PNB 2XA3 - Chps 1,3,4,5

Cognitive psychology

Scientific study of acquisition, retention, and use of knowledge

Did introspection have to be trained?

Yes --> taught vocabulary to give standard for describing what was observed

What is introspection limited by?

Unconscious thoughts --> cannot be explained or described by individual/no way to study this

Does introspection have a way to test its claims?

NO --> derived from personal experience and therefore have to just trust the person's description of their conscious thoughts

Did behaviourism reject introspection?

Yes --> since someone's beliefs, hopes, thoughts are not observable and therefore is not objective data which can be recorded and measured

Transcendental method

Begin with observations and work backwards from the observations to determine how they come about, asking what process must have occurred in order to bring about these effects --> inference to best explanation

Kantian logic

where you begin with a problem and generate a hypothesis about unseen mental events which make the performance possible in order to explain visible data --> continue a loop of testing and hypotheses until the hypothesis fits the pattern of evidence

Edward Tolman

Said learning is not only change in behaviour but also acquisition of knowledge

Tolman's rat maze

Rats left for 10 days in empty maze w/ no motivation --> once food placed in the maze the rats found it very easily due to predetermined knowledge of the maze's layout

Frederic Bartlett

Studied how people fit their experiences and memory into schema and rely on schema to guide understanding + memory

Gestalt psychologists

Argued behaviours and ideas are organized in a way where elements take on meaning as part of a whole --> psychology must understand nature of the whole

What does gestalt theory reject?

Element-by-element analysis

Donald Broadbent

Focused on how we can improve communication between airplanes and control towers --> determined that computer science can be used a language to explain human cognition

Does modern behaviourism consider unobservable entities to be acknowledged in psychology?

YES

2 approaches for data collection

1. Changing conditions and variables then observing how people perform differently

2. Measuring speed --> response time measured to determine what happens in the mind

Which societies is data collected from mostly (fails to reflect properties of overall human species)?

W.E.I.R.D --> Western, Education, Industrialized, Rich, Democratic

Akinetopsia

Inability to perceive motion even though other aspects of vision are normal --> seeing nothing in between when object goes from one position to another (cause it moves, but can't see motion) --> "static photographs"

Layer 1 of retina

Photoreceptors --> rods and cones

Layers 2 & 3 of retina

2 = bipolar cells 3 = ganglion cells

Rods

sensitive to low levels of light + colour-blind

Cones

less sensitive, need more light to operate and sensitive to diff colours, + can perceive fine details

3 types of cones

1. Prefers short wavelengths --> blue/purple perception
2. Prefers medium wavelengths --> green perception
3. Prefers long wavelengths --> red perception

How does pointing to an object help perceive it better?

Focuses the image onto the fovea where cones are concentrated --> cones perceive details well --> gives image in detail

Movement of light stimuli through retina to brain

Photoreceptors --> bipolar cells --> ganglion cells --> converge axons onto optic nerve (fibre bundle) --> LGN (Lateral Geniculate Nucleus) --> primary projection area for vision in occipital lobe

Lateral inhibition

Pattern where cells when stimulated inhibit activity of neighbouring cells --> distorts incoming visual info pattern and produces exaggeration of areas of contrast in visual field

Why do sailors look away from stars in night sky?

To concentrate star image onto rods outside of fovea (in eye's periphery) since rods are used for vision in dim lighting

Mach Bands Illusion

• Cell B is inhibited by neighbouring cells A and C and Cell C is inhibited by neighbouring D cell --> causes less colour perception (i.e. white colour), creating ombre colour gradient - - Cell D inhibited by cell C --> leads to darker colour perception since cells D, E, & F are receiving only moderate input initially

Single-cell recording

Record moment by moment pattern of electrical changes in single neuron

Receptive field

size and shape of visual area where certain cell responds maximally

Types of receptive fields

1. Dot detectors
2. Edge detectors
3. Movement detectors
4. Angle detectors
5. Corners and notches

1. Dot detectors

Cells that fire at max rate when light is presented in a small circular area in a specific position in the field of view

