Empathy for Pain Involves the Affective but Not Sensory Components of Pain

What This Research Found

Tania Singer and colleagues’ landmark 2004 study provided the first robust neuroimaging evidence for how the human brain generates empathy for another person’s pain. Using functional magnetic resonance imaging (fMRI), the researchers examined what happens in the brain when people watch someone they love experience suffering—establishing that empathy operates through the bodily simulation of emotional states in the observer’s own affective brain circuits.

The experimental design captured authentic empathic concern. Singer recruited romantic couples for the study, recognising that genuine emotional connection was essential for studying real empathy rather than merely cognitive awareness of another’s state. Female participants lay in the MRI scanner while their male partners sat nearby. Both received painful electric shocks to their hands in an alternating paradigm: sometimes the woman in the scanner experienced the shock directly; sometimes she watched while her partner received it. This design allowed direct comparison of brain activation during first-person pain and observation of a loved one’s pain.

The affective pain matrix activated for both self and other. The study’s core finding was that observing a loved one’s pain activated the same brain regions responsible for the emotional, distressing aspects of one’s own pain. Specifically, the bilateral anterior insula (AI) and rostral anterior cingulate cortex (ACC) showed robust activation both when participants experienced pain themselves and when they merely watched their partners suffer. These regions comprise what researchers call the “affective pain matrix”—the circuitry that generates the unpleasant emotional experience that makes pain feel bad. The finding demonstrated that affective empathy involves the literal simulation of emotional states in one’s own brain.

Sensory regions showed a critical dissociation. While the affective regions activated for both direct experience and observation, the sensory pain regions showed a different pattern. Primary and secondary somatosensory cortices, posterior insula, and caudal ACC—regions that process where pain is located and how intense it feels—activated only when participants experienced pain directly themselves, not when they observed their partner’s pain. This dissociation demonstrated that empathy involves sharing the emotional component of another’s experience while not sharing the physical sensations. We feel that their pain feels bad, but we do not feel their pain in our bodies.

Individual differences in empathy corresponded to brain activation. Participants who scored higher on self-report empathy questionnaires showed greater activation in the anterior insula and ACC when observing their partner’s pain. This correlation established a direct link between people’s subjective sense of feeling with others and the activation of specific neural structures. It also suggested that variability in empathic response reflects genuine differences in how the brain processes others’ emotional states—a finding with profound implications for understanding empathy deficits in conditions like narcissistic personality disorder.

How This Research Is Used in the Book

Singer’s 2004 study appears throughout Narcissus and the Child as foundational evidence for understanding the neurobiological basis of empathy deficits in narcissism. The research is particularly central in chapters examining the neural architecture of the narcissistic brain and the development of empathy capacity through early relational experience.

In Chapter 6: Diamorphic Agency, Singer’s research is cited directly in explaining how emotional empathy develops—or fails to develop—through early childhood interactions:

“The anterior insula (the Translator) develops its capacity for emotional empathy through thousands of interactions in which other people’s emotional states are perceived, shared, and responded to. The child sees the mother’s face of distress; and the child then feels an echo of that distress in their own body. Through this the child learns that others have inner states and that these states matter. The child discovers that one can and should respond to them.”

The book uses Singer’s framework to explain how contemporary social interaction increasingly substitutes cognitive empathy for emotional empathy—understanding that someone is sad versus actually feeling an echo of their sadness—with concerning implications for empathy development in the digital age.

In Chapter 10: Diamorphic Scales, Singer’s work grounds the detailed examination of how the empathy circuit is selectively pruned in narcissistic development:

“The anterior insula (AI) is the brain’s translator—the structure that converts perceived emotion into felt emotion. When you see someone in pain and wince yourself, that’s the AI at work. When a friend’s grief makes your own chest tighten, that’s translation happening in real time.”

The book explains that affective empathy is not simply a choice or personality trait but depends on specific neural structures that can be damaged by early adversity or lack of attuned caregiving.

In Chapter 7: Architecture Structures, Singer’s research helps explain the structural deficits found in the narcissistic brain:

“Empathy requires simulation: to understand others’ pain, one’s brain must generate a shadow of that pain in one’s own body-sensing regions. When one sees someone suffer, the anterior insula activates as if the observer is suffering themselves, a process called embodied simulation. Witnessing distress can feel physically uncomfortable because the insula is literally simulating a version of what the other person feels.”

Throughout the book, Singer’s distinction between affective and sensory components of pain processing illuminates why narcissists can cognitively understand suffering while remaining unmoved by it—they process the information without generating the corresponding felt experience.

