Brain and cognition (year one) Flashcards
What is an automatic interpretation?
- brain automatically recognises objects, even if they do not exist (if it looks like x, it is x)
- What are info processing systems?
a system that has specific channels to process info to produce a conscious representation
what is the order of info processing systems?
- input, mechanisms, output
What levels of explanations does a theory need to cover?
- behavioural indicators, functional explanation, biological explanation
- what does our interpretation of reality depend upon?
our knowledge
What percentage of cells in the brain are neurons?
10%
Describe the structure of a neuron
- Soma containing nucleus, ER
- Dendrites
- Axon wrapped in Schwann cells
- Nodes of Ranvier in myelin sheath gaps
What is myelination?
Schwann cells wrapping around the axon
What are nodes of Ranvier?
gaps in the myelin sheath, which conduction passes between
What is the resting membrane potential?
-70mv
What is depolarisation?
positive sodium ions flow into the cell as ion channels in dendrites open
What affect do excitatory and inhibitory neurotransmitters have on ion channels?
- EN : depolarizes (sodium ions flow in)
- IN : hyperpolarises (chloride ions flow in)
What is the CNS made up of?
Brainstem, midbrain, forebrain
What is the brain stem made up of?
Spinal cord, medulla, pons, cerebellum
What is the midbrain made up of?
Tegmentum, tectum
What is the tegmentum made up of?
red nucleus, substantia nigra, reticular formation
What is the tectum made up of?
superior colliculi, inferior colliculi
What is the forebrain made up of?
diencephalon, telencephalon
What is the diencephalon made up of?
thalamus, hypothalmus, pituitary gland
Describe the structure of the brain
- two hemispheres, connected by corpus callosum
- hemispheres seperated by longitudinal fissure
What does the cerebrum do?
thinking, learning,emotion, consciousness
What does the medulla do?
respiration, swallowing , vomiting, vasodilation
What does the pons do?
controls sleep/arousal
- connected to the cerebellum
What does the cerebellum do and what does it consist of?
- fine motor skills
- accounts for 10% volume in brain
- 50% brains neurons there
What does the superior colliculus do?
rapid eye movements
What does the inferior colliculus do?
auditory reflexes
What is the tegmentum?
lots of nuclei
What does the reticular formation do?
supervises sleep / arousal of cortex
Which nuclei in the reticular formation releases serotonin?
raphe nuclei
Where do the cranial nerves leave from?
midbrain
What does the substantia niagra do?
voluntary action
What cells are damaged in the substantia niagra in Parkinson’s disease
- dopaminergic cells
What does the thalamus do?
- relay station/ control station
What is the medial geniculate nucleus?
structure between inner ear/ primary auditory cortex
What is the lateral geniculate nucleus?
relays between eyes and primary visual cortex
What does the hypothalamus do?
- four Fs
- feeding, fighting, flighting and sex
What does the pituitary gland do?
- monitors body growth, endocrine glands
- connected to hypothalamus
What does the HPA axis do?
Part of fight or flight response
What is the telencephalon?
- a neocortex with four lobes
What does the telencephalon consist of?
basal ganglia, limbic system, cerebral cortex
What does the basal ganglia do?
regulates movement that is a timed action
- nuclei on side/ top of thalamus
What does the basal ganglia consist of?
caudate nucleus, putamen, globus pallidus
What does the limbic system consist of?
hippocampus, amygdala, fornix/olfactory bulb, cingulate cortex
What does the amygdala do?
emotional processing (fear condition, strong emotional responses)
What does the hippocampus do?
learning, spatial navigation, long term memory
What does the cerebral cortex consist of?
cingulate cortex, frontal lobe, temporal lobe, parietal lobe, occipital lobe
What does the cingulate cortex do?
detection of conflict, reward anticipation
What does the fornix and olfactory bulb do?
processing of smells (connecting smells to episodic memories
What is the neocortex?
- 3-5mm layer on surface
- mostly made of glial cells (gray matter)
What is white matter?
- bundles of axons connecting neurons in grey matter to other parts of the brain
What are the grooves and buldges called?
grooves - sulci
bulges - gyri
What does the occipital lobe do and contain?
visual inputs from lateral geniculate nucleus, processes visual signals
- • Contains retinotopically organised visual maps
What does the parietal lobe do?
info from visual cortex, generates spatial representations for motor control/ action planning
What is the dorsal pathway?
• Dorsal pathway in the two visual streams model
What are sensory homunculi?
: info from touch receptors, which are processed by “maps” on one side of the lateral fissure
What does the frontal lobe do?
planning, working memory, decision making
What does broca’s area control?
speech production
What does the temporal lobe do?
contains ventral visual pathway
What is the fusiform gyrus?
: bottom of temporal lobe, facial recognition
What is Wernicke’s area?
language comprehension
What is a sensory modality?
: dedicated part of the nervous system that responds to physical inputs
Name the sensory modalities and what they respond to
Visual system : light
Auditory system : vibrations in air
Olfactory system : chemicals in nose
Gustatory system : chemicals in mouth
Tactile receptors : mechanical deformation of skin
Nociception : skin damage
Vestibular system : gravity/ acceleration of the head
What is exteroception?
signals about outside world
What is interoception?
signals about inside body
What is transduction
Converts physical energy into action potentials
What does the JND stand for?
just noticable distance
What is Weber’s law?
- JND increases with stimulus intensity
What scaling does time perception show?
weber- fechner type scaling
What is steven’s power law? give the equation and relationship between variables
p(I) = k(I)^a
P = subjective experience of intensity
I = objective intensity
K = proportionality constant
A = exponent (changes depending on stimuli)
- Relationship between physical stimulus and sensation (often non-linear)
which researchers contributed to gestalt psychology?
- Wertheimer, Koffka, Kohler
what is figure ground segmentation?
we segment the scene into figure and ground
Figure owns the borders, ground doesn’t
Bi-stablility : allow us to feel brain switch between interpretations
What is a monocular cue?
work with one eye shut
What are some monocular cues?
Occlusion : near things occlude far things
Linear perspective : parallel lines converge on vanishing point
Texture gradients : texture becomes finer further away
Height : far things are higher
Motion parallax : Near things move faster than far things
Familiar size : things have a standard size
What is a binocular cue
binocular disparity is useful
Only works over 10s of meters
3D cinemas e.g present different images to each eye and create artificial disparity, making image look very deep (touchable)
What area is sensitive to faces
fusiform face area
What is the phenomena of seeing faces in patterns?
pareidolia
What did Galton find regarding the “average face”
- Galton : average face is attractive
Average face may generate a maximal brain response
May be that average faces are healthy which is adaptive
how did important thinkers see our visual experience?
direct and active
how do we now see vision?
active and indirect
how does sensory transduction occur in the eye?
: light reflected into eye is converted to neuronal electrical signals via photoreceptors
what do ganglion cells do?
: output cells which send signals from optic nerve to the brain
name two types of ganglion cells found in the retina and describe their functions
P cells (parvocellular): carry info about colour and fine detail
M cells (magnocellular) : carry low resolution information and detect motion
describe the fovea
high acuity vision and has highest density of cone cells
describe the blind spot
contains an optic nerve bundle and has rod cells
describe how visual information is altered
- Info from left and right visual fields are transmitted to opposite sides of the primary visual cortex (posterior of brain)
- Information in the upper part of visual field is transmitted to lower part of visual field
what is the pathway of information from the retina?
- Info transmitted from ganglion cells via optic nerve, via laterial geniculate nucleus and superior colliculus
o LGN feeds into primary visual cortex
what is retinotopic mapping?
mapping of visual space onto v1
explain cortical magnification
o Disproportionate amount of cortex represents fovea (centre) causes distortion in the cortical representation
o Cortical magnification : gives high amount of space to centre of images
what are ganglion cell receptive fields? explain on and off cells
- Ganglion cell receptive fields : refer to region in visual field that produces neural response in cell
o ON cell :- ON area is in the centre
light in the centre of receptive field will produce a large neuronal response vs light shone in periphery
o OFF cell : - ON area is in the periphery
Light in periphery (receptive field) will produce a large neuronal response vs light shone in centre
what are alternate names for the primary visual cortex?
area v1, striate cortex
how many layers of cells are in the primary visual cortex?
