Cognition Flashcards
What are automatic vs controlled processes in visual attention?
Automatic:
- Fast
- Doesn’t require attention
- Doesn’t reduce capacity for performing other tasks
- Parallel processing
- Unavailable to consciousness
- Unavoidable
- Very hard to modify once learned
Controlled:
- Slow
- Require attention
- Take up processing capacity
- Serial processing
- Available to consciousness
- Can be used flexibly in changing circumstances
- e.g., the Stroop test
What is the psychological refractory period (PRP)? What is a potential explanation for it?
- Psychological refractory period (PRP)
o 2 stimuli
o 2 responses
o Respond to each stimulus as rapidly as possible
o Second stimulus presented shortly after the first
o A marked slowing to the second stimulus
o Response to the second stimulus is slowed because ppts are still processing the first stimulus
o Results delayed in response time when asked to divide attention
What are the limitations of the Multiple resource model?
o Models really focus on visual and auditory inputs but tasks could be presented in other modalities
o Coordination between tasks not really considered
What is the supporting evidence for the multiple resource model?
- Supporting evidence – more interference when tasks share a modality or type of response
What is selective inference and which model does it refer to?
- Selective interference – it is the type of resource that tasks require that determine whether they will interference with one another
- Refers to the multiple resource model
What is the multiple resource model? And what are the 3 successive stages of processing?
- separate sources of capacity may be specialised for particular processes
- 3 successive stages of processing
o Encoding – perceptual processes (visual/auditory)
o Central processing - (spatial/verbal cues)
o Response – voice or manual
What did Bourke et al find (central capacity models)
– 4 tasks which they hypothesised required different amounts of the “central capacity”
o Tested ability to do 2 of these tasks at once – one primary task one secondary task
o Findings consistent with the central capacity theory
o Task which most loaded on the central capacity interfered most
- Theory tends to be rather descriptive and circular
What are central capacity models?
- Assume there is some central capacity which can be used flexibly across a range of activities
- Central capacity has limited resources, so how successfully multiple tasks can be performed depends on the drain of each task on these resources
- Any two tasks will interfere provided they require more resource than is available I the total capacity
What three factors can affect whether or not we can multi-task?
- Task difficulty - harder when task is harder
- Task similarity - harder when tasks are too similar
- Practice - some research shows that practice can improving multitasking but there are limitations to this
What is the threat superiority effect?
– we are faster to notice something potentially threatening compared to something positive or neutral
What are the limitations of feature integration theory?
- Assumption that visual search either all serial or parallel too strong
- Search for conjunctive targets faster than predicted by feature integration theory
- Nature of non-targets in display also important (similarity among distractors speeds search)
- Not all features are equal
What is the serial process (“focussed attention stage”) in feature integration theory?
– requires attention
o Features can be combined by focused attending to the location of the object, which allows search for targets define by a conjunction of features
o “Illusory conjunctions” (random combination of features) can occur due to problems in combining features to form objects at a relatively late stage
o Slower than the parallel process, and depends on set size
What is the rapid initial parallel process (“preattentive stage”) in feature integration theory?
– doesn’t require attention
o Process physical characteristics – “features”
o Targets that are identifiable by simple physical characteristics “pop out”
What is feature integration theory?
- Treisman & Gelade
- Object –> preattentive stage (features separated) –> focused attention stage (features combined) –> perception
What type of processing can visual search tasks suggest?
- If response time is affected by display size – suggests serial process
- If reaction time not affected by display size – suggests parallel process
- Pop-out – no effect of display size
How do visual search tasks work on neglect patients?
- Set size
- Target presented on half of the trails
- Decide if target is present
What is visual search?
- Looking for something in a cluttered visual environment
- Find the target among the non-targets (distractors)
What is happening in neglect patients?
- Exaggeration of normal functioning – we choose what to attend to
- Presence of extinction strongly suggests that there’s some sort of competition
o Presence of a more salient stimulus on the ipsilesional side of space capture attention and hinders perception of the stimulus on the contralesional side
o Can be alleviated by presenting related stimuli, one to each side, as this is more likely to be seen as an integrated whole than as two separate stimuli
True or false, can hemispatial neglect affect memories?
TRUE
Is anything processed if there is neglect?
- Can process some of the unattended information semantically – e.g., Marshall & Halligan (1988)
- So, there is some processing of information presented to the neglected side, but the patient is not consciously aware of the neglected stimuli
The border between neglected space and non-neglected space is:
- Not sharp and absolute as in a primary visual field deficit such as hemianopia
- Doesn’t necessarily align with the vertical midline or meridian
- Can vary with the number of distractor items in the display
What does neglect do to patients?
