The Neural Basis of Empathy

Raising empathetic children requires more than telling children to “be kind.” Organizations that offer guidance about social and emotional health in young children advise educators to model respect and empathy, and to actively teach skills using literature, planned activities, and behavioral cues.15, 21 These practices are based on the concept that empathy is the product of minds open to the ideas and emotions of others—a cognitive choice. Yet, because empathy is an emotional practice, it is also driven by complex biological and environmental influences, which are helpful to understand. This section of the website offers some common definitions of empathy, explains how the brain processes empathy, and identifies the regions of the brain that are at work when we share another person’s pain.  For those who want to learn more about specific regions of the brain, Cold Spring Harbor Laboratory offers a useful tool.8

Many preschool curriculums take the approach that empathy is a curriculum skill that can be taught and mastered. We can certainly encourage children to make empathetic and open-minded choices when relating to their peers. But the best strategies for developing empathy rely on creating environments that encourage the brain to do the work on its own. From an early age, children display a natural tendency toward kindness, which is as rooted in biology as the tendency toward self-preservation. When children consistently lack empathy, their behavior is often “disordered,” suggesting that neurological influences may be at work. It is not necessary (or even practical) to perform a brain scan on every child who has disordered behavior.  Even functional MRI (fMRI) technology, which indicates the regions of the brain that activate during tasks and emotions, will not show all the complex interactions that occur in a child’s brain.10 

A more realistic strategy is to studyneuroplasticity.  Most commonly, the empathetic process is described by educators as a product of cognitive control: we decide we want to care about another person’s feelings, and so we do. Although it is true that researchers have identified different physical reasons for cognitive control to become disordered that is only one part of the story. In the explanations below, boldface type distills the key insight from the technical descriptions.

Common Definitions of Empathy

There are three general approaches to understanding empathy:3, 19

  • A cognitive response (feelings are understood)
  • An emotional response (feelings are shared)
  • A compassionate response (feelings are recognized)

Each of these approaches relies on distinct, but interactive, neurological processes. As it does for all responses to stimuli, the brain generates electrical and chemical reactions that travel across brain tissue and structures. These neural impulses travel in millimeters per millisecond along patterned networks.2 Empathy is also a byproduct of pain processing.9

So, in place of the word ‘feelings,’ substitute ‘pain,’ and the whole process may be easier to understand. Pain is understood, pain is shared, and an effort is made to respond to pain.

Physical Pain

Until recently, research about empathetic responses has largely focused on physical pain. When we are injured, or when we watch someone get injured, sensory input travels from the source of pain (our eyes if we are watching it happen) to various related regions of the brain. It moves from the brain stem (bottom) up.

The brain uses the geniculostriate pathway2 to calculate what is happening (back to the occipital lobe, down to the temporal lobe, and forward to the cognitive regions of the brain) and where/how (back to the occipital lobe, up to the parietal lobe, and forward to the cognitive regions).11 The visual input goes straight to the thalamus, by way of the tectopulvinar2pathway The response is completely primal. It is also fast, so we can react promptly in unsafe situations.

It is interesting to think about empathy in relation to chronic pain. Recent chronic pain research has focused on interoceptive processes (top of the brain down), where the cognitive regions of the brain use predictive coding. In this process, cognitive regions of the brain instruct the sensory receptors to feel physical pain, even when actual pain is not present.2, 9, 18 The body may be healed, but the brain says it isn’t.

A similar situation can occur when study participants look at images of other’s people’s injuries.9 For example, if we see someone else’s fingers closed in the hinge of a car door, our pain network responses can still be seen on fMRI. We do not have to sense pain to experience it. This concept is called the interpersonal sharing effect.6

Scholar. Source: Canva.comI suspect that chronic pain healing processes may prove to be instrumental in healing emotional pain, if it is studied more carefully by future scholars. Psychologists use the powerful tool of cognitive reappraisal, which allows us to detach our practiced emotional responses to a situation, and then to reframe our thinking about it for the future. But does this process work well in populations that have emotional responses that are already detached? Are all neural processes operating in a healthy way when children have conduct disorders? Is there a physical process that can be enacted to generate new, neural connections leading to empathetic behavior? These questions do not have definitive answers, but it may be that research is beginning to clarify what happens in the brain in response to pain.