Why This Matters for Survivors

If you were in a relationship with a narcissist, Singer’s research validates something you experienced but may have struggled to articulate: the profound wrongness of their non-response to your obvious suffering.

Your pain should have echoed in their brain—and it did not. When you cried, when you pleaded, when you showed unmistakable signs of distress, a normally functioning brain would have generated an unpleasant experience in response. Singer’s research shows that seeing a loved one in pain activates the anterior insula and anterior cingulate cortex—the same regions that make our own pain feel bad. This activation creates genuine discomfort that motivates compassion and caregiving. The narcissist’s failure to respond to your suffering was not simply a lack of caring in some abstract sense. It was a failure of the brain systems that would normally make your pain feel aversive to them.

Their indifference was neurological, not a reflection of your worth. Survivors often internalise the narcissist’s non-response as evidence of their own inadequacy. If only you had explained better, been more deserving, communicated more clearly—surely then they would have cared. Singer’s research suggests the problem was never your communication. The anterior insula is supposed to generate the felt experience of another’s emotion automatically, below the level of conscious choice. If this mechanism is not functioning properly, no amount of explanation will create the visceral discomfort that motivates genuine empathic response. You were not insufficiently eloquent. Their empathy circuit was not translating your distress into felt experience.

Understanding the mechanism can help you stop trying. Many survivors exhaust themselves in endless attempts to make the narcissist “get it”—to finally understand how their behaviour hurts. Singer’s research suggests why these attempts almost always fail. Affective empathy is not a cognitive operation that can be induced through argument. It is a visceral, embodied process that either activates or does not. The narcissist may cognitively understand that you are suffering—indeed, their preserved cognitive empathy may make them quite good at reading your emotional states. But understanding is not feeling. Knowing you are sad does not make them feel uncomfortable in the way that would motivate change. This knowledge can free you from the exhausting cycle of trying to find the right words.

The dissociation between cognitive and affective empathy explains their manipulation skills. Survivors are often confused by the apparent contradiction between the narcissist’s profound empathy failures and their skillful emotional manipulation. Singer’s distinction between affective empathy (feeling with others) and cognitive empathy (understanding others’ mental states) resolves this paradox. The narcissist can be exquisitely skilled at reading your emotions, predicting your reactions, and knowing exactly which buttons to push—this is cognitive empathy. What they lack is the automatic generation of an uncomfortable emotional response to your suffering—affective empathy. They understand what you feel; they do not feel it themselves. This combination makes them both capable of calculated exploitation and immune to the distress that would normally make such exploitation uncomfortable.

Clinical Implications

For psychiatrists, psychologists, and trauma-informed healthcare providers, Singer’s research has direct implications for assessment and treatment of both narcissism and its aftermath in survivors.

Assessment should distinguish cognitive from affective empathy. Singer’s work established that these are partially dissociable capacities with different neural substrates. Narcissists typically retain cognitive empathy while showing deficits in affective empathy. Assessment instruments that conflate these capacities may miss the specific pattern of narcissistic empathy impairment. Clinicians should separately evaluate patients’ ability to understand others’ mental states (cognitive empathy) and their tendency to feel emotional resonance with others’ experiences (affective empathy). Self-report measures, behavioural observation, and—where available—physiological measures of empathic responding can help characterise the specific empathy profile.

Survivors’ difficulty trusting after abuse reflects rational pattern recognition. Singer’s research shows that normally functioning humans generate visceral discomfort in response to others’ suffering. Survivors who struggle to trust again have learned that some people lack this basic safeguard. Their hypervigilance to emotional authenticity, their scanning for signs that empathic displays are performative rather than felt, reflects valid learning about human variation. Clinicians should validate this protective vigilance while helping survivors develop nuanced capacity to assess others’ empathic functioning.

Therapeutic approaches for narcissism should target the anterior insula. If affective empathy deficits reflect reduced anterior insula function, interventions that enhance interoceptive awareness and emotional embodiment may indirectly strengthen empathic capacity. Mindfulness-based approaches that increase awareness of bodily sensations, body-based therapies like somatic experiencing, and compassion training protocols that explicitly engage felt emotional states may be more effective than purely cognitive interventions. However, clinicians should maintain realistic expectations: the morphing of these circuits occurred over years of development, and reversing it—if possible at all—requires sustained, intensive effort.

