tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre...
Transcript of tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre...
![Page 1: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/1.jpg)
Inferior Colliculustonotopic organization
Low freq High freq
![Page 2: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/2.jpg)
Structure of Inferior Colliculus
![Page 3: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/3.jpg)
Laminar Structure of Central Nucleus
Disc shaped cells
arranged in
laminae
Stellate Cells cut
across laminae
Input fibers from
lateral lemniscus
![Page 4: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/4.jpg)
Inferior Colliculus: Convergence and Reorganization
• Parallel processing pathways originating in the cochlear nucleusconverge onto the central nucleus of the IC. Some of the projectionsare direct (monosynaptic); others are via intervening synapses inthe SOC and/or the nuclei of the lateral lemniscus.
• IC performs considerable processing: its responses are morecomplex than those of its inputs.
![Page 5: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/5.jpg)
Inferior Colliculus Projects to Medial Geniculate
![Page 6: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/6.jpg)
• Anterior (A), primary (AI), posterior (P) andventroposterior (VP) auditory cortices each are tonotopic.
• Receptive fields of individual cells also include selectivityfor sound source location and sound intensity.
Multiple tonotopic fields in cat auditory cortex
Reale & Brugge
![Page 7: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/7.jpg)
EI
Binaural & frequency
organization in cat AIEE
![Page 8: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/8.jpg)
Classification of binaural interactions
Simple scheme (Goldberg and Brown, 1968)
Two letter code, indicating the predominant effect of each ear on neural response, eitherE (excitation), I (inhibition) or O (no effect). Usually, the contralateral ear comes first.
OE Responds to ipsi ear only (e.g. VCN)
EO Responds to contra ear only (e.g. MNTB)
IE Ipsi ear excites, contra ear inhibits (e.g. LSO)
EE Binaurally excited (e.g. MSO)
Detailed scheme (Irvine, 1986)
Two letters followed by slash and third letter (e.g. EO/S). First 2 letters characterize thepredominant effect of each ear alone (contralateral first). Third letter gives the type ofinteraction observed when both ears are stimulated: facilitation (F), suppression (S), orno interaction (O)
EO/S Responds to contra ear, NR to ipsi, binaural response smaller than contra.
EE/F Responds to each ear alone. Binaurally facilitated
OO/F Responds only to binaural stimulation
![Page 9: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/9.jpg)
Discharge Rate as a function of SoundLevel: Monotonic or Nonmonotonic
![Page 10: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/10.jpg)
Parvalbumin immunoreactivity defines the auditory core
RTR AI
low
highhigh?
![Page 11: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/11.jpg)
Organization of auditory cortexin nonhuman primate (NHP)
Hackett and Kaas
Core auditory cortex is locatedin the ventral bank of thelateral sulcus in NHPs (inHeshl gyrus in humans)
![Page 12: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/12.jpg)
Cat auditory cortex
Schematic summary of cortical areas related to the processing of auditory,auditory plus visual, and visual stimuli. Based on 2-DG studies. (Porembaet al. 2003)
![Page 13: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/13.jpg)
Auditory cortical subdivisions & processing streams
Kaas & Hackett 2000Macaque
![Page 14: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/14.jpg)
aaa
PFC
STGr IPL
PB AL CL CM
ML
R AI
MGm MGv MGd
Auditory cortical processing
Patterns Space
Adapted from Rauschecker, 1998
thalamus (MGN)
core auditory cortex
belt & parabelt
prefrontal cortex
rostral auditory cortex –>temporal & prefrontal
caudal auditory cortex –>parietal & prefrontal
Prefrontal
Belt & parabelt
Core
Thalamus
PATTERN SPACE
Auditory Cortical Processing Streams
![Page 15: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/15.jpg)
The belt and parabelt:
! have additional inputs fromother MG nuclei and other partsof the thalamus.
! are implicated in integrativeand associative functionsinvolved in pattern perceptionand object recognition.
MGB Ventral nucleusPrincipal division of the auditory thalamus
CoreCochleotopically organized fields that generally
display clear responsive to pure tones
BeltMultiple fields less precisely cochleotopic and
more responsive to complex stimuli than tones.
Parabelt
Rostral and caudal divisions lateral to the belt
Kaas, J.H., Hackett, T.A., and Tramo, M.J. Auditoryprocessing in primate cerebral cortex, CurrentOpinion in Neurobiology,1999, 9:164-170.
Temporal, parietal, &
frontal cortex… which mediate space perception,
auditory memory and other functions
![Page 16: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/16.jpg)
Human auditory cortex
![Page 17: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/17.jpg)
Primaryauditory
cortex
Secondaryauditory
cortex
Corresponds toapex of cochlea
Corresponds tobase of cochleaareas 41 & 42
![Page 18: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/18.jpg)
• Connection (arcuate fasiculus) between Wernicke & Broca’sareas enables patients with transcortical sensory aphasia torepeat words despite not understanding them.
• Damage to this connection disrupts speech, but speechsounds, articulation, and comprehension are retained(conduction aphasia).
