Tania Singer

Tania Singer

University Professor, Director at the Max Planck Institute for Human Cognitive and Brain Sciences, Department Social Neuroscience, Leipzig

What defines humans: self-interest, self-focus, empathy or compassion? While “exogenous” sciences like economy and sociology failed to give by themselves an exhaustive definition of human nature, producing partial concepts like the “homo economicus”, the answer might come from an “inside-out” approach. Taking the neurons as the starting point of a journey through psychology, economics, philosophy and evolutionary anthropology, Tania Singer is one of the few scientists in the world with the right titles to take neurosciences beyond their commonalities. This French-German social neuroscientist is director at the Max Planck Institute for Human Cognitive and Brain Sciences, recipient of the Otto Hahn Medal of the Max Planck Society for best dissertation of the year in 2000, and has gained a rare international cursus honorum in organizations like the Wellcome Department of Imaging Neuroscience and the Institute of Cognitive Neuroscience in London, the University of Zurich, the University La Sapienza in Rome. A drama, film, music and dance student, Tania Singer, who organized a conference on altruism in economic systems with the Dalai Lama, keeps on exploring the possible trainability and cultivation of empathy and compassion, and encourages us to rediscover and instill important but forgotten human capacities.

Breaking the Wall between People. How Social Neuroscience Can Help us to Understand How we are Understood


I was in Marburg when the wall came down, and three hours later I was on my way to Berlin. I remember the East German border guards hugging everyone coming from the West.

Dear Ladies and Gentlemen, I first want to apologise if you cannot understand me properly sometimes, as I have a cold and thus my voice is not fully there.  I am very honoured to speak at this wonderful conference today, especially as a first speaker.

I will speak about breaking the wall between people. I brought you some pictures to actually remember these extraordinary moments when the wall broke down 21 years ago. As you just heard, I was on my way to Berlin and literally had to walk into Berlin as there was such a traffic jam that people could not drive into Berlin. I remember that the guards on the border were actually really hugging every person walking through the border into Berlin. It was a real festivity. There was not only a wall breaking down in terms of bricks being destroyed and the wall falling itself, but also walls in the minds of people and in the mental states of people in these days. So, what I really experienced as extraordinary in these days was that people were not assuming any difference anymore between the “me” and the other, between Ossies and Wessies, which are the terms for East and West Germans in Germany. They were also not really appreciative anymore of ingroup or outgroup dynamics, but there was this ubiquitous sense of empathic joy and contagious emotions of joyfulness spreading everywhere during these days. I think everyone who lived through these days will remember them forever. Today, I want to speak about these emotions, social emotions, but from the perspective of social neuroscience.

So what is social neuroscience? It is a very young and emerging field, which, in contrast to the classical cognitive neurosciences, focuses on the understanding of not just a single brain and the mind in isolation, but the interaction between minds and the communication between brains. The main question is: how do we actually come to understand other minds and other brains even though they are not our own? How do I know what you feel? How do I know what you think, what you believe, what your intentions are?

Let me first introduce you to some important concepts. Although everyone thinks they know what empathy, compassion, or emotion contagion is, if you look more closely at existing definitions, you will notice that there is actually not a single and clear-cut definition of empathy. So let me tell you what we as social neuroscientists and psychologists mean when we talk about emotion contagion, empathy, and compassion.

Emotion Contagion is a precursor of empathy, or perhaps a sister of empathy, so to speak. You can already find emotion contagion very early in life, just after babies are born, when they are still quite little. For example, one baby in the hospital starts crying, and then all the other babies in the hospital start crying as well. This happens a long time before infants actually begin to develop a “self-other” distinction and know that there is another person out there which is separate from the Self.

I just want to show you one example, because I am an identical twin myself; and I love this movie: here you see identical quadruplets lying in the arms of their mother, all of them starting to laugh at the same time; which is a great example of laughing contagion. (to the audience) So you see, you all laugh now; you are already “contagioned”. You see, it is amazing, the quadruplets stop, and then oops they all start laughing simultaneously again. It goes on and on. I will stop here; it is probably a fun time to be the mother of these four babies.

Now, we have also discovered that we do not only have these contagion phenomena on the level of laughing or crying, but, for example, if we measure the pupil of a person when they look at the face of another person who is feeling sad, their pupil will become contaged, because their pupil will follow the changes of the pupil of the other person they are looking at.  This happens—of course—without them being consciously aware of it. We have many more of these examples showing that unconsciously we are actually in resonance with other people’s mental states much more than we are aware of.

Emotion Contagion, however, is not empathy yet. For empathy to emerge, you need to have developed the capacity for the “self-other” distinction. So, you will need to know that you are now experiencing an emotion, which is not initially yours but which you vicariously feel for another person. So, if you are in pain (pointing to someone in the first row), I feel your pain, but I know at the same time that it is not my own pain. I am empathising with your pain.