The emotional labour of being with empathy-deficient patients deserves acknowledgment. Clinicians working with narcissistic patients often experience what feels like one-way emotional flow: they generate empathic responses to the patient that are not reciprocated. Singer’s research explains why this experience is so draining. The therapeutic relationship normally involves bidirectional empathic resonance. When one partner in the dyad is not generating the corresponding activation, the other may feel unseen, used, or depleted. Clinicians should have support, supervision, and self-care practices that address this asymmetric emotional demand.

Neurobiological findings do not preclude accountability. Singer’s research explains mechanisms but does not eliminate moral responsibility. Many people with neurological differences do not exploit or abuse others. Understanding that the narcissist’s empathy deficits have biological components should inform treatment approaches and help survivors depersonalise the abuse, but should not be presented as excusing harmful behaviour. Adults retain responsibility for their actions regardless of whether those actions feel aversive to them.

Broader Implications

Singer’s research on empathy for pain extends far beyond individual relationships, illuminating patterns across families, institutions, and society.

The Foundation of Human Moral Behaviour

Singer’s finding that observing another’s suffering activates one’s own pain circuitry provides a neurobiological basis for the philosophical claim that empathy underlies human morality. If causing pain to others produces an unpleasant echo in our own brains, harmful behaviour carries inherent costs. The anterior insula functions as a kind of moral brake, making cruelty feel bad to the perpetrator. This has implications for understanding moral development in children: attuned caregiving not only builds secure attachment but also develops the neural capacity that will later constrain harmful behaviour through embodied discomfort. Moral education that relies solely on rules and consequences may be insufficient if the underlying empathic circuitry is impaired.

The Selection of Narcissistic Leaders

Singer’s research helps explain why narcissistic individuals often rise to positions of power. Making difficult decisions that harm some people for ostensible greater goods is easier when you do not feel an echo of the suffering you cause. Firing employees, implementing austerity measures, sending soldiers into battle—these actions create genuine distress in leaders with intact affective empathy. Leaders with reduced anterior insula activation during others’ suffering can make such decisions more easily. This is not always pathological: some contexts genuinely require the capacity to cause necessary pain without being paralysed by empathic distress. But it explains why narcissistic traits are overrepresented in positions of power, and why organisations must develop external ethical constraints that do not rely on leaders’ empathic responses.

The Architecture of Institutional Harm

Institutions can be designed in ways that either engage or bypass human empathy. When decision-makers are removed from those affected by their decisions—through physical distance, bureaucratic abstraction, or statistical framing—the anterior insula activation that would normally accompany witnessing suffering is prevented. The bureaucrat who denies healthcare claims never sees the patients who die as a result. The algorithm that makes lending decisions does not feel the foreclosed family’s distress. Singer’s research suggests that preserving human connection in institutional decision-making is not merely humane sentiment but neurobiologically necessary for ethical functioning. Policies that insulate decision-makers from consequences may inadvertently create institutional psychopathy even when individual decision-makers have intact empathy circuits.

The Design of Digital Environments

If affective empathy requires activation of the anterior insula through observation of others’ emotional states, digital communication may systematically impair empathic processing. Text lacks the facial expressions, vocal prosody, and body language that normally activate the empathy circuit. Seeing “I’m sad” in a text message does not produce the same anterior insula activation as seeing a friend’s face crumple in grief. Singer’s research suggests that the shift toward digital interaction, especially during the developmental windows when empathy circuits are being constructed, may have consequences for empathic capacity at population level. This offers a neurobiological framework for understanding rising rates of cyberbullying, online cruelty, and digital harassment.

The Intergenerational Transmission of Empathy Deficits

Singer’s research, combined with developmental neuroscience, illuminates how empathy deficits transmit across generations. If empathy capacity develops through experience-dependent maturation of the anterior insula during early childhood, and if this maturation requires caregiver responsiveness to the child’s distress, then parents with impaired affective empathy cannot provide the attuned responses that would develop healthy empathy in their children. The narcissistic parent’s failure to feel the child’s suffering becomes the child’s failure to develop the capacity to feel others’ suffering. The intergenerational transmission of narcissistic traits is not mysterious; it follows from the neurobiological reality that empathy is constructed through relationship. Breaking these cycles requires interventions that provide the attuned responsiveness the parents could not offer.

The Limits of Compassion Training

Singer’s subsequent research examined whether empathy can be enhanced through training. Compassion meditation protocols have shown increases in anterior insula activation and subjective empathic response. However, these studies typically involve motivated participants engaging in intensive practice. The application to narcissism is uncertain: narcissistic individuals rarely seek treatment for empathy deficits, may not sustain the practice required for neural change, and may have more severe baseline deficits than normal-range participants in meditation studies. Singer’s research offers hope that empathy is trainable, but realistic assessment of what this means for narcissism treatment remains essential.

