[MUSIC] So, standard decision series predicts that people are risk averse. And so, people should never buy lotteries, for example. But they do buy lotteries. So, how can we explain this mixed tendencies, of the risk averse in some situations or risk seeking in other situations? Perhaps we can use some modern economic series of prospect theory. So today, we will discuss the major aspects of this theory but unfortunately we do not have time to discuss all details of the prospect theory. So I strongly recommend you to read papers written by, first, Kahneman or really nice books of Daniel Kahneman to get more details of the prospect theory. So prospect theory suggests that expected values are the products of their probability weighting function and a value function. So probability of weight, weighting function is a subjective estimate of probabilities. So you see graphically, on the lower part of this, graph. So the probability weighting function overestimates small probabilities and underestimates large probabilities. It captures our tendencies, our behavioral tendencies to over estimates most probabilities and underestimates large probabilities. At the same time, prospect theory suggests that the values are processed by value function that is highly asymmetrical. So you see here that the value function is much steeper for losses, than for gains. It basically means, that if you lose $50, it triggers much stronger pain than if you win $50. So, importantly, prospect theory suggests that both probabilities and values are subjective. And they are processed during the decision making process, using probability weighting function and the value function. So prospect theory suggests that people apply nonlinear decision weights to objective probabilities. And small probabilities are typically overweighed while high probabilities are typically underweighted. So, prospect theory is very much different from the expected utility theory. And for example, expected utility, theory suggests that we process probabilities in an objective fashion. But prospect theory suggests that instead of objective probabilities, we use decision weight functions. And we overestimate small probabilities and underestimate large probabilities. So, can we find, in our brain, some traces of the overestimation of the small probabilities and of the underestimation of large probabilities? A nice, newer economic studies address this particular question. So, at the beginning of each trial, subjects were exposed to lottery. So, they were exposed to the probability of, of, of an outcome and to the amount of the outcome. So, at the end of the trial, subjects selected between this lottery and another lottery, which expected value was very close to the first lottery. So here, the probabilities of outcomes and the amount of outcomes from manipulated, so neuroeconomics were able to determine brain activity well with why the options was different probabilities and with different outcomes. So in this study, we are able to identify the decision weighting functions for each individuals. On the left side of this graph, we see the results of the previous studies, many previous behavioral studies that indicate that normally, subjects overweights more probabilities than underweight large probabilities. On the right side of this graph, we see the results for this particular neuroeconomic study. You see that majority of subjects, overweight small probabilities and underweight large probabilities. So, let's make a look now, to the activity of the ventral striatum. So, we see the activity of this striatum here. And, you see that this activity is plotted by the blue color. And as we clearly see from the graph, the activity of the ventral striatum overweights small probabilities and underweights large probabilities. So here in the activity of the ventral striatum, we see a specific shape of the activity predicted by the prospect theory. So activity in the striatum during valuation of monetary gambles is nonlinear in probabilities. And this pattern is predicted by prospect theory. Interesting because the degree of nonlinearity of the activity of the ventral striatum on different individuals correlated with the activity of the ventral striatum. So, it looks like we can even find some traces of the patterns predicted by the prospect theory in the activity of the valuation regions in the brain. So prospect theory predicts that people overweights small probabilities and underweight large probabilities. And interestingly, we see the same pattern in activity of the ventral striatum. So, it overweights more probabilities and underweight large probabilities. But prospect theory also predicts that we can be a risk seeking when we face losses, and we can be a risk averse when we face gains. So let's make a choice between two options, you can get 3,000 Euros for sure or you can win 4,000 Euros with probability of 0.8. So which option would you select? Eighty percent of people choose 3,000 Euros in this situation. This makes sense. But now, letâs select between two different options. You can lose 3,000 Euros for sure, or you can lose 4,000 Euros with the probability of 0.8. In this case, 92% of people chose the lottery. So, people are risk seeking in the loss domain. And people are risk averse in the gain domain. These are very similar situations. But in one situation we face gain, and other decision we face loss. And in this loss situation, people start to be risk seeking. So the prospect theory effectively explain why people like risk-seeking ends or loss domain and risk-aversion gain domain. I will not go into details. I will just mention that the value function is really asymmetrical. So you see that the value function is steeper for losses than for gains. So the loss