Questions & explanations
1. What happens to discounting when certain brain areas are impaired?
When the prefrontal cortex is damaged, people show higher temporal discounting — they want immediate rewards much more. Without the prefrontal cortex, the brain cannot plan for the future effectively. Damage to the prefrontal cortex is linked to impulsive behavior in conditions like addiction. On the other hand, damage to the amygdala or ventral striatum can reduce the pull of immediate rewards, making someone more patient. However, this can also reduce motivation. Overall, the balance between these brain areas determines how much you discount the future. Impairments in these regions can lead to extreme impatience or extreme patience, both of which can be problematic.
2. What brain area helps us decide to follow the group or stick with our own opinion?
The dorsolateral prefrontal cortex (dlPFC) plays a key role in resolving conflict between personal and group views. When its activity is high, people are more likely to stick with their own opinion, even if the group disagrees. This is because the dlPFC helps maintain personal goals and override social influence. On the other hand, the ventromedial prefrontal cortex (vmPFC) integrates social information into value signals, often leading to conformity. The balance between these two areas determines whether you follow the herd or stay independent. People with stronger dlPFC connectivity to other regions tend to be more resistant to group pressure.
3. Give an example of how the orbitofrontal cortex encodes subjective value.
Imagine you are very thirsty and someone offers you a glass of water and a chocolate bar. Your orbitofrontal cortex (OFC) gives the water a high value because your body needs it, and the chocolate a lower value. But if you just finished a big drink, the water's value drops and the chocolate becomes more appealing. This shows subjective value: the same reward has different value depending on your state. Brain studies show OFC neurons fire more for the better option in that moment. Also, if you later remember that a certain fruit made you sick, the OFC lowers its value. So the OFC constantly updates value based on internal needs and memories.
4. How does temporal difference learning use prediction errors to learn?
Temporal difference (TD) learning is a way the brain learns from ongoing events. Instead of waiting until the end, it updates predictions step by step. At each moment, the brain compares its current prediction of future reward with the new prediction plus any reward just received. The difference is a prediction error. This error is used to adjust the predictions for the current state. For example, if a light predicts food, the brain learns the light's value by noticing the error between expected and actual reward. Over many trials, the predictions become very accurate. TD learning explains how we learn to predict rewards from cues in time.
5. How does the prefrontal cortex interact with other brain areas to make reward decisions?
The prefrontal cortex (PFC) talks to many brain areas when making reward decisions. It gets information about the value of rewards from the orbitofrontal cortex (OFC) and about bodily feelings from the insula. The PFC also receives signals from the anterior cingulate cortex (ACC) about how much effort is needed. Then the PFC uses this info to choose the best action. It sends commands to motor areas to act and to the reward system (like the striatum) to control how strongly you want the reward. For example, when deciding to work for money, the PFC takes input from the OFC (value of money) and ACC (effort of work), then decides yes or no.
6. Give an example of a gender difference in a decision-making task studied in neuroscience.
In the Iowa Gambling Task, where people choose cards from safe or risky decks, females often learn faster to avoid risky decks that cause big losses. Brain scans show that females have more activity in the orbitofrontal cortex, which processes negative outcomes. Males sometimes continue choosing risky decks even after losses, with more activity in the reward system. Another example: in a trust game, females are more likely to cooperate with a partner they have seen be fair, while males cooperate more if they think it will bring a future reward. These tasks highlight average differences in how genders weigh risk and social information.
7. Compare brain activity during a lottery with known probabilities versus a lottery with unknown probabilities.
During a lottery with known probabilities, the ventromedial prefrontal cortex (vmPFC) and ventral striatum show increased activity, reflecting value calculation. During a lottery with unknown probabilities, the amygdala and dorsal anterior cingulate cortex are more active, indicating conflict and fear. The lateral prefrontal cortex also activates more to try to resolve ambiguity. The known-probability condition feels more predictable, so emotional areas are calmer. The unknown condition triggers a 'ambiguity response' that often leads to avoiding the choice. This comparison shows how the brain distinguishes risk from uncertainty.
8. Give an example of a scenario where the leaky competing accumulator model would predict a different outcome than a simple race model.
Consider two similar options: one starts with a slight initial advantage. In the race model, that advantage persists independently, and the leading accumulator wins if its drift is higher. In the LCA model, mutual inhibition amplifies the small initial difference: the leading accumulator suppresses the other, making the gap larger. So LCA predicts a faster and more exaggerated victory. For example, if two brands are equally liked but one has a small extra cue, LCA says the inhibition will quickly make that cue dominant, while race model would give a closer race. This matches real decisions where small early preferences snowball.
9. How does the orbitofrontal cortex help us compare different options?
The orbitofrontal cortex (OFC) compares options by representing the value of each one in a common 'currency' in the brain. It activates more for higher-value items, like a preferred drink over a less liked one. This allows you to quickly see which option is better. The OFC also keeps track of current needs, such as being thirsty, to adjust values accordingly. It sends these comparison signals to other brain areas that make the final choice. Damage to the OFC makes it hard to choose, as values become jumbled. For example, a person with OFC damage might pick a rotten apple over a fresh one because the value signals are mixed up.
10. How do social decisions differ between male and female brains according to neuroscience?
In social decisions, female brains tend to show more activity in areas involved in empathy and understanding others, like the anterior cingulate cortex. Male brains may rely more on the reward system when competing. For example, when deciding to trust someone, women often consider emotional cues more, while men might focus on potential gains. However, these differences are small and heavily shaped by social learning. Both genders can be cooperative or competitive. Brain scans show that during a trust game, women's amygdala, which processes emotions, is more active, while men's striatum, which processes rewards, lights up more.
11. What is the actor-critic model?
The actor-critic model is a reinforcement learning architecture with two separate parts: an actor that learns a policy (which action to take) and a critic that learns a value function (how good the current state is). The critic computes the TD error, which is then used to update both the critic's value estimate and the actor's policy. This separation allows the actor to directly learn the best action, while the critic provides a feedback signal. It is more efficient than pure value-based methods. The actor-critic model has been linked to different brain areas: the actor to the striatum and the critic to the ventral striatum.
12. How does reward prediction error help explain learning from outcomes?
Reward prediction errors (RPEs) guide learning by telling your brain what to adjust. Positive RPEs make you want to repeat actions that led to better-than-expected outcomes. Negative RPEs make you avoid actions that led to worse-than-expected outcomes. Over time, these errors shrink as your predictions become accurate. This process is called reinforcement learning. For instance, if you touch a hot stove, the unexpected burn (negative RPE) teaches you not to do it again. If you find a new route to work that is faster (positive RPE), you learn to take that route. RPEs are the brain's way of saying 'surprise' to drive learning.