Lec 9/ TB Ch 9&10

Question 1

How is sound created?

direction of motion

W/ a tuning fork, how can we change the amplitude/freq

Answer 1

What is sound?

• Sounds are created when objects vibrate
  
  • Vibrations of object cause molecules in object’s surrounding medium to vibrate as well, which causes pressure changes in medium

Ex. membrane of speaker pushes on sound molecules, causes increase and decrease of air pressure, which in turn propagate across this “tube”

• Longitudinal direction
  
  • Motion of air molecules is longitudinal
  • Speed of sound depends on density of medium
    
    • Air: ~340 m/s; water: ~1500 m/s
    • L vs R
      
      • Amplitude is the same
      • RS: more frequent

Frequency and amplitude

• We can create sounds using tuning fork
  
  • If we hit the tuning fork harder, (wave 2), the amplitude double, frequency is the same
  • If we have a smaller tuning fork, amplitude is the same as original by freq is doubled

Question 2

• amplitude
• intensity
• decibels
• 6 dB
• loudness
• What is the ratio b/w faintest and loudest sounds
  *

Answer 2

What is sound cont?

Basic qualities of sound waves

• Amplitude: Magnitude of displacement of a sound pressure wave
• Intensity: Amount of sound energy falling on a unit area
  
  • Measured in decibels sound pressure level (dB SPL)
• Decibels: Ratio between the pressure of some sound and the pressure of a reference sound p0 (~minimum energy).
  
  • dB = 20*log(p/p0); e.g. 20*log(2) = 6
    
    • need to find the ratio p to po
    • Ex. current sound is 2x as much energy than reference sound -> log (2)
    • 6 dB = something has 2x as much energy than reference sound
    • 12 dB = 2x as much energy than 6 dB
    • 18 dB = 2x as much energy than 12 dB
• Loudness: The psychological aspect of sound related to perceived intensity or magnitude

Intensity of env sounds

• Humans can hear across a wide range of sound intensities
• Ratio between faintest and loudest sounds is more than one to one million
  
  • Ex. 120 loudest, 0 faintest
  • Every 6 dB = doubled energy
  • 120 dB /6 = 20 times (the sound doubled 20 times)
  • 2^20 = 1 million
• 6 dB = double amount of pressure
• 20 dB = falling leaves
• 30 dB = bedroom
• 50 dB = wind turbine
• 60 dB = home
• 70 dB = office sounds
• 90 dB = inside car

Question 3

• freq
• pitch
• amplitude & freq interaction grph
  
  • white = ?
    
    • L vs R
  • light blue = ?
    
    • L vs R
    • Top vs bottom
  • dark blue = ?
  • black line
• What other graph does it resemble?
• blue graph
  
  • pain threshold
  • high risk threshold

Answer 3

What is sound? Cont

• Frequency: For sound, the number of times per second that a pattern of pressure change repeats
  
  • Measured in 1/s = Hz
  • Frequency is associated with pitch
• Pitch: Psychological aspect of sound related to the fundamental frequency (low: tuba, high: piccolo).
• (Phase)
  
  • Recall sine waves for vision
  • It is related to Fourier transform
  • Return to this later
• X
• Interactions between amplitude and frequency
• Human hearing uses a limited range of frequencies: From about 20 to 20,000 Hz
  
  • Vertical axis: sound pressure lv SPL (dB)
  • X-axis: frequency (Hz)
  • White = speech bubble
    
    • Normal range of speech (whisper and yelling)
      
      • Left = male adult
      • Right = bb
      • Everyone else is in b/w
  • Light blue = music instruments
    
    • L: low voice music; R opp
    • Top: loud, bottom: soft
  • Darker blue: audible range
    
    • Black line = absolute threshold
• This is similar to the contrast sensitivity fx, but we need to flip the graph upside down (so the stuff we cannot hear/see is at the bottom)
• 
  
  • For spatial freq: 7 cycles/degree is the best
  • For temporal freq: 3000-5000 Hz (hearing is the best)
• Back to blue graph
• Audition: cannot hear freq less than 20 Hz, are those greater than 20k Hz
• Upper part of blue bubble
  
  • There’s so much energy that you can’t perceive sound and only perceive pain: “pain threshold”
  • “high risk threshold”: your are going to damage your ears

Question 4

• what analysis can explain complex sounds?
• spectrum = ?
• Fourier analysis graph
  
  • top graphs 1&2
  • graph 3
  • graph 4

Answer 4

Sine waves and complex sounds

One of simplest kinds of sounds: Sine wave, or pure tone

• Sine wave: Waveform for which variation as a function of time is a sine function
• Most sounds in world: Complex sounds, (e.g., human voices, birds, cars, etc.)
• All sound waves can be described as some combination of sine waves

Spectrum

• Complex sounds can be described by Fourier analysis
  
  • Spectrum: A representation of the relative energy present at each frequency
  • LS graph: result of Fourier analysis
    
    • Y-axis: how much energy
    • X-axis: freq
    • Top graphs 1&2: pure tone = sine wave, it has only 1 freq -> there’s only 1 bar
      
