Psychophysics, Perception and Decision Making, Bayesian Concepts - - PowerPoint PPT Presentation

▶
psychophysics perception and decision making bayesian
SMART_READER_LITE
LIVE PREVIEW

Psychophysics, Perception and Decision Making, Bayesian Concepts - - PowerPoint PPT Presentation

May 05, 2023 6 likes •801 views

Psychophysics, Perception and Decision Making, Bayesian Concepts Pascal Mamassian Laboratoire des Systmes Perceptifs CNRS & Ecole Normale Suprieure a few key questions in visual perception Newton (1730) stated The rays are not

slide-1
SLIDE 1

Psychophysics, Perception and Decision Making, Bayesian Concepts

Pascal Mamassian Laboratoire des Systèmes Perceptifs CNRS & Ecole Normale Supérieure

slide-2
SLIDE 2

a few key questions in visual perception

  • Newton (1730) stated “The rays are not coloured”. So why do we see colour?
  • Can gene therapy cure colour blindness? Could it help see more colours if

we were able to add a fourth cone?

  • Is the primary visual cortex (V1) necessary for visual awareness?
  • Why are we over-confident for things happening in peripheral vision?
  • Does language influence visual perception?
  • How does vision interact with the other senses?
slide-3
SLIDE 3

Psychophysics, Perception and Decision Making, Bayesian Concepts

Pascal Mamassian Laboratoire des Systèmes Perceptifs CNRS & Ecole Normale Supérieure

slide-4
SLIDE 4

Rock, I. (1983)

The Logic of Perception Cambridge, MA: MIT Press

slide-5
SLIDE 5

visual perception is uncertain

slide-6
SLIDE 6

Visual uncertainty: problem 1. limited hardware

slide-7
SLIDE 7

Thus, the response can be identical for:


  • a weak light at the wavelength of peak sensitivity

(few incident photons, a large fraction of them absorbed)


  • a strong light at a wavelength of lower sensitivity

(many incident photons, a small fraction of them absorbed)

Univariance Principle

Wavelength (nm) Relative sensitivity

“The output of a receptor depends upon its quantum catch, but not upon what quanta are caught.”
 (Rushton, 1972)

slide-8
SLIDE 8

! !

Sinha, P . & Adelson, E. (1993)

Visual uncertainty: problem 2. lack of information

slide-9
SLIDE 9

Hubel, D. H. & Wiesel, T., N. (1968)

Receptive fields and functional architecture of monkey straits cortex Journal of Physiology, 195, 215-243

Visual uncertainty: problem 3. neural noise

slide-10
SLIDE 10

Aperture problem (barberpole illusion)

Wallach, H. (1935)

Ueber visuell wahrgenommene Bewegungsrichtung Psychologische Forschung, 20, 325–380

slide-11
SLIDE 11

Lorenceau, J. & Shiffrar, M. (1992)

The influence of terminators on motion integration across space Vision Research, 32, 263-273

slide-12
SLIDE 12
slide-13
SLIDE 13

Weiss, Y, Simoncelli, E. P ., Adelson, E. H. (2002)

Motion illusions as optimal percepts Nature Neuroscience, 5, 598-604

Physical direction

  • f motion

Perceived direction

  • f motion
slide-14
SLIDE 14

Weiss, Y, Simoncelli, E. P ., Adelson, E. H. (2002)

Motion illusions as optimal percepts Nature Neuroscience, 5, 598-604

slide-15
SLIDE 15

Weiss, Y, Simoncelli, E. P ., Adelson, E. H. (2002)

Motion illusions as optimal percepts Nature Neuroscience, 5, 598-604

slide-16
SLIDE 16

von Helmholtz, H. (1910)

Handbuch der Physiologisehen Optik (Dritter Band) Hamburg und Leipzig: Verlag von Leopold Voss [Helmholtz’s Treatise on Physiological Optics (vol. 3), J. P . C. Southall (Ed.), The Optical Society of America, 1925]

Visual uncertainty: solution: use of priors

slide-17
SLIDE 17

According to the Bayesian framework, all visual perception can be seen as the resolution of an inference problem: What is the most probable world scene that is responsible for the retinal image?