6 layers of cells
what are ocular dominance columns?
each cell prefers either left or right eye input
what are orientation columns?
columns of cells with receptive fields that monitor similar retinal position
what are dorsal and ventral streams involved in?
- Ventral stream (inferotemporal): identity of objects/faces/scenes
- Dorsal stream (posterior parietal) : visual control of action/motion and spatial processing
list and explain some common types of perceptual dysfunction
- Apperceptive agnosia : recognise visual shapes
- Associative agnosia : recognise objects
- Achromatopsia : recognise a colour
- Prosopagnosia : recognise familiar faces
- Alexia : recognise texts
- Social emotional agnosia : interpret nonverbal social behaviour
Describe the case study of DF
DF : VISUAL FORM AGNOSIA
- Damage in inferotemporal cortex
- Very poor shape discrimination, but acuity and colour vision normal
- Task : distinguish square from rectangle (only 80% correct at 2:1 ratio)
- Poor at copying images but can accurately draw from memory (long term memory unaffected)
- Made recognition errors : from a photo of glasses, described the circle and crossbar but guessed bicycle
- Can orientate a stimulus e.g. can post a card manually but can’t match the slots orientation to a picture
- Damage to ventral stream
Describe the case study of RV
RV : OPTIC ATAXIA
- Damage to posterior parietal brain damage
- Problems with visually guided action, visual perception normal
- Problems inserting hand into orientated slot – would orientate hand wrong but COULD report orientation
- Does not grasp objects properly with thumb/ index finger
- Dorsal pathway impaired
what is area v4 involved in?
- Area v4 : involved in colour, orientation and spatial frequency
Shows tuning for shapes of immediate complexity
what is area v5 involved in?
perception of motion
what is contained within the inferotemporal cortex?
- Number of different gyri : (inferior, middle, superior temporal gyrus), fusiform gyrus
what areas aid in facial recognition?
- Areas found in ventral stream aid in face recognition
o Occipital face area, fusiform gyrus
o Cells have different tuning so will show differing activity to moving stimuli
what cells are involved in motion processing and where?
- Magnocellular ganglion cells involved in motion processing in dorsal pathway
- V5/MT important for motion processing
what is the function of the posterior parietal cortex
- Planned movements, spatial reasoning, attention
what is unilateral hemispatial neglect?
impairments in directing attention to parts of the visual field
what wavelength of light is important for vision?
- 400nm (blue) to 700nm (red)
what is achromatic and chromatic information?
- Achromatic info : important for fine detail, depth
- Chromatic info : important for colour
what are metamers?
physically different lights that appear identical
Can happen as there’s only 3 types of cone cells
what is the function of rods?
achromatic night vision
what is the function of cones and what are the three types?
daytime, achromatic and chromatic vision
Red cone : long wavelength sensitive
Green cone : middle-wavelength sensitive
Blue : short wavelength sensitive
what cells are mostly found in the periphery?
rods
what is the principle of univariance?
two different lights resulting in the same cone outputs
o Once absorbed a photon produces the same change in photoreceptor output whatever its wavelength
what do photoreceptors measure?
how many photons have been absorbed
what is protanopia?
: No L cones
Red/ greens not visible
what is deuteranopia?
: No M cones
Red/greens not visible
what is tritanopia?
: No S cones
Mostly grey/ purple visible
what does colour opponent theory state?
- Red and green, yellow and blue cannot be perceived at the same time : opponent colours
- Some ganglion cells are cone opponent
what does perceived colour depend on?
on temporal/ spatial factors (colour contrast/adaptation
- Colour and cognition are also connected (stroop effect)
what is functional localisation?
finding which components of the system are impaired
describe the case study of IES
- 75 year old male
- Bilateral posterior cerebral artery stroke involving ventral occipital lobes including left fusiform and lingual gyrus and right fusiform gyrus
- Also damage to left hippocampus and primary visual cortex
- Upper right quadrant visual field loss : quadrantanopia
- Visual object agnosia and prosopagnosia WITHOUT alexia
- VOA : recognising objects
- Prosopagnosia : face recognition
- Alexia : problems with reading
- Achromatopsia : colour blindness
- No language comprehension/ production problems
- No loss of semantic knowledge
- Named 32.8% of drawn images (BORB)
- 60% recognition of real objects (photorealistic)
- 3% success in famous face naming
- Perceptive integrative agnosia
- Poor ability to line length match and size match = deficit in low level feature perception
- Evidence of damage to lowlevel feature processing, perceptual integration, object constancy, accessing intact stored object knowledge
what is achromatopsia?
colour blindness
describe visual and semantic object naming errors. explain how these errors may arise in relation to brain damage
- Visual error : where response is visually similar to target
- May arise from damage to perceptual system BEFORE access to semantic info, e.g structures supporting colour, shape, low level feature processing, perceptual integration, object constancy
- Semantic : response is semantically related to target, but not visually similar
- May arise from damage to perceptual system AFTER access to semantic info, e.g structures supporting LTM representations of objects, semantic system
what is integrative agnosia and what damage may cause it?
- Difficulty with putting object parts together to see a whole spatial configuration
- Inferotemporal lesions (following PCA stroke)
name and briefly explain the forms of attention
- Active attention : top down, endogenous
• Controlled by intensions/goals/expectations
Passive attention : bottom-up, exogenous
• Controlled by external stimuli - Focused attention : selective
• Directed to one stimuli, amongst distracting stimuli
Divided attention : directed to more than one stimulus at a time - External attention : directed to selection of sensory info
Internal attention : directed to internally generated info - Overt attention : eyes directed to the object that is under focus of attention
Covert attention : focus of attention is independent of where we are looking
what is sound segregation and sound source of interest?
sound segregation : listener has to decide which sound to attend to
sound source of interest : once sound is detected attention must be directed/ kept towards source
what is the cocktail party situation?
when having a conversation at a party, background noise must be filtered out so attention can be focused on the conversation
what is the dichotic listening paradigm?
different auditory message is presented to each ear and must focus attention then repeat the words
- For the two messages to be heard there needs to be
• Clear physical difference, e.g location, pitch
explain Broadbent’s filter model and its critiques
- Series of channels which process different types of info
- Info passes through the channels and reaches a buffer store
- Filter then selects what info is to be further processed (based on properties e.g voice frequency, tone, pitch)
- Selected stimuli reach consciousness, unselected stimuli are rejected
CRITIQUES
o Inflexible
o Salient information can breakthrough into conscious processing
o Breakthrough errors : heard words from irrelevant message if they were probable in context
describe the attenuator model and give critiques
- Anna Treisman
1. Info reaches buffer store by going through channels
2. The filter attenuates instead of eliminating unattended material. All stimuli pass through the attenuator, but not all have the same level of importance
3. Some stimuli reach threshold for detection more easily, so goes to semantic processing and memory level - Expected stimuli have low detection thresholds (top-down influence)
- Unattended stimuli have high thresholds : get filtered out
o Some unattended stimuli have lower thresholds
Words with important meaning
• Words already processed in attended channel
• Words primed by the content from attended channel - doesn’t explain how semantic analysis works
give examples of early selection models
attenuator model
filter model
give examples of late selection models
deutsch and deutsch (1963)
norman (1968)
explain deutsch and deutsch’s model
- Deutsch & Deutsch’s (1963)
o All stimuli reach full processing level and get recognised
o Then selection is made based on semantic relevance for a reaction to be initiated
o Only this info reaches consciousness
explain norman’s model and give critiques
- Filter is as higher preconscious level
- Unattended stimuli can influence behaviour of attended stimuli
ADVANTAGES
No info is missed
DISADVANTAGES
Very demanding, a lot of brain power wasted
explain the perceptual load model and how the study was conducted
- Sees attention as a system with limited capacity
- Demanding tasks : lots of resources needed, so fewer resources will be left for other stimuli
- Combines early/late selection models
Early selection : demanding task, high load
Late selection : easy task, low load
TASK
o Number of letters was manipulated (low vs high load) and interference created using distractor letters
o Interference was greater in low load than high load
High load = no resources left, so distractor is not processed - More flexible and fluid
- Capacity is determined by task demands, characteristics, previous experiences
- Early selection : keeps assumption that perceptual capacity is limited
- Late selection : assumption that perceptual processing is automatic
give examples of models for automaticity
shiffrin and schneiders model (1977)
explain Shiffrin and Schneider’s model
- Controlled processes : effortful, slow/error prone, controllable, attention required
- Automatic processes : effortless, fast/few errirs, outside awareness, uncontrollable
- Both controlled/ automatic processing are based on same cognitive processes and practice increases efficiency
- Controlled to automatic processing is a gradual transition
explain the instance theory of automaticity
- New unpractised tasks – controlled process
Rely on slow, effortful strategies
General algorithm used to solve problems - Familiar practiced tasks – Automatic process
Relevant memory traces are stored and reinforced
These can be retrieved when same stimulus is encountered - Example : learning timestables
explain norman and shallice’s theory
- Supervisory attentional system modulates application of a schema to schema control units
- ## Very detailed explanation
what does retinotopic mean?
the relative spatial structure of the input is preserved
what does contralateral and ipsiateral mean?
