High level perception

Question 1

Difference between low level and high level perception

Answer 1

Low-level perception involves processing raw sensory data, like color and shape, while high-level perception interprets and makes sense of that data at an abstract, conceptual level.

Imagine you’re looking at an object. Low-level perception would detect the edges and colors of the object.
High-level perception would then recognize that those edges and colors belong to a specific object, like a chair, based on the overall shape and context.

Question 2

High level perception

Answer 2

It involves understanding and interpreting the sensory data to create a sense of the world, like identifying objects, scenes, and behaviors.

Question 3

Object recognition

Answer 3

The ability to know what an object is - it involves identifying the shape of the object (despite changes in sensory input) and retrieving information from the LTM about the object (e.g. its function, size, colour, etc)

Question 4

How long does object recognition take

Answer 4

Around 200ms - very rapid due to this neurological machinery firing away to allow you to recognise objects without putting in effort

Question 5

Why is it important to study object recognition? Brain damage

Answer 5

Object recognition is driven by a complex neurological system that is hard to pull apart. Until we have some in depth knowledge about how the neurological system works it is really hard for us to understand what is going on clinically for someone who has brain damage

This has a knock on effect - means it is challenging to develop new treatments and rehabilitation techniques that aim to target those areas of the brain and its function

Question 6

Object constancy

Answer 6

Challenges the brain has to overcome to accomplish object recognition

Defined as the ability to recognise objects across variation in sensory input caused by changes in light (shadow), scale (size), viewpoint and occlusion

Question 7

Object recognition - contextual clues

Answer 7

We make use of contextual cues - despite all these changes in sensory input, the brain can recognise objects

It is possible that we store long-term memory (LTM) representations of objects in a highly abstract form, which enables us to match incoming perceptual information to these representations and thereby recognise the object we are viewing

Question 8

Object constancy: the problem of shadow

Answer 8

When there is a shadow it is challenging to identify the edges that define the object shape

Shadows themselves create edges. The brain has to distinguish between what the edges of the shadow are and what the edges of the object are

The visual system must work to figure out which edges belong to the object and which belong to the shadow

Question 9

Area V1

Answer 9

A very good edge detector

But we live in a 3D world, so objects are often covered by other objects

The brain has to reconstruct all of the missing info

Question 10

Object constancy: the problem of the variations in scale (size)

Answer 10

Object size changes on the retina depending on how far away they are

1. Representations must be abstract enough to allow you to recognise the object that is independent of the size on the retina
2. But at the same time you must have some stored info about how large objects should typically be

Question 11

Object constancy: variations in spatial location

Answer 11

Object position on the retina changes as objects move about

Question 12

Object constancy: the problem of occlusion

Answer 12

Caused by scene clutter - foreground objects partially occlude background objects

A lot of the time we have missing info

The brain doesn’t need all of the complete info to recognise things.

We can still recognise objects despite incomplete sensory input

The brain has to reconstruct all of the missing info

Question 13

Gestalt Laws of perceptual organisation - meaning

Answer 13

A set of principles proposed by Gestalt psychologists in the early 20th century to explain how we naturally organize visual elements into groups or unified wholes when perceiving complex scenes.

The visual system uses ‘bottom up’ processes to group image features into shapes and forming a whole

These laws reflect the idea that “the whole is greater than the sum of its parts” — our brains tend to perceive structured, organised patterns rather than random arrangements.

We tend to group things together using certain principles - the gestalt laws

Question 14

What are the gestalt laws

Answer 14

Law of similarity

Law of simplicity

Law of proximity

Law of continuity

Law of closure

Question 15

Law of closure

Answer 15

We tend to fill in missing information to perceive a complete, whole object, even when parts are missing.

We are able to a modally complete this image - in the absence of complete info of an object, we tend to see it as a complete object

Question 16

What two different ways do we organise sensory input

Answer 16

Bottom up processing is used for our sensory information and top down processing is used for our previous knowledge about things, to organise sensory input

Question 17

What does perception of shape involve

Answer 17

Both ‘bottom up’ and topdown’ knowledge

Question 18

What pathway is associated with object recognition

Answer 18

The ventral pathway

Question 19

What will damage to the ventral stream lead to

Answer 19

All sorts of impairments e.g. object recognition problems but not face recognition problems or word recognition problems but not object or face recognition problems

Question 19

What is the ventral pathway associated with

Answer 19

It is associated with p cells - cells that rapidly respond to high spatial frequency (detailed images and detailed info such as edge info)

Question 20

Main regions of the brain that have been linked to the visual perception of different types of objects

Answer 20

We have specialised neural areas in the brain for recognising different types of objects

FFA

PPA

LOC

Body area

Question 21

FFA

Answer 21

Fusiform Face Area

Much of what we know about face processing involves the FFA. It is located in the inferior temporal cortex. This part of the brain gets activated when you show faces

Question 22

PPA

Answer 22

Parahippocampa Place Area

Gets very excited when you show images that depict scenes
e.g. beaches or city scenes or outdoor landscapes

Question 23

LOC

Answer 23

Lateral Occipital Complex

Located on the lateral surface of the inferior tempal cortex

Underpins the ability to recognise three dimensional objects - so everyday objects

Question 24

Evolution - specialisation for object recognition

Answer 24

It seems that ovewr the course of evolution, the brain has thought the most optimal way to recognise things is to have some degree of specialisation - one bit of brain tissue that is focused on recognising different sensory info - specific brain tissues that recognise objects, a specific tissue that recognises faces etc

