T3: Sensation and Perception

Attention

preferentially process some parts of a stimulus at the expense of processing of other parts of the stimulus

Why is attention needed

perceptual system has limited capacity - can’t process everything simultaneously

• helps us avoid being overwhelmed

Overt attention

looking directly at an object

covert attention

looking at one object but attending to another object

fixation

when we look at an object

saccades

the ballistic (very fast) eye movements bw fixations

• jumping around when looking

Two processes direct our attention

• Initially, fixations are often involuntary

  • attentional capture

• Then, the fixations are voluntary, goal directed and are influenced by our expectations

Involuntary direction of attention

Attentional capture: when scene initially presented, fixations captured by salient parts of scene - INVOLUNTARY

What influences attention?

the salience of a stimuli - what captures our attention

• Regions of colour contrast or luminance contrast

• Regions of size contrast

• Regions orientation contrast

• Regions of motion/flicker contrast

Voluntary direction of attention

After initial first few fixations, you can direct your fixations towards goals

What influences the voluntary part of fixations

determined by cognitive factors such as the observer’s goals and expectations

Expectations influence on fixation

If an object is unexpected, you will fixate on it for longer and fixate it more often

• eg. semantically inconsistent: not on theme/doesn’t belong

  • eg. laminator on stove

• syntactically inconsistent: doesn’t make sense

  • eg. pan hovering in thin air

(Vo & Henderson, 2009)

What are the effects of attention

• Attention speeds responses

• Attention can influence appearance

• Attention can influence physiological responding

Attention speeding responses evidence

Posner (1978)

• Valid = arrow points to where X is and you attend there so the time taken to spot X is quicker

• Invalid = throws you off as you are attending to opposite side so slower

• Neutral = No indication so takes longer

Attention changing contrast of an object

Carrasco et al. (2004)

• attention makes perception more vivid - appearance of stim. changes

• makes objects appear to have higher contrast

Attention influencing physiological responding

neurons in the brain respond more strongly to attended stimuli than to unattended stimuli

The Binding problem

Different aspects of a stimulus are processed independently, often in separate brain areas

• eg. motion is processed by the dorsal stream and form is processed by the ventral stream

The issue of how an object’s individual features are combined (i.e. bound) to create a coherent percept is known as the binding problem.

• eg. how can we see a red vertical and green horizontal? how do we associate correct colour with correct object?

Feature integration theory (FIT)

suggests that the binding problem is solved by attending to only one location at at time.

• only features assoc. with that location are processed and so only those bound tgt

Illusory Conjunctions

Treisman & Schmidt (1982) said that binding of two diff objects mixed up creating percept that does not exist

• Their goal was to attend to the digits so they weren’t attending to the letters. Therefore, they would get the colours and letters right but mix up which colour for which letter

• This is because they weren’t attending to them → creating an illusory conjunction

Balint’s syndrome

RM has parietal lobe damage and when multiple objects present, he finds hard to focus attention on single object

• when shown 2 letters with diff colours - reported wrong 23% of time even with 10secs to see them

• prone to illusory conjunctions bc could not focus attention on just one object

Visual search

Binding might be necessary for visual search if target has same features as distractors

Conjunction search

If target differs from distractors only by particular conjunction then conjunction search

• FIT predicts that in this type of search - attention needs to apply to each object one at a time to determine whether or not the attended object is targer

• very slow search

Feature search

other type of visual search where target contains feature that distractors dont contain

• binding doesn’t need to occur and don’t need to look at each on at a time

• faster

Key idea about visual searches

• Visual searches that require the binding problem to be solved (e.g. conjunction searches) are predicted to be slow

• Visual searches that don’t require the binding problem to be solved (e.g. feature searches) are predicted to be fast.

Change blindness

Attention can also determine what we remember.

• If you don’t attend to it, chances are you won’t remember it.

• We can only remember some parts of a scene at one time

  • if one of those parts change, you notice, if other part changes - you won’t notice = change blindness

Motion transients - why change blindness doesn’t occur all the time

• changes cause motion transients that draw attention to location change = so its easier to spot change

  • when blank screen in bw - motion transients occur for EVERY part of image not just changed parts so no guide for attention to change

Change blindness TUTORIAL

• Central Interest (CI) vs Marginal Interest (MI)

  • Changes to objects people considered important/central to the scene's meaning (CIs) were detected much faster than changes to peripheral/unimportant details (MIs) — even when the MI changes were objectively larger in size.

