Neuroethology Exam #1

Front 1

Tinbergen's 4 Questions

Back 1

Causation (P)
- How does the animal perform the behavior?
- Mechanisms

Individual Development (P)
- How does the behavior change w/ age and what experiences are necessary for this change to occur

Evolution (U)
- How does the behavior compare across species/how might it have arisen
- Phylogeny

Function (D)
- Of what use is the behavior to survival/reproduction
- Adaptive

Front 2

Fixed Action Pattern
- eg: "eyebrow flash"

Back 2

- Sequence runs to completion
- Stereotypic motor sequence
- Triggered by SIGN STIMULUS (male stickleback fish red swollen baby)
- NOT a reflex--> released by stimulus, but not controlled by it
- Sensory feedback NOT required
- Common to all members of a species

Front 3

Supernormal stimulus

Back 3

- Herring gull: red stick

Front 4

Social Behavior as a Sequence of FAPS

Back 4

Interlocking releasers

- FAP of A serves as sign stimulus for B
- FAP of B serves as sign stimulus for A

ex: - Male fish does zig-zag dance
- Female displays belly
- Male goes to nest
- Female goes in, male pokes her--> egg laying

Front 5

Problems w/ ethograms

Back 5

- Some behaviors are rare
- Some behaviors are graded in intensity or frequency
- Some behaviors vary btw judgement

Front 6

Critical periods vs. Sensitive periods

Back 6

- Sensitive periods are a less constrained period of time

Front 7

FAP and motivation

Back 7

- Sign stimuli are not always effective at releasing FAPs
- Depends on internal state of animal - e.g., hunger and reproduction state

Front 8

Vacuum acts

Back 8

- FAPs are sometimes produced in absence of appropriate sign stimulus
- Due to increasing "build up of motivation"

Front 9

Displacement Activity

Back 9

- FAPs inappropriate to particular sign stimulus
- When appropriate response in prevented or when multiple stimuli are in conflict

Front 10

FAPs and Conditioning

Back 10

Classical: creates new sign stimulus for FAP

Operant: changes motivation to perform a particular FAP (b/c of consequences)

Front 11

Lorenz'z Hydraulic Model of Behavior

Back 11

Faucet: energy input
Valve: innate releasing mechanism
Weights: sign stimulus
Consummatory response: FAP

Front 12

Problems with FAP

Back 12

- Behavior is not always fixed
- Many behaviors can be graded in intensity or frequency
- There is often individual variation in behavior
- Behavior can be modified by experience
- Behavior can be modified by immediate environment

Front 13

Neuron Doctrine

Back 13

Consilience: low level theories absorbed into higher level theories
- Nervous system composed of discrete cells (neurons) that communicate w/ each other

Alternative: reticular theory: nervous system is a continuous meshwork of cytoplasm

Front 14

Golgi

Back 14

- Proponent of reticular system
- Problem: available techniques made it impossible to see membranes btw cells

Front 15

Cajal's Modern Neuron Doctrine

Back 15

- Methods for intensifying staining in reaction--> PROOF
- Basket endings--> fibers envelope cell body--> shows discrete boundary

- Neurons are discrete morphological units: each element is autonomous
- Neurons make intimate contacts with each other: contiguous, but not continuous
- Cell bodies, dendrites, axons are conductors
- Dynamic polarization: information is one way
- Axons arise in development by neurite outgrowth

Front 16

Sherrington's contributions

Back 16

Worked on Spinal Reflexes
- Unidirectional flow of info (polarized)
- Synaptic delays
- Coordinated action of many synapses
- Excitatory and inhibitory actions
- Only excitatory actions propagated, while inhibitory actions stabilized the neuron

Front 17

Nerve (def)

Back 17

a bundle of peripheral axons

Front 18

Afferent

Back 18

toward the CNS

Front 19

Efferent

Back 19

away from the CNS

Front 20

Ganglion vs. Nucleus vs. Laminae

Back 20

G: cluster of cell bodies
- can be autonomous; grasshoppers not as centralized as vertebrates
N: clustering of neurons
L: layers of neurons

Front 21

Homology

Back 21

- Similarity due to shared ancestry
- Some version of the trait found in a commo ancestor (eg: olfactory bulb; cerebrum)

