Decision making is of course a huge area. Here, we will focus on research studies to gain an understanding of the factors that influence decision-making.
Value and Expected Value
At the simplest level, decision-making comes down to selecting between a series of two or more options. There is a long-standing notion that we do this through value (or in an economics framework utility). Simply put, in a given situation we select the highest value option available to us. It is important to note that value does not have to reflect financial cost but instead reflects the internal worth of an option. Extending this a bit further, typically we chose the option with the highest expected value. An expected value is equal to the value of an option times the probability of getting that options.
1. This paper by Platt and Glimcher 1999 examines how expected value is encoded within the brain. Summarize the results of the study. What does it tell you about how expected value is encoded in the brain?
2. In humans, fMRI can be used to show similar findings. For example, Knutson et al. 2005 examined this very issue. Summarize the results of this study? How are these results similar/different from those reported by Platt and Glimcher? Why?
Exploration versus Exploitation
So, the decision-making story is over. Simply find out which option has the highest expected value and choose it? Right? Not so easily done. Sometimes we know that people deliberately take lower value or unknown value options. Why? One reason is exploration. To be an effective decision-maker one needs to know the correct expected value of all of the options available to oneself. For example, you decide to order in pizza - which restaurant do you choose? The one you like the most, the one with the highest expected value. However, what is the restaurant that you have never ordered from has better pizza? This is in essence why one needs to explore sometimes as opposed to exploit (choose the highest expected value).
Read this review paper by Cohen et al. 2007.
3. Ensure you can define the explore - exploit dilemma. DO NOT get caught up in the modelling of this behaviour (i.e., the math).
4. What is the difference between a stationary and a non-stationary environment? Whys is exploration critical in a non-stationary environment?
5. How does uncertainty factor into the explore - exploit dilemma?
6. This paper by Daw et al. 2006 is referenced in the review. What are some of the potential neural substrates that govern the explore - exploit dilemma?
Emotion
Now the story is clearly over. Always choose the highest value option but sometimes explore, especially if you are in a non-stationary environment. One small problems. We are humans, and being humans, we have emotions.
7. Review this paper by Sanfey et al. 2003. What is the ultimatum game? How do these results highlight the role of emotion in decision-making? What neural regions support / govern decision-process, at least in terms of this paper?
8. Some interesting work by the Cohen group at Princeton is in line with Sanfey et al.'s findings but in a more complex decision situation. Review this paper by Greene et al. 2001. What is the key finding and how does it align with Sanfey's work? Another way to put this would be what differences were observed in brain function when appropriate versus inappropriate decisions were made (see Figure 3). What does this mean?
Risky Decisions
One last thing to consider is risky decisions making. This is a huge area so we will keep it simple for the purposes of NEURO 500.
9. Consider this paper by Sohrabi et al. 2015. What are "risky" decisions and what brain areas were activated during risky decision making in this study? What does this result potentially mean?
Exam Question
9. How do we make decisions? Take into account the notions of expected value and exploration versus exploitation (and their neural basis). Also review how emotion and risk might impact the human decision-making process.
Value and Expected Value
At the simplest level, decision-making comes down to selecting between a series of two or more options. There is a long-standing notion that we do this through value (or in an economics framework utility). Simply put, in a given situation we select the highest value option available to us. It is important to note that value does not have to reflect financial cost but instead reflects the internal worth of an option. Extending this a bit further, typically we chose the option with the highest expected value. An expected value is equal to the value of an option times the probability of getting that options.
1. This paper by Platt and Glimcher 1999 examines how expected value is encoded within the brain. Summarize the results of the study. What does it tell you about how expected value is encoded in the brain?
2. In humans, fMRI can be used to show similar findings. For example, Knutson et al. 2005 examined this very issue. Summarize the results of this study? How are these results similar/different from those reported by Platt and Glimcher? Why?
Exploration versus Exploitation
So, the decision-making story is over. Simply find out which option has the highest expected value and choose it? Right? Not so easily done. Sometimes we know that people deliberately take lower value or unknown value options. Why? One reason is exploration. To be an effective decision-maker one needs to know the correct expected value of all of the options available to oneself. For example, you decide to order in pizza - which restaurant do you choose? The one you like the most, the one with the highest expected value. However, what is the restaurant that you have never ordered from has better pizza? This is in essence why one needs to explore sometimes as opposed to exploit (choose the highest expected value).
Read this review paper by Cohen et al. 2007.
3. Ensure you can define the explore - exploit dilemma. DO NOT get caught up in the modelling of this behaviour (i.e., the math).
4. What is the difference between a stationary and a non-stationary environment? Whys is exploration critical in a non-stationary environment?
5. How does uncertainty factor into the explore - exploit dilemma?
6. This paper by Daw et al. 2006 is referenced in the review. What are some of the potential neural substrates that govern the explore - exploit dilemma?
Emotion
Now the story is clearly over. Always choose the highest value option but sometimes explore, especially if you are in a non-stationary environment. One small problems. We are humans, and being humans, we have emotions.
7. Review this paper by Sanfey et al. 2003. What is the ultimatum game? How do these results highlight the role of emotion in decision-making? What neural regions support / govern decision-process, at least in terms of this paper?
8. Some interesting work by the Cohen group at Princeton is in line with Sanfey et al.'s findings but in a more complex decision situation. Review this paper by Greene et al. 2001. What is the key finding and how does it align with Sanfey's work? Another way to put this would be what differences were observed in brain function when appropriate versus inappropriate decisions were made (see Figure 3). What does this mean?
Risky Decisions
One last thing to consider is risky decisions making. This is a huge area so we will keep it simple for the purposes of NEURO 500.
9. Consider this paper by Sohrabi et al. 2015. What are "risky" decisions and what brain areas were activated during risky decision making in this study? What does this result potentially mean?
Exam Question
9. How do we make decisions? Take into account the notions of expected value and exploration versus exploitation (and their neural basis). Also review how emotion and risk might impact the human decision-making process.