- Contralateral : on the opposite side
- Ipsilateral : on the same side
what is unilateral hemispatial neglect?
- Tendency to ignore objects in hemispace opposite to the lesion (contralesional)
what diagnostic tests can be used to diagnose neglect syndrome?
line bisection test, line cancellation, visual search tasks
what is space centred neglect?
- Space centred neglect : affects perception of locations in space
what is object-based neglect?
- Object-based neglect : neglect of an object, even if position has changed
what is perceptual neglect?
- Perceptual neglect : neglect of stimuli in the perceptual input
what is representational neglect?
- Representational neglect : neglect of internally generated image
explain the early views of memory
- Plato : memory is like a tablet of wax
- Seen as a passive store for centuries
- In the last century research has identified : encoding, storage, retrieval, info processing
what is ablation?
removal or destruction of a body part
describe the HM case study
o HM : ablation of right/left medial temporal lobe (including two thirds of hippocampus/amygdala)
o After surgery HM couldn’t store new memories into LTM, could perform tasks requiring STM
explain Murdock (1962) research into short term memory
Murdock (1962)
o Used a free recall test – participants had to remember lists of 10/15/20/30/40 words
o Presentation time for each was 1s/2s
o Recall was immediate
o Found a serial recall curve with primacy/recency effects
explain postman and phillips (1965) research into short term memory
o Free recall with lists of 10/20/30 words
o Two main conditions : no retention interval, recall after 15/30 second delay filled in with counting task
o Presence of retention interval led to recency effect being eliminated
define architecture and processes
- Architecture : way the memory system is organised
- Processes : activities within memory system
describe the cognitive architecture of Atkinson and Shiffrin’s model
- Input is received
- Input in turn generates cognitive processing
- Some behaviour is produced (output)
- Control processes regulate/control flow of info between components
give the components of atkinson and shiffrin’s model
Sensory register : Long term store : Control processes : Response generator Short term store
explain the flow of information through the modal model
stimuli input → sensory register → short term store (rehearsal buffer) → long term store → response generator → response output
what is echoic and iconic memory?
o Iconic memory : visual information
o Echoic memory : auditory information
describe the long term store of the modal model
- All info acquired through learning
- Info stored permanently
- Infinite capacity
- Insensitive to interference
describe the short term store of the modal model
- Processes info from all modalities
- Organises info to produce behaviour
- Sensitive to interference
- Needs constant rehearsal
- Forgetting is rapid without rehearsal
describe control processes in the modal model
- Control input of the short term store (filter amount/quality of information coming from sensory registers
- Control how much info/which info is retrieved from long term store
- Selects type of cognitive operations applied to info in short term store
explain the incorrect assumptions of the modal model
- Items are encoded in LTM as a result of stay in STM
o Items can be processed in STM without moving to LTM
o Type of rehearsing is important, not amount - Short term store considered to process unitary information
o Short term deals with at least two types of information
Verbal info and visuospatial info - ST store can be considered a unit
- Resources engaged by one process are not available to another process (not the case)
Baddeley (1986) : concurrent visual task doesn’t impair performance of verbal task
state some theories of forgetting in short term memory
- Decay theory : fading of memory traces over time
- Interference theory : due to interferences caused by neural events
explain proactive and retroactive interference
o Proactive : previously learned items interfere with learning of new items
o Retroactive : opposite influence from new items to already leant items
Most likely to occur when items are semantically similar (McGeoch & McDonald, 1931)
describe the phonological store
PHONOLOGICAL LOOP
- Speech based info
- Two components : phonological store
o Phonological store
Holds items in a passive way, either acoustic or phonological format
Activation of items decreases over time
Information lost if not refreshed (happens after 2 seconds)
describe the rehearsal process
- Sets items activity back to maximum
- Auditory or visual input
o Visual input : recoded into phonological equivalent by an orthographic to phonological translator
o Auditory input : passes a phonological detector in charge of detecting words - Output : spoken words
describe the visuospatial sketchpad
- Holds/manipulates visuospatial information
- Only ¾ items held simultaneously
- Items can decay
- Two subcomponents : visual cache and inner scribe
o Visual cache : holds visuospatial information
o Inner scribe : carries out cognitive operations on the information (e.g rotating oject) - Separate mechanisms for visual/spatial format information
describe the episodic buffer
- Added in 2003
- Replaced central executive for coordinating between the two slave systems (VSS, PL)
- Aims to explain we perceive things as wholes, not discrete parts
- Integration of info in single percepts is done here
- Integrates info from LTM
- Place where integrated info is held/ available
- to consciousness
- Span of episodic buffer reflects performance of working memory
state the main components of cowan’s working memory model
central executive, long term memory, short term memory
o Central executive : allocates attention to task relative stimuli/ monitor voluntary processing
o Long term memory : central component, memory records have a level of activation which reflects accessibility to consciousness
o Short term store : set of records available for processing by CE
describe cowans three-level view of memory processing
- Set of long term memory records that are not activated are outside the scope of CE active processes
- Set of activated memory records that constitute the ST store that are not focussed on
- The set of items under focus of attention that benefit from cognitive operations
explain the sensory memory, STM and LTM
SENSORY MEMORY - direct represenentation huge capacity - capacity : less than 500ms for visual, 2s for acoustic - decays over time
give the processes which allow information from the sensory memory to reach STM and LTM
SM - STM : explicit encoding
SM- LTM : implicit encoding
give the name of the rehearsal type needed to pass information from STM to LTM
elaborative
explain the central executive
- Monitors attentional processes (dividing/switching/focussing attention)
- Controls info flow within/between PL and VSS
- Controls info arriving from senses and supervises retrieval from memory
explain sensory registers
- Keep trace of sensations until info moves to ST store
- Holds modality specific information for few hundred milliseconds
o Iconic memory : visual information
o Echoic memory : auditory information - Info not accessible
- Large memory capacity
give findings for categorising in the real world
- in lab experiments/ real voting behaviour, physically attractive candidates get more votes
- people infer invisible characteristics (trustworthiness, competence, intelligence) from features (height, symmetry, facial cues - blemishes, jaw) - “vote for me” photos looked more at the person than “don’t vote for him” photos
- manipulates voters
explain rosch’s experiments 1-3
E1 : more attributes were provided for categories at the basic and subordinate levels (best at nonbiological taxonomies)
- E2 : motor programmes : ppts coded the motor patterns used to interact with the same 90 objects) - hard to code people’s responses
- E3 : visually similar category members : high overlap for basic level category (dog), low overlap for superordinate level category (animal)
explain rosch’s experiments 4-6
- E4 : identifying averaged shapes : superordinate categories : difficulty in producing stimuli/ inshape variability
- E5 : looked at whether hearing the name of the category aided in picture detection
- decided if a picture of an object was shown on right or left side, abstract picture was shown on the other side
- basic and subordinate condition had significantly better detection - E6: participants heard a label at the superordinate, basic or subordinate then saw two images
- made a speeded decision about whether the pictures were identical or two different objects
- priming was only significant for basic/ subordinate level names
- superordinate were no help at all
explain rosch’s experiments 7-9
E7 : easier to verify the basic level name for pictures of objects
- saw a label at any 3 of the levels then a colour photo - made decisions about whether label/photo matched or belonged to different categories
- reaction times in were faster for basic level labels for true/ false decisions - E8 : adults/ children chose the pair of pictures that went together
- sorting at the basic level was perfect from age 3-4 up
- took longer for superordinate level to be acquired : 4 years - E9 : children/ adults sorted 16 pictures from 4 basic level categories into groups that went together
- 5/6 year olds : superordinate sorting was worse than basic level sorting
- 8 year olds : superordinate/ basic level sorting was perfect
explain rosch’s experiments 10-12
- E10 : what level do adults prefer to use
- named some or all of the 54 pictures from taxonomies used in E1
- adults prefered to name at the basic level - E11: analysed which concrete nouns were used by a child from her speech corpus
- prefered to label at the basic level - E12 : level most likely to have a sign assigned in ASL
- level most likely is basic level
describe the structure of language
- language has many levels of analysis
- structure : medium of transmission
- phonetics
- phonology
grammar
- morphology
- syntax
meaning (semantics)
- lexicon
- discourse
is language constant?