Front 22

Homoplasy/analogy

Back 22

- Similarity due to convergent evolution
- Similarities result from similar selective pressures
- No version of trait found in common ancestors
- E.g.: fins of sharks and dolphins; SA and Afr. Electric fish sensory structure for electricity

Front 23

Model Systems Approach

Back 23

Why study a particular brain/animal?
- Simplicity: easy to study
- Experimental advantages: IDed neurons, genetic tools, accessibility
- Homology to humans
- Homoplasy to humans
- Behavioral specializations

Front 24

Galvani vs. Volta

Back 24

- Electricity is generated by animal tissue
- Electricity is used by the nervous system and muscle to communicate
- BUT electricity is NOT a fluid substance and is NOT different from other types of electrity
- In Galvani's exp: metals served as a battery that provided a stimulus to elicit a natural cascade of nerve-muscle electrical activity

Front 25

Summation

Back 25

Spatial: from 2 diff synapses activated at the same time

Temporal: 1 synapse in rapid succession

Analog --> Digital (all or nothing)

Front 26

Neuron Logic

Back 26

AND: output only if 1+2 are both active (relatively high threshold)
OR: output if either 1 or 2 are active (relatively low threshold)
NOT: output if 1 or 2, but not 3

Front 27

Short term plasticity

Back 27

Synaptic responses can depend on recent history of synaptic activity

- Information processing

Front 28

Long Term plasticity

Back 28

Synaptic responses can depend on the ancient history of synaptic activity

- Info storage / circuit rewiring

Front 29

Sensory transduction

Back 29

Process by which a physical stimulus is converted into an electrical signal in the nervous system

Front 30

Necessity & Sufficiency

Back 30

Correlation: activity is correlated w/ behavior

Sufficiency: stimulating the neuron elicits the behavior

Necessity: silencing the neuron prevents the behavior

Front 31

Spallanzani (1793)

Back 31

- Bats avoid obstacles, even in complete darkness/eyes removed

Front 32

Jurine (1795)

Back 32

- Bats fail to avoid obstacles if ears are plugged
- Concluded bats can 'see' through ears

Front 33

Cuvier (1795)

Back 33

- Disagreed, asserted that touch receptors in the wing membrane were responsible for orientation in dark
- B/c bats flew in silence

Front 34

Maxim (1912)

Back 34

- Hypothesized that bats emitted low-frequency sounds and detected the echos

Front 35

Hartridge (1920)

Back 35

- Hypothesized that bats emitted high frequency sounds (ultrasound) and detected the echos

Front 36

Why Ultrasound?

Back 36

1) Low frequencies (long wavelgths) "bend" around small objects
--> high frequencies deflected by small objects--> wants echos to bounce back

Front 37

Mega vs. microchiroptera

Back 37

Mega: DO NOT echolocate, rely on vision; non-moving prey

Micro: 813 species; all echolocate; eat a variety of things

Front 38

Specializations in Bat anatomy for echolocation

Back 38

Nose: specialized for directional sound emission
- parabolic
- concentrates sound and sends out like beam
Pinna: specialized for sound reception
Tragus: dampens sound directly
- only in front
- directional info for where sound in coming from

Front 39

Phases of bat hunting

Back 39

- Detection
- Gain control
- Decision
- Terminal

- Increase rate of echolocation
- want time info when get close--> rapid updating

Front 40

Phylogenetic bats (U)

Back 40

- Earlier thought evolved from primates; but first common ancestor is too far back

- There are other echolocating mammals--> evolved multiple times
- Toothed whales echolocate
- Use mellon (of fat) to focus sound; transmit vibrations from lower jaw

- Echolocation in birds
- Oilbirds in Northern SAm
- Live in caves; echolocate w/ sharp audible clicks (1-15 kHz); broadband clicks

- Swiftlets in S. Asia, Pacific Islands, NE Australia
- Live in caes
- Echolocate w/ audible pulse pairs (3-10kHz, 1-3ms interval)

Front 41

Sound

Back 41

- Vibration in an elastic medium
- Propagates as a wave of compression and rarefaction
- Particles of the elastic medium move only small distances to create changes in pressure

Front 42

Sound Intensity (sound pressure level)