      • There’s x amount of energy for freq x only
      • Graph 2: higher frequency
        
        • Yellow graph, the bar is shifted to the right
      • Lower amplitude
        
        • The bar is shorter (less energy)
    • Graph 3: complex sounds w/ 2 sine waves
      
      • Yellow graph: 2 bars (same as b4)
      • IOW: this complex sound is composed of 2 pure tones
    • Graph 4:
      
      • Need infinite # of sine waves -> a bunch of bars in the yellow graph

Question 5

• characteristic in square graph
• water in bucket analogy
  
  • what happens when rock drops in it
  • How is it similar to guitar string
  • modes 1-5
    
    • wavelength and length relationship
    • Connection to freq

Answer 5

Sine waves and complex sounds

• Square graph abv: in it’s yellow graph, we see the bars are separated by the same amount of distance
• Ex. drop a stone, waves propagate to the boundaries of the bucket, then travels back and forth within the boundaries (this is similar to a guitar string)
• • Guitar string plucked: the string’s displacement moves left & right, ‘travelling waves’
    
    • Waves reflected at ends; waves travel back and forth
      
      • a guitar string is mounted on both ends
      • When we pluck the string, the waves propagate to the boundary, then back (just like the waves in the bucket)
      • When we add everything together -> we get a standing wave
    • Left & right travelling components add to a standing wave
    • The standing waves can be decomposed (like fourier transform)
      
      • We get Modes
      • Mode 1:
        
        • The wavelength is 2x longer than the length of the string
        • String = 10 cm
        • Wavelength = 20 cm (there’s only half of the wave on mode 1)
        • 2/1 x L = wavelength (wavelength is 2x the length of the guitar string)
      • Mode 2: 2/2 x L = wavelength (the wavelength = the length of the string)
      • Mode 3: 2/3 x L = wavelength
      • Mode 4: 2/4 x L
      • Mode 5: 2/5 x L
      • Freq = 1/ wavelength
        
        • NOTE: 1/L = G
        • Then we cancel the ½ -> F
        • The frequencies are whole integer multiples
          *

Question 6

• Harmonic spectrum & its components
• Fundamental f
  
  • aka?
  • 1st, 2nd, 3rd, 4th harmonic
• top graph
• graph 2
• bars 1&2
• Tiimbre

Answer 6

Complex sounds

• Harmonic spectrum: Typically caused by simple vibrating source, (e.g., string of guitar).
• The frequencies of its components are integer multiples of lowest frequency.
• Fundamental frequency: Lowest frequency component of a sound (aka 1st harmonic)
  
  • 1^st harmonic = 262
  • 2^nd harmonic = 524
  • 3^rd harmonic = 786
  • 4^th harmonic = 1048 Hz
• Graphs show the results of fourier analysis
  
  • Top graph: pure tone only has 1 freq
  • Graph 2: use tenor saxophone to play the same pitch as the pure tone
    
    • Bar 1: same as pure tone (aka fundamental freq or 1^st harmonic)
      
      • C4 on piano = 262 Hz
    • Bar 2: 2^nd harmonic
      
      • Notice the distance b/w 0 to 1^st harmonic, and 1^st to 2^nd harmonic is the same
      • This shows that the 2^nd harmonic is an integer multiple of 1^st harmonic
    • NOTE: across the instruments, the 1^st and 2^nd harmonic are the same
    • But their indiv profile is different across the instruments
• Timbre: Sounds w/ same pitch and loudness may still sound different (piano vs. guitar)

Question 7

• Auditory transduction
• overall path (4 parts)
• Main purpose of canal

Answer 7

Video

• Auditory transduction: ear converts sound waves into electrical impulses, which are interpreted by the brain
• Sound -> external auditory canal -> tympanic membrane
  
  • Tympanic membrane vibrates
    
    • High A = more vibration; Lower A = less vibration
    • High f = faster vibration
  • Tympanic membrane -> vibrate the ossicles, it has 3 bones (malleus, incus, stapes)
    
    • This passes info of freq and A
• pinna → ear canal → Tympanic membrane → ossicles

Basic structure of mammalian auditory system

Outer ear:

• Sounds are first collected from environment by the pinna
• Sound waves are funneled by the pinna into ear canal
• Length and shape of ear canal enhance sound frequencies
• Main purpose of canal is to insulate structure at its end: Tympanic membrane, vibrates in response to sound

Question 8

Middle ear

• problem: someone is yelling you but you are under the water → So how do we get sound from air through fluid in the cochlear?
• Why do the bones work?
  
  • define lever principle & application
  • define SA & application
• middle ear - 3 bones
• Purpose of muscles b/w these bones (esp when they are stiff)

Answer 8

Middle ear

• # 1: what is the problem w/ audition?
  • we need to hear what others are saying
  • Sound travels and enters our ear, and needs to be transformed into neural energy in the cochlea
  • The cochlea has fluid, that is higher density compared to air
  • Scenario: someone is yelling at you, but you are diving in water
    
    • It is difficult to hear what the person is yelling about b/c most of the sound energy will simply bounce off the surface of the water
    • So how do we get sound from air through fluid in the cochlear? 3 bones in middle ear
    • • The 3 bones are derived from gills of some odd fish back then
• Step 2: physics refresher
  
  • How do the bones work
  • Thought exp 1:
    
