the grade only
1
Sensation and Perception
Chapter 11: Auditory Brain and
Perceiving Auditory Scenes
Most images © 2014 Worth Publishing. Most images from Yantis (2014)
Lecture Outline
• Ascending and descending auditory pathways – Cochlear nucleus – Superior olivary complex – Inferior colliculus – Medial geniculate body /
nucleus
• Auditory cortex – Core, belt, parabelt – Tuning curves – “What” and “where” pathways
• Sound localization – Azimuth
• ILD & ITD • Front-back confusions • Cone of confusion
– Elevation • Spectral shape cue
– Distance • Inverse square law, echoes,
Doppler effect
– Ventriloquism effect
• Auditory scene analysis – Grouping by harmonic
coherence – Grouping by synchrony or
asynchrony – Grouping by frequency
similarity – Grouping by temporal proximity
• Perceptual completion of occluded sounds
• Perceptual organization
2
Ascending Pathway
Descending Pathway
• Information also flows from the cortex to the cochlea
– Modulate the motile response of the outer hair cells
– Attention – block task-irrelevant ascending auditory signals
– Activate the acoustic reflex
Auditory Cortex
• Top of temporal lobe
• Core – A1
• Pure tones
• Belt – Complex sounds
• Parabelt – Complex sounds
• Tonotopic organization
3
Auditory Tuning Curves
“What” and “Where” Pathways
Sound Localization
4
Perceiving Azimuth
• Interaural Level Differerence (ILD; IID; IAD)
• Most sounds are closer to one ear than the other
• Inverse square law
• Acoustic shadow
ILD
• Greatest for sounds directly to the left of the left ear or to the right of the right ear.
• Smallest for sounds anywhere on the median plane
• Works best for higher frequencies
• Processed by the lateral superior olive
ITD
• Interaural Time Difference (ITD)
• Greatest (~600 μs) when sound directly to left of left ear or to the right of the right ear
• 0 for sound anywhere on the median plane; front-back confusions • Minimum audible angle (MAA) – 1° to 10 °
5
Neural Basis of ITD
Cone of Confusion
Elevation Localization
• Spectral shape cue
6
Elevation Localization
Distance Localization
• For sounds of familiar loudness, inverse square law
• Inverse square law is greater for higher frequencies than lower frequencies; results in “blurring”
– Blurring can be used to determine distance
Distance (m) p (μP) 20 log10(p/p0)
1 900,000 / 12 = 900,000 93 dB
2 900,000 / 22 = 225,000 81 dB
3 900,000 / 32 = 100,000 74 dB
5 900,000 / 52 = 36,000 65 dB
10 900,000 / 102 = 9,000 53 dB
20 900,000 / 202 = 2,250 41 dB
Distance Localization
• With echoes, compare direct sound intensity to reflected sound intensity
– Direct > reflected sound is near
– Direct < reflected sound is far
• Changes in intensity
• Doppler effect – when a sound source is moving, frequency is higher in front of the source than behind the source
© 2007 Charly Whisky; http://en.wikipedia.org/wiki/File:Dopplerfrequenz.gif
7
Ventriloquism Effect
• Sounds tend to be localized on the basis of visual cues when visual and auditory cues conflict
– Visual and auditory events must occurs close together in time
– Visual and auditory events must be plausibly linked
– Visual and auditory events must be close in space
Auditory Scene Analysis
• Auditory stream
• Auditory scene
Grouping by Harmonic Coherence
• In (a) the frequencies are all harmonics (integer multiples of fundamental)
– Grouped as a single
auditory stream
• In (b) the harmonics are grouped as one auditory stream and the non-harmonic is perceived as a separate auditory stream
8
Grouping by Harmonic Coherence
• In (a), two separate sound sources produce harmonics of a single sound
– Perceived as a single stream
• In (b), two separate sound sources produce harmonics, but not of a single sound
– Perceived as two auditory streams
Grouping By Synchrony or Asynchrony
Grouping by Frequency Similarity
9
Grouping by Frequency Similarity
Grouping by Temporal Proximity
Perceptual Completion of Occluded Sounds
https://www.youtube.com/watch?v=ZyvyGMkzNQc
10
Perceptual Organization
• Effect of past experience
• A melody is played with notes alternating between octaves
• Listeners find it difficult to identify the song
• But after they hear the normal melody, they can then hear it in modified version using melody schema