Limitations and Considerations

Singer’s foundational study has important limitations that inform responsible interpretation.

Sample specificity. The study examined romantic couples, ensuring genuine emotional connection but limiting generalisability. Empathy for strangers, out-group members, or people perceived as different may involve different neural mechanisms or magnitudes of activation. Subsequent research has confirmed that empathy is modulated by perceived similarity, group membership, and whether the other person is seen as deserving—factors that may be particularly relevant in narcissism.

The challenge of reverse inference. While the study demonstrates that anterior insula and ACC activate during empathic pain observation, inferring that this activation constitutes the experience of empathy requires caution. These regions serve multiple functions; their activation is consistent with but does not prove empathic experience. The correlation with self-reported empathy scores strengthens the inference but does not establish it definitively.

The role of attention and instruction. Participants were explicitly attending to their partners and anticipating when shocks would occur. Natural empathic processing in everyday life may differ. Some research suggests that narcissists can show normal empathic activation when explicitly instructed to attend to others’ emotions, suggesting that automatic versus deliberate empathy may be differently affected in narcissism.

Individual differences beyond empathy. The study focused on empathy but did not examine how factors like attachment style, affect regulation capacity, or trauma history might modulate the relationship between observation and anterior insula activation. Subsequent research has shown that these factors matter, and survivors’ own empathy processing may be altered by their experiences.

Translation to clinical practice. While the research establishes neural mechanisms, translating these findings into therapeutic interventions remains challenging. Knowing that the anterior insula underlies affective empathy does not immediately suggest how to strengthen its function in those with deficits.

Historical Context

Singer’s 2004 paper was published in Science, one of the world’s most prestigious scientific journals, reflecting its significance as a breakthrough in understanding the neural basis of empathy. The study arrived at a moment of intense interest in the social brain, following the discovery of mirror neurons in macaque monkeys in the 1990s and growing evidence that humans possess similar systems for understanding others’ actions and intentions.

Prior to Singer’s work, empathy was studied primarily through behavioural measures and self-report questionnaires. The neural mechanisms remained poorly understood, and whether observing another’s emotional state actually activated similar neural systems in the observer was empirically undemonstrated. The philosophical question of whether we can truly share another’s experience had no neurobiological answer.

Singer’s elegant experimental design—using romantic couples to ensure genuine emotional connection, comparing self-experienced and observed pain in the same participants, and identifying both shared and distinct neural activations—provided the first robust evidence that affective empathy involves actual simulation of emotional states in the observer’s brain. The finding that cognitive understanding of pain and affective sharing of pain involve partially distinct systems proved particularly influential, helping to resolve debates about the nature of empathy and establishing the anterior insula as a key hub for embodied emotional understanding.

The study launched a productive research program. Singer and colleagues subsequently investigated how empathy is modulated by perceived fairness, in-group/out-group membership, and characteristics of the observer. The research extended to examine compassion as distinct from empathy, leading to training studies showing that compassion cultivation can enhance positive approach responses to suffering while reducing the burnout risk associated with excessive empathic distress.

The paper has been cited over 5,000 times and remains foundational for understanding empathy deficits in autism, psychopathy, narcissistic personality disorder, and other conditions characterised by difficulties in emotional resonance with others.

Further Reading

• Singer, T., & Lamm, C. (2009). The social neuroscience of empathy. Annals of the New York Academy of Sciences, 1156, 81-96.
• Singer, T., Seymour, B., O’Doherty, J.P., Stephan, K.E., Dolan, R.J., & Frith, C.D. (2006). Empathic neural responses are modulated by the perceived fairness of others. Nature, 439, 466-469.
• Klimecki, O.M., Leiberg, S., Ricard, M., & Singer, T. (2014). Differential pattern of functional brain plasticity after compassion and empathy training. Social Cognitive and Affective Neuroscience, 9(6), 873-879.
• Fan, Y., Duncan, N.W., de Greck, M., & Northoff, G. (2011). Is there a core neural network in empathy? An fMRI based quantitative meta-analysis. Neuroscience and Biobehavioral Reviews, 35, 903-911.
• Decety, J., & Jackson, P.L. (2004). The functional architecture of human empathy. Behavioral and Cognitive Neuroscience Reviews, 3(2), 71-100.
• Hepper, E.G., Hart, C.M., & Sedikides, C. (2014). Moving Narcissus: Can narcissists be empathic? Personality and Social Psychology Bulletin, 40(9), 1079-1091.