    • Man trying to lift a rock
    • He uses a lever
    • Lever: can help create more force
    • The middle ear uses this principle
    • • Though exp 1:
    • When person applies X amount of pressure on top of your hand, it isn’t too painful
    • Then the person applies X amount of pressure (via a pen) on top of your hand, it will hurt
    • Rationale: Surface area
      
      • SA of pen is smaller than that of a palm -> stabbing is more painful
• Step 3: solution
  
  • SA applies to sound
    
    • Tympanic membrane & ossicles = applying force via a pencil
    • It helps focus the energy onto the oval window (which has a smaller SA as well)
  • Lever principle
    
    • Ossicles are arranged similar to a fulcrum/lever ->
    • This explains how sound energy is increased and travel from air to liquid despite an increase in energy
• Middle ear: 3 ossicles: malleus, incus, stapes;
  
  • smallest bones in body
  • Middle ear helps Enhance sound (this is based on the lever mechanism + focusing pressure on smaller area)
  • Stapes (last of ossicle) transmits vibrations of sound waves to oval window
  • Loud sounds: muscles
    
    • Stiffen the middle ear
    • Ex. when you hear a loud sound, the muscle stiffens, so the sound is less loud
    • No need to worry about the muscle names
      *

Question 9

inner ear

• define cochlea
• the 3 canals
• which canal is surrounded
• organ of corti
  
  • basilar vs tectoriical membrane: where is it relative to organ of corti?
• which canal is connected to oval window
• which canals are separated by basilar membrane

Answer 9

Inner ear

• Inner ear – contains the cochlea
  
  • LS: whole cochlea
  • RS: cut through the whole structure
    
    • There are 3 canals
    • They are separated by membranes
    • So stapes hit the oval window -> sounds travels through cochlea (the tubes) and reaches the apex; if there is enough energy, the vibrations (rewinds) and hits the round window
    • • Fine changes in sound pressure are translated into neural signals
• Cochlea (snail): oval/round window, three canals

Cochlea

Cochlear canals and membranes

• Cochlea: Spiral structure of the inner ear containing the organ of Corti (it sits on 1/3 membranes)
  
  • Cochlea is filled with watery fluids in three parallel canals (middle, vestibular, tympanic)
    
    • Middle canal is surrounded by the Vestibular canal + tympanic canal
    • Canals are separated by membranes (ex. basilar membrane)
      
      • Organ of Corti (converts sound to neural signals) sits on top of basilar membrane, covered by tectorial membrane
• Image: cross section of cochlea
  
  • Vestibular canal (base of cochlea) is connected to the oval window
  • Sound pressure travels through liquid in vestibular canal -> hits the apex of the canal -> if there’s enough energy, sound travels back and hits the round window
  • Membranes separate the canals: Reissner’s membrane, and basilar membrane
  • Basilar membrane*: separates middle canal and tympanic canal
  • Organ of corti: on top of basilar membrane
  • Tectorial membrane: on top of organ of Corti

Question 10

• pathways of extremely intense sounds (7 parts)
• specialized neurons in organ of corti
• What terminates at the base of hair cells?
• Purpose of tectorial membrane
• Arrangement of hair cells - how many rows?
  
  • inner hair cells: purpose? # of rows?
  • outer hair cells: purpose? # of rows?
  • What can outer hair cells do additional do?
  • aka??? in stirring wheel
  • 3 steps
• Vibration and tectorial membrane
  
  • resting position
  • sound-induced vibration
    
    • 2 steps
    • upward phase
      
      • which direction is the shearing force?
      • What does this cause?
    • downward phase
      
      • which direction is the shearing force?
      • What does this cause?
        *

Answer 10

Basic structure of the mammalian auditory system cont

• Vibrations transmitted through tympanic membranes and middle-ear bones cause stapes to push and pull flexible oval window in and out of vestibular canal at base of cochlea
  
  • If sounds are extremely intense, any remaining pressure is transmitted through helicotrema and back to cochlear base through tympanic canal, where it is absorbed by another membrane: Round window

Organ of Corti

• Sound waves transformed into movements of the ossicles …
• Movements of cochlear partition which then are translated into neural signals by structures in the organ of Corti; extends along top of basilar membrane (these 2 structures extend to the apex)
• Made up of specialized neurons called hair cells (stereocilia), dendrites of auditory nerve fibers that terminate at base of hair cells
• Also made up of scaffold of supporting cells

Cochlea

• Tectorial membrane: Extends atop organ of Corti; gelatinous structure
  
  • It brushes through the hair cells/stereocilia (via shearing forces)
• Hair cells in each human ear: Arranged in four rows that run down length of basilar membrane
• Inner hair cells: 1 row, afferent auditory information
• Outer hair cells: 3 rows, efferences, feedback system
  
  • Can contract itself, creates own mechanical force
  • Service system
    
    • 1: tectorial membrane brushes outer hair cells
    • 2: Outer hair cells are stimulated, it contracts and relaxes to create more shearing force
    • 3: cause inner hair cells to be even more stimulated
  • Similar to steering wheel (there’s another system that detects steering and amplifies it)