- language is not constant : words are invented and used in different frequencies over time
- also a change in the meaning of words : e.g gay = happy, prostitute in 1800s, homosexuals in 1920s (1200-2000)
describe the six types of grammar
descriptive grammar : used in language
pedagogical grammar : used for language teaching
prescriptive grammar : rules for correct use of grammar
reference grammar : comprehensive description for researchers
theoretical grammar : universals in grammar
traditional grammar : greek and roman grammar
describe views on prescriptivism
- correct way to speak
- influenced by vocab, grammar and accent
- 1794 : Murrey : published grammar prescribing the use of English
- 1761 : Priestley : custom of speaking is the original and only just standard of any language
describe syntactic structure
- word order is given in each language e.g english = subject verb object
describe views on word meaning
- views of word meaning
- iconic relationships : words - objects
- symbolic relationships : words - concepts - objects
- behaviouralist view : stimuli - word - responses - the same words can mean very different things : Ogden and Richards (1923) found 16 meanings of mean
how many new words in english are created a year
- 5400 words in english created a year
what are some forms of communication in animals
- use of signaling and chemistry, e.g honeybee recruitment dancce
- communication by voice e.g parrots and speech prodcution, dogs/cats can understand speech, whales/dolphins have complex communication
- communication with obvious purpose e.g predator call by vervet monkeys differs depending on danger
define the auditory vocal channel
- language is spoken and heard
define broadcast transmission and directional reception
- those in range can hear the vocalisation and are able to pinpoint where the speaker is
define rapid fading
- spoken language fades not long after it is produced (doesn’t include tracks, recordings or writing)
define the interchangeability
- humans can produce/receive any message
- courtship behaviours are sex dependant
define total feedback
- humans are capable of monitoring and reflecting on our speech
- animals cannot monitor their courtship behaviour
define specialisation
- humans have a dedicated communication system
- e.g dog panting can communicate that it is hot, but it is a physiological signal and not one purely for communication
define semanticity
- language conveys a specific meaning that is associated to real world concepts or situations
- alarm call associated with predator
define arbitrariness
- nature of reality and signal are not linked
- no reason “dog” is word for dog, no reason why a cry in vervet monkeys signals a snake
- but different groups all use the same signal e.g honeybee waggle-dance with direct reflection of physical parameters
define discreteness
- speech uses a small set of discrete tokens
- speech is perceived categorically : 40 ms timing difference in vocal cord vibration is the only different in pin and bin
- waggledance has a specific speed
define displacement
- language can deal with concepts that are not the present tense
- vervet monkeys do not communicate about yesterdays eagle, waggledance is only displaced in space not time
define productivity
- humans can use a set of symbols to create an infinite range of meaning
- can also invent new words
define traditional transmission
- language is taught, not inherited
- bee dances are innate
define duality of patterning
- sounds in language must be combined to form units which have meaning
- bee dances/ animal calls don’t combine meaningless elements to make a larger meaningful element
explain the difference between index and symbols
INDEX VS SYMBOLS (DEACON)
- indexical communication (alarm calls) : a sound or sign represents an object or schema
- symbolic communication in humans : meaning is extracted from a combination of signals, not singular signs
give differences between vocalisations and language
VOCALISATION LANGUAGE
Indexical symbolic
- given sound stands for something - given sound stands for many things
- monkey call indicates predator - explicit, learnt code for entity
- laughter indicates mental state - context dependent (grammar)
Involuntary Voluntary
- oops - have to learn language
- impossible to fake - have complete control
- contagious
- easy to learn - hard to learn (index based)
- limited to scope - unlimited scope
give examples of primates being taught language
- chimps cannot be taught human speech as they dont have suitable vocal tracts but can be taught sign language (Gardner and Gardner 1969)
- can put signs together productively : used signs for water bird as swans
- however takes generous interpretation of gestures to say chimps have learned a lot of words - Nim Chimpsky : chimp could put together two-word sentences that followed a predictable word-order (e.g more drink)
- 85% of sentences with more followed this order
- however isn’t evidence of productivity : nim couldve learned the sequences as wholes from his trainer
- longer utterances don’t obey rules of syntax - kanzi : bonobo, reared in a human like environment and had a keypad to communicate
- good comprehension, could comprehend novel sequences
- production could be subject to over-interpretation
- 95% of utterances were requests for food/activities
explain why human language is advanced
- humans have evolved an innate universal grammar, teaching chimps to use a few words doesnt mean they’re doing something similar to us
- built in mentalese : humans have a language of thought that came before spoken language
- language is our ability to translate mentalese into strings of words - only humans take part in collaborative acts that require understanding other’s intentions
explain what makes up words
- many graphemes that describe distinct sounds
- phoneme : speech sound
- some phonemes always need letter groups e.g th, ng,sh
- often the same letters represent multiple sounds
describe transparency
- how words are used in a word
- English is non-transparent
explain reading experiment 1
- hypothesis : if we read left to right, we should be able to observe eye movements going left to right
- method : tracked eye movements while bilinguals read aloud words in native language L1 and second language
- stimuli : 160 words (60 pseudo words)
- results : 25-30% of the letter in looked at first
- German is more transparent than French
- French L1 learners look 40-45% in, French L2 look around 25%-30%
- German L1/L2 learners look 25%-35% way in
- look a quarter into words, not at first letter
explain reading experiment 2
- rayner et al
- need 10 fixations to read a sentence containing 40 plus letters
- changing the beginning letter had the highest number of fixations (13)
- only look at about 10-13 out of 45 letters
what are some of the suprasegmental features of speech?
- suprasegmental features of speech
- intonation, stress/accent, tone, pauses, rate, duration
- phonemes are basic building blocks
give some information about intonation
- changes in pitch (fundamental freq) over time
- a statement will have a falling pitch contour
- a question has a rising contour
- intonation is language specific
give information about stress in language
- refers to length/pitch/volume of syllable
- isolated words un declarative intonation have higher pitch peaks on the stressed syllable
describe accented and tone languages
- particular syllables may have accents which change meaning of the word : accented languages like japanese
- tone language : each syllable is marked by one of four/five tones with distinct meanings
give information about pause, rate and duration
- pause
- segmentation cue
- strong stress signal
- rate
- the average number of speech segments per unit time (4 syllables a second)
- duration
- relative duration of segments shows important semantic meaning
give information about phonemes
- distinct sounds in a language that permit contrasts
- abstract representations (e.g categories) but not actual sounds
- e.g phoneme /p/ is a category of similar p sounds (phones)
- allophones : phones that sound slightly different but swapping them doesn’t make it a different word
- grapheme : unit of written language
- we group sounds that are objectively different into classes and ignore within class differences (very effective, we cannot hear them)
- we also learn unconscious rules for combining phonemes into words
what are the 3 main types of phonemes?