People always think that empathy is already a good thing, but if you have too much empathy, for example, in caregiving professions, this can develop into a problem. You can experience the phenomenon called “burn out”. So, the crucial thing is to have empathy, but to also be able to transform it into what we call compassion or empathic concern or sympathy, which includes a concern for the other and also the motivation to help and increase the well-being of the other. Compassion is also associated with a feeling of warmth and love. I will tell you in a minute how we can actually start training these emotional states.

Social Neuroscience has further shown us that there are different routes underlying the understanding of other people’s mental states. There is an affective one (empathy) and a cognitive one, which we call “theory of mind” or “cognitive perspective taking”. The latter is not an affective route to the understanding of the mind of others, but a purely cognitive one. Perhaps the best way to understand this distinction is, for example, to imagine psychopaths, who lack empathy—that is a defining criterion of psychopathy. However, they actually still have a preserved theory of mind and cognitive perspective taking ability. As you know, psychopaths are extremely good at manipulating other people, so they know exactly what the needs and beliefs of other people are, but they lack empathy; by consequence this will lead them to be cruel as opposed to being pro-social, to say the least.

Now one important finding is that social neuroscientists were able to identify different networks in the brain which underlie these different routes to social cognition. So, you see on the slide that the areas marked green on this brain refer to the theory of mind or cognitive perspective taking network. The orange ones identify the empathic network, which develops much earlier than the other one, and is a cortical representation of our emotions and feelings.

As an example, I will show you now how we actually measure these social emotions in the very unsocial environment of a fMRI scanner—of an imaging machine. We invented an empathy for pain paradigm in 2004. The question was: can we collect evidence that if we empathise with the pain of another person, we actually share this pain by activating a circuitry in our brain that underlies the processing of our own experience of pain.

With the following movie, I would like to illustrate how such a paradigm is implemented.  This is a scanner, and we invited couples to come to our laboratory. The wife lies in the scanner. This (pointing to the husband sitting next to the scanner) is the husband sitting next to the scanner close to the wife, who is in the scanner, and there are pain electrodes attached to both of them. For one second, you may receive pain via these electrodes—the pain is actually not so bad; people come back to our lab. But, the pain feels bad enough to ensure that you really care about it. We then measure the brain response of the wife while she is really experiencing pain, or just knowing that her partner is in pain. Here, I show you a short movie with bad quality as the scanner is a big magnet and you cannot do fancy movies in it. But, you’ll see in this movie that we managed to create a very social, real-life situation. First, the couples can talk, and when the experiment has begun, flashes on the big screen point to her hand or to the hand of the husband (shows them on the screen facing the subject lying in the scanner). A mirror system allows her to see whether she will receive pain, and then of course she experiences it, or whether he is getting pain and thus she is not experiencing anything, but knows that her husband is instead suffering pain. This allows us to compare the networks in the brain which underlie both processing of your own pain, and empathising with the pain of your partner. This is what we call the shared network (the film comes to the end and shows the words “happy hands”). This is the phase at the end of the experiment. We call that “happy hands”, because after experiencing pain on your hands for quite a while, you are happy that it is over.

On the next slide, you see what we usually find. You see an overlapping activation in the so-called interoceptive cortex; this is a part of our brain which codes for our emotional and bodily states. This brain region is crucial to understand our own emotions, but also to create a model about the emotions of other people. Thus, it is crucial for empathy. So, you could say that we embody the other’s pain—that is, we relive the pain of the other—in terms of activating the brain circuitry which underlies our own feelings.

Now, at that time, we were all very excited and said, “My God, if we are hardwired to resonate with other people all the time, we should also empathise all the time.” But, obviously our society is, rather, challenged by a lack of empathy than by having too much of it. So, we asked ourselves at the time: what are the factors, which actually block these automatic responses? We asked whether we could turn—even in a laboratory setting—these empathic responses to what the Germans call “Schadenfreude”. As lots of nations do not have a word for that, I brought you this movie (shows a funny short movie in which one penguin throws another penguin into the water). So I heard everyone in the audience laughing about this poor penguin.  The feeling you just all displayed is actually Schadenfreude; this is the shameful joy expressed in response to a negative outcome of the action of another person; in this case, laughing about the suffering of this poor penguin.

Now, what we did is to induce these kinds of opposite emotions to empathy in the laboratory. We asked two actors to come in, who would engage in economic game play. These games were inspired by game theoretical models in economy, in which people exchange real money in the laboratory. Now, one player always reciprocated trust that had been extended to him by the scanned subject previously and send money back to him. Another player would always be egoistic, defect, and kept all the money instead of reciprocating trust by sending money back. These games work wonderfully, because they are based on the principle of fairness, which is really inbuilt in us. Thus, violation of fairness induces very strong emotions. We very much like fair players and strongly dislike unfair players.  In a second step, subjects would then get into the scanner again and undergo the empathy for pain paradigm I explained before. But now there would not be your husband sitting next to you but a previously fair or an unfair player sitting on the left or right of you and all three would receive painful stimulation to their hands. Now, you could measure and compare the empathy-related brain responses elicited when you see someone suffering who has been fair or unfair to you before. You can see that the women show empathic responses to fair as well as unfair players.  Men show—this is good news—empathy for the fair player, but no empathy-related brain signal in interoceptive cortex for the unfair player when they observe him being punished.