Vibration and tectorial membrane

• Resting position: basilar membrane horizontal
• Sound-induced vibration: sound enters, and cause basilar membrane to move (up and down = upward and downward phase)
  
  • Upward phase: basilar membrane is pushed against tectorial membrane, this causes a shearing force outwards and bends the stereocilia -> inhibition of inner hair cells
  • Downward phase: vv; causes shearing force inwards -> activation of inner hair cells
  • IOW: mechanical force of sound is converted into neural signals

Question 11

• 2 ways freq is encoded
• # 1: place code definition?
  • basilar membrane: freq @ base vs apex
  • base vs apex: Thickness & width of basilar membrane
  • How is it similar to visual system
  • x
  • The auditory nerve (AN) & cochlear partition
  • Frequency selectivity - when is it the clearest?
  • Threshold tuning curve

Answer 11

Basic structure of the mammalian auditory system cont

• Coding of amplitude and frequency in the cochlea
  
  • Amplitude: pushed up and down a lot -> lots of shearing force; vv
• 2 ways freq is encoded (place code, and temporal code)
• # 1: Place code: Different parts of cochlea tuned to different frequencies; information about the frequency of an incoming sound is coded by place along cochlear partition, i.e. at the location with greatest mechanical displacement
• 
  
  • Cochlea: base -> apex (& helicotrema)
    
    • Basilar membrane
    • Base: high f; apex: low f
  • This has to do with the Thickness & width of basilar membrane
    
    • Base: membrane is narrow and thick -> resonate higher f
    • Apex: membrane is wider and more floppy -> resonate low f
    • IF you have several pure tones -> stimulate diff parts of basilar membrane (it looks similar to a Fourier analysis) -> stimulate organ of corti
  • Similar to the visual system
    
    • Light coming from different directions fans out on the retina
    • Light from LS goes into one part of the left eye, and another part in the right eye?
• Inner hair cells are connected to auditory nerve fibers
• The auditory nerve (AN)
  
  • AN fibers sensitive to certain frequencies depending on their place along the cochlear partition (base vs apex)
  • Frequency selectivity: Clearest (can see clear results) when sounds are very faint
  • When we use faint sounds, we can create a threshold tuning fx
  • Threshold tuning curve: Map plotting thresholds of a neuron or fiber in response to sine waves with varying frequencies at lowest intensity that will give rise to a response
  • Characteristic frequency

Question 12

• based on the 6 graphs, what type of hair cells are they connected to?
• LS top vs RS (bottom)
  
  • which part of the cochlea is it connected to?
  • What type of freq is it sensitive to?
• y-axis?
  
  • higher vs lower values?
• x
• Two-tone suppression
• Rate saturation
• isointensity curves
• what the red curve means
• purple curve
• Does frequency selectivity of AN fibers change with intensity?

Answer 12

Threshold tuning curves

• LS (top): auditory nerve fiber connected to an inner hair cell at the cochlea’s apex
  
  • very sensitive to low f, AN fiber fires even when that freq has very little energy
• RS (bottom): auditory nerve fiber connected to an inner hair cell at the cochlea’s base
  
  • Very sensitive to high f
• Other graphs: in b/w
• Y-axis: threshold intensity
  
  • Higher vales (ex. 80): not so sensitive
  • Lower vales (ex. close to 0): very sensitive
• X-axis: f
• NOTE: This is the case for pure tones
• X
• For complex sounds (multiple tones)
• Two-tone suppression: Decrease in firing rate of one auditory nerve fiber due to one tone, when a second tone is presented at the same time
  
  • Due to mechanical properties of the basilar membrane
  • When tone 1 is presented -> displaces the basilar membrane
  • When tone 2 is also presented, the basilar membrane is already displaced
• Tuning curve, showing sensitivity at 8 kHZ
  
  • LS bubble: tone 2
  • RS bubble: tone 1
• X
• Rate saturation: Point at which a nerve fiber is firing as rapidly as possible and further stimulation is incapable of increasing the firing rate
  
  • Isointensity curves: Chart by measuring an AN fiber’s firing rate to a wide range of frequencies, all presented at same intensity level (aka all sounds are played at same intensity)
  • Figure diff isointensity curves
    
    • # 1: red curve (faint tone)
      • play a bunch of pure tones (low and high f), all the tones have the same amount of energy (20 dB)
      • depending of freq, the rate of AP (aka discharge rate = y-axis) differs
      • It peaks at around 200 Hz
    • # 2: purple curve; play a bunch of pure tones @ 40 dB (more energy)
      • Here, the amplitude of the curve increased (makes sense)
      • For the purple curve, the peaks are wider than that of the red curve
    • # 3 & #4: the peaks for 60 dB, 80 dB are even wider than that of the 20 dB & 40 dB curves
      • We also see that the amplitude no longer increases for the 60dB and 80dB curves -> rate saturation (or ceiling effect)
      • Here, the neurons cannot fire faster than 250 spikes/s
  • Does frequency selectivity of AN fibers change with intensity?
    
