ANIM3320 Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

Humans have three cone photoreceptors:

A
  • Humans have three cone photoreceptors:
    o Each one has a spectral sensitivity -> absorbs wavelengths differently
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
1
Q

What is colour

A

light….
colour is … and ….
colour signal is affected by…
why does a banana appear yellow…

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Spectral sensitivity of a photoreceptor

A

photoreceptors
single photo receptor…
two photoreceptors… and neutral point

three photoreceptors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Principle of univariance

A

The principle of univariance means that any single cone system is color blind in the sense that different combinations of wavelength and intensity can result in the same response from the cone system. This implies that color vision depends critically on the comparative inputs of the different cone systems.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Testing for colour vision is all about controlling intensity

A

o you can have achromatic contrast only, where the circle differs from the background due to brightness
o you can have chromatic contrast only, where the background differs by colour
o and you can also have both which is the background differs from the circle due to colour and brightness
o To show colour vision one needs to eliminate the brightness contrast meaning the brightness is controlled

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Achromatic and chromatic contrast

A

o viewed through brightness channel, in other words viewed by a colourblind system-> Won’t be able to detect the circle with chromatic contrast only
o viewed through spectral (colour) channel only -> won’t be able to detect the circle with achromatic contrast only

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q
  • Colour constancy: colours appear to be the same despite a change in illumination
A

simple systems may achieve colour constancy,
simply by receptor adaptation to the mean light
intensity—known as a von Kries mechanism . This allows the animal,
with minimal neural processing, to recognize an object in
sunlight or shade and even in a patchy light environment

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Dual forced choice discrimination task between two coloured lights of varying intensities (brightness)

A

task between two coloured lights of varying intensities (brightness)
o This experiment shows that the animal has colour vision but doesn’t say what their colour vision is like
o if they can correctly pick two different stimuli regardless of intensity they have colour vision

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Evidence for dichromacy is the existence of a neutral point -> this is the null point for dichromats only

A

o when the ratio of the coloured line is about the same as the ratio of the white light that is where the null point is -> cannot distinguish lights
o this is the point along the wavelength spectrum where a dichromat cannot distinguish a single monochromatic colour from a broadband light (usually white). Trichromats do not have such a point
o Dichromats have a narrow range of which they can discriminate colours
* the null point experiment is a conclusive way to show an animal is a dichromat

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Behavioural experiment for trichromacy:

A

colour threshold experiments and colour mixing experiments can be used to infer dimensionality of an animal’s colour space

using additive colour mixture
experiments, where choice
between a coloured light (training
wavelength) and an additive
mixture of two different coloured
lights (primary wavelengths) is
based on differences in chromatic
content and occurs regardless
of relative brightness

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Become familiar with the structure of invertebrate eyes

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q
  • Ommatidia
A
  • Ommatidia: makeup compound eyes
    o crystalline cone focuses light onto rhabdom (light sensitive structure)
    o each ommatidium contains several photoreceptor cells
    o each photoreceptor cell consists of two main sections, the cell body and the rhabdomere
    o the cell body contains the nucleus, while the rhabdomere is made up of toothbrush like stacks of membrane called microvilli
    o the membrane of the rhabdomere is packed with millions of visual pigment molecules
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q
  • advantages of compound eyes
A

o wide and flexible field of view projected onto small sheet of receptors
o large depth of focus sensitive to movement at any distance with small lenses so they don’t need to accommodate
o short path length of light through the ommatidia reduces loss of UV radiation, allows for wide spectral range for vision
o spherical and chromatic aberration negligible due to short focal length
o polarisation vision

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

disadvantages for compound eyes

A

o diffraction by small lens facets limits spatial resolution
 blurs image a lot since the lenses are too small
o small aperture/lenses limit sensitivity
 sensitivity depends on opening of its lens, so many islets cause small aperture

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

polarisation vision

A

Polarization vision is the ability of animals to detect the oscillation plane of light and use it for behavioral responses.