Speech processing — the Wernicke-Geshwind model
Primarymotor cortex
Arcuate fasciculus
Broca's area
A1Wernicke's area
angulargyrus
V1
Broca's areaSpeech production and articulation• stores information needed for speech
production
• responsible for programming the motorcortex to move the tongue, lips andspeech muscles to articulate words.
Wernicke's areaLanguage comprehension• Region of angular gyrus specialized for
processing human speech, and receivinginput from belt/parabelt auditory cortex.
• Stores information for arranging learnedvocabulary into meaningful speech
• Damage leads to Wernicke's Aphasia —speech may be fluent yet confused andnonsensical. Speech comprehension isalso impaired.
• Damage limited to the posteriorlanguage area leads to transcorticalsensory aphasia (patient can’t understandbut can repeat words).
![Page 19: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/19.jpg)
Amplitude-modulated tones
• Percepts generated by AM sounds vary with modulation frequency (recall Spottiswoode’s description of the sound quality of beats)
![Page 20: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/20.jpg)
Representation of AMauditory nerve
![Page 21: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/21.jpg)
Transformations in representation of AMauditory nerve (dashed lines) and cochlear nucleus (solid lines)
enhancement of envelope synchronizationand extended dynamic range
Emergence of bandpass tMTFs.
![Page 22: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/22.jpg)
Transformations in representation of AMBrainstem to midbrain
![Page 23: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/23.jpg)
Transformation in representation of AMThalamus to cortex
![Page 24: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/24.jpg)
Topography in representation of AM
Langner and Schreiner 1988,
Schreiner and Langner 1988:
![Page 25: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/25.jpg)
Langner and Schreiner 1988 (first of two papers)
![Page 26: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/26.jpg)
Langner and Schreiner 1988 (first of two papers)
![Page 27: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/27.jpg)
D
M
R
Schreiner and Langner 1988 (second of two papers)
![Page 28: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/28.jpg)
How can masking occur?
• Excitation: Swamping of neural activity due to masker.
• Suppression: Reduction of response to target due to masker.
If the ear is exposed to two or more different tones, one tone may mask
the others. Pure tones, complex sounds, narrow and broadband noise
all show differences in their ability to mask other sounds. Masking of one
sound can even be caused by another sound that occurs a split second
before or after the masked sound.
Masking — an upward shift in the hearing threshold of a weaker tone by a louder
tone,dependent on the frequencies of the two tones.
![Page 29: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/29.jpg)
Frequency (kHz)
incre
ase
in
te
st-
ton
e t
hre
sh
old
(d
B)
60
0.1 0.2 0.5 1 2 4 10
basilar membranebase
(high)
apex
(low)
8 kHz 2 kHz
![Page 30: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/30.jpg)
Masking - Psychophysical Tuning Curves
• present a pure tone at 10 dB
• mask it with noise of varying centre frequencies
• At what intensity will the noise prevent
• detection of the pure tone?
0.1 0.2 0.5 1 2 4 10
Frequency (kHz)
0.1 0.2 0.5 1 2 4 10
![Page 31: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/31.jpg)
Frequency (kHz)
0.1 0.2 0.5 1 2 4 10
30
40
50
60
70
inte
nsity o
f m
ask (
dB
SP
L)
Masking - Psychophysical Tuning Curves
![Page 32: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/32.jpg)
Frequency (kHz)
0.1 0.2 0.5 1 2 4 10
30
40
50
60
70
inte
nsity o
f m
ask (
dB
SP
L)
Frequency (kHz)
40
50
60
70
2 105 20
thre
shold
(dB
SP
L)
neuron's
characteristic frequency
30
Masking - Psychophysical Tuning Curves
![Page 33: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/33.jpg)
20
40
60
80
100
120
dB
(S
PL)
20 100 500 1000 5000 10000
Frequency (Hz)
0
![Page 34: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/34.jpg)
Some conclusions from Masking Experiments
• Pure tones close together in frequency mask each other more
than tones widely separated in frequency.
• A pure tone masks tones of higher frequency more effectively
than tones of lower frequency.
• The greater the intensity of the masking tone, the broader the
range of frequencies it can mask.
• Masking by a narrow band of noise shows many of the same
features as masking by a pure tone; again, tones of higher
frequency are masked more effectively than tones having a
frequency below the masking noise.
• Masking of tones by broadband ("white") noise shows an
approximately linear relationship between masking and noise
level (that is, increasing the noise level 10 dB raises the
hearing threshold by the same amount). Broadband noise
masks tones of all frequencies.
![Page 35: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/35.jpg)
Critical bands demonstrated by masking
Results suggest that we can “tune into” the region around 2 kHz, andthat only masker energy around 2 kHz affects our ability to perceive thetone.
The bandwidth of effective masking is the critical band (Fletcher, 1940).
Frequency
Le
ve
l (d
B S
PL
)
Signal
Masker(Band-pass noise)
Auditory filter
2000 Hz
![Page 36: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/36.jpg)
Masking — an upward shift in the hearing threshold of a weaker tone by a
louder tone,dependent on the frequencies of the two tones.