Back 42

- Amplitude
- RELATIVE (dB RE:...)
- 6 dB = 2x as big
- Lp = 20log10(p/p0)
- Typically p0 is the threshold of human hearing = 20 muPa

Front 43

Sound frequency: Fourier Transform

Back 43

All fcts can be represented as sum of sine waves w/ different frequencies and amplitudes
--> break down complex wave forms

Front 44

Harmonics

Back 44

- Integer multiples of the F0 (fundamental freq--> lowest, baseline freq of noise)

Front 45

KNOW SONOGRAM vs. SPECTROGRAM

Back 45

Sonogram: y axis is relative amplitude
Spectrogram: y axis is frequency

x axis is always time

Front 46

Echolocation: Distance info (range)

Back 46

DELAY btw sound and echo

R = (c*t)/2
R: distance (m)
c: speed of sound (m/s)
t: time delay (s)

Front 47

Echolocation: size info

Back 47

Amplitude of echo + info re: distance

- Larger objects make stronger echos
- Amplitude alone is ambiguous, since amplitude increases as object gets closer
- Amplitude corrected for distance gives estimation of size

Front 48

Range accuracy

Back 48

Improves w/ signal bandwith (how broad the frequency content of the emitted sound is)

deltaR = c/(2*beta)

delta R: minimum discernible range difference (m)
beta: bandwidth (1/sec)
c: speed of sound (m/s)

Front 49

Some bats produce broadband clicks, but this is rare b/c...

Back 49

- they are short
- total sound energy is proportional to call duration
--->amplitude is weak; not a lot of energy
--> don't travel as far; the echo becomes too soft at longer range

- sound pressure decreases as 1/R^4

- High frequency sound is absorbed and scattered more easily
--> smaller wavelength means that sound is absorbed by smaller objects
- Small targets can reduce echo amplitude

- Amplitude is reduced by sounds not directly in front of the bat
- Sound output (nose leaf) and receivers (pinna & tragus) are highly directional

Front 50

Frequency Modulation (FM)

Back 50

- Call sweeps across frequencies
- Frequency range is continuous
- Produces strong broadband sound
- Better for range estimation
- because there is a large correlation value btw output and echo only at the actual delay, NOT at multiples of the period
- narrow window of delays that give you a strong correlation

- Worse for velocity estimation
- Frequency content of outgoing sound and returning echo are mostly overlapping

- Worse for longer distances
- Energy is distributed across frequencies
- Less energy at each frequency
- Each returning frequency can only be evaluated for a fraction of a millisecond

Front 51

Constant Frequency (CF)

Back 51

- Energy is concentrated over a narrow range of frequency
- Not a pure tone, but discrete frequencies (harmonics)
- harmonics are louder than f0

- worse for estimating range b/c there is cross-correlation at several delays
- strong correlation at the actual delay (delta t), but also at multiples of the period (T+ delta t)
- wide window of delays that gives you a strong correlation

- Better for velocity
- Frequency content of outgoing sound and returning echo are mostly distinct

- Better for object "flutter"

- Better for longer distances
- Energy concentrated at a specific frequency
- Returning frequency can be evaluated for duration of the call

Front 52

Cross correlation

Back 52

Correlation as a fct of the relative delay btw 2 waveforms
- For echolocation, the 2 waveforms are the outgoing sound and the returning echo; the delay is the delay btw the sound output and echo return

Front 53

Echolocation: Velocity Info

Back 53

Doppler shift: frequency shift proportional to: v/c
v: relative velocity of object (m/s)
c: speed of sound (m/s)

receding object: lower frequency than output
approaching object: higher frequency

Also used to analyze object "flutter"
- Echo has high overall frequency (bat is approaching insect)
- Echo frequency is also modulated up and down w/ wingbeats (glints)
---> Frequency modulation can be used to get info about an objects ID

Front 54

Species combining CF and FM

Back 54

- CF during searching b/c detects are away object, and determines velocity and ID (flutter)

- Switch to FM as target approached--> determine precisely where the target is located as you close in on it

Front 55

Ecology of CF and FM

Back 55

CF and CF-FM calls:
- species foraging wing-beating insects in dense vegetation
- b/c doppler shift detects ID (distinguish from background clutter)
- Longer range critical to combat effects of scattering and absorption