So we asked ourselves, perhaps males show Schadenfreude or experience something like a reward signal in the brain when they see the unfair player being punished. And indeed, men—but not women—showed an increase in activation in nucleus accumbens, an area in the brain associated with the processing of reward. After scanning, we asked all participants whether they had a desire for revenge, or thought that it was just fair to punish an unfair player. All the men gave very positive responses and indicated a very strong desire for revenge: The stronger the signal in this reward-related area, the more they expressed desire for revenge.

In a second, follow-up study, we asked whether not only perceived fairness but also perceived ingroup or outgroup membership may be a very potent factor for modulation of empathic brain responses. What is the best example of ingroup-outgroup perception in Germany and in other countries? Football. We invited football fans into the scanner, and they could now see an ingroup or an outgroup member from the rival team suffering pain. As this experiment was performed in Zürich, one football fan would be from FC Zürich and the other from FC Basel; they don’t like each other. Importantly, we wanted to know whether these competing motivational signals actually predict real helping. So, in a second session, subjects could choose to either share the pain with the other or to escape the situation by watching a football video. A third option was to just watch the other suffering; this was the sadistic option. The first option was what is called altruistic helping. The good news is that most of the people actually helped each other by sharing pain. And that is really something, because here, you have to endure real pain for helping.

But you see in this graph that participants were willing to help much more the ingroup than the outgroup member. In contrast, if subjects chose to watch the other suffering, this was mostly chosen to see outgroup but not ingroup members suffering. What you also see is that these empathic brain responses in interoceptive cortex are much bigger for the ingroup member than your rival football team member. And remember, subjects did not know the two persons before. They were only told that one is from the rival football team and the other from their own team. What we also see is that the higher the empathic signal, the more you will help. The larger the difference between empathy for ingroup compared to outgroup member, the more you will only help ingroup members, but not outgroup members later. The third finding showed that the higher the Schadenfreude signal—that is, the reward-related signal in the brain—was, the lower the empathic brain response in interoceptive cortex: it seems that these two antagonistic motivational signals are competing. When one is high, the other is low. Whereas empathy-related activation predicts helping, reward-related activation predicts a lack of helping behaviour. These data show us that within the same person, just by tuning a little bit the contextual factors, we can reverse empathy into exactly its opposite motivation. And the motivation signal you measure determines whether you will help or refrain from helping later.

In a next step, we asked ourselves: can we train compassion and positive emotions, like empathic concern? Usually, if you do plasticity research in neuroscience, you ask experts first. What do the brain networks look like in people who train some special ability over thousands of hours? The next question was: Who are the experts in the world who train hours and hours of compassion and positive emotions? These are, for example, Buddhist monks, among others, of course. We cooperate here with the Dalai Lama and some of the monks around him. Some of them train several hours every day over more than thirty years.  They are really like expert pianists; they are expert compassionists. And nowadays we are so lucky, as some of them are even willing to come into our labs, and be tested in a scanner.

Here, you have a picture of Matthieu Ricard in our scanner, who himself was a researcher in Paris when he was young. We asked him not only to go into states of compassion, or disgust, or empathy, but also to do it in a very regulated fashion. So, we asked him: could you—similar to a cooking oven where you can adjust the intensity of the flame—go into a compassionate state but just with 30%, or 60%, or the full-blown intensity? We asked him to regulate the intensity of compassionate states as we are also interested in our ability to regulate our emotions. We also need to learn how to regulate our emotions in everyday life. This was easy for Matthieu Ricard. We also asked him to repeat the same task with real-time fMRI, that is, Biofeedback from his brain; this allowed us to observe online what is going on in certain regions of his brain while he was going in and out of different mental states. In rt-fMRI, we feed back the activity of his brain so that he can learn to up- and down-regulate his own brain activity. We also realised that he was actually perfectly able to perform this task.

Now, of course, we don’t all want to become monks. Thus, we wanted to know if we can use this technique and this knowledge to actually train people like you and me, who have never heard of these mental practices. This is Susanne Leiberg—she had never done real-time fMRI before. You see that after one week of training compassion-enhancing techniques and some biofeedback training, she was able to reproduce this compassionate network; but after the first in-scanner training day, this was not achieved in a very regulated manner. But as you can see (pointing on the bar graphs), after another, second really intense training day, she was already beautifully producing these mental states in three different intensities. So, the good news is you don’t have to sit in a cave your whole life to become a better person. We are now conducting quite a lot of compassion training studies in Leipzig and Berlin; if you want to participate, you are invited in Berlin and in Leipzig, where we will train people over several months. We will not only measure the brain, but subjective well-being, cortisol, health, immune system, and so on, to see how these types of mental training can affect your health and regulate your stress system.

Thank you very much for your compassionate attention this morning.