(s) if there is sufficient context from the remaining parts. Illusion of continuity.
Vision: illusory contours are often seen when one object obscures another.
Hearing: a tone that is alternated with intense bursts of noise may be perceived as continuing through the noise.
82
What can act as a constrain for our perceptual system in speech production?
Regularities in stimulus - speech production is constrained by anatomy.
83
Spectrograms
Shows the pattern of sound in frequency and time. Take short samples of sound to determine the frequency content. (x = time; y = frequency; light/dark = energy).
84
What is the segmentation problem?
Individual words have to be derived from the continuous speech stream.
85
What is the problem of lack of invariance?
The relationship between between the acoustic characteristics of speech and the sounds that are perceived is highly variable.
86
What are the 4 sources of variability in speech?
1. Coarticulation. (Preceding and following sound affect the characteristics of current speech sound).
2. Poor articulation. (People speak sloppily).
3. ID between speakers. (Accents, pitch).
4. Reflections from wall and background noise.
87
Additional information to help understand speech
1. Words of our own language (lexical rules).
2. Sequences in which speech sounds can occur in our own language and the possible sounds that are used to convey meaning in our own language (phonetic rules).
3. Grammatical rules of our own language.
4. Context of conversation
5. Meanings of words that have already been interpreted.
6. Probability of occurrence of different words in our own language.
7. Accents of familiar talkers.
All of these sources of knowledge are used together with the acoustic information to arrive at an interpretation of an utterance.
88
What is phonemic restoration?
Sometimes, we hear a speech sound even when it is completely removed from a sentence and replaced with a cough or burst of noise (using context).