  • works functionally similar to colour vision, and found in many invertebrates
  • visual pigment chromophores preferentially absorb light polarised parallel to their long axis (Z-axis)
    o Higher sensitivity of polarised light along the longitudinal axis of the microvilli
    o all microvilli in the photoreceptors are in the same direction
     Photoreceptor becomes sensitive to that particular orientation of polarisation
  • light from the sun is unpolarized, light reflected in the atmosphere and directed towards the earth becomes more polarised
    o You get polarisation pattern in the sky that is symmetrical around the sun, so you can always see where the sun is if you have polarisation vision
  • Light reflected from a surface such as a body of water is polarised parallel to that surface
  • polarisation compass: used as a compass for where the sun is
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Critical structural components for polarisation vision:

A

o the ability of a chromophores to absorb light depends on its orientation relative to the lights plane of oscillation
o chromophores therefore need to be aligned rather than randomly oriented (microvilli)
o Compare the output signals from photoreceptors with different polarisation sensitivity
 similarly to colour vision

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Advantage of polarisation:

A

o position of the sun or moon
o helps with removing light scatter underwater
o helps with reducing glare, seeing into the water
o provides contrast
o helps identify or find water

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q
  • eye space is limited so…
A

o so light sensitivity and spatial resolution are traded off in all eyes. Colour vision and polarisation vision also make things worse
o large eyes can facilitate lots of photoreceptors and more ommatidia
o To improve both spatial detail and sensitivity you need bigger eyes
 however this is not possible so there is a trade off

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q
  • spatial resolution: sampling resolution
A

o Eyes with high sampling resolution are large, have large lenses (aperture), narrow receptors and a long focal length
o In compound eyes, aperture is too small -> so you get lots of diffraction so the image gets blurred
o sampling resolution depends on the number of pixels that sample the retina

o Poor sampling resolution looks like pixels
o is determined by number of receptors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q
  • Spatial resolution: optical resolution
A
  • Spatial resolution: optical resolution measures optical quality of the eye, not sampling density
    o wider apertures give better optical resolution: the ability to perceive 2 points of light as distinct entities
  • Optical resolution:
    o poor optical resolution results in blurring of vision
    o determined by receptive field size
    optical resolution determines how good the retinal picture is
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q
  • What happens if sampling resolution and optical resolution are very different
A

o you either waste energy densely sampling a poor image or you have gaps in your visual field and will be subject to aliasing

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What determines light sensitivity

A

o eyes with high light sensitivity are large, have large lenses (aperture), long receptors with a large diameter and a short focal length
o big aperture for high light sensitivity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q
  • Sensitivity vs resolution: trade off
A

o in all eyes there is a trade off between resolution and sensitivity
o wider receptors have bigger acceptance angle and catch more light but at the cost of resolving power

23
Q
  • The visual world of fiddler crabs:
A
24
Q

three pillars of experimental design

A

randomisation, replication and control

25
Q

why do we need replication
replication

A

builds confidence in the results, increased power in analysis -> detect smaller differences between treatment groups

estimation of error, increase in precision, larger range of observations (larger sphere of inference)

ensures that measurments are INDEPENDENT of each other
estimate of mean becomes much narrower

26
Q

why do we need randomisation
randomisation

A

To avoid experimental bias
assures equal probability for assignment of treatment groups to test subjects
small sample size or large effect size design may end up very unbalanced if we completely randomise the design-> solution is blocking

27
Q

why do we need controls

A

a control tries to isolate the treatment from its associated effects, used to minimise the effect of variables (that are associated with the treatment but of no interest to the study)

negative control: no response expected of
positive control: a treatment different to the experimental treatment that is expected to produce a response

28
Q

balanced experiment

A

balanced experiments are easier to analyse when we want to explore other (non experimental) influences on the response

balancing an experiment reduces the cross talk (correlation) between explanatory variables (fixed effects)

in completely balanced experiments-> the effect of one variable does not change the effect of another UNLESS there is an interaction between them (interaction meand they are not independent and they affect each other)

29
Q

why do we have to think about noise in our data

noise and replication relationship

A

the structure of the noise determines the analysis
eg. Gaussian, poisson, binomial (data has a distribution and so does the residuals)

independence of measurements means independence of noise

30
Q

variance in the dataset vs residual variance

A

In summary, while variance in the dataset measures the overall spread of the data, residual variance specifically focuses on the unexplained variability remaining after fitting a statistical model to the data.

31
Q

blocking

A

Blocking isthe arranging of experimental units in groups (blocks) with the aim to keep conditions within blocks as similar as possible to help remove variation between treatments caused by the blocking factor (in our case time).
Blocking makes the treatments within a block more comparable.

