Introducing What We Value: The Neuroscience of Choice and Change

in their security infrastructure. Although Jenny sometimes uses physical force, lock picks, or computer code, her main tools come from psychology. She can read a person or a situation and predict how someone or a group of people will respond to her depending on what she does. Then she can create a situation that moves her toward particular goals and outcomes. This is just what she did when she was hired

to break into a bank in Germany. Her mission was to enter the bank during business hours, get past security, and locate a particular office where she was to plug a USB drive the company had given her into a computer. A program preloaded on the drive would then install itself on the computer, letting the company know that Jenny had successfully penetrated their security. The morning of the big job,

Jenny ready to costume in props. She wrapped her hand and wrist in a bandage, figuring that people might be more likely to hold doors open for her if she appeared to be injured. She brought a big file box full of papers to occupy her hands, further increasing the odds that people might hold doors for her. Thus prepared, she went into the bank, walked into the grand lobby furnished with leather sofas and approached huge doors blocking access to the employee's only portion of the bank. Those doors

presented Jenny's first of many obstacles. They were operated by fingerprint scanners, and of course, Jenny's fingerprint wasn't in the bank system. She wasn't an employee. She was pretending to be one. But she walked over to the fingerprint scanner and put her finger on the pad. Anyway, it beaped no luck. She hadn't expected the censors to let her in, but as a penetration tester performing a security audit, it was still part of her job to check. At this point,

Jenny had choices. She could ask the security guard on duty in the lobby to let her in, but what incentive would he have to do that. It was his job to keep strangers out. So instead she did the obvious thing. She swore really, really loudly. Just as Jenny had planned, the security guard came over to see what was happening. You don't have to work on the lock, Jenny leader explained, work on the person behind the security. It doesn't matter what they put in place. If someone's

got access, then I can access them. And then were down to me versus the person. When the guard approached, Jenny said, impatiently, this isn't working. It was working yesterday. The secure part guards suggested she tried the fingerprint censor again. She made a big show of being annoyed, cursing once more and awkwardly balancing her big box of papers on her bandaged hand. She tried again. The machine beeped again.

Maybe she wasn't pressing hard enough, the guard ventured. She grudgingly placed her finger on the sensor again, at which point the guard took her hand and tried to help her press her finger onto the machine. Jenny yelped in apparent pain and swore loudly. Once again. She made a point of dropping the filebox, which scattered papers everywhere, and made a big show of trying to pick them up, all while swearing away. Now she had drawn attention to herself.

People in the lobby were looking, for God's sake. Go in, the guard said, and beeped her through the doors. Thank you, Dunkashane, Jenny replied, and she was on her way down the hallway to the designated office, where she inserted the USB key she had been given. What happened here? Making a big commotion like Jenny did might not work for every person in every situation. For one thing, some people might be more influenced by being buttered up or feeling like

they're doing someone a favor. For another, the same actions can be interpreted as more or less threatening depending on the characteristics of the person doing them in the environment therein. But in this case, Jenny felt confident that causing a scene would help her break into the bank because she knew that in Germany people generally feel highly embarrassed by a scene, and based on her gender and the way she looks, she wasn't likely to be perceived as a

physical threat or a computer hacker. Under these conditions, making the commotion the most prominent thing in the guard's mind would tip the scales of his decision making. She figured that the guard would perceive her as low risk and would rather buzz her in than deal with the discomfort and disturbance of a spectacle, and she was right. Maybe you feel tempted to harshly judge the guard for letting Jenny in. The bank's rules no doubt emphasized that he

should not let strangers through the door. If Jenny had been a malicious hacker. The USB drive she plugged in could have uploaded a computer virus that stole customer's personal information in life savings, or taken down important parts of the bank's infrastructure. But the truth is that many of us would do the same thing in that situation. We want to see ourselves as helpful, kind people, and much of the time other people aren't trying to deceive us.

If Jenny had been an injured employee simply trying to get into her office to do her job, the guard's actions would have been helpful to the bank, not harmful, for better or worse. Jenny's understanding of these decisions making mechanics, the sometimes unconscious, near instant calculus we perform when choosing between options and how they can be influenced, enabled her

to break into the bank. Recent advances in neuroscience allow us to understand more about the underlying systems in the brain that allowed her to do this and that might allow others to resist, including one that scientists call the value system. As we begin to explore the value system, which brings together many different types of information to guide our decisions, it may be helpful to imagine the thought process of the security guard when he was confronted with Jenny.

