Task 5: Success, failure & the cerebral cortex Flashcards
A1: What part of the cortex is specialized for deciding on preferences for objects in environment based on their value?
1) lateral PFC -> state representation
- State + utility
2) OFC -> value representation
- Utility (of action options)
3) ACC -> outcome evaluation
- Action options + utility
Describe Lee et al.’s three-way division of the cerebral cortex in terms of goal-oriented decision making
1) lateral PFC -> state representation
- accumulates sensory evidence
- transforms perceptual info into motor output
- state is maintained in working memory -> delay activity -> expected outcomes
- > represents various states of environment & their utility
2) OFC -> value representation
- Expected reward/outcomes for different stimulus options
- > allows switching to better action choice when values change
- responds to expected reward, like lPFC but without state encoding
- based on learned S-R associations
3) ACC -> outcome evaluation
- Represents possible actions & their value/utilities
- Key role when environment is uncertain/dynamic
- > updating strategy after error
- > encoding/updating utility
A1: Explain what utilities are and why “object selection” processes are interesting to neuroeconomists.
Utility:
- assigned value to possible actions
- Highest utility -> optimal choice
- Preference -> most popular choice
- Rationality: people prefer optimal choice (in reality, biases, etc.)
Object-selection:
- represents a decision-making process which is what economists study
- neuroeconomics tries to find the neural correlates of decision-making
- selection of objects based on utility comparison & updating in order to maximize outcome
A2: Discuss Ruthworth’s 3 distinctive steps of goal-oriented decision-making (2nd model)
lOFC: Value assignment
- Learning & updating S-R associations
- Activity
- > when new reward cue are learned, feedback-phase
- > when values of S-R associations change (e.g. sensory-specific satiety)
mOFC/vmPFC: Reward representations + comparison
- Represents rewards, but not cues
- involved in weighting the value of available stimulus options & choosing the one with the highest subjective value
- Activity
- > proportional to reward expectations
MCC: action selection
A2: Briefly name the difference between Rushworth’s 1st and 2nd model
1st model:
lOFC -> value assignment
vmPFC/mOFC -> value expectation
MCC -> value comparison
2nd model:
lOFC -> value assignment
vmPFC/mOFC -> value expectation + comparison
MCC -> action selection
A2: Compare/combine Lee’s & Rushworth’s divisions
Stimulus-reward associations (S-R):
- Lee
- > lPFC (state + reward)
- > OFC (utility based on S-R)
- Rushworth
- > lOFC (S-R value assignment)
- > vmPFC/mOFC (S-R value comparison)
Action-Outcome associations (A-O):
- Lee
- > ACC (encoding & updating A-O)
- Rushworth
- > MCC (action selection)
A3: Define the phenomenon of sensory-specific satiety
The decrease in reward value of a food which is usually greater for the food eaten to satiety than for other foods
- Sensory-specific: specific to certain food that were just eaten
- Eating food to satiety -> reward value for that food becomes 0 -> decreased pleasantness -> decreased expected activity related to value of that food
- Example of change in sensory context -> value updating necessary (hence we want food variety)
A3: Describe the O’Doherty (2000) study on sensory-specific satiety
Measurements before vs. after eating banana
- Pleasantness ratings for banana vs. vanilla
- > equal pre-satiety
- > both lower post-satiety
- > larger decrease for banana
- Intensity ratings
- > no significant changes
- fMRI BOLD signal in lOFC related to olfactory sensory-specific satiety
- > decreased activity to odor of eaten food (banana)
- > no decrease for not-eaten food (vanilla)
- -> OFC activity is related to sensory-specific satiety
- -> reward value of food changed after eating it -> OFC important for updating reward values
A3: connect the results of O’Doherty with Lee & Rushworths’ models
Rushworth: - lOFC for S-R value assignment (& updating/"reassignment") - mOFC for S-R value comparison Lee - OFC for utility based on S-R
E.g.
lOFC assigns value: I love bananas!
-> mOFC/vmPFC compares value: I like bananas more than vanilla!
-> ACC for A-O: if I eat banana, I’ll feel good
-> MCC selects action, etc. : we eat banana
-> lOFC needs to update value: I don’t like bananas as much right now because I just ate one
A4: What role does the cerebral part of reward system (i.e. vmPFC) play a role in personality?
- self-referent effect
- judging personality
- describing self/others
vmPFC combines information in relation to self -> self-referential processing
- Essential for appropriate social behaviour
(e. g. Phineas Gage’s OFC) - Self-referent effect
- > Connecting info to oneself increases depth of processing –> improves storage of information
- > “self” is extreme/deep/best point of memory processing
- > anything about ourselves is more engaging/interesting/relevant
- Study on judging personality adjectives
- > better memory to judgments about self vs. others
- > unique activation of mPFC
- Description of self relies on “summary” of traits, not specific events
Description of others focuses on specific behaviours/related to specific traits
-> Dense amnesia -> intact ability to describe oneself - Frontotemporal dementia (vmPFC, OFC regions affected)
- > acquired sociopathy, antisocial acts, immoral emotions, … (specifically vmPFC)
- > personality change with almost intact cognition
A4: How accurate are our self-perceptions? And which areas are involved in self-perceptions? What happens when they are lesioned?
Usually, tendency for positive illusion about selves
- mPFC: default network -> self-referential processing
-> vACC: positive self-perceptual bias
o distinguishes positive from negative self-relevant info -> allows focus on own positive features
o involved/dysfunctional in depression
o active when thinking about positive future events/judging positive adjectives
-> OFC: maintain relatively accurate insight into own behaviour
o Needed for social functioning
o Self-imaged is biased but not delusional/detached from reality
o Lesion -> inappropriate conversation topics, unaware of social mistakes
A5: Present the Default Mode Network (DMN)
1) Affective node: vmPFC/mOFC
- > Self-referential information-processing -> core role in goal-oriented decision-making
2) Memory-related node: PCC (limbic system)
- > Retrieval of episodic memory
- > judgment about other people
Function:
- Activity when brain at rest
- Deactivated when engaged in a task
- > daydreaming, mind-wandering
A5: Why is the affective node (vmPFC) of the Default Mode Network (DMN) relevant for personality & goal-directed decision-making??
vmPFC/OFC
- Rushworth: S-R value comparison
- Here: Reward comparison in relation to self -> value for the self
- > self-referential processing & self-perception are important for decision-making
- > damage -> changes in personality & impaired social interactions
A5: How does the Default Mode Network’s deactivation reflect decision-making
mPFC:
- Rest -> higher metabolic rate
- Self-referential judgments -> less deactivation
- Default mode: self-referential processing, daydreaming
- Cognitive tasks -> more deactivation
A6: Give evidence for Rushworth’s claim that goal-directed decision-making requires division between object selection & action selection.
vmPFC activity
-> value difference signal: Bigger vmPFC signal if reward signal of 2 stimuli become dissimilar
mCC
-> inverse value difference signal: Bigger mCC signal if reward signal of 2 stimuli become more similar
Lesion
- vmPFC/mOFC -> value comparison impairments
- MCC -> disrupted A-R association, still S-R association
(- vmPFC/OFC + lPFC -> no S-R association, still A-R association; think back to Lee)
- -> Double Dissociation between MCC & vmPFC/OFC
- vmPFC/mOFC doesn’t decide about action but selects reward goal (stimulus-reward association)
- MCC decides about what action to perform to obtain reward (action-reward association)