- 3 main types of phonemes
- ) Vowels
- periods of resonance (uninterrupted air flow out of mouth in a steady way for long period)
- all vowels are oral in english
- sound determined by tongue position/ lip rounding
- always voiced - ) Consonants
- produced quickly/ dynamically
- closure/ partial closure of airway - ) diphthongs
- speech gestures involving the transition between two vowels
- ) Vowels
what are the four types of consonant production?
1.) - bilabial stop : with lips
velar stops (soft palate)
alveolar stops (ridge behind upper teeth)
2.) - fricatives : partially obscured airflow to produce rushing sound
- place of articulation
3.) affricatives : mixture of stop followed by fricative
4.) nasals : used in languages like french
how do we percieve isolated vowels?
“We identify vowels by their formant frequencies (Primarily F1 and F2)
BUT because the actual frequency is determined by sex of the speaker, size of the vocal tract etc…, it’s the RELATIVE position of the formants to each other that is the basis for recognizing the vowel sound”
- Vowels can be perceived in isolation, based on their formant frequencies
- For consonants this is virtually impossible; they’re produced much more briefly in broad-band “bursts” of acoustic energy.
- If we take a sound such as /ba/ and isolate the consonant, it doesn’t sound like a /b/, but a “chirp”
define parallel transmission
- Most of the information about a consonant is, in fact, encoded in the vowel that follows
explain the problems of non-invariance/ contextual properties
- Every instance of a phoneme encountered will be slightly different
1) Due to differences between speakers (sex, age, etc…)
2) Because of contextual variation: A given phoneme is produced differently depending on the vowel that follows it (e.g., /d/ before i or u – see next slide). - Despite this, we perceive phonemes CATEGORICALLY (i.e., we perceive phoneme as either /d/ or /t/ never “something in between”.
define categorical perception
- Excellent discrimination across boundary
- Poor discrimination within boundary
what acoustic cues give rise to categorical perception?
- Voice Onset Time (VOT) How can you tell the difference between tIp and dIp?
There’s a gap (voice onset time) between the end of the t or d and the beginning of the voiced vowel I.
- This gap (VOT) is about 60ms longer after an unvoiced consonant (t) than a voiced consonant (d)
- Word initial voicing in oral stops /di-ti/ : VOT
duration of preceding vowel longer for voiced
- Word final voicing in oral stops /ab-ap/ : duration of preceding vowel
- Place in nasal vowels /ma-na/: start /direction of second formant
start /direction of F2
- Voicing in final fricatives /as-az/: duration of preceding vowel
duration of preceding vowel (shorter = voiceless)
- Place in fricatives /sa-sha/ : noise frequency
noise frequency (/s/ about 4Khz, /sh/ about 2.5)
- Liquids /ra-la/: F3 frequency . . .
F3 frequency of preceding vowel lower for /r/ than /l/)
explain context dependent words
- Likely to “hear” what you expect (“top-down processing). Phonemic restoration effect (Warren & Warren, 1970):
- It was found the [COUGH]eel was on the shoe
explain Cattell’s view on the word superiority effect
- Cattell (1886) : takes twice as long to read words that have no connection
- context effect helps us read sentences/syllables/words faster
explain gough’s model of word recognition
- Gough’s model : explains sequence of events during the first second of reading
- key features : linear model (visual input - character - phonetic tape - lexicon - semantics - understanding)
- built around character recognition
- key features : linear model (visual input - character - phonetic tape - lexicon - semantics - understanding)
- model is wrong : don’t look at every single letter
- we can read words that are partially occluded as we use context to fill in the “caps” to read the words
explain Rumelhart’s explanation
- word level = letter level = feature level - visual input
- bottom up model would predict that each interpretations of word are not equally likely when word is ambiguous
- rumelhart : instead of going bottom to top, there are multiple levels of representation and all levels are looked at simultaneously
- we simultaneously activate brain areas that encode high level representations : semantics (e.g there is a car) and low level visually driven input
- later on, neurones linking low level and high level features activate as they get positive input
- then neurones with net positive input turn on and turn off alternatives
explain the key points of rumelhart’s explanation
- perception is both partially top down and bottom up
- perception is driven by external input and expectation
explain modularity in the brain
- Dorsal pathway : perception for action - how and where
- Ventral pathway : perception for recognition
- however we don’t know how taxonomies are processed/ represented in the ventral visual cortex
describe the method and results used in E1 by iordan et al (2015)
Method
- Study 1 : 2 behavioural experiments as precursors to 3rd fMRI experiment
- taxonomies : natural and man made
- 32 subordinate level categories
- 32 examples of each subordinate level category - procedure : compared speed and accuracy when word is labeled at superordinate (e.g biological), basic (e.g flower), subordinate level (e.g orchid)
- results : reaction times faster at the basic level (replicating rosch)
describe the methods and results used in E2 by Iordan et al (2015)
- study 2 : speeded decision is 2 pictures show an object from the same subordinate level or different
- on half the different trials thje objects different at only the basic level (e.g flower and dog). on half they were from different superordinate categories (e.g biological flower and shoe)
- results : in two-dimensional solution each of basic categories formed separate clusters
- first MDS dimension separated natural and manmade (superordinate)
- second MDS dimension separated dogs and shoes from flowers and planes
describe the fMRI test methods and results in Iordan (2015)
- study 3 : fMRI test
REGIONS OF INTEREST
give the regions of interest in Iordan’s fMRI study - 4 early visual areas
- lateral occipital complex
- transoccipital sulcus
- parahippocampal place area
- retrosplenial cortex
- fusiform face area
- results : strong subordinate/ basic level representations found in all regions of interest
- especially at subordinate level in V1/V2/V3v and at basic level at later stages of visual processing in lateral occipital cortex
explain the MVPA technique and results in Iordan (2015)
define haemodynamic response
- also conducted multivoxel pattern analysis (MVPA) of fMRI data
- used correlation classifiier to try to predict categorises from the haemodynamic response (the patterns of activation detected by fMRI) at each level
- category info was significantly above chance at all taxonomic levels with highest decoding accuracy (ability to predict the category shown based only on haemodynamic data)
- decoding was better at basic level in LOC, FFA and RSC
describe the method and results of the repeated study by Iordan (2015)
-repeat : repeat the first study but with three new taxonomies with more appropriate superordinate categories (natural/ man made not sufficient) and background was removed from images
- taxonomies : vehicles
furniture
musical instruments
- 36 subordinate level categories, 40 examples
- match to category method used (same as E1)
- results : categories identified faster at basic level
- same-different subordinate level categorisation method used (E2)
- fMRI also replicated
- results : evidence for subordinate stronger in V1/V2/V3v/hV4
subordinate and basic similarly strong in later higher level visual areas
MVPA : decoding accuracy was better at basic level only in object selective LOC
explain the conclusion of Iordan et al
- basic level may be favoured as it is an emergent property of visual system
- emergent property : something a complex system has but individual components of a system have
give findings for categorising in the real world
- in lab experiments/ real voting behaviour, physically attractive candidates get more votes
- people infer invisible characteristics (trustworthiness, competence, intelligence) from features (height, symmetry, facial cues - blemishes, jaw) - “vote for me” photos looked more at the person than “don’t vote for him” photos
- manipulates voters
explain rosch’s experiments 1-3
- E1 : more attributes were provided for categories at the basic and subordinate levels (best at non biological taxonomies)
- E2 : motor programmes : ppts coded the motor patterns used to interact with the same 90 objects) - hard to code people’s responses
- E3 : visually similar category members : high overlap for basic level category (dog), low overlap for superordinate level category (animal)
explain rosch’s experiments 4-6
- E4 : identifying averaged shapes : superordinate categories : difficulty in producing stimuli/ in shape variability
- E5 : looked at whether hearing the name of the category aided in picture detection
- decided if a picture of an object was shown on right or left side, abstract picture was shown on the other side
- basic and subordinate condition had significantly better detection - E6: participants heard a label at the superordinate, basic or subordinate then saw two images
- made a speeded decision about whether the pictures were identical or two different objects
- priming was only significant for basic/ subordinate level names
- superordinate were no help at all
explain rosch’s experiments 7-9
- E7 : easier to verify the basic level name for pictures of objects
- saw a label at any 3 of the levels then a colour photo - made decisions about whether label/photo matched or belonged to different categories
- reaction times in were faster for basic level labels for true/ false decisions - E8 : adults/ children chose the pair of pictures that went together
- sorting at the basic level was perfect from age 3-4 up
- took longer for superordinate level to be acquired : 4 years - E9 : children/ adults sorted 16 pictures from 4 basic level categories into groups that went together
- 5/6 year olds : superordinate sorting was worse than basic level sorting
- 8 year olds : superordinate/ basic level sorting was perfect
explain rosch’s experiments 10-12
- E10 : what level do adults prefer to use
- named some or all of the 54 pictures from taxonomies used in E1
- adults preferred to name at the basic level - E11: analysed which concrete nouns were used by a child from her speech corpus
- preferred to label at the basic level - E12 : level most likely to have a sign assigned in ASL
- level likely is basic level
define conditioning and give some learning mechanisms
- some stimuli trigger automatic, stable, reproducible responses e.g babies grasping
- learning mechanisms :
- classical conditioning
- instrumental conditioning
- non-associative learning
- consolidation
explain unconditional stimulus response
- innate, doesn’t depend on prior learning
- tightly associated with the unconditional stimulus
- elicits an unconditional response
briefly explain classic conditioning in relation to Pavlov’s dogs
- bell becomes conditioned stimulus
- salivating becomes conditioned response
- associative learning
what conditions must be met for classical conditioning to take place?