FM calls:
- Hunt insects in open air
- Provides precise distance info during flight

Broadband clicks:
- Species that pick up prey from substrates or visit flower/plants
- Broad spectral energy allows bats to differentiate surface texture

Front 56

Signal "Design" of Echolocation

Back 56

- Calls are high frequency b/c short wavelengths get strong echoes off small targets
- Call frequency varies w/ environment and prey item; also w/ species--> recognition, reproductive character displacement
- Call frequency varies btw individuals--> private channels of communication

Flexibility
- Can change calls to suit environment
eg: high frequencies, shorter durations, steeper FM sweeps in clutter environments

Harmonics
- Higher harmonics have stronger amplitude than F0
- F0 is only heard on outgoing pulse; echo is weak--> other bats can't hear it
- Each individual has slightly different F0--> serves as key; selective responses to bat's own call--> only listen to sounds that come right after a sound at your F0

Front 57

Tympanic membrane

Back 57

Vibrates b/c of pressure diference
- bones move and oval and round windows in cochlea push/pull against each other

Front 58

Scala vestibuli

Scala tympani

Back 58

Connected to oval window

Connected to round window

Vibration at tectoral and basilar membrane

This is the endpoint where external vibrations become internal vibrations

Front 59

Vibrations of the cochlea

Back 59

Tectoral membrane
- At the tip vibrates at low frequency
- At the base vibrates high frequency

- Inner hair cells send info to the brain
- Afferent axons: from cochlea to the brain
- Efferent axons: from brain out to cochlea--> increase or dampen sensitivity

Front 60

Frequency tuning

Back 60

- sharp V shaped tuning curve (threshold pressure)
- BF = best frequency--> for a specific neuron--> max sensitivity
- BW = bandwith (difference at 30 or 10 dB above threshold)

Q10 = BF/BW--> sharpness of tuning
- Most mammals 10-20
- CF bats 200 (more sharp, b/c smaller bandwidth)

Eg: compare guinea pigs' and mustached bats' tuning curve
- Mustached bats have sharper tuned curves for many neurons; the neurons are not even spread--> concentration around 61 hz (calls)--> smaller range

Front 61

Acoustic fovea

Back 61

Much sharper and more precise sensitivity to a specific frequency

- Horseshoe bat cochlea has a much longer length for high frequencies--> overrepresentation of freq (calls are @ 83-86 khz)

ONLY PRESENT IN CF BATS
- auditory afferents are concentrated at a narrow BF and have high Q10 values
- FM bats: more like other mammals--> broad distribution of BFs and smaller Q10 values

Front 62

Problem with the acoustic fovea of CF bats?

Back 62

Doppler shift can cause echo frequency to be outside fovea--> echo will be at high frequency if moving toward object

SO

Doppler-shift compensation
- Lower freq of call if moving forward--> no shift for backswing b/c doesn't fly backwards

Front 63

Bats: Central Auditory Pathway

Back 63

Cochlear Nuclei (CN)--> Inferior Colliculus (IC)--> Auditory Cortex (AC)

Front 64

Central Auditory Pathway: Cochlear Nuclei

Back 64

- In hindbrain; receives info from cochlea
- Binaural processing: convergent input from 2 ears (different in time and sensitivity from 2 ears)

Front 65

Central Auditory Pathway: Inferior Colliculus

Back 65

- Latencies are constant across a wide range of amplitudes
- Unusual for sensory neurons: latencies usually increase with decreasing amplitude
- Neurons are sharply tuned to a particular frequency, and they have a low threshold for action potential

- This allows for faithful representation of the timing of both the pulse and the echo
- Response of different neurons tuned to different frequencies are combined for analysis in the cortex

Front 66

Central Auditory Pathway: Auditory Cortext (of cerebral cortex)

5 Maps

Back 66

1) Tonotopic Map of Frequency- spatial map of frequency

2) Doppler-Shifted CF Area
- Central representation of the acoustic fovea
- radiates outwards for higher freq and for sound intensity
- 30% of auditory cortex
- Neurons respond solely to echo, regardless of the frequency of the emitted pulse
- Map of frequency and amplitude
- Fine frequency distinctions