Social Pain

In the field of psychology, empathy is often considered to be a component of social cognition. That’s because neural activity in the frontal lobe elements (such as the anterior cingulate and the dorsolateral prefrontal cortex) can often contribute to an intellectual response that regulates your social responses and behavior. So, for example, you have an instinctive negative response to another person’s comment—but understand how to “bite your tongue.” Or you feel sympathetic to a person’s plight, but you offer a response that is decidedly neutral.  The social cognition perspective is accurate, important, and based on observations about behavioral activity when the orbitofrontal cortex (OFC), the inferior frontal cortex, or the anterior cingulate cortex are impaired.2 But social cognition places an emphasis on thinking, which is not always evident in poor behavior.

When empathy (or lack of it) is understood to be a response to another’s pain, emotional responses take on more nuanced characteristics. Brain researchers have focused on three neural pathways, or a ‘network of networks.4, 14 Jointly, they are known as The Empathy Network, which includes:

  • The Affective Network, which activates in response to human interactions,
  • The Saliency Network, (amygdala-based, fight or flight reactions and other primal emotional pain responses),
  • The Mentalizing Network, also called the Theory of Mind Network (which manages taking the perspective of other’s experiences, and has cognitive characteristics).

Scholar. Source: Canva.comTheorizing that there is a difference between physical pain processing and emotional pain processing, researchers have moved toward exploring the neural experience of “social” pain processing. Three key resources,4, 12, 13 extend the wide field of earlier research. The most recent conclusions suggest that there is an interesting, perhaps even binary and reciprocal, relationship between physical pain and emotional pain. It is here, I believe, that a solution to the problem of antisocial behavior may emerge. We must study it to learn how to control the empathy network and to reinforce positive neuroplasticity.

There seems to be a distinct neural pathway for social pain. Early adopters of the idea,12 used fMRI to evaluate participants who watched a social situation in which someone was excluded. They expected to see activation in the ventromedial, medial, and dorsomedial prefrontal cortices (VMPFC, MPFC, DMPFC), the posterior superior temporal sulcus (pSTS), and other regions associated with mentalizing in physical pain. (That is, taking another’s perspective.) However, they also hypothesized that fMRI would reveal activation in the dorsal anterior cingulate cortex (dACC) and the anterior insula, which are typically linked with social exclusion.

Interestingly, activation in the MPFC, which contributes to the ability to perceive another as being similar to self, was the only area that was statistically significant for the mentalizing network. It was more active during empathetic responses to social pain. But that was not significantly different than what could be expected for responses to physical pain.

The study found that the affective network (the dACC) was more closely linked to social exclusion than was recorded for the insula. But this was only true in highly trait-empathetic individuals, as assessed by raters on a 7-point scale. Therefore, the researchers12 concluded, it is likely that physical pain and social pain are processed differently. And that there is likely a distinct neural pathway for empathy relating to social pain.

Can Empathy Be Controlled?

When research participants exercise control over their empathetic responses, fMRI has also shown distinctive differences in neural activation.4 E. Bruneau et al. demonstrated that the amygdala can be regulated deliberately when evaluating social pain, but not physical pain, at least when reading about it.4 So, deliberate control of emotional pain decreased activity (bilaterally) in the amygdala and other regions, but did not decrease activity in any region when participants read about physical pain. In both cases, if there were any voxel peaks (activation in three-dimensional space), it only occurred significantly in the left amygdala, suggesting that this region may contribute more to social pain.