Mamassian, P ., Landy, M. S. & Maloney, M. S. (2002)

Bayesian modelling of visual perception In R. Rao, B. Olshausen & M. Lewicki (Eds.) Probabilistic Models of the Brain. Cambridge, MA: MIT Press

slide-18
SLIDE 18

Kersten, D., Knill, D.C., Mamassian, P . & Bülthoff, I. (1996)

Illusory motion from shadows Nature, 379, 31

slide-19
SLIDE 19

Kersten, D., Knill, D.C., Mamassian, P . & Bülthoff, I. (1996)

Illusory motion from shadows Nature, 379, 31

Mamassian, P ., Knill, D. C. & Kersten, D. (1998)

The perception of cast shadows Trends in Cognitive Sciences, 2, 288-295

slide-20
SLIDE 20
slide-21
SLIDE 21

! !

!

slide-22
SLIDE 22

thin strips in relief thick strips in relief

Mamassian, P . & Goutcher, R. (2001)

Prior knowledge on the illumination position Cognition, 81, B1-B9

slide-23
SLIDE 23
  • • •

Mamassian, P . & Goutcher, R. (2001)

Prior knowledge on the illumination position Cognition, 81, B1-B9

slide-24
SLIDE 24

"thin" score

.2 .4 .6 .8 1

Mamassian, P . & Goutcher, R. (2001)

Prior knowledge on the illumination position Cognition, 81, B1-B9

slide-25
SLIDE 25

Johannes Vermeer (1668)

De astronoom (The Astronomer) Oil on canvas, 51 x 45 cm, Musée du Louvre, Paris

Eugène Delacroix (1830)

La liberté guidant le peuple (Liberty Leading the People) Oil on canvas, 260 x 325 cm, Musée du Louvre, Paris

Analysis of paintings in the Louvre museum

slide-26
SLIDE 26

Mamassian, P . (2008)

Ambiguities and conventions in the perception of visual art Vision Research, 48, 2143-2153

Analysis of paintings in the Louvre museum

slide-27
SLIDE 27

Gregory, R, L. (1966)

Eye and the Brain: The Psychology of Seeing London: Weidenfeld and Nicolson

slide-28
SLIDE 28

Is Bayesian inference
 always the right framework?

!

Adelson, E. H. (1995)

slide-29
SLIDE 29

Is Bayesian inference
 always the right framework?

!

slide-30
SLIDE 30

Is Bayesian inference
 always the right framework?

!

!

Mamassian, P . (2000) Adelson, E. H. (1995)

slide-31
SLIDE 31

psychophysical methods

(

slide-32
SLIDE 32

Accuracy vs. Precision

sensitivity = precision bias = lack of accuracy In Signal Detection Theory, sensitivity: d’, area under ROC bias: criterion

slide-33
SLIDE 33

Psychophysical Tasks: discrimination

A B Is stimulus ‘B’ larger than stimulus ‘A’? A B Repeat for multiple values of ‘B’, keeping ‘A’ constant.

slide-34
SLIDE 34

Kingdom, F. A. A. and Prins, N. (2010)

Psychophysics: A Practical Introduction Academic Press: London

Psychophysical Tasks: discrimination

PSE PSE: Point of Subjective Equality psychometric function discrimination threshold

slide-35
SLIDE 35

Sensory Evidence (s)

Each stimulus category ‘A’ and ‘B’ is represented as a probability distribution along a sensory continuum. On each trial, it is assumed that the sensory evidence is a sample from

  • ne of these distributions.

Signal Detection Theory: Discrimination

slide-36
SLIDE 36

−4 −2 2 4 0.1 0.2 0.3 0.4 Type 1 Evidence Probability Density Type 1 Probability Distributions

Signal Detection Theory: Discrimination

criterion

Sensory Evidence (s)

Each stimulus category ‘A’ and ‘B’ is represented as a probability distribution along a sensory continuum. On each trial, it is assumed that the sensory evidence is a sample from

  • ne of these distributions.