His brain's value system would compute the value of different possible choices, allow the swearing woman to continue making a scene or buzz her in, select the one with the highest value, buzz Jenny in, and then track how rewarding the choice is. Now the scene is quiet, and I feel good that I helped an injured person. Much of the time, this value calculation happens quickly and seamlessly. Importantly, as Jenny understood so well, its outcome depends on what

our brains pay attention to in the moment. In that split second, The value calculation can be shaped by any number of factors, our own goals, how we feel, our identities, what we think others will think and feel, other people's actions, cultural norms and expectations, are social status, and much more. Jenny used her implicit understanding of the value calculation to gain access to the bank, as she had been hired to do.

and resist future attempts to hijack it. In that way, where the bank might provide more opportunities for security guards to get to know the other bank employees so that it would be clear when a new employee joined, as well as who was a stranger. Of course, to think of all these options requires thinking along a number of different dimensions, checking in with the bank's bigger picture goals, the security guard's goals, and where there might be room

for greater possibility in the overlap. So what options or combinations of options would make it more likely that the security guard chooses differently next time. How might we become more aware of when our value calculations are being shaped by people who don't have our best interests at heart. To figure this out, it's helpful to know what's going on in our brains when we're confronted with choices kool

aid or peppermint tea. One remarkable power of the value system is that it allows our brains to take complicated, messy, real world decisions and boil them down into comparable quantities. Thus simplified, our brains are able to choose between options, often almost instantaneously, and with a fair amount of internal consistency. I find it useful to think of the value calculation

as a hidden game of would you rather? You're probably familiar with this common icebreaker, in which one player offers two ideally silly choices and other players say which they would prefer. Would you rather have a cat's tongue or roller skates for hands? Would you rather be able to speak every language or have the most beautiful singing voice on earth? Would you rather live alone on a desert island with all the movies and books ever made, or

with one other person you choose but no media. When you think about it, it is borderline magical that you can answer would you rather questions comparing alternatives that differ in so many ways from low stake situations like playing the game? Would you rather at a party? To the decisions that determine our actual behaviors each day. Our value systems help guide us to our choices. But how does the brain do this? For a long time, no one knew the answer. Did the brain have different systems that

each monitored different dimensions of a choice? How much sugar or salt is in each food we're choosing between how hot or cold is each food, how green is each food?

Or were there different brain systems that would handle choices in different domains, a brain system that decides what kind of foods we want to eat, a different brain system that keeps track of how much fun each of our potential dinner companions is, and a third that handles the financial decision about whether we can afford to eat out the foundations of how we currently think about the neural underpinnings of this kind of decision making were laid in

the nineteen fifties by researchers who mapped a set of brain regions that tracked simpler types of rewards and that guided animals behavior to maximize those rewards, even if choosing the reward was objectively bad for the animal's well being

in the longer term. James Olds and Peter Milner, scientists at McGill University in Canada, discovered that when given the chance, rats repeatedly pressed a lever that triggered electrodes that stimulated particular parts of their tiny rat brains that made them feel good. In other words, the rats found it rewards to stimulate these parts of their brains, and scientists at the time began to think of the regions being stimulated as the reward system. It turned out that stimulating this

reward system had powerful consequences for the rats behavior. For example, when rats were given the chance to press a lever that stimulated these reward regions, they would even forego food they needed to stay alive. And it wasn't just rats. Scientists soon found parallel reward systems in rhesus monkeys and eventually came to learn that all mammals had similar infrastructure

in their brains across species. When scientists stimulated neurons, the cells that transmit messages through the nervous system, deep in the brain in a region called the straatum and in certain regions in the front of the brain frontal cortex, the animals seemed to experience reward, as evinced by their tendency to seek out the stimulus over and over. Like humans. Some animals also displayed facial expression or made sounds showing