In addition, activity in the amygdala was positively associated with the Theory of Mind network, and negatively associated with physical pain, implying that the amygdala could be regulated when taking another’s emotional perspective. The authors theorize that as concern for another person’s emotions increases, the brain removes distraction of attention towards the own body’s emotional and physical pain. They learned:

  • The anterior superior temporal sulcus (STS) is believed to contribute to regulation because it “was the only region where activity reliably preceded and predicted subsequent amygdala response to emotional pain stories (p. 117). 4
  • When participants were instructed to remain objective (to control the response) the related brain regions showed increased activity.

More intriguingly, there appears to be an inverse relationship that toggles between bottom up (primal brain-stem to cognitive) reactions and top-down regulation of primal reactions by cognitive structures. This suggests that only one region of the brain can maintain control at a time—either the cognitive region or the instinctual region. The balancing act in healthy minds is reminiscent of the autonomic (unconscious) processes in the default mode network that shuts down when cognitive processes are engaged.1 Again—only one region of the brain maintains control at a time. The speculation is, then, that emotional dysregulation is a product of disrupted—or co-existent—states, meaning that this toggling feature is either not well regulated, or that both regions of the brain are competing for control.22 In daily observation, this makes sense: what were they thinking when they engaged in that behavior?

Yours, Mine, Our Pain

Our past personal experiences also contribute to our understanding of another’s pain. The Social Context Network Model (SCNM) draws upon group membership, existing attitudes, threat information, and emotional cues to consolidate the three Empathy Networks (Saliency, Pain Matrix, and Mentalizing)14 Under the SCNM, affective arousal is stimulated first (bottom up by the amygdala), followed by emotional understanding (a sensory experience leading to cognition), and trailed by emotional regulation (a cognitive process). For true social empathy to occur, the stimulating event must be undeterred through all three networks—and be supported by compassionate action.

Sensorial Reactions

Various areas of the brain possess specialized sensory neurons called von Economo neurons (VENs). These cell bodies are especially large, with very long nerve branches (dendrites) and they assist in the bottom-up collection of sensory input, during the processing of experiences.14 The insula—a primal structural that is on the brain stem, but tucked immediately under the more evolved sensory and cognitive portions of the brain—contains VENs. The insula is also part of the Social Context Network Model, because it integrates details about experiences with insight about sensory conditions14

In 2012, Xiaosi Gu et. al7  used meta-analysis to study the insula further. The researchers concluded that both the anterior insula (sensory input, from the brain stem) and anterior cingulate (sensory output, and aversion to stimuli) are active and critical to empathy processing. However, empathy cannot be processed if sensory input is not received by the insula.

Scholar. Source: Canva.comIf this insight is supported by additional research, then it would be helpful for a child’s support network (parents and teachers) to understand that a child may not be choosing not to care about others, but rather is limited by biology. (Again, teaching empathy may require more than instructing children how to be kind.) And, if there is a physical limitation, what public (and independent) school accommodations must be made to meet the requirements of Title II of the American’s with Disabilities Act?20

Motor Reactions

Sensory processes are critical, but motor processes also play a role in behavior. There is some evidence (and debate) that motor neurons may act as sensory neurons, using ‘mirror-based action understanding’, within the parieto-frontal circuit.5, 16  This process contributes to the practice of perspective taking.

When we see other’s actions (or facial expressions) we see not only what, but make judgments about why. For example, we have a motor (and sensory) memory of what it feels like to trip and fall on concrete. If we watch someone else trip and fall, we might have an empathetic response “from the inside” of our minds.16 While it is true that this response is still only based on our personal experiences, it brings us closer to understanding others.

Compared with the aforementioned (motor-neuron-absent) superior temporal sulcus (STS), Rizzolatti and Sinigaglia (2010) felt the mirror-neuron-rich parieto-frontal cortex and primary motor cortex both offered more in the way of perspective taking and in making predictions.17 In the tripping example, we might expect the person to be in pain, or to cry out. In an individual whose abilities are compromised physically, the event is only perceived as unrelated motor actions. And there may not be any emotional response at all.

After considering the definitions of empathy, how empathy is processed, and the regions of the brain that are involved when empathy, we can start to apply these concepts when understanding the behavior of children who have conduct disorders.