  • This shows attention is guided by high-level interest/meaning, not just low-level visual salience.

• A → A’ one flick trial

The Verbal Cue Experiment

• When participants were told in advance what part of the scene to watch (a valid cue), detection sped up dramatically — for both CI and MI changes. Invalid cues slightly slowed things down.

• Why this matters: it proves the difficulty isn't about visibility (the information was always there) — it's about knowing where to direct attention

Why some best computer algorithms misclassify objects

common objects presented at unusual angles misclassified - so object perception is hard

• The stimulus on the retina is ambiguous

• Objects can be hidden or blurred

• Objects look different from different viewpoints and in different poses

Difficulty 1: stimulus on retina is ambiguous

Difficulty 2: objects partially occluded or blurred

Difficulty 3: objects look diff from diff viewpoints

Machines find it hard to recognise objects when they appear in unexpected poses or are viewed from unexpected angles

How do humans succeed?

• Structuralism

• Gestaltism

Structuralism

distinguishes bw sensations and perceptions

• sensations: elementary processes that occur in response to stimulation

• perceptions: conscious awareness of objects and scenes

Conscious awareness is the sum of elementary sensations

• contains nothing that was not already present in elementary sensations

Gestaltism

directly contradicts Structuralism - conscious awareness is more than the sum of the elementary sensations

• conscious awareness can have characteristics not present in any of the elementary sensations

Gestaltism evidence

There are two main pieces of evidence that support the claim that conscious awareness can be more than the sum of the elementary sensations

• Apparent motion

• Illusory contours

1. Apparent motion

observer sees two stationary dots flashed in succession and although stationary we perceive motion

• conscious awareness/percept has character (MOTION) not present in elementary sensations (bc both stationary)

2. Illusory contours

• Illusory contours are seen in locations where there are no physical contours.

• The conscious awareness of the illusory contour is constructed – there is no physical contour at these locations.

Final gestaltism vs structuralism

There is plenty of evidence that conscious awareness is constructed and can contain characteristics not physically present in the image.

• eg. motion can be perceived when there is no motion in the image (e.g. apparent motion)

• eg. contours can be seen when there are no contours in the image (e.g. illusory contours)

This evidence argues against Structuralism but in favour of Gestaltism.

Perceptual organisation

Together, grouping and segregation allow a scene to perceptually organised into its constituent objects thereby allowing observers to make sense of the scene.

Grouping

process by which parts of an image are perceptually bound together to form a perceptual whole (e.g. the perception of an object)

Segregation

process by which parts of a scene are perceptually separated to form separate wholes (e.g. the perception of separate objects)

Gestalt principles of grouping

Grouping is governed by 5 key principles. The more of these principles that apply, the more likely components of an image will be grouped together to form a perceptual object.

Original Gestalt principles

• Good continuation

• Prägnanz

• Similarity

• Proximity

• Common fate

Two additional ones (added later)

• Common region

• Uniform connectedness

Good continuation

Aligned (or nearly aligned) contours are grouped together to form a single object.

• This is why contour A is grouped with contour B, instead of with contours C or D

Prägnanz: principle of good figure

Essentially, groupings occur to make the resultant figure as simple as possible.

• In the figure you see a panda, not a collection of unconnected splotches.

Similarity

The more similar objects are, the more likely they will be grouped together.

• In a), all the dots are the same colour so it is unclear whether things are organised vertically or horizontally.

• In b), colour similarity groups the dots into columns

Proximity

The closer the dots are, the more likely they are to be grouped together

• In b), grouping by proximity forms horizontal rows

Common fate

Things that are moving in the same way are grouped together.

Common region

Elements that are within the same region of space tend to group together (Palmer, 1992)

Uniform connectedness

Connected regions with the same visual characteristics (e.g. colour) tend to group together

Segregation - not just within

not enough to group components of an image together to form an object, you also need to segregate the different objects in the scene from each other and also segregate the objects from the background

Figure-ground segregation

objects are normally perceived as “figures” and the background is typically perceived as the “ground”

• Consequently, if you can identify what the figure is, you can typically identify the objects.