their pleasure. But although it was clear early on that stimulating specific reward regions caused animals to want things, it took several decades for scientists to understand how this translated into more complex decision making in humans. Why would a system that tracks how much food you want or how much you want to press a lever have anything to do with whom you want to be president or which

movie you want to see? Could a single brain system really handle comparing choices that take place at various points in time now versus later, concrete rewards like which snack to eat, and abstract questions about society and morality. A series of important insights about how brain systems make more complicated calculations about the relative values of a wider range of goods and ideas came in the mid two thousands,

one of them through offering kool aid to monkeys. Camillo Padua Sciopa and John Hasad were researchers at Harvard Meadow School studying decision making and economic choices when they wondered whether the reward system discovered in rats and other animals could also help monkeys make somewhat more complicated decisions, and if so, how, On the one hand, they reasoned, it was possible that regions of the reward system might respond

to objective properties of different potential rewards, like the amount of sugar in a juice. This might be the case if a particular nutrient like sugar or fiber had been important to the survival of the species in the evolutionary past, and a physical feature of the food like color or firmness, was a good indicator of how much this nutrient was present in it. If so, there should be a tight correspondence between certain biological or chemical properties of foods and

the response of the reward system. On the other hand, what if the reward system could take a wider range of things into account to make more subjective calculations. Did it explain why a monkey might have different food preferences at different times, or even predict what a monkey was in the mood for. In their experiments, Camillo and John would present a monkey, let's call him Gizmo, with a series of choices while recording the activity from neurons in

his brain. Would Gizmo like one drop of lemon cool aid or two drops of peppermint tea, five drops of milk or one drop of grape juice. Gizmo would look left or right to indicate his decision. After many of these choices, the researchers could calculate how much value Gizmo assigned to each drink relative to the other drinks what

neuroscientists now call its subjective value. We say the value is subjective because it turned out not to be fixed to some objective quality like the density or overall amount of sugar present in each liquid, the exact temperature, the quantity of liquid, and so on. The scientists found that Gizmo and other monkeys generally preferred to have more to drink if possible, but like humans, they liked some drinks, specifically lemon, kool aid and grape juice, more than others.

Depending on the offer, the monkeys would sometimes choose a smaller amount of their preferred drink over more of one they liked less. By offering the monkeys the drinks in different ratios, Camillo and John could arrive at a mathematical

description of the monkey's preferences in each session. For example, if Gizmo was really in the mood for grape juice in one session and chose one drop of it over up to three drops of water, then Camillo and John could say that one drop of grape juice was worth three points, while one drop of water was worth one. While hanging out with the monkeys. Camillo and John also found that subjective value was influenced by the context within

which the decisions were made. The monkey's drink preferences, that is, the relative value of one drink to another, varied from day to day, even for the same monkey. Imagine that you yourself are at someone's house and they offer you a cup of coffee or a cup of lemon ginger herbal tea. Your decision depends partly on stable preferences you have. You typically like coffee more than lemon ginger tea, but also on the situation it's late and you worry that

caffeine might make it harder to sleep. Similarly, on Tuesday, Gizmo might prefer grape juice to water three to one, but on Friday he might feel less strongly because he's already had plenty of fruit and may prefer the grape juice to the water only two to one. This is what subjective value means. Different aspects of a situation change how much something is worth to someone at a given

time in a given situation. When Camillo and John looked at the data from the monkey's brains, they discovered that neurons in the front and center. Specifically, a region called the orbit offrontal cortex fired in response to each monkey's overall subjective preferences for the juices. The activity in these neurons correlated with the overall ratios Camillo and John had

calculated based on the monkey's decisions. When the monkey preferred one option three times as much, these neurons fired correspondingly. More interestingly, the firing didn't seem to depend on objective aspects of the choice, such as the specific ingredients of

the drink. If, as you might think, there were neurons tracking the amount of sugar, which side of the screen showed the offer if neurons here kept track of what motion the monkey needed to perform to get the juice, or how many drops of juice were offered in total, if more is always better. Instead, the neurons tracked the overall subjective value, and this subjective value was tied to

the choices the monkeys made. Just by seeing what was happening within Gizmo's orbitofrontal cortex when he was shown the different options, Camillo and John could predict which choice Gizmo might make with re markable accuracy. In other words, the monkey's brains were computing subjective values for each option on a common scale that allowed them to make decisions and compare apple juice and orange juice. But what about humans.