- conditions must be met :
- temporal simultaneity : presentation of the neutral and unconditional stimulus
- necessary to repeat presentations
- can take quicker under some conditions e.g emotions
explain operant conditioning
- learning through trial and error
- also called instrumental conditioning
- reinforcers : stimuli that strengthen probability of behaviour
- often material rewards, can be social
explain primary reinforcers
- primary reinforcers : stimuli whose meaning is not dependent on prior learning
- determined genetically/ inherited
- can effect response of learning
- used to teach phonemes to autistic children
explain punishers
- punishers : stimuli that lower probability of behaviours occurrence
- can be material/social
explain discriminative stimuli
- discriminative stimulus : pressing lever only has an effect on environment when specific conditions are met (red light is on)
explain positive and negative reinforcement
- positive reinforcement : give good things to increase probability of behaviour occurring
- negative reinforcement : take away bad things to increase probability of behaviour occurring
explain positive and negative behaviour
- positive punishment : give bad things to decrease probability of behaviour
- negative punishment : take away good things to decrease the probability of behaviour
explain research into conditioning in practice
- homme et al : rewarded desired behaviour (staying quiet) with running around - class was calm a few days after
explain the subtle influence of conditioning
- skinner (1948)
- pigeons delivered food ever 15s
- pigeons began displaying weird behaviours : turning around in circles, brushing head
- associated behaviour they had at the first delivery of food
- how superstitions are born
explain implicit learning
- implicit memory : graf and schacter (1985) - knowledge without awareness
- previous experiences facilitate performance on a task that does not require intentional recollection of the experiences
- defined by the task : observation of improved performance without declarative knowledge is explained by implicit memory
explain covariance learning and research
- relations within the stimulus material affect performance, participants not aware
- lewicki (1986) : participants get 6 pictures with descriptions of depicted person
- people described as kind and helpful or capable and effective (description correlated to hair length)
- decisions in line with previous experience need more time
- participants learned the co-variations
define and explain research studies for artificial grammar
- central task in learning to read : associating graphemes with phonemes
- artificial grammar : rule used to construct the stimuli (unknown to participant)
- participants shown strings of letters that are constructed based on the rule
- second phase : asked to identify whether stimuli are constructed according to rules or not
- dyslexic children performed less well than typically developing children
explain sequence learning
- acquisition of procedural knowledge without intention to learn
- participants not aware they have acquired this
- show strong performance effects
- participants performance increases without them knowing why : brain learns rule governing stimuli
explain autobiographical memory
- memory for events of ones life
- autobiographical memory and episodic memory overlap
explain the stages of memory
- infantile amnesia : for first 3 years
- reminiscence bump : between 15 and 25 (novelty and stability)
- retention function : for memories up to 20 years olds, with older memories less likely to be recalled
explain flashbulb memories
- special neural mechanism may be activated by such memories
- may be result of ordinary memory mechanisms with more efficient encoding
What is non declarative memory?
- refer to actions/[performance
- only require a few resources
- operate automatically
What is habitation
- cessation of behaviour in response to a repeated, harmless stimulus
- form of discriminative learning : occurs when a stimulus is not relevant/ harmful
- Colombo et al. : attention derived from habituation predicts performance
- sensitisation : a magnification of reflexes in response to a meaningful stimulation
What is sensitization?
a magnification of reflexes in response to a meaningful stimulation
What is perceptual learning?
- improving ability to distinguish between objects
- oscillatory gamma band activity : neural synchrony that allows neurons to co-ordinate activity to perform a perceptual task
- perception GBA : correlates with recognition of complex patterns
- musician’s brains respond differently to timbres : brain rewires itself to better process the stimuli
Explain research findings around perceptual learning and sleep
- experimental performance improved after each session but the largest improvement was found after a night’s sleep
- chronotype did not affect the memories after a night’s sleep
Give research findings for practising motor skills
- Baddeley & Longman (1978) : performance in a task (incl nights sleep in between tasks) increases with hours of practice
Explain the power law of learning and give research findings
- performance follows a pattern as a function of trials (rapid improvement at first, then slower improvement)
- research shows signs of recognition of mother’s voice at early age (ability increases with time)
describe TMS
- Transcranial magnetic stimulation (TMS)
- uses magnetic field to change electric activity in neurons
- creates phantom sensations
- creates virtual lesions
describe fNIRS
Functional near infrared spectroscopy (fNIRS)
- BOLD signal captured by absorption of infrrared light
- can show higher level functions : morale, values, decision making
- can see results in real time
explain the assumptions of similarity based accounts
- assume all previously experienced exemplars of a category are stored - no abstraction of a central tendency
- may be large storage requirements in LTM compared to prototype accounts
explain how similarity based categorisation works
- a stimulus is usually compared to the most similar saved exemplar
- if it is sufficiently similar to one/more stored exemplars it is categorised as a category member
- representation of stimulus is then added to category
- slower naming of atypical examples
give weaknesses of exemplar theories
- typicality doesnt always co-vary with membership as predicted by theories
- typicality effects have been found for fuzzy categories AND rule-defined categories - not clear how categories represented as exemplars can be organised into hierarchies (these theories don’t show basic level advantage)
explain the assumptions of explanation based accounts
- assume that categories have a core definition that explains appearance/function etc of category members
- categories are defined by common sense explanations
- explanations include perceptual info (e.g dogs bark) and general semantic knowledge (dog barks are warnings)
- we choose the theory that best explains the pattern of features of an item
- explains fuzzy category boundaries as some exemplars fit well into more than one category theory
give advantages and disadvantages of explanation based theories
+ stimulated useful experimental research and show important background info that is ignored by other theories
- more vague/ less explicit than prototype/ exemplar theories so are harder to test
explain the difference between expertise and problem solving
- expertise : people being good at solving problems similar to those they have experienced before
- problem solving : involves people trying to solve novel problems that rely on domain-general knowledge such as flexible, general heuristics (e.g using the shorter of two alternatives)
give research findings around the tower of hanoi
- tower of hanoi : people do quick moves then pause, so they divide the problem into subgoals with pauses between
- egan and greeno (1974) : people were trained on 3 disk Hanoi problem then transferred to 4 disk problem
- initially they used general heuristics (don’t reverse previous move)
- with more experience they developed specific strategies for dealing with tower (learned specific heuristics)
explain the limitations of the hanoi problem with regard to Newell and Simon’s problem-space theory
- problems need to be well-defined to be solved using problem-spaces
- background knowledge is often imprecise to hard to analyse problems this way - often real word problems need lots of general background knowledge and have ill defined goals
give research findings into expertise
- chi, feltovich and glaser (1983) : physics experts were faster and more accurate than novices at solving physics problems
- experts spent longer analysing/understanding problems
- experts categorised problems based on deep similarity (theroetical concepts), novices categorised on surface similarity
explain how chess experts are good examples of expertise
- initial state (position of pieces at start) and goal are well specified but there are many vast numbers of intermediate states
- chess experts make better moves : faster at recognising the type of problem, have more solutions stored
- win games by being quicker at retrieving better moves from memory
give research findings of chess
- degroot (1956) : experts recalled chess piece positions from real games much better than novices (only in actual configurations, not random spaces)
- chase & simon (1973) : experts needed less viewing time to remember position of pieces and had larger encoded chunks (But do not have better general memory)
- gobet & simon (2000) : learn many configurations from previous games and reproduce this info when playing
- brilliance at chess is domain specific : weak correlation between chess ability and intelligence, amount of practice more important
describe the three main components of ACT
- declarative memory : used by novices, used to solve novel problems, explicit, flexible, slow
- feeds into working memory - procedural memory : productions aquaried with practice, domain specifc, implicit, fast
- becomes more important than declarative memory as we become experts - working memory : stores active info from declarative/procedural memory
describe skill acquisition in ACT
- skill acquisition = 1. gradually storing/ finetuning chunks of domain specific knowledge by combining info from declarative memory to procedural memory
2. composition - series of productions combined into a single, large production
what happens as information becomes less declarative according to ACT?
