Lecture 1: Thinking Before Doing
Mental representation
any mental content or operation that stands for something else in the world
- categories, symbols, memories, truth values, probabilities …
Function
- classification
- additional attributes
- steering attention and interpretation
- communication
- thinking
Availability & accessibility
- eg. many files: availability is all the files, accessibility is opened file in front of you
- accessibility is activation potential of available knowledge
How is knowledge stored?
- associative network models
- schema models
- predictive coding
- connectionist models
- multiple format models
- embodied cognition
- situated cognition
Associative network models
conceptual representation that views memory as consisting of set of nodes and interconnecting links where
nodes represent stored info or concepts and links represent the strength of association between this info or
concepts
- parallel (ie. spreading activation) and serial (ie. intentionally performing search) versions of this theory
Bruner’s Schema models
- operating like a lens and thereby directing attention, memory and judgment
- no nodes, but about relationship about situation and schema that we have (eg. relationship between
present objects and people)
Predictive coding
- related to schema models
- prior = prediction about something
- affects perception
- posterior = comparison between perception and prior
- sees brain as machine that predicts future and tests whether prediction fits actual situation
- if not, change prior
- mars bar always tastes same → strong prior & apple does not → less strong prior
Connectionist model
connectionism = parallel distributed processing
- nodes do not have semantic meaning (like in associative network models)
- facilitative and inhibitory links
- concepts exist by means of dynamic interplay of distributed elements
- representations are seen as emergent patterns of connected nodes
- different representations are only different activity patterns in same nodes; semantic info is not
‘stored’ anywhere
- input, connection weights (hidden) and output elements
- cells that fire together, wire together
, - local connectionist models = same as connectionist model, but some nodes do have semantic
meaning
Multiple format models
separate systems → semantic, procedural and affective memory system
Embodied cognition
do mental representations extend outside mind, both to body and external environment?
- eg. when thinking about coffee, not just associations like ‘caffeine’ and ‘hot beverage’ come to mind,
also smell and taste → sensory info in addition to knowledge
- info about what it is like to experience something as being in possession of human body
- cognition is automatically linked to bodily experience because through this knowledge is
gathered in the first place
- representations are modality-specific
- stored ‘in the body’
- partial re-experience
Situated cognition
mental representation result form dynamic interactions between brain, body and environment
Impulsive vs. reflective behavior: ‘Four horsemen’ of automaticity
1. Level of consciousness
2. Level of intentionality
- more intentional is less automatic
3. Level of controllability
- more controlled is less automatic (eg. breathing is not controlled)
4. Level of efficiency
- amount of mental resources something costs
Reflective impulsive system
Two systems working in parallel that elicited behavior through different processes:
- impulsive → always activated
- reflective → needs motivation and opportunity to be activated
- uses knowledge representation to make propositions based on associations
Impulsive processes
- purely based on associations
- behavior within impulsive system is based on spreading activation (like associative network models) of
knowledge to motor representations according to ideo-motor processes (= thinking about behavior is
enough to trigger behavior)
- activated associations directly influence behavior
- eg. seeing older person → activate associations → affects behavior: walking more slowly
Reflective processes
- behavior in reflective system is based on choices
- choices may, by means of intentions, activate motor-representations in impulsive system
- choices can based on full consideration of pros and cons, intuition or anything in between
Strack & Deutsch: Only if there is sufficient processing time, intention and cognitive capacity to extract
meaning of a negation, will reflective system be engaged and task be successfully completed
- instruction not to perform behavior certain behavior may strengthen link to behavior schema that
produces undesired behavior
- eg. seeing ‘do not litter’ without enough processing time, intention and/or cognitive capacity →
‘do not litter’ is not negated and will be interpreted by impulsive system as ‘do litter’
Understanding priming effects
- empirical (in)consistency → failure to replicate findings
, - theoretical aspects → what models do we use to predict the effect of a prime?
Empirical (in)consistency
- conceptual priming seems very robust phenomenon
- items with similar meaning are primed
- behavior priming effects seem difficult to replicate
- exposing people to external stimulus activates mental construct associated with stimulus →
affect behavior without actor necessarily being aware of influence
- many studies were underpowered
- behavior priming effects: false positive?
- despite underpowered, many empirical evidence for small behavioral priming effect
Theoretical aspect
- process is critical, not prime itself
- moderators (eg. value, self-relevance)
- new models: how does prime affect behavior?
- eg. situated inference model
Situated inference model
1. Prime exposure
- primes produce general construct accessibility which is used flexibly
- priming stimulus → general construct accessibility → …
2. Misattribution
- accessible content is misattributed to one’s own response
- … accessible content is assumed to be caused by perception of focal target → …
3. Afforded questions
- misattributed content is used to answer question afforded by focal target
- … person, object or event (‘who is that?’) → construal priming
- … appropriate behavior (‘what will I do?’) → behavior priming
- … current motivation (‘what do I want?’) → goal priming
Different from reflective impulsive model where association directly leads to behavior (= ideo-motor idea)
- need to misattribute prime to own perception first!
Lecture 2: Attitudes in Action
Attitude
psychological tendency that is expressed by evaluating particular entity with some degree of favor or disfavor
- tendency → doesn’t have to be stable; context dependent
- evaluating
- particular entity → eg. things, products, persons, ideas, behaviors
Functions of attitudes
- knowledge & utility function → eg. having an allergy
- value-expressive function → similar attitudes help to bond (social adaptive function)
Measuring attitudes
- explicit measures
- semantic differentials → using contrasting adjectives about certain object
- implicit measures
- affective priming procedure → classify positive or negative adjective and measure response
time
How are attitudes mentally represented?
