figure 4 responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this...

49

Post on 22-Dec-2015

215 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting
Page 2: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting
Page 3: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting
Page 4: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting
Page 5: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting
Page 6: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting
Page 7: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this

response to progressively earlier reward-predicting conditioned stimuli with training (middle). The bottom record shows a control baseline task when the reward is predicted by an earlier stimulus and not the light. From Schultz et al. (1995) with permission.

Page 9: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting
Page 10: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Odor Selective Cells in the Amygdala fire preferentially with regard to outcome or reward value of an odor prior to demonstration that the animal has learned this outcome or value.

Odor Selective Cells in the Amygdala fire preferentially with regard to outcome or reward value of an odor simultaneous to demonstration that the animal has learned this outcome or value.

Page 11: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Cells in Orbitofrontal Cortex (OFC) show less selectivity to outcome, in rats without an amygdala. This

demonstrates a role for the amygdala in conveying motivational/reward information to the OFC.

Page 14: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Dopamine, reward processing and optimal prediction

ONLY AS A REFERENCE FOR THOSE WHO ARE INTERESTED IN BEGINNING TO CROSS THE NEUROBEHAVIORALCOMPUTATIONAL DIVIDE – Maybe after the Exam??

Page 15: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Human dopaminergic system

Page 16: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Cortical and striatal projections

Schultz, 1998

Page 17: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Koob & Le Moal, 2001

Page 18: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Schultz, Dayan & Montague 1997

Page 19: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Expected Reward

v = wu

v : expected reward w : weight (association) u : stimulus (binary)

Page 20: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Rescorla-Wagner Rule

Association update rule: w w + αδuw : weight (association)α : learning rateu : stimulus

Prediction error: δ = r - vr : actual reward

v : expected reward

Page 21: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Rescorla - Wagner provides account for:

Some Pavlovian conditioningExtinctionPartial reinforcement

and, with more than one stimulus:

BlockingInhibitory conditioningOvershadowing

… but not

Latent inhibition (CS preexposure effect)Secondary conditioning

Page 22: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

A recent update: uncertainty (i²)

Kakade, Montague & Dayan, 2001

Page 23: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Kalman weight update rule:

wi wi + αi δ

With associability:

αi = i² ui

jj² uj +E

Page 24: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

An example:

Page 25: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

U1 U2 U3 U4 U5

U(t)

input

Page 26: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

U(t)

input

r(t)

Page 27: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

U(t)

input

r(t)

w(t)

Page 28: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

U(t)

input

ŵ(t)

v(t)

Page 29: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

U(t)

input

r(t)

ŵ(t)

v(t)

Page 30: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

U(t)

input

r(t)

ŵ(t)

v(t)

δ(t)

Page 31: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

(t) = r(t) - v(t)

Error Rule

Page 32: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

U(t)

ŵ(t)

v(t)

inset

Ui -input

i wi

-uncertainty -weight

Page 33: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Uncertainty

Page 34: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Kalman learning & associability

weight update rule:

ŵi (t+1) = ŵi (t) + α i (t) δ (t)

associability:

αi(t) =i(t)² xi (t)jj(t)² xj (t)+E

Page 35: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting
Page 36: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Stimulus uncertainties

Page 37: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Reward prediction

Page 38: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Predicting future reward

single time steps:v = wu v : expected reward

w : weight (association)

u : stimulus

total predicted reward:

v(t) = w(τ) u(t - τ) t : time steps in a

trial τ : current time step

t τ=0

Page 39: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Sum of discounted future rewards:

With 0 ≤ γ ≤ 1

In recursive form:

Schultz, Dayan & Montague, 1997

Page 40: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Exponential discounting, γ = .95

0 10 20 30 40 50 60 70 80 90 1000

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

TIME STEPS

RE

WA

RD

VA

LUE

Page 41: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Temporal difference rule

Total estimated future reward: v(t) = r(t)+ γv(t+1) r(t) = v(t)-γv(t+1)

Temporal difference rule: δ = r(t)+γv(t+1)-v(t)

(With single time steps: δ = r - vr : actual reward

v : expected reward )

Page 42: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Temporal difference rule

Total estimated future reward: v(t) = r(t)+v(t+1) r(t) = v(t)-v(t+1)

Temporal difference rule: δ = r(t) + v(t+1)-v(t)

(With single time steps: δ = r - vr : actual reward

v : expected reward )

Page 43: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Schultz, Dayan & Montague, 1997

Page 44: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Schultz, 1996

Page 45: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Anatomical interpretation

Schultz, Dayan & Montague, 1997

Page 46: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Temporal Difference Rule for Navigation

between successive steps u and u’

δ = ra (u) + γ v(u’)-v(u)

Page 47: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Behavior evaluation Hippocampal place field

Foster, Morris & Dayan 2000

Page 48: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Spatial learning

Foster, Morris & Dayan 2000

Page 49: FIGURE 4 Responses of dopamine neurons to unpredicted primary reward (top) and the transfer of this response to progressively earlier reward-predicting

Conclusions

• Behavioral study of (nonhuman) neural systems is interesting

• Neural processes amenable to contemporary learning theory

• .. they may play distinct roles a normative framework of learning

e.g. vta, hippocampus, subiculum, also- Ach in NBM/SI, NE in LC, 5-HT, ventral striatum,

lateral connections ,core/shell distinctions of the NAAC, patch-matrix anatomy in basal ganglia, the superior colliculus,

psychoalphabetadiscobioaquadodoo