    • Yes
    • When the sounds are louder (more intenset), we get broadening of frequency selectivity

Question 13

Temporal code for sound freq

• phase locking
• what does the top graph indicate?
• What does the 2nd graphs indicate?
• Based on the red lines, why do neurons skip some peaks?
• What does the bottom graph mean?
• x
• Temporal code
  
  • what freq it best works for?
• volley principle
  *

Answer 13

Temporal code for sound freq

• Phase locking: Firing of a single neuron at one distinct point in the period (cycle) of a sound wave at a given frequency.
• – Volley principle
• Top graph: sound-pressure wave of a pure tone
• Bottom 2 graphs: graphs that show how the AN fibers respond (APs)
• Note: the APs align with the sound-pressure wave (top) (red lines) -> Phase locking
• IOW: the APs occur at the peak of the sound-pressure wave
  
  • So APs correspond to the freq of sound pressure wave -> temporal code
• NOTE: the neuron skips some peaks in b/w
  
  • This happens more often when the sound pressure wave has a high freq
  • Why? Neurons have a refractory period and cannot fire that often
  • IOW: you can’t really tell the freq of sound based on the APs of 1 single AN fiber
  • But in our ear, we have groups/populations of auditory nerve fibers, and they are all stimulated by that particular freq
    
    • So when one AN fiber takes a break, others do not
    • In essence, it works in teams -> volley principle
  • Bottom graph: firing rate of a population of AN fibers that are connected to the same part of the basilar membrane
    
    • We see that the overall pattern of APs reflects the original freq in the sound-pressure wave (top graph)
• X
• Temporal code: Tuning of different parts of the cochlea to different frequencies, (Low f stimulate apex of cochlear; vv) in which information about the particular frequency of an incoming sound wave is coded by the timing of the firing of one or multiple neurons (volley principle) as it relates to the period of the sound
  
  • **Works for frequencies less than 1000 Hz
    
    • It is important to allow us enjoy music (next wk)
• The volley principle: multiple neurons can provide a temporal code for frequency if each neuron fires at a distinct point in the period of a sound wave but does not fire on every period
  
  • Phase locking happens, but the neuron does skip some phases
  • Since you have a pop of neurons, you have volley principle

Question 14

• Cochlea - 3 components
• pathway - 7 parts
  
  • Visual pathway vs auditory path
    
    • inferior colliculus vs superior colliculus
    • geniculate nucleus

Answer 14

Auditory system pathways

• Cochlea (has basilar membrane, organ of corti and inner hair cells), inner hair cells are connected to AN fibers (which form auditory nerve) -> AN fibers connect to cochlear nucleus -> connect to superior olive on both sides -> inferior colliculus on both sides
  
  • Recall Primary visual pathway: inferior colliculus = superior colliculus, superior colliculus has a main role in visual input, but it is not part of the primary visual pathway
  • In contrast, the inferior colliculus is part of the auditory pathway
• -> inferior colliculus on both sides -> thalamus (medial geniculate nucleus/complex)
  
  • Visual system: it uses the lateral geniculate nucleus
  • Auditory: uses the medial
• -> thalamus (medial geniculate nucleus/complex) -> primary auditory cortex
• Auditory brain structures
  
  • AN (cranial nerve VIII) carries signals from cochlea to brain stem
  • There, all AN fibers initially synapse in cochlear nucleus
• Superior olive, inferior colliculus, and medial geniculate nucleus all play roles in auditory process
  *

Question 15

• for A1, where does it receive input from
• 2 parts A1 is surrounded by
• Tonotopic & retinotopic organization - how does it grow?
  *

Answer 15

Basic structure of mammalian auditory system (cont)

• TOP: RH of human brain
  
  • Black oval = temporal cortex
  • Blue = auditory cortex
  • Primary auditory cortex: on top of temporal cortex but tucked away; below the parietal cortex (triangle)
  • If you vertically cut away the top of parietal cortex -> see primary auditory cortex (A1) in green that receive input from medial geniculate nucleus
  • A1 is surrounded by belt and parabelt regions
• A1 has tonotopic organization
  
  • Recall: visual system has retinotopic organization
  • This has to do w/ dev of embryos
    
    • Retina has ganglion cells coming out of it
    • The axons grow along growth cones
    • They grow in parallel and make connections with the lateral geniculate nucleus
    • Since the ganglion axons grow in parallel and connect w/ the neurons in the LGN, and the LGN connect to the primary visual cortex and the cerebral cortex
      
      • Since these happen all in parallel, the topography of the retina is preserved all the way through in the LGN, area V1, V2, V3
  • This is also the case in the auditory system
    
    • AN fibers grow in //, tonotopic org extends from basilar membrane to A1
    • Ex. in auditory cortex, there is hAI, hR
      
      • Has tonotopic organization
• Tonotopic organization: An arrangement in which neurons responding to different frequencies are organized anatomically in order of frequency
• frequency composition determinant of how we hear sounds
  
  • Maintained in A1
  • A1 neurons -> belt -> parabelt area

Question 16

• psychoacoustics
• audibility threhols/absolute threshold
• equal loudness curves
  
  • how is method of estimation used here
  • describe the points on @10
  • orange marks
  • blue marks
• Why is distance b/w @10 & 20 curve = distance b/w @60 & 70?
• How is this related to Fechner’s law?
• How does duration of sound impact perception of loudness?