The observer places a criterion along her continuum and decides to respond ‘A’ whenever the sample is to the right of the criterion (and ‘B’ if left).

slide-37
SLIDE 37

Signal Detection Theory

−4 −2 2 4 0.1 0.2 0.3 0.4 Type 1 Evidence Probability

Hit: p(Resp = A | Stim = A)

−4 −2 2 4 0.1 0.2 0.3 0.4 Type 1 Evidence Probability

False Alarm: p(Resp = A | Stim = B) criterion

Sensory Evidence Sensory Evidence

Sensory Evidence (s)

slide-38
SLIDE 38

0.25 0.5 0.75 1 0.25 0.5 0.75 1 Type 1 FA Rate Type 1 Hit Rate −4 −2 2 4 0.25 0.5 0.75 1 Type 1 Criterion Type 1 Correct

Signal Detection Theory

criterion

probability correct (depends on criterion) Receiver Operating Characteristic (ROC) Area under ROC is a criterion-free measure of sensitivity

Sensory Criterion False Alarm Rate Probability Correct Hit Rate

Sensory Evidence (s)

slide-39
SLIDE 39

0.25 0.5 0.75 1 0.25 0.5 0.75 1 Type 1 FA Rate Type 1 Hit Rate

Signal Detection Theory

d’ (in σ unit)

p(False Alarm) p(Hit)

d’=0 d’=1 d’=3

Sensory Evidence

False Alarm Rate Hit Rate

Receiver Operating Characteristic (ROC) Area under ROC is a criterion-free measure of sensitivity d’ (“d-prime”) is another criterion- free measure of sensitivity d’ = z(Hit) – z(FA) z(.) = norminv(.)

slide-40
SLIDE 40

psychophysical methods

( )

slide-41
SLIDE 41

Tse, P . (2005)

Voluntary attention modulates the brightness of overlapping transparent surfaces Vision Research, 45, 1095-1098

An example of top-down effect

slide-42
SLIDE 42

Thompson, P . (1980)

Margaret Thatcher: a new illusion Perception, 9, 483–484

Thatcher illusion

“There was once an urban myth that the illusion only worked for Margaret Thatcher’s face.” –Peter Thompson, The Guardian, 19 September 2016

slide-43
SLIDE 43

Tangen, J. M., Murphy, S. C., & Thompson, M. B. (2011)

Flashed face distortion effect: Grotesque faces from relative spaces Perception, 40, 628-630

Flashed face distortion effect

slide-44
SLIDE 44

colour adaptation produces complementary colour afterimages

slide-45
SLIDE 45

colour adaptation produces complementary colour afterimages

slide-46
SLIDE 46

! !

shift of the equilibrium point “white” towards adapted colour a physical white is now perceived greenish physical colours perceived colours before adaptation adapted colour perceived colours after adaptation

colour adaptation produces complementary colour afterimages

slide-47
SLIDE 47

Rob van Lier, Mark Vergeer & Stuart Anstis (2008)

colour adaptation produces complementary colour afterimages

slide-48
SLIDE 48

Jeremy Hinton (2005) Lilac chaser

slide-49
SLIDE 49

John Sadowski (2006)

colour adaptation produces complementary colour afterimages

slide-50
SLIDE 50

time when the fixation dot turns red, is the stimulus oriented more clockwise (R) or counter-clockwise (L)? not to scale

slide-51
SLIDE 51

The probability of seeing “Left” at time t depends on
 both recent stimuli (5-10 sec) and older stimuli (5-10 min).

Chopin, A. & Mamassian, P . (2012)

Predictive properties of visual adaptation Current Biology, 22, 622-626

Proportion “Left” seconds ago Proportion “Left” minutes ago Proportion “Left” perceived now Proportion “Left” perceived now

The new percept makes the recent and global statistics more alike.

slide-52
SLIDE 52

Bistability: perceptual dynamics

slide-53
SLIDE 53
slide-54
SLIDE 54
slide-55
SLIDE 55

Mamassian, P . & Wallace, J. (2010)

Sustained directional biases in motion transparency Journal of Vision, 10(13):23

N=688 (with Mark Wexler) N=34

slide-56
SLIDE 56

Wexler, M., Duyck, M., & Mamassian, P . (2015)

Persistent states in vision break universality and time invariance Proceedings of the National Academy of Sciences USA, 112(48), 14990-14995

days bias (deg)

Initial Bias Bias 1 Year Later

Stability over 1 year

slide-57
SLIDE 57

McNeill, C. (26 February 2015)

The Dress Tumblr

(non-) universality of perception

slide-58
SLIDE 58

Mamassian, P ., Landy, M. S. & Maloney, M. S. (2002)