3) Distance coding
- FM/FM area
- 3 separate areas: each respond selectively to FM1 of pulse and FM2, FM3, or FM4 of harmonic
- Each region has a map of the time delay btw the pulse and echo (distance)
- FM1 is weakest--> only you can hear it for your call--> then selectively sensitive to own echoes

4) Velocity Coding
- CF/CF area
- 2 regions: each respond selectively to CF1 of pulse and CF2 or CF3 of echo; both regions have a map of frequency diff btw pulse and echo (velocity)

5) Azimuth Coding
- Neurons sensitive to relative amplitudes of sounds at the 2 ears
- Inner to outer (radiates)--> where a sound is coming from

- in FM/FM area: vertical echo track--> same echo delay for columns

Front 67

Combination sensitivity

Back 67

Neuron only responds to FM components of pulse + echo

Front 68

Delay tuning in FM1/FMx

Back 68

- Neurons only respond to delays within a narrow limit
- Tuned to longer delays at more posterior locations
- Map of echo delays (distance)
- Closed tuning profiles--> only respond to objects of a certain sizes at a given distance

Front 69

Big picture for auditory cortex

Back 69

Process info in different ways--> extract different features
- hierarchical organization of sensory system

Front 70

Auditory system can be modified by experience

Back 70

- Repeated presentation of a tone a single frequency--> short term changes in cortical and subcortical areas--> frequency b/comes overrepresented
- Longer term changes when paired w. shock (elicits fear)

Front 71

Corticofugal feedback in auditory processing

Back 71

- Drives plasticity
- Auditory pathway has both ascending and descending projections

Front 72

MOTHS FIGHT BACK.

Back 72

- Can detect ultrasound--> drops to evade
- 15 independent origins of tympanal hearing organs--> located in different parts of the body

Front 73

Trigona clicks

Back 73

- When muscles contract they cause the tymbal organ to buckle = active half-cycle
- When muscle relax, unbuckling = passive half-cycle
- Broadband clicks--> cover freq range of bat's call
- Click more in terminal phase of bat

Front 74

Sound localization:

How do we localize azimuth?

Back 74

Interaural Intensity/Level Differences (IID or ILD)

Interaural Timing Difference (ITD)

At low freq, use ITD; at high freq use ILD
- poor localization at dividing line

Front 75

How do we localize elevation?

Back 75

- Anatomy of outer ear causes frequency filtering that varies with elevation
- Highly individualistic
- Frequency bouncing off object has to have low wavelength--> higher frequencies bounce off pinnae
- Pinna shadows sound

Front 76

Interaural Timing Difference

Back 76

- Works well at low frequencies
- Phase ambiguity when wavelength is smaller than the difference btw the ears
- otherwise, the sound seems to be in phase at the two ears...but it is really off by one cycle--> seems like sound is coming from straight ahead

- Humans can detect as small as 13 microseconds BUT action potentials last 1 ms, and synaptic delays are 1 ms...so how can a neural circuit detect diffs in timing that are 2 orders of mag small tan timescale of neural activity?

Front 77

Interaural Level Difference

Back 77

- Works well at high freqs
- Otherwise, long wavelengths refract ("bend") around the head
- Head is small compared to wavelength--> sound will bend around the ears

- Humans can detect as small as 0.5dB

Front 78

Jefress Model

Back 78

DELAY-line COINCIDENCE detection
- delays: due to axonal conduction; linear array
- coincidence: output cells only fire if they receive coincident inputs from 2 ears

- Map of sound source azimuth

In medial superior olive (brainstem)

Front 79

Sound localization in barn owls

Back 79

- Passive listening to detect and localize prey
- Owls turn head toward sound, leave perch and attack sound source (paper)
- Precise to w/in +/-2 degrees azimuth and elevation

- Head turning doesn't require feedback
- for short sounds, head turn begins after sound has ended (can compute after stimulus ends)
- Magnitude of turn is correct for sounds coming from different directions
- Ballistic movement (all at once)--> owl doesn't move head until sound in the two ears is the same

Front 80

Neuroethological Approach

Back 80

1) ID the problem: mechanisms of sound localization
2) Find an animal that is either: very good at performing the behavior or amenable to study
3) Characterize the behavior: describe it; determine how it is performed
4) ID the neural mechanisms: trace pathways; determine how info used for behavior is processed

Front 81

Barn owls coding elevation and azimuth

Back 81

ILD for Elevation
- Right ear opens upwards; left ear opens downward
- If plug right ear, err downwards; if plug left ear, err upwards

ITD for Azimuth
- Onset ITD and Offset ITD have no effect
- Ongoing ITD is used (use headphones to manipulate 3 timing cues)--> sensitivity to as small as 10 microseconds

Front 82

Behavioral map of auditory space?