Answer 16

The function of hearing

• How we perceive loudness and pitch
  
  • What is psychoacoustics?
  • What is the difference btw intensity and loudness?
• Psychoacoustics: The study of the psychological correlates of the physical dimensions of acoustics; a branch of psychophysics
• Audibility threshold: A map of just barely audible tones of varying frequencies
  
  • Aka absolute threshold (bottom black line/ red line in the graphs)
• RS: graph also have a bunch of blue curves -> equal loudness curves
• Subjectively perceive sounds to be equally loud
• Ex. method of estimation
  
  • On a scale of 1-10 how painful is the headache -> on a scale of 1-100, how loud is this sound
  • We can see along which combination of frequency and sound levels do ppl report the same number
  • Ex. @10 (magnitude estimation value = sound is very faint)
    
    • 4500 Hz (5dB) = 20 Hz (60 dB)
  • Ex. orange marks
    
    • 1000 Hz (50 dB) & 100 Hz (60 dB)
    • They lie on the same equal loudness curve (@50), but don’t have the same energies
  • Ex. blue marks
    
    • Both have 80 dB, but we perceive them as having diff loudness
      
      • 4500 Hz is perceived as louder than the 1000 Hz one even though both are 80 dB, this is b/c we detect 4500 Hz better than 1000 Hz
• Equal-loudness curves (note the roughly regular distances thanks to dB)
  
  • i.e. distance b/w @10 & 20 curve = distance b/w @60 & 70
  • Why? The graph is log transformed
    
    • Loudness follows Fechner’s law: log relationship b/w how loud you perceive (based on the freq) and the energy
      
      • When energy increases, loudness increases on a logarithmic fashion
      • When you convert the vertical axis to log -> steps will look equal
    • More physical intensity (sound lv, dB) -> increase in subjective perception of loudness
    • There’s also an interaction w/ freq (swiggly lines)
• Sound played at a constant level is perceived as being louder when it is of greater duration
  
  • Ex. gun shot is extremely loud; but since it is so brief, we don’t perceive it as loud as it should be
  • indicative of temporal integration.

Question 17

• 3 main causes of hearing loss
• Example of ear canal
• conductive hearing loss
  
  • What is it
  • what doe MD do to relieve this
  • otoscelerosis
• Sensorineural hearing loss
  
  • what is it
  • cause
• common hear loss cause
• Easter island study
  
  • results

Answer 17

Hearing loss

• Hearing can be impaired by damage to any of structures along chain of auditory processing
  
  • –Obstructions of the ear canal
    
    • Ex. you go swimming, water enters your hear -> muffles hearing
    • curable
  • Conductive hearing loss: ossicles & tympanic membrane (more serious)
    
    • e.g., ear infections (ex infect middle ear)
      
      • MD make a cut in tympanic membrane -> release puss
    • Otosclerosis: abnormal growth of ossicles; can be remedied by surgery
• Can be more serious
• Sensorineural hearing loss: Common, serious auditory impairment. Due to defects in cochlea or auditory nerve; when hair cells are injured
  
  • Drugs can cause deafness: (antibiotics or cancer drugs)
• Common hearing loss: Damage to hair cells due to excessive exposure to noise
  
  • Young (15 yo): Range of 20–20,000 Hz
  • Old (≥ college age): 20–15,000 Hz etc.
• Easter islanders: place that is very quiet
  
  • Graph: hearing ability of those who stayed in easter island vs those who left
  • Y-axis: how good your hearing is (higher = better)
  • X-axis: f
  • Those who never left = good hearing
  • Those who left for 3-5 years and returned -> hearing declined
  • Those who left for 5+ years -> hearing even worse
  • Those in Toronto -> hearing is shit
    *

Question 18

• echolocation
• Soundscapes with the vOICe -
  
  • how does it work
  • 4 parts of training

Answer 18

Hearing light

• Echo location: send out sounds (clicking tongue) -> use the echoes to locate objects
  
  • Dolphins and Ben Underwood (can use echo location)
• Soundscapes with the vOICe
  
  • Has video camera, converts those signals into sound that are played through headsets
  • # 1: vertical scan the room
  • # 2: Objects higher up in the images are converted into high f; vv
  • Ppl can learn to locate and grasp objects using this system
  • X
  • Training
    
    • Start w/ an image (LS) w/ 1 dot -> listen: you hear 1 blip
    • Next image (RS): has 3 dots
      
      • Dot 1: lower on the visual field -> lower blip sound
      • Dot 2: higher on the visual field -> higher blip sound
      • Dot 3: in b/w
    • Training w/ 1 line (you hear wheeeeeet)
      
    • Training w/ 2 lines (you hear 2 wheeeeets)
      
    • Training w/ gratings (you hear constant rhythm of noise bursts)
      

Question 19

• 5 Sound Localization Cues
• Chirping cricket example
  
  • how is it different to localization of vision
  • how is it similar
• 2 things that are essential to determine audit location

Answer 19

Hearing the env

Sound Localization Cues

• Interaural time difference
• Interaural level difference
• What’s a cone of confusion?
• Cues from the pinna and the head
• Auditory distance perception

Sound localization

• How do you locate a sound? (sorta similar to localization via vision)
  