Figural properties

Regions of the image are more likely to be seen as figure if:

• They are in front of the rest of the image

• They are at the bottom of the image

• They are convex

• They are recognisable.

Rubin vase - figural properties

Regions of an image in front of the rest of the image tend to be seen as figures

(i.e. they are seen as objects)

• Rubin vase is ambiguous as it can be perceived as either a vase or two faces

Lower areas perception as figures

No left-right bias though

• only lower ground = figural properties

Vecera et al. (2002)

Convexity (Peterson & Salvagio (2008))

Convex regions are assumed to be figures (i.e. objects)

Experience effect of determining figure

People also used past experience to segregate overlapping objects

Experience

As a) is in a familiar orientation it is easier to segregate it from the background than in b)

• once we see it, we cannot unsee it

Gist perception

When scenes are flashed rapidly in front of an observer, she may not be able to identify all the objects in the scene.

• she might think that the image shows “a crowded cafe” so overall impression is gist

Gist perception

Potter (1976) studied gist perception using the following paradigm.

• In each trial, the observer was cued with a particular scene description. - eg. bridge

• Then she saw 16 randomly chosen scenes, each for 250 ms.

• Then she was asked if any of the scenes fitted the description.

• Observers were at near 100% accuracy.

• This showed that observers can rapidly perceive a scene’s gist.

Fei-Fei → minimum exposure time req. for gist perception

could start perceiving aspects at 27ms

• longer time = more detailed + accurate descriptions

• THEREFORE, although gist extracted very quick - may not be detailed - very accurate perception in 250ms but gist in 27ms

Function of motion perception

motion attracts attention and help us segregate objects from the background.

 Help break camouflage

 Help attract attention

 Help segregate objects from the background.

 Help us interpret events.

 Help us determine the structure of objects

 Help us determine what actions people are performing

How motion helps interpret events

By seeing how objects interact, you can infer causality relationships and even social relationships.

• gestures etc.

Inferring structure from motion

help us determine the shape of a moving object

• This is sometimes referred to as the “kinetic depth effect”

• eg. seeing smth from one angle and then it turning and you can recognise as 3D

Interpreting actions from motion

While static poses are often ambiguous when someone moves, their actions and intentions are often made clear

Point-light walkers

Point-light walkers are created by placing lights on a person’s joints and having them perform an action (e.g. walking!)

• humans so good with processing stim. that they guess what action

Condition where can no longer perceive motion

akinetopsia

Case study of akinetopsia: L.M.

She had difficulty pouring cup of tea, crossing street, following speech

• could see that things HAD moved but couldn’t see them MOVING

• no motion perception

When do we perceive motion?

• Real motion (smth actually moving)

• Illusory motion (nothing actually moving)

• Motion aftereffects

• Induced motion (moving background causing still object to appear to move)

Types of Illusory motion

Rotating snakes illusion (Kitaoka & Ashida, 2003)

Apparent motion

Rotating snake illusion

Don’t know why static illusion gives impression of motion

• caused by contrast bw the colours but unclear why that leads to percept of motion

Apparent motion

Series of stationary images presented in succession to give impression of motion

• eg. two slides with dots in diff positions looks like dot moving

Korte’s Third Law of Apparent Motion

• only works with dots sufficiently close tgt - if too far apart then no motion perception

• as separation increases alternation rate needs to decrease

Apparent motion and colour

mostly insensitive to colour changes

• but colours changes can make movement appear to be going certain ways

Motion aftereffect

when something is constantly moving in one direction, if you look at smth near it - all of a sudden it will look like its moving the opposite direction

Induced motion A

A nearby object (usually a large one) either affects the perceived motion of a second object (usually a small one) or causes a second object to appear to move

• Large, surrounding objects tend to be treated as the "stationary frame"

• Small, enclosed objects tend to be treated as "the thing that's moving"

Motion induced change blindness

• Normally colour changes attract attention because of the transient signals associated with the change (i.e. the “flicker”)

• However, when things are moving, there are transient signals associated with all objects – not just the ones that are changing

• Thus, attention is no longer drawn preferentially to the changing objects, so the changes are not noticed.