Around the same time that studies on monkeys revealed that their brains responded to subjective rather than objective value, scientists began to find similar responses in the human brain. In the span of a decade or so in the early two thousands, scientists ran hundreds of experiments mapping what happened in people's brains when they made choices based on these

subjective preferences. In one early study, the neuroscientist Hilca Plasmin and her colleagues at Caltech found that when they measured how much human volunteers were willing to pay to eat different snacks, they showed similar activity in brain regions analogous to those the monkeys used to choose between lemonade and

grape juice. The team showed pictures of salty and sweet junk foods like chips and candy bars to hungry humans while scanning their brains using functional magnetic resonance imaging fMRI. This type of brain scan let scientists see when different parts of the brain are active, and then connect this

activation to different psychological processes and behaviors. The volunteers in Hilka study were told they had a specific budget and were asked how much they would be willing to pay for different food items shown as images on a screen in the fMRI scanner. As in the case of Camillo and John's monkeys, brain activity increased the most within a similar region in humans, the venturemedial prefrontal cortex for the

items they rated as most valuable. In other words, there was more activity in response to snacks they were willing to pay three dollars for than snacks they were willing to pay one dollar for or didn't want to buy at all. People's brains kept track of the subjective value to them personally of different foods and chose accordingly. This was a breakthrough, but in daily life we often have to choose between options that are harder to compare than

two kinds of snack foods. Could the same brain regions that decide if you'd rather drink coffee or tea also compare things that are rewarding in very different ways. For example, would you rather drink grape juice or go see a movie? Or did such choices go beyond their role in decision making. To probe this question, a team of scientists at Caltech and Trinity College Dublin designed an experiment that was, in essence,

a variant of the would you rather dilemma. The research team gave volunteers in an fMRI scanner a twelve dollars budget that they could use to bid on different types of goods, from sweet and salty snacks to DVDs, Caltech memorabilia, and monetary gambles. They found that an overlapping area of the venturemedial prefrontal cortex tracked how much people were willing to pay not only for different foods, but also for

products like college memorabilia and DVDs. Around the same time, other groups of scientists were also finding that activity in the human medial prefrontal cortex and other regions like the ventral stratum tracked people's willingness to pay different prices for a range of consumer goods. These findings suggested that a common system was keeping track of the value of a

wide range of different kinds of choices. As this body of research grew, this group of brain regions, including the ventral stratum and venturemedial prefrontal cortex came to be known as the value system. By twenty ten, activity in the value system had been shown to track not only people's decisions about how much money they would pay for different goods,

but other kinds of financial choices as well. For example, would you prefer to take one hundred percent chance of winning ten dollars or a fifty percent chance of winning twenty dollars. Would you rather have ten dollars now or twenty dollars in six months? All of these types of choices seemed to work through a similar mechanism in which the value system identified and assessed the subjective valuevalue of

different choices, compared them, and then acted. By twenty eleven, researchers could even predict, based on activity observed in volunteers value systems while they were looking at different goods what they would later choose, even when they weren't asked to make any choices during the initial scan. In other words, the value system seems to track the subjective value of different things regardless of whether the person is consciously trying

to make a decision about them. When we're in line at the grocery store, our value systems are weighing the value of the candy bars by the register and absorbing information from the news headlines and magazine covers. When we're scrolling through social media passively consuming ads, our value systems are still registering the inputs, even if we aren't actively

paying attention to them. A decade later, it is now more widely accepted that our brains can make calculations using a common value scale that allows us to compare things that aren't inherently comparable. Could probably easily decide if you'd rather snuggle a puppy or have five dollars right now. This is because your value system converts each option onto

a common scale and makes the comparison. Likewise, when Jenny yelled for the security guard, he quickly made the decision to try to help her use the fingerprint scanner rather than demand ID, and eventually to let her through the doors, rather than calling for backup, asking her to leave, or