Back 82

- No map of auditory space on the cochlea
- BUT space-tuned neurons in IC (inferior colliculus)
- Binaural: need info from both ears
- Respond to SPECIF. COMBOS of ITD and ILD
- Spatial receptive fields
- Receptive fields are arranged topographically
- ITD mapped A-P
- ILD mapped D-V
- Center-surround organization: inhibitory region surrounding excitatory space--> better for localization/discrimination
- Sounds played from 2 speakers: 1 in center of receptive field; other in different locations---> sound from 2nd speaker inhibits sounds from the first

THE MAP IN THE IC IS SPECIAL B/C THERE IS NO MAP OF AUDITORY SPACE IN PERIPHERY--> brain constructs/computes map

Front 83

How are ITD and ILD combined to construct the brain's map of auditory space?

Back 83

Top-Down Approach
- Start high up in sensory pathway to determine what features the system is sensitive to, work way toward periphery
- Two parallel pathways from ear to IC, specializations for time-coding
- Angular nuclei (ILD)
- Small axons, typical myelination
- Diffuse dendritic arborizations
- Small synaptic boutons
- Low concentrations of Ca-binding proteins
- Slow acting glutamate receptors and K channels
- Population code: how many neurons are active
- Magnocellular nuclei (ITD)
- Thick axons, heavy myelination (speed)
- Adendritic cells (only 1 input, takes less time)
- Endbulb synapes (engulf the cell--> higher fidelity; maximize prob of postsynaptic cell firing; 1:1 firing
- High concentrations of Ca binding protein (shorter refractory period)
- Fast-acting glutamate receptors and K channels
- Place doe: which neurons are active?

Front 84

Phase-locking of auditory neurons

Back 84

Strong phase locking= high vector strength
Weak phase locking = low vector strength

- Neurons in NM are more phase-locked to stimulus than those in NA
- NM: harder to resolve high freq b/c neurons start to skip cycle
- Regardless of intensity, only care about time

Front 85

Nucleus Laminaris

Back 85

- ITD tuning through binaural convergence
- response of right ear is delayed relative to left ear
- response is greatest when sound hits right ear before the left ear--> b/c of delay, compensate--> responses line up

- as get deeper, prefer sounds hitting the ears at the same time, then when ipsi-lateral is first

- map of ITD, as get deeper, systematic shift in preferred ITD--> Jeffers models

Front 86

PLLN

Back 86

- ILD turning
- Contralateral excitation from NA
- Ipsilateral inhibition from NA

Inferior Colliculus get bilateral input from PLLN
- At high intensities, inhibition "closes" the ILD tuning curve--> neurons respond most strongly to a specific range of ILDs

Front 87

Mammalian sound localization

Back 87

2 Parallel Pathways, but both code for azimuth
- ITD: low frequency (MSO)
- Thick axons, heavy myelination
- Adrendritic cells, no need to integrate
- Endbulb synapses
- High [Ca binding protein] (shorter refractory)
- Fast acting glutamate receptors and K channels (speed)

- ILD: high frequency (LSO)
- Small axons, typical myelination
- Diffuse dendritic arborizations
- Small synaptic boutons
- Low [Ca binding protein]
- Slow acting glutamate receptors and K channels

Front 88

Evidence for and against the MSO in mammals fitting the Jeffress model

Back 88

For:
- most MSO neurons receive binaural input
- MSO neurons are bipolar (dendrites coming out of 2 sides), like in NL
- Overrepresentation of low frequencies in MSO tonotopic map
- Phase-locking enhanced in spherical bushy cells of the VCN that provide excitatory input to MSO
- Most MSO neurons are tuned to particular ITDs; the preferred ITD can be predicted by relative delyars of monoaural input
- Albino cats have atrophy of MSO neurons w/ corresponding behavioral deficits in azumuthal sound localization and reduced ITD sensitivity in IC