- as knowledge becomes less declarative it is
- less verbalisable (explicit to implicit)
- more automatic
- less effortful
how do motor experts differ from cognitive experts?
- experts can read situations quicker and better than novices, and so are better at predicting what might happen so can prepare a good response
explain the occlusion paradigm 1
- later stages of serve in cricket was occluded so experts/ novices could predict where the ball would go
- temporal occlusion : occuled at different times of the serve
explain the occlusion paradigm 2 - spatial occlusion : hid parts of the bowler’s body
give the results of the occlusion paradigm experiment - experts in both racquet sports/cricket used the same cues as novices PLUS early body movements to predict ball trajectory
explain eriksson’s theory of deliberate practice
- eriksson (1994,2006) : deliberate practice was necessary and largely sufficient to become an expert
- deliberate practise : - be repetitive
- be challenging
- be monitored so useful feedback is given
- allow for correction of mistakes - may work by reducing the need to use our limited working memory by improving retrieval and storage from long term memory
when will deliberate practice not occur?
- will not occur if youre
- doing something for fun (not concentrating on task)
- trying to do your best (focussing too much on succeeding) - amount of deliberate practice can predict performance for chess, musicians, sports players
give evidence for the claim that expertise only consists of deliberate practice, not natural ability
- sloboda et al (1996) : best musicians needed as much practisse to reach the same level as weaker muscians (nurture not nature)
- “child prodigies” may just be trained/ encouraged from a young age
explain nature vs nurture on expertise
- balance between nature/nurture depends on environment
- a fairer world with more equal opportunities would allow nature to be emphasized
define discrimination and generalisation
- Discrimination : distinguishing between individual exemplars
- Generalisation : group perceptually different exemplars into common categories
define object constancy and typicality
- object constancy : consistently categorising objects to achieve invariance when the object can appear different from a different view
- examples of variation in input : depth rotations, plane rotations, size, position, lighting changes
- typicality : distance in representational space of an exemplar to the category prototype
give the types of theories of categorisation
- four types of theories of categorisation :
- classical/defining features theories
- prototype theories
- exemplar theories
- explanation-based theories
explain the assumptions of classical theories
- assume concepts are defined by a rule specifying the necessary features which are enough to decide if an item belongs to a category
- membership to the category is binary : either belongs to category or doesn’t at all
give research findings for classical theories
- Bruner et al (1956) - important role of memory and strategy
- memory : people were better at working out rules if they could see all cards at once or could use pen/paper
- role of strategy : people try different strategies depending on preference /abilities/ time available/ short term memory constraints
- people learnt explicit rules that defined the concepts - disadvantages : findings don’t generalise - concepts were predefined
discuss issues with the classical features theory
- may be grey areas where although someone may fit the category, they would not be defined as such (e.g the pope fits in to the category of “bachelor”, but would not be defined as such)
discuss Rosch’s findings as an issue for classical theories
- may be hard to identify the necessary defining features for a category
- Rosch and Mervis (1975) : members of a category just resemble each other and the boundaries are fuzzy
- McCloskey and Glucksberg (1978) : people are inconsistent with categorising objects
discuss categories in the external world as a problem for classical theories
- categories do not stably exist in the external world
- people disagreed with each other and categories were inconsistent
discuss hierarchical networks as an issue for classical theories
- assume concepts are arranged into hierarchy
hierarchical networks assume :
1. category inclusion relations - a category at one level can only belong to a single category at the next higher level
2. transitive relations - a category membership at a higher level dictates membership at lower levels
- however people don’t agree that a butter knife is a tool, but a knife is a tool
explain hierarchical categorisation
- highest level : subordinate level (granny smith apple, Siamese cat)
basic level (apple, cat)
lowest level : superordinate level (fruit, animal) - the basic level is preferred by adults : objects are also named faster at the basic level than at the subordinate/ superordinate level
ASIC LEVEL (Rosch)
- cognitive economy : exemplars of basic level categories share specific features, which distinguishes them from similar categories at the basic level
- the most efficient way for the visual system to initially identify objects
- labels give most important information for little effort - superordinate level labels contain too little information
- subordinate level labels distinguish between very similar categories (e.g using appearance and behaviour to distinguish between Siamese and Persian cats) but extra work is needed for very little useful information
explain the effect of expertise on labelling
- Tanaka and Taylor (1991) : experts (but not novices) were as fast as identifying even typical category members at the subordinate level as at the basic level
- also listed more features of subordinate categories than novices and were more likely to use subordinate names than novices
Define explicit and implicit coding
Explicit coding : consciously think about something
implicit coding : learning without consciously thinking about something
define consolidation
- consolidation : process where a neural trace is strengthened until it becomes stable
- factors for consolidation : exposure, intension, depth of processing, skill acquisition, interferences
define exposure and research findings
- the number of times something has been exposed to a person is not enough for coding e.g coin details
- penny test : only 41% got the coin design right
define intention and give research findings
- Hyde and Jenkins (1973) : half of participants were told they would have to later recall words from a list
- what’s important is the level of processing words, not intention
define levels of processing and give research findings
- Craik & Lockhart (1972) : make distinction between shallow (superficial) and deep processing (activates related knowledge)
- Craik & Tulving (1975) : participants processed lists of words in different conditions, follows by unexpected recall task
- structural : capital letters etc
- phonemic : rhyming words
- category
- sentence : does it fit in a sentence- structural task had worse recognition percentages, sentence and category had the highest
- shows importance of depth of processing over intention / exposure
. define the power law or practice
- follows power law of practice (learning curve)
- tells us performance relates to training
- more and more training is needed to keep learning at the same rate
define retroactive and proactive interference
- retroactive : new memory interferes with old info
- proactive : old memory interferes with new info
give research findings for acoustic/semantic interference
- interference can be due to formal - acoustic confusion
- interference due to semantic content - closely-related concepts
- Lutz et al : no distractors, phonetic distractors and semantic distractors used
- interpolated task used to stop rehearsal
- delays varied between study and test
- phonetic distractors more efficient at interfering in short periods
- Lutz et al : no distractors, phonetic distractors and semantic distractors used
explain retrieval
- info in working memory accesses LTM
- neural evidence shows activation of knowledge above a threshold is important
RECOGNITION
- means to know again : ability to match perceptual input with knowledge
WORKING MEMORY
- conscious request of info
give research findings for recognition and retrieval
- shephard (1967) - recognition task : identification of present item as known/unknown
- performance in recognition tasks usually better than in recall tasks
- recognition performance remains high even after long time (e.g recognising class mates after 20 years) - issues : distractors can falsify recognition, new stimuli may be considered known if they are similar to old stimuli
explain and give research findings for cued recall
- tulving & thomson (1973) : encoding specificity principle
- cues used during intial learning can help later retrieval more than novel cues as they were stored alongside with memory of experience - godden & baddeley (1975) : words learned onland are better recalled on land, words learned underwater are better recalled underwater