Question 25

Body Area

Answer 25

Located on the lateral region of the inferior temporal cortex. This part of the brain responds to anything to do with the human body - if you showed someone the heads, shoulders, arms etc the Body Area is involved in classifying things as the human body

Question 26

Body Area - evolution

Answer 26

Important for us to recognise people and sitinguish between people and other animals for survival purposes. Seems we haver some specialised area in the brain that is focused on classifying things as the human body

Question 27

Why do we have patients with object recognition problems but not face recognition problems and vice cersa

Answer 27

Asymmetry in the brain E.g. FFA seems to be more lateralised to the right cerebral hemisphere PPA activation in both cebral hemispheres - could argue it is more lateralised to the left cebral hemisphere but it is not as striking as the FFA We do have asymmetry across the ` and left cerebral hemispheres so to some degree we have unequal distributions of functions across the right and left cerebral hemispheres

Question 28

Importance of edges - what shape information is used to represent objects

Answer 28

Low-level image features such as edges and vertices (intersections between edges) provide important shape information

Question 29

Attneave (1954) 'Sleeping cat'

Answer 29

Line drawing of a sleeping cat can still be identified when the smoothly curved contours are replaced by straight line segments Even with most of the visual detail removed, people could still easily recognise the image as a cat. This suggests that we don’t need every detail to recognise objects — our visual system is highly efficient and uses salient features (like edges and curves) to identify things quickly. It supports Gestalt ideas like simplicity, closure, and good continuation — our brains “fill in the gaps” to form a complete perception from minimal information.

Question 30

Biderman (1987)

Answer 30

Biderman presented objects with different amounts of edge contour and vertices deleted and measured recognition accuracy The results showed that deletion of vertices affects recognition more than deletion of other edge contour Vertices carry more information about a shape than edges

Question 31

Surfaces

Answer 31

Give us information about texture, colour and patterns of objects

Question 32

Volumetric parts

Answer 32

Need to be able to identify parts of an object to identify it as a whole - spoken about more in theories of object recognition

Question 33

What are the problems of face recognition

Answer 33

Face recognition presents a difficult computational problem for the visual system: Object constancy problems still apply to face recognition as well, but face recognition also poses other problems - faces are visually similar - requires subordinate level classification - perceptual input highly variable across viewpoint - this means that we tend to interact with people from all kinds of viewing positions - faces are dynamic, moving stimuli - faces are variable because of emotional expressions and speech

Question 34

Subordinate-level classification

Answer 34

Occurs when we have to distinguish one category member from another (e.g. the faces of two different people or two different kinds of a car).

Question 35

Basic-level classification

Answer 35

Recognition of a particular category e.g. distinguishing a table from a chair

Question 36

An information processing model of face recognition

Answer 36

Bruce & Young (1986) The model argues you must have some sort of processing module in your head that allows you to identify the face of a person that is independent of the emotional expression that the face is showing. Two systems independent of each other - one for recognising the face of someone and another to recognise some sort of emotional expression that is shown on that face - face identify and expression might be separately stored in memory

Question 37

Evolutionary origin of face recognition

Answer 37

Important to distuinshish between other people and animals for survival purposes - might help to explain why we have these specialised brain areas - FFA Can help to explain why we can see faces in random configurations of things - that is because your brain is so adapt to recognising faces that it can quickly put things together and recognise it as a face The brain automatically uses principles of perceptual organisation to impose order on sensory information.

Question 38

Abstract representations of face identify

Answer 38

In order for us to see faces among random objects, we must have some sort of very abstract representation of face identity

Question 39

Face inversion effect

Answer 39

Shows how faces are different to objects - we find it harder to recognise inverted faces than inverted objects

Question 40

Processing of faces vs objects

Answer 40

Research has argued that we process faces and objects fundamentally differently with objects, we care moe about the individual parts, but we process faces more configurally, this means that instead of the indivudal parts we care more about the relationship between the parts So it is really important for us to identify for example the handle of a cup but less important to identify someone's nose - what we care more about is the relationship between the parts

Question 41

The 'Thatcher' illusion

Answer 41

Face perception is 'tuned' to the upright orientation - we do not notice feature inversion in the upside down image This suggests that face processing relies on configurable or 'holistic' information about the whole stimulus, rather than an 'analytical' processing of each face feature

Question 42

What does the Thatcher illusion tell us about the differences between object and face recognition

Answer 42

Can use the thatcher illusion as evidence to demonstrate that we process objects and faces fundamentally differently Shows us that we don't care about indidvcual parts of a face - what we care more about is the relationship between the parts

Question 43

Passive viewing of faces and objects during fMRI scanning - study

Answer 43

The Fusiform Face Area showed greater activity for faces than objects The Lateral Occipital Complex showed greater activity for objects than faces

Question 44

Bentin et al, 1996

Answer 44

If you show someone a stimuli of a face and measure the electrical activity, you can observe a phenomenon that is often referred to as the N170 The N170 is a face-specific EPR component The N stands for negative and It represents this negative deflection that we get after the onset of a stimulus The 170 records roughly the time at which we get this negative deflection (happens at around 170ms)

Question 45

The Perceptual Expertise hypothesis

Answer 45

It states thee is no domain-specific processing for faces. Rather, the face-specific effects previously found in other studies solely reflect subordinate-level processing of faces Argues that N170 is not face specific and you would get this if you got people to distinguish between two visually similar things