Against
- Only weak evidence for map of ITDs in MSO
- A-P axis: short-long ITDs; based on combining data across animals
- Maps seen in OT (SC) but not IC--why lose it and then construct it?
- Some projections to MSO match delay line concept
- Contralateral projections to MSO have A-P delay line config
- Ipsilateral projections have less steep P-A delay line config
- Only true for some fibers
- Evidence for maps & delay lines is only from large mammals
- Peak ITD often outside behaviorally relevant range (greater than 90 degrees)
- Strong phase-locked inhibition to MSO via MNTB/LNTB...why?

Front 89

Maybe a population code for ITD in mammals?

Back 89

Steepness of ITD curves is maximal across the behaviorally relevant range of ITDs
- Could be represented by relative amts of activity across the population of L and R MSO neurons
---> Temporal code converted

Precisely timed inhibition serves to adjust the gain of this curve so that the max change in response occurs over the behaviorally-relevant range of ITDs
--> entire curve moves up; peak to beh-rel range--> not as big a jump from min to max in beh-rel range

Front 90

Different solutions for ITD in birds and mammals

Back 90

Owls
- Place code
- Spatial map of ITDs
- Excitation
- Delay-line coincidence detection

Mammals
- Population code
- No clear map of ITDs
- Excitation and inhibition
- Inhibitory adjustment of coincidence-detection

WHY?
- Small mammals vs. big owls
- Small differences in ITDs w/ azimuth due to small head size may be better coded as the balance of activity across a large number of neurons, rather than location of neurons tuned to a narrow range of ITDs---> BUT chickens use owl's strategy and have a similar head size to gerbils
- Frequency differences
- NL neurons in owls are tuned to much higher freq than MSO neurons in mammals
- Tuning to low freq result in broader ITD tuning curves--> may not be precise enough--> BUT hcickens are tuned to lower freqs than owls
- Independent origins
- May not be homologous pathways; may have evolved independently
- Different strategies
- Mammals use ITDs at low freq and ILDs at high freq, w/ some overlap at extremes--> may require a common currency
- Owls only use ITDs for azimuth--> can be unique code for only ITDs

Front 91

Ochracea: parasitoid fly

Back 91

How can such a small animal localize a cricket using sound?
- Distance btw ears is .5mm
- ILDs are undetectable (no sound shadowing)
- Maximum ITD: 1.5 microseconds

- Direction and magnitude of turns vary w/ sound azimuth; significant for even 1% deviation from midline

Special ear: Mechanical coupling of the two ears like a see-saw--> amplified ITD and adds in ILD
- Converts ITDs 1.5 microsecond to 55 microseconds
- Creates < 10dB difference in vibration amplitude

- Receptors fire a single spike in response to each chirp at a fixed latency (greater for contralateral responses than ipsilateral)
- Latency decreases w/ increasing intensity
- Difference in intensity btw 2 ears caused by mechanical coupling--> further increases the effective ITD

Front 92

Visual Calibration of Auditory Map

Back 92

Plasticity of Sound Localization
- Plugging ears leads to systematic errors in sound localizations--> young owls adjust to errors in 2-3 weeks--> removing plugs leads to errors in opposite direction
- Post removal errors are reversible, but not if owl is blinded or vision is displaced by vision--> VISION GUIDES DEVELOP. OF AUDITORY MAP

- Prisms immediately shift visual orienting responses
- Auditory orienting responses slowly shift to match th visual responses
- Removing prisms causes visual orienting responses to immediately return, but not auditory responses

Front 93

Visual and auditory pathways converge in...

Back 93

Optic Tectom

Plasticity of auditory receptive fields
- shifts in auditory receptive fields occurs in optic tectum map
- V & A rec. fields aligned after ear plug has been left in
- Removal of ear plug leads to shift in auditory receptive field

ITD tuning is adjusted to match visual receptive field

Front 94

How is the ITD map reorganized?

Back 94

- No physical changes in ICC, but changes in ICX
- Caused by reorganization of projections from ICC to ICX
- Mediated by feedback from OT
- Persistence of cnnxns may allow plasticity in adults