Basilar membrane tuning produces asymmetry: a tail extends toward the high-
frequency end. Thus it is easier to mask a tone of higher frequency than one of
lower frequency.
As the intensity of the masking tone increases, a greater part of its tail has
amplitude sufficient to mask tones of higher frequency. This "upward spread" of
masking tends to reduce the perception of the high-frequency signals that are so
important in the intelligibility of speech.
Forward: A tone can be masked by a sound that ends a short time (up to about
30 ms) before the tone begins. Forward masking suggests that
recently stimulated cells are not as sensitive as fully-rested cells.
Backward: A tone can be masked by a noise that begins up to 10 milliseconds
later, although the amount of masking decreases as the time interval
increases (Elliot, 1962). Backward masking apparently occurs at
higher centers of processing where the later-occurring stimulus of
greater intensity overtakes and interferes with the weaker stimulus.
![Page 37: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/37.jpg)
• Determining the source of a sound depends on analytic processing to segregatespectral components according to source.
Auditory Scene Analysis(sound source determination)
• An auditory stream refers to the perception of a series of sounds as the output ofa single sound source. Separation of the auditory streams is affected by timeand frequency separation.
• A single source can stimulate a wide region of the cochlea, with multiple sourcessimultaneously stimulating the same peripheral frequency channels.
• Sounds from multiple sources arrive at the auditory periphery as one complexsound field. Individual sources are then resolved from spectro-temporal patterns.
![Page 38: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/38.jpg)
Waves from environmental sound sources minglebefore reaching our ears
Time
Fre
qu
en
cy
![Page 39: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/39.jpg)
Auditory system disentangles the sources
Time
Fre
qu
en
cy
![Page 40: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/40.jpg)
Yost calls this “Sound Source Determination”
![Page 41: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/41.jpg)
Mechanisms
Level
Freq
ClarinetVoice
Simultaneous SequentialAuditory ‘streaming’
Time
Fre
quency
Harmonicity,
Onset/offset,
Co-modulation
![Page 42: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/42.jpg)
Deviations from harmonicity promote segregationIf a harmonic is mistuned by > 2-3%, it stands out perceptually
(Moore et al., 1985, 1986; Hartmann et al., 1990 )
Le
ve
l
200 400 600 800 1000 1200
Frequency
200 400 618 800 1000 1200
stimulus percept
1 soundpitch = 200 Hz
2 soundsHarmonic (pitch = 200 Hz) +
Pure tone (618 Hz)
![Page 43: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/43.jpg)
Onset asynchronies promote perceptual segregationF
req
ue
ncy
Time
stimulus percept
1 sound
2 sounds
![Page 44: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/44.jpg)
Co-modulation promotes perceptual segregation
Fre
qu
en
cy
Time
stimulus percept
2 sounds
Coherent FM promotes the fusion of harmonics
Darwin et al. (1994)
![Page 45: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/45.jpg)
When all the tones frequencies aresimilar, you can hear a single streamof sound, with a galloping rhythm.
Auditory streaming
repeating triplet of tones.
But when the "A" and "B" tones arefar apart, you hear 2 streams and thegallop disappears.
![Page 46: tonotopic organizationdavid/courses/sm12/Lectures/semple/NS2_07_A… · Profile analysis & Timbre ¥Sound spectra are characterized by their intensity variation as a function of frequency.](https://reader036.fdocuments.us/reader036/viewer/2022062921/5f03217c7e708231d407b097/html5/thumbnails/46.jpg)
Spectral separation• Sometimes two or more sounds having different spectral
components can be resolved by frequency selectivity.
Temporal separation & Asynchrony• Multiple sound sources are often asynchronous (onset
or offset) & temporally complex, aiding segregation.
Cues for sound segregation
Profile analysis & Timbre• Sound spectra are characterized by their intensity
variation as a function of frequency.
• The peripheral auditory system is sensitive to changes inspectral profile despite variation in overall sound level.
• Timbre allows discrimination of one sound from anotherwhen both have the same pitch, loudness, and duration.Different spectra give rise to different timbres.
Spectral Modulation• Changing a frequency modulation or modulating a
complex set of harmonics facilitates segregation.
Spatial separation (e.g., cocktail party effect)
• IID and ITD are cues for sound source determination,particularly with competing signals.
• Increasing separation between masker and signalimproves discrimination. e.g., Two individualsspeaking simultaneously at various locations.
Informational masking• The inverse relationship between pattern
complexity and target resolution is calledinformational masking.
• Variation in target location in a pattern, totalnumber of elements in the pattern, and stimulusuncertainty affect the information content of asequence.
Temporal modulation• Coherence of change among spectral components
may help the auditory system to identify thosecomponents as originating from a single source (canbe AM or FM).
• A coherently modulated cue band offers relief frommasking (comodulation masking release).
• A modulated masker can impair detection of amodulated signal (modulation detection interference).
Harmonicity• A sound consisting of harmonics is usually perceived as a
single pitch equal to the fundamental frequency even ifthe fundamental is absent in the sound spectrum.
• Fusion of spectral components into an image or entity isdescribed as a pitch.