Lecture 5 Flashcards
Causal Question’s explainations in the role of DA in reward:
- “liking”
- Learning
- “wanting”
Experimental approaches to determine what DA does in reward: what reward function is _______
- Necessary cause (lost)
- Sufficient cause (enhanced)
- coding of function (coded)
Answers of Causal DA Question (1/4)
- Activation sensorimotor hypothesis of effort, arousal and response vigor
Answers of Causal DA Question (2/4)
- Hedonia hypothesis of reward pleasure
Answers of Causal DA Question (3/4)
- Reward learning hypothesis of associative stamping-in, teaching signals and prediction errors
Answers of Causal DA Question (4/4)
- Incentive Salience hypothesis of reward “wanting”
(A: 1/4) Activation-sensorimotor hypothesis
DA mediates general functions of: action generation, effort, movement, general arousal/behavioural activation.
(A: 2/4) Hedonia hypothesis
DA in NAc is a “Pleasure neurotransmitter”: Mediates +ve reinforcing effects of reward stimuli. Supress DA~Anhedonia
DA and Hedonic Rxn’s in rats
DA reduction does NOT decrease “liking”
DA activation does not increase “liking”
5 different brain manipulations insufficiently enhancing hedonic impact of reward
- Hyper-DA mutation (KO mice)
- AMPH injected in NAc
- AMPH systemic administration
- Sensitization
- Electrical brain stim reward
DA and Hedonic Rxn’s in humans
DA does NOT mediate subjective ratings
DA does NOT mediate subjective ratings w/ two methods:
- Patients with Parkinsons show normal liking ratings
2. DDS patients have compulsive actions due to more L-Dopa administered.
Nonhedonic hypotheses of reward (two main)
- Reward learning
2. Incentive Salience
(A: 3/4) Reward Learning hypothesis
DA signal modulates synaptic plasticity in target neurons, psychologically suggesting DA acts to etch links (in neostriatum and NAc) b/w S-S and S-R events.
Reward-Learning (DA) Hypotheses (3 types)
all say DA mediates learning but in different ways
- DA signals stamp in S-R and S-S associations whenever reward follows
- DA activation causes new habit learning (ergo habit performance)
- DA mediate teaching signals via US prediction errors
Reward-Learning (DA) Hypothesis (1/3): Associative stamping in
DA acts as US reinforcement signal to learn associations about preceding reward stimuli. Based on Law of effect (thorndike)
Reward-Learning (DA) Hypothesis (1/3): Associative stamping in: Thorndike’s Law of effect (3 components)
- Context (S)
- Instrumental response (R)
- Response outcome (O)
Reward-Learning (DA) Hypothesis (1/3): Associative stamping in: Thorndike’s Law of effect (summary)
if R in presence of S followed by satisfying event~ S-R association strengthened
If R in presence of S followed by unsatisfying event~S-R association weakened.
Reinforcer (O) “stamps” S-R association
Reward-Learning (DA) Hypothesis (1/3): Associative stamping in: Supporting Evidence
- Extinction-mimicry
- DA modulating mechanisms of LTP and LTD
- DA manips right after learning trial alters consolidation.reconsol of memories
Reward-Learning (DA) Hypothesis (2/3): Habit Learning
DA causes new S-R habits to be learned &/or modulates strength of learned S-R habits
Reward-Learning (DA) Hypothesis (3/3): Prediction Error Learning Model(s)
suggest DA embodies specific parameters of learning equations by acting as a teaching signal.
- Rescorla-Wagner Model
- Temporal Difference Model
Reward-Learning (DA) Hypothesis (3/3): Prediction Error Learning Model (a) Rescorla-Wagner Model
suggests DA boosts might increase predictions of future reward (V) to a (CS). Increase in DA elevates prediction error (lambda-V) generated by hedonic impact of US (lambda) at moment of reinforcement, boosting amount of learning on the trial (deltaV).
Reward-Learning (DA) Hypothesis (3/3): Prediction Error Learning Model (b) Temporal Difference Model
time more explicit in expectations of reward as series of future events. V becomes more accurate thru learning about actual rewards via prediction errors that modulate synaptic weights in circuits involved in future predictions.
DA makes _____ contributions to both ____ & ____, but not directly causing _______
Indirectly, learning, learned performance, formation of new reward associations
Evaluation of DA’s direct roles (divided into 3 questions)
- is DA a necessary cause?
- Is DA a Sufficient cause?
- DA neuronal firing patterns?
Is DA necessary cause?
unneccessary for normal reward learning, ergo not a cause. DA contributes to learning as teaching signal, prediction error or Stamping-in mechanism
Is DA a sufficient cause?
showed increased “wanting” but not “liking”.
DA neuronal firing?
DA neurons may code informational consequence of learning signals. Hypothesized to be incentive motivation.
(A: 4/4) Incentive Salience hypothesis: central premise
Reward is composite of wanting, liking and learning.
DA only mediates “wanting” portion.
Incentive salience is NOT:
a) hedonic “liking, nor sensory pleasure
b) reducible to learning
Incentive salience is:
a motivational transformation of a reward-related neural representation (ie: percieved or recalled CS or US).
Incentive salience value of stimulus is generated by
mesolimbic systems each time reward stim occurs and pre-existing reward related associations
DA= _____
Incentive salience
Important that Incentive salience combines:
Mesolimbic dopamine system mechanisms AND Physiological states
Motivational magnets
CS produced by Incentive salience.
Incentive Salience theory step 1:
- CS “wanting” assignment based on “liked” US. predictors go from neural to desired. The hedonic activation by new US.
Incentive Salience Theory step 2:
- CS “wanting” reboosting. Steps are appetitive. Associative-learning between, strengthens the incentive salience assignment to rewarded stimuli.
Reboosting
keeps up motivation for long-term. Keeps it wanted for later trials. Reboosting keeps the reward relevant. counters extinction mimicry.
Extinction Mimicry
gradual falling off of consummation of a well-established reward
Incentive Salience Theory step 3:
Attribution of “wanting” to a CS. making the CS more “wanted” by route of the reboosting of stage 2 and adds physiological states of each reward introduction.
Incentive Salience of CS becomes a conditioned motivational response