  • Chirping cricket example
    
    • See 2 crickets on the ground A and B
    • You hear them chirping
    • You determine that cricket A on ur left is chirping
• This Problem is dissimilar to determining visual direction
  
  • To tell that cricket A is on the L, light from cricket fall on the red part (red dashed arrows) of the retina compared to cricket B
  • For audition, it is computed different b/c the sound from cricket A travels and enters both ears
    
    • WE can tell the direction of sound based on the time difference
    • Sound of cricket A is closer to left ear -> sound of cricket A arrives to left ear microseconds earlier compared to right ear (temporal disparity)
• Recall: we use binocular disparity for stereovision
• Sensing direction of sound is a problem similar to determining how far an object is based on vision (how we perceive depth) several cues
  
  • Ex. retinal disparity is one binocular depth cue for stereovision
  • There are several visual depth cues; we also have several auditory depth cues
• 2 ears (& their shape): Critical for determining audit. Locations

Question 20

• ITD
  
  • what is it?
  • Azimuth?
  • 0 vs 180 deg
    
    • implication of distance the sound travels
  • sounds coming at an angle
    
    • what happens

Answer 20

Sound localization Cues

• Cue 1: Interaural time difference (ITD): The difference in time between a sound arriving at one ear versus the other
  
  • Azimuth: Used to describe locations on imaginary circle that extends around us, in a horizontal plane
    
    • 0 deg = in front
    • 180 deg = directly behind
  • Sound coming directly in front of us will have to travle equal distance to the L and R ear
    
    • IOW: the difference is = 0
    • The distance doesn’t matter
  • For sounds that are coming in at an angle (ex. 60)
    
    • It reaches the left ear faster than the right ear (by 480 micro s)
    • NOTE: if your head is smaller, the time difference is also smaller
    • NOTE: the distance of the object doesn’t matter (1cm vs 40 km) if the angle is constant (ex. 60 deg)
    • NOTE: if sound is coming from the L = -ve sign; vv

Question 21

• ILD
  
  • what is it
  • why does the incoming angle of sound doesn’t matter as much in sound intensity?
  • Define “sound has shadows”
  • Which ear is sound more intense for?
  • What degree is ILD largest?
  • What special property can sound waves do when it comes to obstacles?
    
    • This is particularly true for?
    • So ITDS is best for what type of f?
• What is the limitation of ITD and ILD?
• Cone of confusion
• Example w/ ITD 60 deg, 120 deg
• How many cones of confusion do we have?
• Where is the biggest cone of confusion?
• 2 ways to overcome cone of confusion
  
  • what happens when you move your head?
  • Frog case
  • Why do Sound from ear phones seems to come from inside the head?
  • What do the ridges of our pinna do?
  • How can we simulate the effects our pinna do?
• Head-related transfer function (HRTF):

Answer 21

• Cue 2: interaural level difference: The difference in level (intensity) between a sound arriving at one ear vs. the other
  
  • Ex. If the sound source is located at an angle (ex 45 deg), it is closer to the left than the right ear
    
    • But that doesn’t make a huge difference in sound intensity
    • NOTE: sound has shadows
      
      • Here, the head is obstructing how sound travels
  • • -> Sounds are more intense at the ear closer to sound source (this is b/c the head is in the way)
      
      • ILD is largest at +/-90 degrees
        
        • Aka 90 deg of the LS or RS
      • ILD related to angle of sound source
      • In the image, the sound waves can wrap around the head/obstacle, this is especially the case for low f
      • IOW: ITDs works best for higher f
• X
• The prev 2 cues (ILD and ITD) are based on the fact that the ears are mounted on both sides of the head
• Limitation: cone of confusion
• Cone of confusion: Regions of positions in space where all sounds produce the same time and level (intensity) differences (ITDs and ILDs)
  
  • Ex. for ITDs
    
    • Sound coming at 60 deg & 120 deg from the left -> both cause ITD = 480 micro s
    • This is b/c they both lie on the same cone
      
      • This is not only true on the azimuth plane, it also happen in other heights (on any point of the cone)
      • IOW: on any point of the cone, a sound source located there can create identical ITD
    • • We have many cones of confusion
        
        • Ex. one for -60 and -120 (- 480 micor s)
        • Ex. another one for -20 and -160
        • Ex. another one for -45 and -135
    • • The cones are stacked
  • The biggest cone of confusion is b/w 0 and 180 deg, where both ITD = 0 micro s
  • • If you move your head, the cones of confusion moves with you
  • Since the sound sources do not move with your ear, this means that the sound source b4 you turn your head is on a cone of confusion; when you move your head, the cone of confusion for that sound source changes
  • IOW: the cones of confusion are different b4 you move your head vs after
  • IOW: head motion is a way to overcome cones of confusion
• Another way: Pinna

Solution 1 for cone of confusion – move your head

• Moving the head rules out the green frog
• LS: blue frog = 60 deg; green frog = 120 deg
  
  • For both frogs, the ITD = -480 µs
  • IOW: in this case, we cannot be fully sure which frog is crooking
• • If you turn your head, the blue frog is on a cone of confusion, green frog is on another cone of confusion
    
    • IOW, if the green frog is crooking, we know it is NOT from the blue frog
      
      • If the blue one is crooking, we may mess it up w/ the red frog
• • So when 2 sound sources are sitting on the same cone of confusion, once you turn your head, you can determine which is the actual sound source
• X

Solution 2 for cone of confusion – pinna

• Sound from ear phones seems to come from inside the head. Why?
  