Patient fails to act upon, identify, or acknowledge contralesional (on the side opposite the lesion) stimuli and acknowledge only stimuli which are ipsilesional (on the same side as the lesion).
What commonly causes neglect?
A stroke
What side is neglect most common on?
The right hemisphere of the human brain
What is hemispatial neglect?
a relatively common neuropsychological condition in which patients ignore one side of space.
What are multiple attentional spotlights? Who provides evidence for this?
- Awh & Pashler (2000)
- Ppts are asked to report 2 numbers in a 5x5 grid filled with 23 letters
- Cued (80%) valid to two squares. Interested in what happened on the invalidity cued trials
- Zoom lens predicts space between cued locations should be included in the focus of attention – but this was not the case
- Possibly have multiple attentional spotlights instead of one lens that can change
What is the zoom lens model of attention? Who provides evidence for it?
- Evidence from LaBerge
o Press button when letter S (cue) is followed by letter R (target) - “Letter condition”:
- Probe appeared in different spatial locations
- Letter task attempted to focus ppts attention on centre of display. Faster reaction time when probe within central attentional beam
- In word task, ppts required to categorise entire word = broader attentional beam
What are the endogenous systems (cues)?
- Controlled by the individual’s intentions and expectations
- Involved when central cues are presented
What are the exogenous systems (cues)?
- Automatically shift attention
- Involved when uninformative peripheral cues are presented
- Stimuli that are salient or that differ from other stimuli are more likely to be attended to
How can we control where our attentional spotlight is?
- Disengagement of attention from a given stimulus
- Shifting of attention from one target stimulus to another
- Engaging/locking attention onto a new visual stimulus
What is location-based attention?
- “Attentional spotlight”
- Can be focused on a particular location
- Posner (1980) suggested that the attentional spotlight can shift to a different visual location without eye movements (covert attention)
How do you process task irrelevant stimuli under Lavie’s perceptual load theory?
o If the perceptual load is high: early selection
o If the perceptual load is low: late selection
- Stage of selection is variable, and task-irrelevant info can be processed if there is spare capacity
What is the prediction of Lavie’s perceptual load theory?
o There will be less interference from the distractor on the more difficult task relative to the easier task (because there’s less attentional resource left over for processing it)
What is perceptual load?
o Perceptual load – number of units in the display and the nature of processing required for each unit (difficulty)
What is Lavie’s Perceptual Load theory?
- Everyone has a limited attentional capacity
- Total available capacity is always allocated
- Spare capacity that’s not required for primary task is automatically allocated to irrelevant stimuli
- Attentional capacity allocated to main task depends on perceptual load
What research isn’t explained by Deutsch & Deutsch’s late-selection theory?
Treisman & Riley – dichotic shadowing task but instructed to shadowing and to tap the desk whenever they heard a target word in either message
- More target words detected in attended than unattended
- Wouldn’t have expected this under D&D theory as all stimuli are processed equally – would’ve expected equal number of words detected in either message
What is Deutsch & Deutsch’s late-selection theory?
- Suggest filtering occurs late – so that all stimuli are analysed equally and the most relevant stimulus determines the response
- Accounts for the extensive processing of unattended stimuli just as well as Treisman’s Attenuation theory
What is a problem with Treisman’s attenuation theory?
it’s a bit circular – noticed unattended message because it was relevant, relevant because we noticed it (more descriptive than explanatory)
What is Treisman’s attenuation theory?
- Unattended input not completely rejected, but filter attenuates (reduces) analysis of the unattended signal
- The threshold for processing stimuli is consistent with the context or our expectations lowered
o Accounts for breakthrough effects - Partially process stimuli in the unattended message sometimes exceeds the threshold for conscious awareness
- Making it more likely that we process task-related info e.g., like our own name at a cocktail party
What is “Breakthrough” as described by Treisman?
- When ppts say a word that was presented in the unattended message in a dichotic listening task
- Occurs more commonly when the word is highly likely given the content
What did the MacKay study show?
- Dichotic listening task
- At the end they had to paraphrase the attended message
- Despite being unable to report the content of the unattended message, it still affected performance: more likely to interpret the attended message in a way that was consistent to the information presented in the unattended message
What did the Von Wright study show?