Centre-surround cells (dot detectors)

Light presented to central region of receptive field has one influence whereas light presented to the surrounding ring has the opposite influence

2. Edge detectors

Cells that fire at max only when a stimulus containing an edge of right orientation appears in receptive field --> farther the edge is from preferred orientation = weaker signal

3. Movement detectors

fires strongly if stimulus moves across receptive field

4. Angle detectors

Cells that fire at max rate only if an angle of particular size appears in receptive field

5. Corners and notches

Cells that fire maximally in response to detecting corners and notches

What stimulus do cells in Inferotemporal (IT) cortex respond to?

Human hand

Parallel processing (of visual system)

Simultaneous occurring analysis of information, quicker speed --> neither system working at same time gets priority over other (they work concurrently)

Visual agnosia

Inability to recognize visually presented objects including common items --> can recognize visual orientation --> due to lesions in ventral system

What do lesions in dorsal (where) system lead to?

Inability to reach for objects, cannot determine object's spatial position

Inferotemporal cortex located?

Lower edge of temporal cortex

Posterior parietal cortex located?

Rearmost portion of parietal cortex

3 elements solving binding problem?

1. Spatial position
2. Rhythm
3. Attention

1. Spatial position

Reassembling information can be done via reference to if stimuli is being processed in same position of the brain

2. Rhythm

If info is processed synchronously by neurons then these attributes/info of the object is registered to belonging to same object

3. Attention

To bind together separate features of a stimulus by processing the features together while ignoring background interference

Conjunction errors

Correctly detecting features present in visual display but making mistakes about how the features are bound together --> error in attention

Steps of perception?

1) Info gathering --> collecting info of visual features 2) Interpretation --> deciding on perceptions and interpreting raw data

Can the steps of perception happen out of order?

YES --> interpretation can happen before you start cataloguing basic features --> ex. features in images can depend on how form is organized by viewer

Perceptual constancy

Perceiving constant properties of objects even though sensory info received changes whenever viewing circumstances change

Size constancy

Sizes of objects correct/constant despite changes in retinal-image size depending on viewing distance

Shape constancy

Correctly perceive shapes of objects despite changes in retinal image created by the viewing angle changing

Brightness constancy

Correctly perceive brightness of objects in any illumination --> sensitive to surface orientation relative to available light source to estimate how much light reaches the surface

Helmholtz

Said there's an inverse relationship between distance and retinal image size --> if object doubles in distance than the image is halved in size (principles of optics)

Distance cues

Features of stimulus that indicates object's position

Binocular disparity

Difference between two eyes' views --> leads to depth perception even if no other distance cues are present

Monocular distance cues

Depth cues depending only on what each eye sees individually

Types of monocular distance cues

1. Adjustment/accommodation
2. Pictorial cues
3. Interposition
4. Linear perspective cue
5. Texture gradient cue
6. Motion Parallax

1. Accommodation

Adjustment the eye's lens must make to see an object depends on the object's distance --> lots of adjustment = nearby object

2. Pictorial cues

Relies on physics principles to give impression of depth on a flat surface

3. Interposition

Blocking your view of one object by some other object --> b/c light cannot travel through an opaque object

4. Linear perspective cue

Pattern where parallel lines seem to converge as they get further away from viewer

5. Texture gradient cue

Pattern of change where elements of texture grow smaller and less detailed as they become distant (only if suitably uniform texture in view)

6. Motion parallax

Projected images of nearby objects more quickly and further away objects move more slowly

Ocular dominance

Individual relies on one eye far more than the other --> if you close one eye then the other, will see that one eye has a more centre focus of an image than the other

Associative agnosia

Failure to link what the person sees with their basic visual knowledge --> cannot link perception with stored knowledge, therefore they cannot understand the meaning of objects

