Reward Processing

Question 1

What is reward prediction error?

Answer 1

When a predicted reward does not equal the reward obtained (unpredicted or surprising)

Question 2

What are the consequences of prediction error?

Answer 2

Positive = new learning; zero = no new learning; negative = extinction

Question 3

Neurons in which brain region produce dopamine, which help us encode reward processing?

Answer 3

Ventral tegmental area (in the mid brain)

Question 4

Which parts of the frontal lobes are associated with reward processing?

Answer 4

OFC, medial and dorsal PFC

Question 5

In which regions does the dopaminergic pathway travel through?

Answer 5

Between midbrain, striatum and cortex (ventral tegmentum; OFC; PFC; basal ganglia + nucleus accumbens + globus palidus in striatum; substantia nigra; down to brain stem

Question 6

Dopamine neurons (involved with reward prediction and error detection) travel specifically to medial temporal cortex, dlPFC, premotor, parietal cortex, OFC, striatum, and amygdala. What are each of these regions associated with?

Answer 6

Medial temporal - reward detection/prediction; dlPFC, premotor and parietal - goal representation; OFC - relative reward value/reward expectation; striatum - reward detection/goal representation; amygdala - emotions/conditioned effects

Question 7

What occurs in the brain when performing a sequence of steps leading to a rewarding outcome, such as the process of making a coffee?

Answer 7

There’s an increasing amount of dopamine concentration; there’s an inherent reward leading to the outcome, even though the process may not be rewarding in itself

Question 8

Compare what happens in the synapses when taking cocaine to methamphetamine

Answer 8

Cocaine blocks the dopamine transporter so hinders reuptake in the pre-synaptic neuron; concentrations are increased in the synaptic cleft, leading to a continuous feeling of reward; Meth passes through the terminal buttons and gets mixed in with the dopamine, passes across the membrane and blocks pre-synaptic cleft, also blocking reuptake

Question 9

When measuring the activity of a ventral tegmental neuron in a monkey brain, we see an increase of action potentials firing in this region straight after a reward (apple) is obtained, and no activity when there is no reward and they only touch a wire. What happens when the reward is delayed?

Answer 9

Whilst doing a picture discrimination task, the monkey’s learnt the reward will be delivered after 1 sec, but when they make him wait, there’s a delayed response in activity; bursting of action potentials changes depending on when reward is delivered (temporally precise)

Question 10

What was discovered when measuring a putamen neuron’s activity in a monkey brain based on expectation during learning, and what does this suggest?

Answer 10

The monkey executed a movement to obtain a reward and showed an increase in the build up period (anticipation); also rewarded for inhibiting a movement (same effect); when movement is unrewarded, shows initial trial and error (neurons generating signals) until he learns there’s no reward and anticipatory signals disappear; suggests anticipation shapes behaviour

Question 11

Describe how activity of an orbitofrontal neuron in a monkey’s brain was associated with relative reward preference

Answer 11

During a pattern discrimination task, the monkey showed less activation for reward B (apple) when compared to A (raisin); but activations increased for reward B (as much as it had for A) relative to reward C (cereal); showing preference changes according to relativity

Question 12

Compare the caudate activity in a monkey’s brain when making an uncued movement for reward vs. non-reward

Answer 12

A large increase in activity in anticipation of reward, during movement onset and at reward onset, but no activity for unrewarded movement

Question 13

Kristjansson et al. investigated how reward influences “pop-out” during a visual search, where Ps had to determine whether the target (green or red diamond) had the top or bottom missing. What were the conditions and results?

Answer 13

Green target = high reward (10 points 75% of time/1 point 25% of time); red target = low reward (1 point 75% of time/10 points 25% of time); RTs much faster for high reward condition; evidence of priming effects

Question 14

In Kristjansson et al.’s visual search task, what was found when looking at priming effects from preceding to current trial, and what does this suggest?

Answer 14

Largest priming effect when high reward was likely (green target) and high received; smallest priming effect when low reward likely (red target) and low received; same effect for high likely and low received, or low likely and high received (in middle); shows what we expect to happen and what actually happens influences behaviour

Question 15

What happened in Kristjansson et al.’s study when they reversed the reward contingencies 3 times per 200-trial block (red - high; low - green), and what does this show about human behaviour?