Explain retrieval as reconstruction and give research findings
- knowledge used to reconstruct events during recall
- mcdermott (1996) : ppts learned word list, then had recognition test with old words from middle of the list and a new word ( sleep)
- delayed recognition test two days later
- in delayed recall there was higher recognition of lure word than studied words
- ppts probably activate the concept sleep when hearing words - loftus and palmer : group who heard the word “smashed” predicted higher speed at collision than the “contacted” group
give research findings for declarative memories
- Collins & Quillian (1969) : hierarchical network
- sub-concepts have all features of super-concepts plus differentiating features
- evidence : “a canary is a bird” (1160ms), “a canary is an animal” (1240ms), “a canary has skin” (1480s) : depends upon distance between nodes in network- problems : typicality effects cannot be explained, additional assumption is necessary
- hierarchy does not always work - ppts are faster to decide an ostrich is an animal than a bird
explain the spreading activation model and it’s limitations
- semantic networks
- concepts in memory linked to each other
- concepts are the nodes
- semantic relations are the links - problems : vague assumptions
only conclude that there are links, if one unit primes another unit
activation spread not clarified : after some time is whole network active
activation may fade with distance to source
Define the two types of reasoning
- Inductive reasoning : probable conclusions taken from available statements
- Deductive reasoning : conclusions follow from the premises
Describe Wason’s selection task
- Wason (1966,1968) : selection task
o Most chose E or 4 (4 doesn’t bring new info)
o Choose 4 as if the rule is correct, there will be a vowel on the other side (however, doesn’t necessarily work in both directions, e.g vowel with even number does mean there may be an even number with a vowel
o Need to turn over 7 to falsify the rule (e.g if there’s a vowel, the rule is wrong) - Shows confirmation bias : human tendency to confirm views rather than falsify
Describe the concrete version of Wason’s task
- If someone is in the bar drinking beer, they must be over 18
- Four people : 1 drinking beer, 1 drinking coke, 1 over 18 and 1 under 18
o Should check person drinking beer and person looking under 18
o ¾ participants get answer, 1/10 get the abstract version
Which two logic rules do conditional problems use? Define and give examples
- Selection task is a conditional problem
- Uses two logic rules : modus ponens and modus tollens
o Modus ponens : if p then q
P is true, so q is true
E.g if it rains, there are clouds
o Modus tollens
If p then q, q is not true therefore p is not true
E.g if it rains, then there are clouds
There are no clouds, therefore it does not rain - Modus ponens/ tollens are the only valid inference rules with conditional problems
Explain Popper and Hume’s ideas of scientific reasoning
- Karl Popper (1968)
o Falsification is only way to test validity of theories - Hume’s problem of induction
o Not possible to prove a scientific theory as future data may refute it
What do we need to conclude causation?
- Aim to prove theory wrong
- If we want to conclude that X causes Y we need
o Rule of agreement: if X is followed by Y, then X is sufficient for Y and could also be the cause
o Rule of difference: if Y does not occur when X does not occur, then X is necessary for Y to occur
Explain Wason’s 2-4-6 task
- Given sequence 2, 4 6 and asked to find rule behind sequence
- Participants formed the rule “sequence of numbers increasing by two”, and can give their own sequences, and corrected by experimenter
- Actual rule : any sequence of increasing numbers
o None tried to falsifiy using sequences like 1, 7, 32
Explain lord et al’s experiment on death penalty
- Two groups : pro and anti death penalty
- Read two studies with two independent variables : design (longitudinal/cross-section) and outcome (supports/doesn’t support)
- Flaws were highlighted
- Participants thought studies consistent with their views had better designs and were more convincing, failed to notice weaknesses
o Increase original belief
Define barnum statements
- Barnum statements: vague statements that can fit almost anyone
Explain expected utility
- Comes from classic economics
- People take all available information into account
- They give a value (utility) to each option and estimate the probability that this will occur
- Expected utility = sum of the utilities of each possible outcome x probability
Explain bounded rationality
- Simon (1957)
- We cannot analyse all options
- People can only carry out highly selective searches due to limitations of cognitive system
- People are satisficers, not maximisers – they use heuristics
Explain the availability heuristic
o Flying is safer than car
o Car accident usually ends up in few casualties
o Plane accidents are more severe
o Plane accidents more available from memory
Explain the representativeness heuristic
o Base conclusions on small samples, but should base on a large number
o E.g gambler’s fallacy
Give a case study to show the application of cognitive psychology in eyewitness testimony
- John Jerome white
- No rape kit, pubic hair and skin used
- Police sketch used; victim picked white from photo array
- Exonerated by DNA analysis in 2007
- Eyewitness testimony
o Poor lighting
o Not wearing glasses
o 6 weeks between attack and line up
o White only person present in both photo array and live line-up
How reliable is eyewitness testimony?
- Victims often report clothing, build, height, age and weapon incorrectly
- Jurors do not understood memory properly, so may think eyewitness testimony is more reliable than it is
Give some problems for eyewitness testimony
- Problems for eyewitness testimony o Cross-racial impairment o Weapon focus effect o Time estimate bias o Confuse confidence with accuracy o Perception is reconstruction, not like a photograph
Give research findings for perception and face recognition
- Megreya & Burton (2006) : matching face experiment, some people particularly poor at it
- Eyewitnesses testimonies are recollected by interviews (implying words translate images, however this is not accurate)
Describe factors in EWT
- Retention period : the more time you have to observe a fact, the better the encoding
- Age : children over 4 are as accurate as adults when target is in line-up (children/elderly less accurate when target is not)
o Children are suggestible in interviews - Witnesses : may not be reliable, confabulation
- Weapon focus : attention is driven to most relevant stimuli (e.g weapon)
- Stress : arousal too low/ high : poor recall
o Emotions have a huge effect on memory storage (flashbulb)
Describe how memory can affect EWT
- Reconstruction of events, not like a video
Give the locations of Broca’s and Wernicke’s areas
- Broca’s area : inferior frontal cortex
- Wernicke’s area : posterior superior temporal cortex
Describe studies into localising language
- Lesion studies : damage to left hemisphere causes language deficits
o 95% of right handers have left lateralised language - Language deficits and RHS motor control commonly associated
Describe how Broca’s area was identified
- Patient with speech production problems but not speech comprehension
- Progressive disease
o After 10 years : paralysed in right arm, leg, intellectual functioning impaired - Lesion in left hemisphere frontal lobe
Describe Broca’s aphasia
- Broca’s aphasia :
o Damage to third convolution of frontal lobe
o Patients can only speak haltingly, miss many function words, abbreviated syntax
o Good comprehension
o Non fluent aphasia : slow and effortful
o Agraphia : difficulties with handwriting (generates commands for motor cortex)
Describe Wernicke’s area and how it was identified
- 1874 : 2 patients
- Difficulties in understanding spoken language
- Paraphasic speech : nonsense speech production
- Lesion of superior temporal cortex in left hemisphere
- Difficulties in speech comprehension
- Receptive aphasia : speech perception impaired
- No trouble in comprehending other’s speech but difficulties monitoring their own and repeating sentences (conduction aphasia : damage to arcuate fasciculus, difficulties in communication between Wernicke’s and Broca’s areas)
Describe the dual stream model and how it differs from previously models
- Hickok (2012)
- Involves dorsal and ventral streams
o Dorsal : upper side of brain
o Ventral : lower/ inferior part of brain - Two parallel streams
o Parallel processing within each stream
o Ventral primarily supports speech comprehension
o Dorsal pathway primarily supports sensory motor integration (speech production) - Network expands past Wernicke’s and Broca’s areas
o Dorsal and ventral premotor areas : most language activity is at boundary between BA 44/45 and premotor cortex (BA6)
o Anterior-frontal areas of superior temporal gyrus and anterior/ posterior middle temporal gyrus - Bilateral processing in ventral streams
- Activity in Wernicke’s and Broca’s areas in speech observation tasks (fMRI) and activated network between (premotor, sensory)
