Notesale: Turn your study into money

Document Preview

Extracts from the notes are below, to see the PDF you'll receive please use the links above

---

Lecture 3 – Visual Cognition
RECAP 1
Linguistics = scientific study of language and its structure
Psycholinguistics = the psychological basis of language and how its formed
Lexicon = collection of words
Syntax = rules and principles that govern sentence structure
Grapheme = smallest meaningful unit i
...
letters
Phoneme = basic unit of language, combine to form words
Orthographic = Vision-based
Phonological = Sound-based








The mental lexicon – a metal storage of information about words (semantic, syntactic, word forms –
spelling and sound)
What is visual word recognition? – Knowing whether a letter string is a real word? Familiar? Regular
consistent words are pronounced most quickly
...

o Interactive-activation Model – extension of logogen showing HOW the word units become
activated
...


What and where are the two basic questions to be answered in visual perception
What (object/face recognition) and is the Ventral Stream
Where (location) and is the Dorsal Stream
Today we are going to focus on the ‘what’ part

Visual Cognition

Object Recognition
It seems clear that, when we recognise an object, some representation based on incoming sensory information
is matched to some stored representation of the object
...

However, in practice, theories of recognition by humans have described the representation in terms of the forms
of (parts of) objects and how they are put together
...
Examples are on the right
...

EDGE EXTRACTION- responsive to differences in surface characteristics, namely, luminance, texture,
colour, providing a line drawing description of the object
Next step, to decide how a visual object should be segmented to establish its parts/components
...
It provides an answer to how we identify objects in spite of substantial
differences among the members of any category in shape, size and orientation
...
In addition, concavities and edges are
of major importance in object recognition
...
it focuses primarily on bottom-up processes triggered by the stimulus input and therefore de-emphasises
that importance of top-down processes based on expectation and knowledge
...
it accounts only for fairly
unsubtle perceptual discriminations (decide whether the animal in front of us is a cat or a dog, but not how we
decide what particular breed it is)
...
the notion that objects consist of invariant geons is too inflexible

Grandmother cells and Ensemble coding (consistent with Hubel & Wiesel – neurons fired when at a certain
orientation):
The grandmother cell is a hypothetical neuron that represents a complex but specific concept or object
...

Not due to visual acuity impairment
Could describe objects – not name them
Could identify using other senses
Used pantomime to aid identification

Agnosia – failure of recognition
Visual agnosia = defined as the inability to identify objects by sight in the absence of any significant visual or
intellectual impairment (not due to memory loss, it is a recognition problem)
...


Apperceptive agnosia – patient fails to identify an object because they are unable to form stable
representations or impairment in perceptual processing – CANT COPY PICTURES

2
...
Simultaneous agnosia (Balint’s syndrome) – inability to perceive more than one object at a time,
competing stimulus
...
Colour agnosia – unable to differentiate between colours or relate colours to objects (intact colour
vision), colour processing
5
...
Auditory agnosia – failure to recognise sounds

Are Faces Special?
How does face recognition differ from the recognition of other objects? Face recognition involves holistic
processing (Young, Hellawell, and Hay, 1987) - strong integration of information from an entire object focussing
on the relationships among features as well as the features themselves
...

Evidence that holistic processing is used more than for objects comes from studies on the FACE INVESION
EFFECT (McKone et al
...


They found that faces are much harder to identify when presented inverted or upside-down rather than
upright
...
BUT Diamond and Carey (1986) found dog
experts had comparable inversion effect for dog breeds and faces
...
But, if the
pictures are inverted, performance on the faces is worse than on the other objects
...
Didn’t find effect in animals/houses – so face is special
More evidence to support that faces are special comes from the PART-WHOLE EFFECT (that it is easier to
recognise a face part when presented in a whole face that in isolation
...
If such patients had great impairments
in object processing as well – would suggest the two types of processing are similar
...
These individuals have PROSOPAGNOISA
...

Thought to identify individual faces by deviations from it (Leopold et al, 2001, 2005)
...


Attractiveness?
Average man and average woman – rated high in attractiveness
Sizes and positions are means of those features in a large no
...


Neural Basis of Face Processing
If faces are processed differently from other objects, we might expect to find brain regions specialised for face
processing
...

Two main lines of research support the involvement of this area in face processing:
-

This area is often damaged in patients with prosopagnosia (Kanwisher & Yovel, 2006)
Neuroimaging studies indicate that activation in the FFA is typically greater to faces than to other
objects (Downing et al
...
, (1969) – neurons in the inferotemporal cortex of the monkey had visual receptive fields
which were very large
...




Perrett et al
...
, (1997)
o

Using fMRI, an area in the Right Fusiform Gyrus (in 12 out of 15 patients) was more active when viewed
faces than when viewed assorted common objects
...


o

BUT only 80% of the participants showed greater activation within the Fusiform face area to faces than
other objects
...

o
o

2% of the population suffer from prosopagnosia
Caused by a stroke or brain damage

o

o

o
o

o

o

The precise problems of face and object recognition VARY across patients – due to brain damage
(ACQUIRED PROSOPAGNOSIA) or in the absence of obvious brain damage (DEVELOPMENTAL
PROSOPAGNOSIA)
...
She didn’t realise until she was in her 30s because she identifies people on their
hairstyle, body type, clothing, voice
Case study – Patient BR YOUTUBE(didn’t know own face, family faces)
Simon et al
...
In spite of that, PS showed more activation in a brain area
associated with face processing (the fusiform face area) when presented with familiar (but not
unfamiliar) faces
...

Farah et al
...
Can
recognise people by what they are wearing, hearing their voice etc
...
(1999) – congenital prosopagnosia without MRI-revealed brain damage

Patients with prosopagnosia often have damage to the occipital face area as well as (or instead of) the Fusiform
Face Area
...


Summary



Visual processing
Theories of object recognition



Visual agnosia
o types of visual agnosia



Facial recognition – is it special?



Prosopagnosia



Overall – what does neuropsychological evidence tell us about a typical working brain? The Fusiform
Face Area is definitely involved in face processing and face recognition for most (but not all) individuals
...
What is more likely is that
face processing involves a brain network including the fusiform face area as well as the occipital face
area (Atkinson and Adolphs, 2011)

---