Randomised block design (blocking)

32
Q

sensory ecology and matched filters

A
  • Matching receivers to signals and matched filters
  • magic word when it comes to matching signals to environments is contrast
    o want to optimise the contrast between signal and background
33
Q
  • sensory diversity: is there because
A

o evolutionary history
o body (size, energy)
o the environment
o signals (communication and others)
o Information- behaviour

  • How do we find out what is driving sensory diversity
    o comparative approaches
    o psychophysics
    o ecology- behaviour
    o robotics
  • information and the organisation of behaviour: sensory behaviour feedback loop
    o Neural mechanisms constrain options of behavioural options/needs
    o behavioural options/needs Affects the timing and the accuracy of information which is then to be processed by neural mechanisms
    o the functional context determines the relevance of information
  • automation drivers:
    o cost, operating environment, size, complexity, task
34
Q
  • body size: high frequency hearing in mammals
A

o allometric relationship Or in its logarithmic form
 allometric relationship is driven by the physics of sound: high sounds get attenuated faster but are easier to separate from each other
o outliers of the allometric line indicates animal doesn’t conform to expected hearing frequency for its given size
 for example moles should have higher frequency but it doesn’t it’s usually worse for their size
o driven by sound localization demands
o Large animals use interaural time difference and the frequency spectrum to locate sounds
 according to the frequency spectrum higher frequencies get attenuated faster
o super small animals have super short interaural time difference
 so sure that the neural system can’t resolve it anymore
o small animals can hear higher frequencies than larger animals
 It is easier to measure the time difference when there is lots of peaks and troughs in the wavelength

35
Q
  • Tuning curves and sensitivity curves:
A

o they measure how strong the signal has to be to trigger a response
o sensitivity curve: how well an animal can hear sounds-> they lower the sounds and plot if the animal can hear it, so it has a peak where there is high sensitivity
o tuning curve: start at a frequency and you make the sound louder until you get a response
 has a trough where there is high sensitivity
o they are predictable because we here under the same physics of sound
 they reflect the physics of sound
 however they do differ because: physics changes the size, there are environmental differences, the biology of what you need to hear differs

36
Q
  • Environmental impacts: evolution of frog calls
  • environmental impacts: variability in bird calls
A
  • Environmental impacts: evolution of frog calls
    o matching signals to environments
    o lives next to waterfall.
    o produces an ultrasonic sound of a very high frequency
     they do this because they live next to waterfalls that have constant noise of water falling down. this allows the signal to be heard by other receivers despite loud noise of waterfall
  • environmental impacts: variability in bird calls
    o correlation between song frequency and ambient noise in urban great tits
    o great tits increase their lowest frequency in noisy environments -> presumably to conserve energy
     minimum frequency increases in loud noise environments, minimum only gets higher in frequency
37
Q
  • Tuning curves for different animal groups
A

o There is more variability in the tuning curve for mammals because there is a bigger size range and they are sensitive to a wider range of noise

38
Q
  • Matching sensitivity and signal: call Frequency and hearing sensitivity in a bat
A

o Bat is extremely sensitive to its own call but not to other sounds
o high contrast as the noise is not contaminated by nearby frequencies so they can hear the sound very easily
o Particularly sensitive to 60 kHz, which is the main frequency of their echolocation call
o Echolocation stands out, you don’t have to extract it from overall noise
 there is a large contrast which simplifies information processing

39
Q
  • Tuning of calling hearing sensitivity: call frequency and hearing in a katydid
A

o example of a dynamic matched filter
o the call is around 5 KH but the hearing sensitivity peaks around 15 KH
o Their call is at lower frequencies than is optimal for hearing
o signalling of the same sensitivity as what is optimal for their hearing would also attract predators and mates
 the solution is to signal at a lower frequency and change their own hearing sensitivity by closing their Spiracle
 closing Spiracle changes hearing sensitivity

40
Q
  • How to measure environment impacts on perception:
A

o electrophysiological recording from neurons in the auditory system of the animal itself
o records neural activity to sounds they hear in their actual environment
o maps animals hearing using the animal itself

41
Q
  • environment: active range and information content of cricket calls
A

o the receivers neural response drops differently with distance depending on habitat
 there is a faster drop as sound diminishes quicker in dense bushland compared to open grassland
o the signals active range increases with the signal’s height above the ground
 if the signaller is further above the ground they can create a longer range
o directional information improves with receivers height above the ground
 higher up the receiver is the clearer the separation between the sounds
o going up higher means that they can hear further and localise sound better