- Paired some words with a mild electric shock to ppts
- Then conducted a dichotic listening task with those words presented in the unattended message
- Ppts demonstrated an increased galvanic skin response (GSR – indicative of emotional response)
- This suggests that there was some analysis of the content of the unattended message, and this affected unconscious physiological measures
What is the “Cocktail party effect”? and which theory doesn’t answer this?
- talking to your friend, actively paying attention
- But if someone says your name from across the room you will notice this even though you weren’t paying attention
- Broadbent’s theory can’t answer this
What is Broadbent’s filter theory?
- Filter prevents overloading a limited-capacity mechanism beyond the filter
- Filter allows one stream of info through
- Other outputs remain in the buffer for later processing
- Theory accounts for many of the findings from the dichotic listening task
In the dichotic listening task, in the non-attended message what do and don’t ppts notice?
Do notice:
o Some basic physical characteristics e.g., if the voice stops or changes from male to female
Don’t notice:
o Any of the content or individual words
o When the language change e.g., English to German or when the speech was spoken backwards
What was the Dichotic listening task?
- Ppts are played two different messages to different ears over headphones.
- Asked to repeat back one of the messages
What did William James say about attention?
- Attention is an active process
- James suggested attention was filtering out what was not relevant
What are the limitations of computational cognitive science?
- explanations without prediction is common
- difficult to falsify
- typically de-emphasise motivational and emotional components of cognition
What are the strengths of computational cognitive science?
- theoretical assumptions are clear and precise
- increasingly used to model the effects of brain damage
What are connectionist models (neural networks)?
– mathematical structures that “learn.” Don’t always know how the model reaches its conclusions. Makes the same mistakes that children do when learning.
How is computational models different to AI?
Computational models:
- Involves programming computers to model human cognition
- Successful models need to perform the way humans do (make mistakes etc.)
AI:
- involves building computer systems that produce intelligent outcomes
- don’t have to work in the same way as humans
- e.g., Deep Blue was able to consider 200 million chess positions per second and so beat the chess grandmaster – a human couldn’t do this
What is computational cognitive science?
– developing computational models to understand human cognition (not the same as AI)
What are the limitations of functional imaging, ERP & TMS?
- correlational data – correlation =/= causation
- over-interpretation of data
- difficulty relating brain activity to psychological processes
- studies are often underpowered with a high likelihood of false positives (e.g., Bennett et al., 2009 – scanning dead salmon and picked up brain activity)
What are the strengths of functional imaging, ERP & TMS?
- variety of techniques offer a range of spatial and temporal resolution
- can study functional specialisation
- rich data allow elaborate assessment of neural functioning
- TMS allows for causal inferences
What is the spatial and temporal resolution of a naturally occurring lesion like?
Naturally occurring lesion likely to affect a region or even the whole brain and has quite poor temporal resolution (hours, days etc.)
What is the spatial and temporal resolutions of a single cell recording like?
A single cell recording has a very fine spatial resolution (cell level), temporal resolution varies depending on set up.
What is Transcranial Magnetic Stimulation (TMS)?
- creates temporary brain “lesions” by running an electrical current through a coil placed near the skull
What is Magnetoencephalography (MEG)?
– measures the magnetic fields produced by brain activity
What is Functional Magnetic Resonance Imaging (fMRI)?
– imaging blood oxygenation using MRI (indexes metabolic activity)
What is Positron Emission Tomography (PET)?
– brain scanning by measuring positrons
What are event related potentials (ERPs)?
– present the same stimulus repeatedly and measure the pattern of electrical brain activity recorded from scalp electrodes
What is a single cell recording?
– use of a microelectrode to record the activity of a single neuron
What is cognitive neuroscience?
– use converging evidence from brain and behaviour to understand cognition
What are the limitations of cognitive neuropsychology?
- don’t usually know what patients were like before brain damage
- patients can develop compensatory strategies to mitigate the effects of their brain damage
- sometimes rely on single-case studies (but are these really necessarily problematic?)
What are the strengths of cognitive neuropsychology?
- double dissociations are very powerful and provide strong evidence for modularity (different parts of the brain are responsible for different things)
- casual links between damage and cognitive function
- can reveal surprising complexities in cognition
- provides insights that cannot be obtained in any other way
Give two examples of cognitive neuropsychology patients:
- Phineas Gage - a pipe went through his brain and altered aspects such as his personality
- HM - issues with memory
What is cognitive neuropsychology?
– study patients with brain damage/neurological disease to understand how these affect cognition
What are the limitations of cognitive psychology?