Bottom-up processing

Data driven and shaped by stimuli

Top-down processing

Concept driven and shaped by previous knowledge

Transcranial magnetic stimulation (TMS)

procedure causing temporary disruption in specific brain sites to interfere w/ recognition of categories of objects for ex. --> to study brain region functions

Tachistoscope

Device designed to present stimuli for precisely controlled amounts of time --> then followed by a post-stimulus mask to interrupt any continued processing of presented stimulus

Repetition Priming

If a participant views a word then soon views the word again, the first exposure primes the participant and makes them quicker at determining the presented word --> frequent words = less activation level

Well-formedness

How well a letter sequence conforms to usual spelling patterns of english --> enables good prediction for word recognition

Feature nets

A network of detectors, organised in layers, with each subsequent layer having more complex, larger scale objects --> allows for more efficient recognition however at the cost of occasional error

Activation level

The level reflecting where the detector is in the moment and how energized it must become to activate

Why are some detectors easier to activate than others?

Due to difference in how activated each detector is to begin with --> either higher baseline or lower threshold level

Will detectors which have fired recently and more frequently have higher activation level?

YES

What is activation level dependent on (principles of what)?

Recency and frequency

Response threshold

the point where the detector will fire, sending a signal to other connected detectors

Bigram detectors

detectors of letter pairs --> triggered by lower-level detectors and send output to higher-level ones

Distributed representation of knowledge

Knowledge about bigram frequencies is contained in the network, knowledge is represented anywhere and doesn't have local representation, simply represented by pattern of activations distributed across network (only detectable when considering the entire network's functions)

McClelland and Rumelhart Model

Proposed a network better able to identify well-formed strings than irregular strings in context vs in isolation

Excitatory connections (part of McClelland and Rumelhart model)

Allows one detector to activate neighbours

Inhibitory connection (part of McClelland and Rumelhart model)

Causes detector to deactivate another

As part of McClelland and Rumelhart model, is there a one way street of information flow (i.e. from lower level to higher level detectors) or not?

NO one way street --> higher level detectors can influence lower level detectors too --> detectors at any level can influence other detectors at the same level as well

Prosopagnosia

Cannot perceive faces but have normal vision

Super-recognizers

very accurate facial recognition --> can produce social awkwardness

Inversion effect (facial recognition)

Viewing faces upside down causes disruption in recognition due to holistic processing of facial recognition

Holistic perception

Facial recognition depends on the face's overall configuration

Composite effect

When asked to identify only one half of a familiar person's face --> you can't easily do so if each diff halves of two diff faces are aligned --> due to face being processed as a whole (holistically)

Do you rely on internal or external facial features when recognizing someone familiar?

Internal features

Fusiform Face Area (FFA)

Specialized region of the brain which is primarily responsible for facial perception/recognition

Unilateral neglect syndrome

Damage to right parietal cortex causes inability to control how a person focuses their attention on any input coming from one side of their body

selective attention

skill enabling focus on one input or task while ignoring other stimuli in a scene

dichotic listening task

individual asked to focus on only one input from one ear (attended info) and ignore the other ear (unattended info), then required to repeat back what was heard from attended ear (shadowing task)

What attributes could be remembered from unattended ear?

Physical traits and important info, NOT semantic content

What does the dichotic listening task demonstrate?

Human attention is a limited resource which acts as a filter, blocking unattended info while only allowing meaningful cues to be perceived and attended to

How is unattended vs attended info filtered?

Filter shields distractors by inhibiting responses to these stimuli while promoting processing of desired stimuli

Cocktail party effect

shows that only highly meaningful cues break through attention filters

Inattentional blindness experiment

Participants asked to look at fixation target in centre screen and press one of two buttons depending on if the "+" sign that appears has a longer line vertically or horizontally, except on trial 4 the fixation target was replaced by a different shape --> participants asked about this change (89% reported not noticing)

Inattentional blindness

Pattern where people fail to see prominent stimulus even if they are staring at it --> perception requires more than what you see