of 3,000 Euros triggers a much stronger pain than the same, the same gain of 3,000 Euros. Can we find this asymmetry at the behavioral level? So this our tendency to be risk seeking in the loss domain and risk averse in the gain domain can be nicely illustrated by the Asian disease problem. So you remember this example. So imagine that you face an unusual disorder that can kill 600 people. So we can select between two programs, A and B to treat the disorder. If Program A is adopted, 200 people will be saved. If Program B is adopted, there is two-third probability that no people will be saved and one-third probability that 600 people will be saved. So in this case, majority of people select Program A. So, they avoid a risk and select the safe option. But now, if you have to select between two other programs, A and B. If Program A is adopted, 400 people will die. If Program B is adopted, there is one-third probability that nobody will die and two-thirds probability that 600 people will die. In this case, subjects robustly select the program B, and they take risk. So when we make a closer look to these options, we see that Programs A in both cases are identical. So to say that 200 people out of 600 will be saved is exactly the same as to say that 400 people of 600 will die. But in first case, this option is framed as a gain. And the second case, this option is framed as a loss. And it dramatically changes the risk-taking behavior of subject. So in first case, subjects take the safe option. And the second case, subjects select the risk options. Does framing of those option, affect also the neuronal processing of the options. So, indeed, the framing of the options severely modulates the processing of the options during the financial decisions. So Benedetto De Martino showed this in a very interesting study published in science. So here in this study subjects participated in two conditions. In Condition 1, they initially received 50 pounds and next they faced a decision, so they can choose between the sure option to keep 20 pounds out of 50 or to take a gamble. So, there's a certain probability that subjects will keep all money or they can lose all money. And the probability of ,uh, losing all money was indicated on the screen by the red color. And in Condition 2, subject also initially obtained 50 pounds and next decided whether to take the sure option of losing 30 pounds of the 50 or to take a gamble where they can, once again, can keep all money or lose all money. And the probability of losing money was indicated by the red color on the screen. If we will make a closer look to these two conditions, we will find that these are two similar conditions. Simply the sure option was framed as a game in the Condition 1 or as a loss in the Condition 2. But these, these are the same options. To say that you will keep 20 pounds out of 50 is the same as to say that you will lose 30 pounds out of 50. But whether the framing of the options change the decisions of subjects to take risks or to take, sure option, and does it modulate also the brain processing of the options. In fact, the results of this study show that if their sure option was framed as a loss, subjects demonstrated the higher percentage of their decisions to gamble as compared to their condition when the sure option was framed as a gain. So framing dramatically modulated subject's risk-taking behavior. If we will make a look inside of the brains of the subjects, we will see that the reason brain activity that reflects the framing effect. So framing effects was specifically associated with activity of the amygdala. So if you will make [UNKNOWN] a graph, to the right side of the graph, you will see that amygdala was particularly active during the loss frame, when subject's decided to gamble. And the amygdala was deactivated when subject's selected the sure option. But this pattern was opposite during the game frame, so amygdala was more active in subject's selecting the sure option. And it was deactivated when subjects selected the gamble. So the activity in these brain regions that processes the anticipated costs is dramatically modulated by the framing of the options. So it gives an additional support to the prospect theory, suggesting that the framing of the options, the reference point of the options, modulates the processing of the options and modulates our risk-taking behavior. So we can summarize the brain activity during the decisions on the risk in the modified version of the graph suggested by Peter Mohr. So, we can suggest that the insular cortex is involved in emotional risk processing. Dorsal anterior cingulate cortex is involved in the cognitive risk process. Ventral striatum contributes to the approach to risk. And the amygdala perhaps modulates the risk-taking behavior by implementing framing effects. So, all together, this information affects the dorsolateral prefrontal cortex that finalizes the decisions. So, we make the first steps in our understanding of decisions under risk. So, right now, we understand some key brain regions that are important for risk-taking decisions. So based on our lecture, we can add some additional functions to the key brain regions involved in the decision making. So nucleus accumbens and ventral striatum will just participate in the calculation of the game magnitude in the expected, anticipated gaming magnitude, but also contributes to their approach to risk. So, amygdala not just calculate the expected costs, but also participates in framing effects, modulating our decisions in risk taking. Insular cortex not just emits an arousal emotional signals that contributes to decisions, but also participates in the risk avoidance behavior. So altogether, these brain regions orchestrate our decisions. [MUSIC]