1. Dual model
2. MODE-model
Mental representation
any mental content or operation that stands for something else in the world
- categories, symbols, memories, truth values, probabilities …
Function
- classification
- additional attributes
- steering attention and interpretation
- communication
- thinking
Availability & accessibility
- eg. many files: availability is all the files, accessibility is opened file in front of you
- accessibility is activation potential of available knowledge
How is knowledge stored?
- associative network models
- schema models
- predictive coding
- connectionist models
- multiple format models
- embodied cognition
- situated cognition
Associative network models
conceptual representation that views memory as consisting of set of nodes and interconnecting links where
nodes represent stored info or concepts and links represent the strength of association between this info or
concepts
- parallel (ie. spreading activation) and serial (ie. intentionally performing search) versions of this theory
Bruner’s Schema models
- operating like a lens and thereby directing attention, memory and judgment
- no nodes, but about relationship about situation and schema that we have (eg. relationship between
present objects and people)
Predictive coding
- related to schema models
- prior = prediction about something
- affects perception
- posterior = comparison between perception and prior
- sees brain as machine that predicts future and tests whether prediction fits actual situation
- if not, change prior
- mars bar always tastes same → strong prior & apple does not → less strong prior
Connectionist model
connectionism = parallel distributed processing
- nodes do not have semantic meaning (like in associative network models)
- facilitative and inhibitory links
- concepts exist by means of dynamic interplay of distributed elements
- representations are seen as emergent patterns of connected nodes
- different representations are only different activity patterns in same nodes; semantic info is not
‘stored’ anywhere
- input, connection weights (hidden) and output elements
- cells that fire together, wire together
, - local connectionist models = same as connectionist model, but some nodes do have semantic
meaning
Multiple format models
separate systems → semantic, procedural and affective memory system
Embodied cognition
do mental representations extend outside mind, both to body and external environment?
- eg. when thinking about coffee, not just associations like ‘caffeine’ and ‘hot beverage’ come to mind,
also smell and taste → sensory info in addition to knowledge
- info about what it is like to experience something as being in possession of human body
- cognition is automatically linked to bodily experience because through this knowledge is
gathered in the first place
- representations are modality-specific
- stored ‘in the body’
- partial re-experience
Situated cognition
mental representation result form dynamic interactions between brain, body and environment
Impulsive vs. reflective behavior: ‘Four horsemen’ of automaticity
1. Level of consciousness
2. Level of intentionality
- more intentional is less automatic
3. Level of controllability
- more controlled is less automatic (eg. breathing is not controlled)
4. Level of efficiency
- amount of mental resources something costs
Reflective impulsive system
Two systems working in parallel that elicited behavior through different processes:
- impulsive → always activated
- reflective → needs motivation and opportunity to be activated
- uses knowledge representation to make propositions based on associations
Impulsive processes
- purely based on associations
- behavior within impulsive system is based on spreading activation (like associative network models) of
knowledge to motor representations according to ideo-motor processes (= thinking about behavior is
enough to trigger behavior)
- activated associations directly influence behavior
- eg. seeing older person → activate associations → affects behavior: walking more slowly
Reflective processes
- behavior in reflective system is based on choices
- choices may, by means of intentions, activate motor-representations in impulsive system
- choices can based on full consideration of pros and cons, intuition or anything in between
Strack & Deutsch: Only if there is sufficient processing time, intention and cognitive capacity to extract
meaning of a negation, will reflective system be engaged and task be successfully completed
- instruction not to perform behavior certain behavior may strengthen link to behavior schema that
produces undesired behavior
- eg. seeing ‘do not litter’ without enough processing time, intention and/or cognitive capacity →
‘do not litter’ is not negated and will be interpreted by impulsive system as ‘do litter’
Understanding priming effects
- empirical (in)consistency → failure to replicate findings
, - theoretical aspects → what models do we use to predict the effect of a prime?
Empirical (in)consistency
- conceptual priming seems very robust phenomenon
- items with similar meaning are primed
- behavior priming effects seem difficult to replicate
- exposing people to external stimulus activates mental construct associated with stimulus →
affect behavior without actor necessarily being aware of influence
- many studies were underpowered
- behavior priming effects: false positive?
- despite underpowered, many empirical evidence for small behavioral priming effect
Theoretical aspect
- process is critical, not prime itself
- moderators (eg. value, self-relevance)
- new models: how does prime affect behavior?
- eg. situated inference model
Situated inference model
1. Prime exposure
- primes produce general construct accessibility which is used flexibly
- priming stimulus → general construct accessibility → …
2. Misattribution
- accessible content is misattributed to one’s own response
- … accessible content is assumed to be caused by perception of focal target → …
3. Afforded questions
- misattributed content is used to answer question afforded by focal target
- … person, object or event (‘who is that?’) → construal priming
- … appropriate behavior (‘what will I do?’) → behavior priming
- … current motivation (‘what do I want?’) → goal priming
Different from reflective impulsive model where association directly leads to behavior (= ideo-motor idea)
- need to misattribute prime to own perception first!
Lecture 2: Attitudes in Action
Attitude
psychological tendency that is expressed by evaluating particular entity with some degree of favor or disfavor
- tendency → doesn’t have to be stable; context dependent
- evaluating
- particular entity → eg. things, products, persons, ideas, behaviors
Functions of attitudes
- knowledge & utility function → eg. having an allergy
- value-expressive function → similar attitudes help to bond (social adaptive function)
Measuring attitudes
- explicit measures
- semantic differentials → using contrasting adjectives about certain object
- implicit measures
- affective priming procedure → classify positive or negative adjective and measure response
time
How are attitudes mentally represented?
1. Dual model
2. MODE-model