asking her on a date. Predicting and learning, it's tempting to think there are good choices and bad choices, but the truth is that these are moving targets, and the value system is dynamic, constantly weighing competing interests, and the context. This means that the choices we make depend on what options we imagine we are choosing between and what dimensions of the choice we focus on. If your kid has never met a male nurse, it might constrain the career

options he imagines choosing to suit his empathic personality. Moreover, the subjective value we assigned to a given choice option can change depending on a variety of factors related to our past experiences, our current situation, and our future goals. If your kid believes you'd like him to get a job that helps a lot of people, that dimension might

weigh heavily as he considers career options. Likewise, if his crush gushes about Austin, Texas, that might cause your son to give weight to the geographical flexibility of different job options. This is one neural foundation of what social psychologists call the power of the situation. Our decisions depend on our current context, which gives certain inputs to the calculation more weight. Let's say you're deciding whether you'd rather eat a salad

or chocolate cake. If your brain only followed objective rules, you might only care about how much the food filled your stomach, or how many calories it offered, which could translate directly to keeping you alive in earlier moments of human evolution. But that's not how it works, as you have no doubt experienced. When you decide what to eat, you might focus on any number of things. How does the food taste, how will you feel after eating it? What is your date eating? Did you just get a

bad doctor's report? Do you have a great metabolism? Is it someone's birthday? How much does each cost? Did you just run a marathon? Are you in a bad mood. Your brain does this quickly and may not even take into consideration all of these dimensions, limiting what it weighs in any given choice. Based on what factors it does weigh, your brain can compute subjective values for salad and cake on a common scale, then choose the higher value alternative.

Once you've made the choice, your value system transmits it to the parts of your brain that help you act on the decision, like reaching out and grabbing your chosen food and eating it. Importantly, your brain's value system then keeps track of how good the decision's outcome was relative to what you thought would happen. In other words, how accurately it guessed how rewarding the choice would be at tracks.

Not only your prediction that cake looks delicious, I remember how much fun I had at birthday parties as a kid, but the prediction error, or the discrepancy between your prediction and the actual outcome. If the choice ends up being more rewarding than you expected, that cake was delicious, totally worth it, your brain generates what neuroscientists call a positive prediction error, seen as an increase in activation within the

value system after the choice. Conversely, if the choice ends up being worse than you thought that cake made me feel gross, your brain generates a negative prediction error, seen as a decrease in activation within the value system after the choice. These prediction errors help you learn for the future, updating how your brain makes the value calculation over time. In some there are three basic stages to what neuroscientists

call value based decision making. First, our brains determine what options they are choosing between, assign a subjective value to each one, and identify the option with the highest value.

longer term well being. This means that there isn't one single right answer, and what our brains perceive to be the highest value option right now might change if considered from other perspectives, for example, when thinking about career goals versus wanting to be a good friend. Finally, when we've made the choice, our brains track how rewarding it turns out to be so they can update how they make

the calculation next time. This means that we often overweight the outcomes of our choices rather than improving our process. This highlight it's at least three places where we can intervene. We can imagine more or different possibilities, consider the existing possibilities from different angles, or pay attention to different aspects of the outcome. We can think again of our security guard.

If as the guard, you buzz in a bumbling person making a scene and it yields a better social reward than you had expected, the person gives you a big, grateful smile and tells you how much she appreciates you. Your brain will generate a positive prediction error. That data will be stored, and in the future you will be

more likely to let in the next bumbling stranger. But if something bad happens and the outcome is worse than you anticipated, the bumbling person turns out to be a security tester, and your colleagues are annoyed with you because now you all have to sit through extra training sessions. Your value system stores that too. Next time, you might think twice before letting in a stranger. But of course,

no one scanned the brain of the security guard. Most of the studies we've explored so far have taken place in highly controlled lab settings, So what actually happens outside the lab in the real world. Can we link activity in the value system to what people do in their day to day lives outside the brain scanner? A great

day for science. I was a budding neuroscientist in the early two thousands when our understanding of the value system first started to take shape, and I was interested in whether brain imaging could give us insight into health decision making. I wanted to help people make choices that would help them live healthier, happier lives. But I also knew that

these choices could be very difficult to make. It's hard to change, and even when we are motivated to change, we don't always take time to figure out why we do what we do in the first place, or know why. Some ways of thinking are helpful in achieving our goals and some aren't. I was thinking about how to make