42
Q
  • grey tree frogs: temperature dependence of calls and hearing
A

o calls changing according to temperature
o there is an overlap in calling in the two species that live in a neighbouring environment which could cause confusion
o frogs changes response rate of females (hearing changes) according to the temperature changing
 to the female frogs it would sound the same and get the same input depending on different temperatures
 hearing changes according to temperature and there is no information processing involved as it is an automatic process

43
Q
  • matched auditory filter in flying crickets
A

o The natural song of a cricket is approximately 4.7 kilohertz
o a bat species uses calling frequencies of more than 15 kilohertz, and these prey on the crickets
o crickets move towards this artificial song playing when it is below 10 kilohertz approximately, but when the song nears 15 kilohertz they move away because their predators have calling frequencies of 15 KH
 this is a simple threshold
o they ran an experiment where they trained to habituate the crickets to one frequency on one side of the threshold
 if they hear the sound from opposite side of the threshold they break habituation
 if they hear sound from the same side of the threshold habituation is not broken
 this shows they don’t know specific frequencies they just know if it is above and below a certain frequency

44
Q
  • How can you use habituation to test whether information is relevant to animals or not?
A

o you can let animals get habituated to a certain stimulus and then change a certain aspect of the stimulus
o if animal detects a clear change they often dishabituate

45
Q
  • risk assessment: a matched filter
    in fiddler crabs
A

o The visual horizon of fiddler crabs can be used to categorise objects
o information processing in this case is dominated by the eye
o they have a social zone which is eye level and anything above the social zone is the predator zone
o decision is egocentric meaning it is relative to the eye position
 only gets a response when it reaches part of the crabs eye that looks above the visual horizon also known as the predator zone

46
Q
  • Burrow surveillance: crabs can measure distance between intruder and burrow
A

o successful burrow defence is crucial for survival
o crabs respond to intruders when they reached a certain distance from the burrow not themselves
 crabs are able to measure distance between burrow and intruder so they can get to the burrow first
o there is a neural hypothesis that assumes crabs keep burrow in the same part of their visual field at all times
 aligning their body axis with their refuge and the eye with the flat environment
 centres the burrow on the most sensitive part of the eye
 allows for simple neural computations

47
Q
  • Why do fiddler crabs keep their eyes on long stalks
A

o expand the angular extent of the ground plane-> measure elevation
o helps to separate predators and conspecifics-> by making sure larger con specifics do not penetrate horizon
o the penalty they pay for this is a loss of stereopsis

48
Q
  • Matched filters in humans:
A

o we use filters to recognise faces
 if an image triggers the filter such as illusions we get a strong sense of a face even if it is completely wrong

49
Q

o The dimensionality of an animal’s colour space is -1 the number of photoreceptor types they have

A

because the colour space does not include intensity differences. They need to be calculated out by the nervous system
* the main mechanism with which it is done is colour opponent ganglion cells

50
Q

trichromatic colour space
dichromatic colour space and Tetra chromatic colour space

A

trichromatic colour space is two dimensional-> hue and saturation
dichromatic colour space is one dimensional-> not hue or saturation and is perhaps a mixture

tetrachromatic-> the extra dimension is unknown to us

51
Q

random noise vs bias

A

random noise:
won’t lead to wrong decision
The noise can come from differences between our subject, measurement error, or random influences from other sources

bias:
will lead to the wrong conclusions
Bias is usually introduced if treatments correlate with some other factor that is not experimentally controlled. The effect of bias is that we see differences between our treatments that stem from a correlated factor rather than the treatment itself.
need to balance out experiments

52
Q

residual data

A

leftover variation in the data that the model couldn’t explain. It represents the random fluctuations or unaccounted factors that influence the data but are not captured by the model. Reducing residual data helps improve the accuracy of the model’s predictions.

53
Q

fixed effect

A

Fixed effects are predictors in a model. They consist of main effects and their interactions. They represent systematic or group-level factors that you want to examine to understand their effect on the dependent variable.
Eg: treatment or drug, temperature

54
Q

random effect

A

Random effects, on the other hand, are variables that represent sources of random (usually uninteresting to your investigation) variation in your data. They are often used to model the variation across specific groups or clusters in your data. Random effects are considered random because their levels are assumed to be drawn from a larger population, and you want to account for the variability associated with these groups.
Eg students from different schools, repeat measurements on individual animals, groups etc

55
Q

why do we need to test for all the different combinations and ORDERS of variables in a generalised linear regression approach

A

to test for correlations between fixed effects
to test for the effects of an unbalanced design