Answer 15

Opposite effects; after the point of change in reward schedule, RTs for red targets got faster, and slower for green targets; Shows how behaviour can track reward contingencies even when unaware of the change; many of our behaviours are motivated implicitly

Question 16

Anderson et al. had Ps train in a visual search task to examine reward-driven attentional capture. In the training phase, they had to search for a red circle among distractors (high reward: 80% chance of 5c/20% chance of 1c) or a green target (opposite probabilities - low reward). Describe the test phase

Answer 16

There was no reward in the test phase; they had search for a target with a unique shape (e.g. diamond) among circle distractors; in some trials, one of the distractors contained a colour associated with reward in the training phase (red/green)

Question 17

What were the results of Anderson et al.’s study?

Answer 17

Ps were slower to respond to the shape target when accompanied by a reward distractor (especially for high); the same time cost was observed even a few later

Question 18

What did Anderson et al. find in relation to working memory as a function of reward-driven attention capture?

Answer 18

People with a high working memory capacity were less influenced by distractors (they were able to block them out); those with lower WM capacity were more susceptible to reward manipulation (showing the most reward driven capture)

Question 19

In an ERP study, Kiss et al. had Ps search for red or green targets amongst grey distractors, counterbalancing reward contingencies (red:5; green:1/red:1; green:5 points), to see whether peaks/troughs would be different as a function of reward. What was found?

Answer 19

Stronger electrode responses contralaterally; N200 (N2pc) responses were larger and earlier in time for high reward targets; SPCN also higher for high reward than low; Individual differences found: N2pc amplitudes correlated with those who showed the greatest reward effects

Question 20

Weil et al. put Ps in a fMRI scanner, told them what they stand to win in a block of trials, and had them perform a grating discrimination task on the screen (determining whether the gratings in a circle to left or right of fixation was the same or different as previously presented). What did they manipulate?

Answer 20

Reward levels (80 pence - high; 10 - low; 0 - error); Level of difficulty (grating similarities); cue or no cue (to left or right)

Question 21

What behavioural effects were found in Weil et al.’s study?

Answer 21

No differences in reward contingencies in difficult trials; higher proportion correct when reward was high in easy condition; no differences in performance when attending to left vs. right

Question 22

Describe the fMRI results in Weil et al.’s study

Answer 22

Stronger responses when attending contralaterally; greater activity in OFC, ventral striatum and higher visual areas (V4)when given feedback that they’d received a reward

Question 23

When Weil et al. investigated how well Ps performed in grating discrimination after receiving rewards (trial by trial effects) what did they find?

Answer 23

Higher proportion correct after being rewarded (2/3 trials ago made no difference - only immediately preceding trial); stronger difference between V1 activity if previous trial had been rewarded

Question 24

Describe what Malhotra et al. did using reward modulation with unilateral spatial neglect patients

Answer 24

They conducted a variation of the cancellation task where Ps were presented with pound coins as targets and blurry coins as distractors, and told the more targets they find the bigger reward they’d receive (up to 15 pounds), compared to a non-reward task

Question 25

What behavioural effects were found in Malhotra et al.’s study, and what differed in 2 of the patients?

Answer 25

After receiving a reward, they performed significantly better the next time they did the rewarding task (attended more to targets on left); the 2 patients who didn’t show these effects were found to have lesions on putamen, caudate, globus pallidus and frontal cortex

Question 26

Lucas et al. had neglect patients and a control group perform an active exploration task to win rewards (point to a target to reveal its reward value). What were the conditions?

Answer 26

Points awarded per target: 0, 5, 10, or 50; Distribution: symmetric (equal rewards on both sides); asymmetric (higher rewards on left than right)

Question 27

What were the results of Lucas et al.’s study?

Answer 27

Controls were sensitive to the spatial distribution of rewards, whether they were aware or not; neglect patients showed a bias towards the right but moved further to the left in asymmetric condition (showing they’re also sensitive to rewards)

Question 28

When Lucas et al. looked at neglect patient’s performances in other tasks (e.g. star cancellation; line bisection; bilateral stimulus detection) pre and post search game session, what did they find?

Answer 28

They showed a greater improvement in performance post task in asymmetric condition than symmetric; suggests simple reward contingencies may show some generalisation

Question 29

What do both Malhotra et al. and Lucas et al.’s study suggest?

Answer 29

Reward can be used to alleviate attentional impairments in patients with unilateral spatial neglect

Question 30

Reward circuitry is widespread in the primate brain, and includes which regions?

Answer 30

Frontal cortex, nucleus accumbens, ventral tegmentum, amygdala and striatum