- ecological validity – this can be mitigated by well-designed experiments to test specific questions
- behavioural evidence offer only indirect evidence about the internal mental process – so researchers must infer what this means - theories can be too general and not make clear predictions
- findings can be paradigm specific
- no unifying model
What are the strengths of cognitive psychology?
- scientific and systematic approach to studying cognition
- provided bedrock for other approaches to build upon
- diverse and flexible – has strongly influenced lots of other areas of psych
Do cognitive processes occur in parallel? Explain
Many processes occur in parallel and there is interaction between cognitive system (e.g., both attention and memory are involved in decision making)
What is top-down processing?
– driven by expectation/knowledge about the world
What is bottom-up processing?
– driven by environment (input produces output) but this is an oversimplification
What does cognitive psychology study and what evidence does it use?
– uses behavioural evidence to study cognition.
What is the Alan Turing Test?
Alan Turing test – converse with a robot without knowing that you are talking to a robot. Difficult to pass but raises the questions about what it would mean to pass this test – is language intelligence or is it just following instructions?
How did the cognitive approach come about?
Cognition came about as a response to behaviourism as many findings can’t be explained by a simple stimulus response
What did Bruner’s research show?
- asked rich and poor kids to estimate the physical size of coins.
- Found that the poorer children overestimated the size. Argued that this was because the coins meant more to the poorer children and this “need” inflated their perception of the physical size of the coin – shows an internal representation and so presents problem for behaviourists
How did Tolman challenge the behaviourist view?
– challenge behaviourist view that nothing happens between S –> R
Rats will learn layout of maze even when not reinforced to do so. Proposed that animal and people’s behaviour is goal directed – represent external world using internal mental representation
How did Chomsky critique Skinner concerning language?
- Language is inherently generative (e.g., can produce unique utterances that they’ve never encountered before)
- Language is learned even where there is poverty of stimulus
What is the issue with behaviourism?
Issue with behaviourism is that it does not study the mental processes between stimulus and response
What did Skinner say about the way we learn language?
Skinner – way we learn language can be explained by behaviourists principles (reinforcement)
What is blocking?
– involves pairing of the CS and US at exactly the same time
- unlikely to lead to an association (CC)
What are the types on contingency?
- positive contingency - so bell signifies a higher probability of seeing meat
- negative contingency - where the bell indicates that meat is unlikely to come.
- Both cases demonstrate clear learning
- According to contingency learning, the animal updates their understanding of the probability of the US following the CS on a trial-by-trial basis
- Learning occurs through things being surprises
What is contingency?
– how often the US follows the CS
What is contiguity?
– when one thing follows another
- In Psychology we think of it as the proximity in which a response follows a stimulus.
- The timing of this is crucial in order to develop a clear, strong association
Explain how CC can relate to diabetes
- Diabetes – inject insulin, to deplete blood sugar levels
o After many injections, sight of needle starts to trigger early physiological response = increase in blood sugar levels
o Body is striving for homeostasis – CR prepares for US (compensatory reaction)
Explain how a CR in CC can involve conditioned emotions?
- E.g., Mr Wolf sees cat shack, then hears footsteps and is then shocked
- Initially, wolf ignores the footsteps, salivates
- In later trails, wolf starts to suppress salivation
- called a conditioned emotional response procedure
What is an inhibitor?
something that stops the CR, may be because the ppt learns that this stimulus is not similar to other stimuli that produce a CR
Can classical conditioning generalise to similar situations?
Yes, if a similar stimulus is seen, the CR can generalise to similar situations, as you go further from these similar situations however, level of CR response decreases
What is re-conditioning?
If CS is the continuously associated with CR, then this is called re-conditioning – not starting from zero, already have been conditioned once so sees a jump
What is spontaneous recovery?
Spontaneous recovery – if CS is accompanied by CR after it not being accompanied by it
– conditioning response is initially not as strong
When does extinction happen?
Extinction happens when CS is not accompanied by CR
What is second order acquisition?
a phenomenon whereby a conditioned stimulus (CS) acquires the ability to elicit a conditioned response (CR) without ever being directly paired with an unconditioned stimulus (US).
True or False? More repeated pairings lead to a greater probability of the CR after the CS?
TRUE
What is classical conditioning?
- Leaning = associations
- Either strengthening or weakening of existing associations
An evaluation of automatic vs controlled processes in visual attention?
- Monitor performance
- Detect conflict and resolve it
- Flexibly switch between responses
- Inhibit inappropriate response
- Keep track of where system is – and was
- Sustain attention on current goals
- But raises the issue of what does the controlling? What chooses?