better health coaching and messaging campaigns. I was also thinking about how we might talk with our family members and friends, roommates, and colleagues to help motivate them to make healthy changes, and even how we might talk to ourselves to make decisions that are more in line with our goals. I wondered if brain imaging could give us a new window

into this decision making. Maybe looking at brain responses to health campaigns and health coaching messages could help us understand what made people change and what would make it easier to work with rather than against our desires. If that were true, maybe it could help us design and select better messaging. I decided to apply to graduate school to work with Matt Lieberman at UCLA. Matt's lab was full of scientists studying how people understood themselves and others and

how they made important decisions. Along with a group of other young faculty, Matt had recently ignited a new field of study that combined social psychology with cognitive neuroscience. Whereas neuroscientists before had focused on topics ranging from vision and memory to reward and motor actions, many fewer had delved into topics that were more at the core of being human, like where our sense of self comes from, how we understand what others think and feel, and how imagination works.

At the time, it felt like a long shot to connect what happened in a neuroimaging lab to real world behavior changes outside the lab, But it also felt fundamental. What good was all this research if it couldn't help us in real life. Luckily, during the years I was in graduate school, we did start to see a connection, a pattern indicating that activity in the brain's value system could reveal who is more likely to change their behaviors in response to messaging and what kinds of messages were

most likely to elicit this kind of activity. The first work we did in this space focused on sunscreen use in Los Angeles, where to sunny almost every day. I had a daily reminder that despite how great the sun feels warming your skin. Sunburns and other invisible damage from UV rays can cause skin cancer. Matt and I designed an fMRI experiment scanning the brains of volunteers while exposing them to messages about the importance of wearing sunscreen every day.

The finding was simple. The more activation we saw in a person's value system, specifically the venturemedial prefrontal cortex, in response to the messages, the more likely they were to increase their sunscreen use in the next week. It suggested that the value system helps guide not only simple choices that people make in the lab, but also real world, consequential behavior change outside the lab. When I saw the data, I started jumping up and down on the lab couch.

My friend and then officemate, Sylvia, claims that I screamed, this is a great day for science. I deny it. While I don't know if non scientists would be this excited about a data plot, it felt like a big moment. And although this initial study relied on what people told us about their sunscreen use, later studies in the lab I now run at the University of Pennsylvania and others

have shown similar results. In people being coached on other health habits where behavior change has been measured more objectively. When sedentary adults were exposed to messages encouraging them to get more exercise, the activity in their value system corresponded with how much exercise they got, later measured objectively using

wrist worn activity trackers. Similarly, smokers whose value systems responded more strongly to messages encouraging them to quit smoking were significantly more likely to reduce their smoking over the following month, which we confirmed using a device that measures how much

carbon monoxide smokers have in their lungs. In fact, our ability to predict how much people would reduce their smoking was twice as good when we included information from both brain responses and self report surveys as when we included only information from the surveys. This suggests that there was useful information that the value system captured that was not fully captured by surveys alone. Figuring out why this is the case and how far in the future we can

predict is a current frontier. Another current frontier involves understanding when and how people make the kind of deliberate decisions that will mostly focus on in this book, compared with other kinds of decisions. For example, it is increasingly clear that a lot of what humans do is guided by habitual routines, which is not the kind of choice we'll be discussing. But some of these habits start with deliberate choices, which is our focus. To illustrate this distinction, let's consider

my walk to work. When I first moved to Philadelphia, I wanted to walk to work rather than drive or take the subway, so i'd get outside more. That was an active choice. I used my phone's map to find the shortest route, and following my phone's map was also an active choice. Over time, as I repeated this walking route over and over, it became a habit, something I could do and did on autopilot, whereas other options like driving, taking the trolley, or even walking a different route require

more conscious thought. In other words, when repeated over and over, what start as goal directed, value based decisions become routine and get handed over to another brain system that supports the kind of automatic pilot I was on. This book explores what happens in the first type of decisions, when we are more deliberately choosing and setting in motion paths

that may or may not eventually become habits. Find what we value at Pushkin, dot Fm, Slash Audiobooks, or wherever you get your audiobooks,