EG. when dots are stationary, its easy to notice them changing colour but when they move its harder to notice

Motion illusions

inform us of the processes underlying motion perception

• eg. the Footsteps Illusion shows us that contrast affects motion perception. As the yellow and blue rectangles traverse the striped background, they appear to speed up and slow down asynchronously.

Explaining contrast in motion illusions

higher contrast = easier to see = appears faster

• As the rectangles traverse the striped background the contrast at the lead edges vary.

• When the contrast is high, they appear to move faster.

• When the contrast is low, they appears to move slower.

• When the contrast the leading edge of the yellow rectangle is high the contrast at the leading edge of the blue rectangle is low (and vice versa)

• Consequently, the yellow and blue rectangles appear to speed up and slow down asynchronously.

Apperture problem

If you can’t see the ends of a line, the movement of a line is ambiguous.

• the motion of the line viewed through an aperture is ambiguous and is “captured” by the movement of the terminators (the points where the line joins the aperture)

• When the terminators move horizontally, the line appears to move horizontally.

• When the terminators move vertically, the line appears to move vertically.

• When we can see that actual ends of the line, we can see its actual motion

Barber Pole Illusion

In this illusion, lines curve around a cylinder creating virtual terminators that move vertically.

Thus, the lines themselves appear to move vertically.

Function of colour perception

• Find things (like berries)

• Determine if fruit is ripe

• Spot (and identify) poisonous animals

• Identify a potential mate

Visible light

• Visible light is electromagnetic radiation with wavelengths varying from about 400 nm to about 700 nm

• White light is a mixture of all these wavelengths

Colour of opaque objects

• An opaque object is an object that light cannot pass through. It is not at all transparent.

• The colour of an opaque object is determined by the light that it reflects

• If an opaque object reflects all light, it will appear white

Colour of Transparent Objects

The colour of a transparent object is determined by the colour it transmits.

• eg, If an object absorbs blue but transmits red, then it will appear red

Mixing paint

• Blue paint absorbs red light but reflects blue and green light.

• Yellow paint absorbs blue light but reflects red and green light.

• Mixing blue and yellow paint results in a mixture that absorbs both red and blue light but reflects green light – so looks green.

  • Blue = short (S) wavelength light, Green = medium (M) wavelength light, Red = long (L) wavelength light

Mixing light

• Mixing red and green light makes yellow light.

• If you mix blue light with yellow light (i.e. light that contains both green and red), you will get light that contains blue, green and red.

• This light will look white because that is what what white light is - a mixture of all three colours.

Munsell colour system

In the Munsell colour system colours are categorised according to:

 Value (lightness)

 Hue (colour)

 Chroma (saturation)

Trichromatic theory of vision

In the retina, there are photoreceptors known as rods and cones.

• Rods cannot distinguish between colours and are active only at low light levels.

• In normal light conditions, only cones are active.

3 Types of Cones

How we distinguish bw colours using cones

By comparing the relative activities of these three types of cones

• Eg. if L cones are most active = light is primarily red

Colour matching

Consider a pure green light that only contains a wavelength of 500 nm

• It will strongly activate the M cones and the L cones but only weakly activate the S cones

Colour matching by adjusting relativities of cone types

Looks identical to the test field even though physically not the same/identical

Metamers

Physically different stimuli that appear the same

By adjusting the amount of blue, green and red light we

can exactly match the cone activations caused by 500

nm test field even though the light patches themselves will be

physically different (i.e. one is pure 500 nm light while

the other is a combination of red, green and blue light).

Types of colour deficiency

monochromatism and dichromatism

Monochromatism

usually have no functioning cones and only have functioning rods

• colourblinded and only see shades of grey, very sensitive to light (affects 1 in 100,000 people)

Dischromatism

Dichromats are lacking one of the three types of cones

3 Types of dichromats

• Protanopes

• Deuteranopes

• Tritanopes

What colours do dichromats actually see?

Colour matching as dichromats

possible to match a patch of any coloured light using just two lights

• just need spotlights matching whichever cones they have