  • Even though some music are recorded by 2 microphones (simulating one ear on each side of your lead) to create ITD and ILD, the music still seems like it is coming from inside your head -> why?
    
    • wrong ITD? – no
    • wrong ILD? – no
    • sound source moving with the head? – not really
  • This is b/c the recording cannot simulate our pinna (each person has their own unique set of pinnas)
  • Our pinnas has different ridges, this causes sound to bounce to around and echoes -> certain freq are enhanced or reduced
  • All of this also depends on the direction of the sound
  • IOW: This causes changes in timbre, and timbre also depends on the direction of the sound and shape of pinna
  • If you have a microphone with an artificial pinna similar to your’s, this can simulate the sound alterations caused by your own pinna (hair cut recording)
• If we record music w/o those pinna modifications, when we listen to music via those headphones, it sounds like music is coming from inside your head
• X
• Shape and form of pinna helps determine localization of sound
  
  • Head-related transfer function (HRTF): Describes how pinnae, ear canal, head, and torso change intensity of sounds with different frequencies that arrive at each ear from different locations in space (azimuth and elevation)

Question 22

• what does the top graph indicate?
• y-axis?
  
  • @5000 HZ vs 10,000 Hz
• Rainbow graph
  
  • red patch vs blue patch meaning/implication?
• What do both graphs indicate about frequency preferences?
• What info do ITD & ILD provide?
• How about HRTR?
• x
• How do listeners know how far a sound is? 3 cues
  
  • Relative intensity of sound
    
    • inverse squared law
  • Spectral composition of sounds
    
    • how is related to aerial perspective from depth perception?
    • what happens to sound when it travels through ear? (hint freq)
    • IOW: if we still hear all the high freq, what does it mean?
  • direct vs reverberant energy
    
    • whisper vs singer far away
    • how their direct vs reverberant energy differ

Answer 22

Head related transfer fx (part 1)

• Here, depending on whether the sound is coming in front of you, LS/RS, or behind you, diff freq get enhanced or decreased (see graph)
  
  • Y-axis: tells you if sounds are enhanced or not
    
    • Ex. enhanced at 5000 Hz, but 10,000 Hz gets sig decreased (dip)
    • If you hear sound, and the brain noticed that in the range 10,000 Hz things are muffled/you don’t here as much, the brain can conclude the sound comes from a certain specific direction
• certain frequencies are
  
  • enhanced when the sound comes from LS bottom (red patch)
  • this freq is reduced when if comes from behind (blue patch)
• Frequency preference varies with elevation & azimuth
  
  • IOW: each direction has it’s own fingerprint (graph)
  • ITD and ILD gives info about the azimuth
  • The HRTR -> elevation info

Sound localization cont

How do listeners know how far a sound is? 3 cues…

• Cue 1: Relative intensity of sound: Intensity decreases with distance (inverse-square law)
  
  • Ex. we can tell app how far a car is based on the relative intensity of sound
  • This is b/c sound travels in expanding spheres
  • As distance increases, this energy gets spread over a wider region (less energy)
    
    • Similar to a pencil vs hand pressing on your hand
• Cue 2: Spectral composition of sounds: Higher frequencies decrease in energy more than lower frequencies as sound waves travel from source to one ear
  
  • This is similar to aerial perspective from depth perception (lec 6)
    
    • Blue mountains: when light travels through air, air is not completely transparent (ex. make hues faded, reduces contrast, shifts all colors towards the colder range) -> blue mountains
  • For sound, air also filters out freq
    
    • Here, Spectral composition of sounds: Higher frequencies decrease in energy more than lower frequencies as sound waves travel from source to one ear
    • Something further away sounds more mumbled than smth that is closeby
    • IOW: if we still hear all the high freq, that means the object/sound source is closer by
• Cue 3: direct vs reverberant energy
  
  • Aka echoes
  • How can you tell the singer on stage is far away from you w/o looking compared to the person behind you who is whispering to you.
  • If the singer has no mic, the singer’s voice will have a similar intensity as the person whispering behind you
    
    • IOW cue 1 on perceiving auditory distance does not work
  • However, you can still tell the singer is further away from you
    
    • This is b/c sounds does not travel in straight lines, it expands in spheres
    • Sound from the singer goes in all directions around her mouth, bounces off some surface (ex. ceiling in concert hall, floor) -> to your ears
    • Since it does not travel in a straight line, the echo is slow (IOW, we here echoes later than the actual sound)
    • Actual sound -> travels in a straight line (singer’s mouth to you)
    • Reverberant energy from the echo is delayed
  • For the whisperer
    
    • Direct energy (straight line) reaches you way faster than the reverberant energy (echo = black crooked line)
    • The echoed sound is much reduced (very faint) compared to the direct sound
  • For the singer
    
    • Direct energy (straight line from singer to you) reaches you about the same time as the reverberant energy from the echo
    • IOW: the direct sound seems to be of similar intensity compared to the echoed sound