Bayesian modelling of visual perception In R. Rao, B. Olshausen & M. Lewicki (Eds.) Probabilistic Models of the Brain. Cambridge, MA: MIT Press

slide-59
SLIDE 59

Crazy Road Traffic Mumbai (2012) YouTube

slide-60
SLIDE 60

time Visual Stimulus T T time Motor Response

(T = 500 msec)

Methods

Stimulus: 3 pairs of dots forming a hexagon, presented sequentially Task: anticipate the occurrence of the 3rd pair of dots Reward: 100 points if timing occurs within a pre-defined interval

slide-61
SLIDE 61
slide-62
SLIDE 62

Reward: + 100 points if on time and – 200 points if a bit too late

slide-63
SLIDE 63

What is the optimal behavior?

slide-64
SLIDE 64

Mamassian, P . (2008)

Overconfidence in an objective anticipatory motor task Psychological Science, 19, 601-606

Participants did not shift early enough their motor response. This is indicative of “overconfidence” in the sense that participants underestimated their motor uncertainty.

slide-65
SLIDE 65

Jazayeri, M. & Shadlen, M. N. (2010)

Temporal context calibrates inteval timing Nature Neuroscience, 13, 1020-1026

slide-66
SLIDE 66

Meta-perception

slide-67
SLIDE 67

courtesy Daniel Simons

slide-68
SLIDE 68

courtesy Daniel Simons

slide-69
SLIDE 69

courtesy Ulrich Neisser

slide-70
SLIDE 70

courtesy Daniel Simons

slide-71
SLIDE 71

courtesy Kevin O’Regan

slide-72
SLIDE 72

courtesy Kevin O’Regan

slide-73
SLIDE 73

m = stimulus strength response “up” response “down”

response “up” response “down” response “first”

stimulus

de Gardelle, V., & Mamassian, P . (2015)

Weighting mean and variability during confidence judgments PLoS ONE, 10(3), e0120870

Confidence forced-choice

Barthelmé, S. & Mamassian, P . (2010)

Flexible mechanisms underlie the evaluation of visual confidence PNAS,107, 20834-20839

slide-74
SLIDE 74

−10 −5 5 10 0.2 0.4 0.6 0.8 1 Orientation ’m’ (deg) Proportion Response ’m > 0’

de Gardelle, V., & Mamassian, P . (2015)

Weighting mean and variability during confidence judgments PLoS ONE, 10(3), e0120870

Confidence forced-choice

Barthelmé, S. & Mamassian, P . (2010)

Flexible mechanisms underlie the evaluation of visual confidence PNAS,107, 20834-20839

slide-75
SLIDE 75

−10 −5 5 10 0.2 0.4 0.6 0.8 1 Orientation ’m’ (deg) Proportion Response ’m > 0’

de Gardelle, V., & Mamassian, P . (2015)

Weighting mean and variability during confidence judgments PLoS ONE, 10(3), e0120870

Confidence forced-choice

chosen declined Confidence The ability of participants to discriminate trials that lead to better performance is a signature of metacognition

Barthelmé, S. & Mamassian, P . (2010)

Flexible mechanisms underlie the evaluation of visual confidence PNAS,107, 20834-20839

slide-76
SLIDE 76

Mamassian, P ., Landy, M. S. & Maloney, M. S. (2002)

Bayesian modelling of visual perception In R. Rao, B. Olshausen & M. Lewicki (Eds.) Probabilistic Models of the Brain. Cambridge, MA: MIT Press

slide-77
SLIDE 77

Mamassian, P ., Landy, M. S. & Maloney, M. S. (2002)

Bayesian modelling of visual perception In R. Rao, B. Olshausen & M. Lewicki (Eds.) Probabilistic Models of the Brain. Cambridge, MA: MIT Press

Valid Percept Confidence

Mamassian, P . (2016)

Visual confidence Annual Review of Vision Science, 2, 459-481

slide-78
SLIDE 78

Valid Percept Confidence

There are multiple sources of uncertainty in visual perception. However, most of the times

  • not only we are not aware of these uncertainties,
  • but we can also make good confidence judgments about our

perceptual performance.

slide-79
SLIDE 79

The end