by Herbert Harris
When we say, “I’m feeling self-conscious,” we usually mean we are uncomfortably aware of being the center of other people’s attention. We might worry about how we look, what we wear, or how we act. While we are, to some degree, concerned with aspects of ourselves, our main focus is on others and what they think about us. “Self-conscious” is an interesting choice of words that might reveal something deep about the nature of self and consciousness.
Two hundred years ago, the German philosopher G.W.F. Hegel introduced a groundbreaking idea: self-consciousness does not arise from introspection, but from mutual recognition. We become aware of ourselves as individuals not by looking inward, but by encountering another mind that perceives us as conscious, and by recognizing that mind in return. Hegel argued that this process of reciprocal recognition is the foundation of personhood. It is only through others that we truly understand ourselves. Hegel linked this movement from recognition to a broader concept of freedom. For him, freedom was not just the lack of constraints, but the achievement of autonomy through mutual recognition. According to Hegel, we become free not by turning inward, but by engaging in relationships that acknowledge and affirm our self-consciousness.
This insight has resonated across philosophy, psychology, and social theory. George Herbert Mead viewed the self as emerging from social roles and symbolic interaction. He emphasized how the self develops by internalizing others’ perspectives, especially through language and shared symbols, which create a ‘generalized other’ that shapes individual identity. Sartre described self-consciousness as an unavoidable confrontation with the gaze of the other. His concept of the ‘gaze’ illustrates how we feel exposed under someone else’s scrutiny and how their judgment influences our self-awareness. Frantz Fanon later developed these ideas, showing how the Black subject becomes aware of themselves through the racialized gaze of a white other, illustrating how power dynamics and social identity impact self-consciousness. Each of these thinkers extended Hegel’s idea of the self as socially constructed and relational rather than autonomous or private.
Although Hegel’s ideas have been highly influential, they would seem to have limited applicability to neuroscience-based explanations of consciousness. The primary explanatory frameworks in neuroscience are bottom-up, beginning with molecules and ending with complex neural network systems. But the Hegelian outside-in, top-down perspective might provide insights that both complement and inform neuroscience.
Recent work by Karl Friston and colleagues introduces a broader perspective that can be applied to interpersonal interactions. Known as active inference, this approach describes the brain as a prediction machine. It creates internal models of the world, compares these models to sensory data, and updates them to minimize prediction errors. Active inference offers a strong framework for understanding perception, agency, and even consciousness.
Anil Seth, for example, has convincingly argued that our sense of being a conscious self comes from the brain’s predictions about internal bodily states. According to this view, consciousness is rooted in interoception—the brain’s ability to interpret signals from the heart, lungs, gut, and skin. This explains why conscious experience feels so immediate and embodied, and why disruptions in interoception are linked to altered states of awareness. Seth refers to this as the ‘beast machine’ theory, emphasizing that we are embodied organisms whose conscious experience arises from the brain’s regulation of bodily states. The theory offers a compelling account of conscious presence, of what it is like to be a body.
But conscious presence is not the same as self-consciousness. We can be consciously present without reflexively knowing ourselves as subjects who are being perceived, evaluated, or judged. We can feel pain without feeling ashamed. We can act without wondering how others will interpret our behavior. Yet many of the most defining moments in human life involve this second layer of awareness. Pride, guilt, embarrassment, envy, admiration, and even love all involve recursive self-modeling in a social space. These are not private phenomena. They are saturated with the imagined perspectives of others.
How might this happen in the predictive brain? Seeing the brain as a prediction machine operating in social settings by modeling ourselves and others as we interact may explain how self-consciousness emerges. We use language, behavioral cues, attitudes, and gestures to understand and predict each other’s actions. But when people interact, the challenge of prediction grows rapidly. While we are building predictive models of others, they are building predictive models of us. As our interactions become more complex, we need to understand and predict how they see us, how they will treat us, and how they will respond to our actions. To do this, we need predictive models that capture how others are modeling us. This involves a concept that a computer scientist might call recursion. It is something like Douglas Hofstadter’s notion of a “strange loop,” where the self emerges from self-referential feedback loops that loop back on themselves. This dizzying process of modeling models is at the heart of our social interactions and may be key to self-consciousness.
Developmental psychology provides strong support for this view. Infants develop a sense of self through social interaction: through joint attention and attunement with caregivers. Starting with the social smile, caregivers act as psychological mirrors that reflect and validate the child’s feelings. For example, when a six-month-old infant smiles and observes their caregiver’s delighted response, they begin to form a mental image of themselves as a happy person. This is more than just a reflective image created by a mirror. Caregivers are constructing models of the child as another mind, and they communicate aspects of these models through actions, gestures, and eventually, language. The child uses this information to develop a self-image. Unlike self-modeling based only on interoceptive data, this social feedback enables the child to create recursive models of how others see themselves. This shift—from embodied awareness to socially mediated self-awareness—may mark the true start of self-consciousness. Theory of mind—the ability to understand others’ beliefs and perspectives—develops alongside this recursive self-modeling. Self-consciousness, therefore, is not just a result of brain development but a product of social experience.
A growing awareness of the importance of social interactions has started to influence neuroscience through new fields like second-person neuroscience and multi-agent modeling. Instead of studying brains in isolation, researchers now focus on brains in interaction: parent and infant, therapist and patient, teacher and student. These studies show that shared brain rhythms accompany mutual understanding, emotional connection, and effective communication. For example, work by Uri Hasson and colleagues at Princeton has demonstrated that during storytelling, the neural activity of speakers and listeners becomes closely linked, effectively aligning their mental representations. In another area, Ruth Feldman and her team have shown that synchronized brain activity between mothers and infants predicts the development of empathy and emotional regulation. These findings indicate that shared consciousness is not just a metaphor.
Computer simulations of interacting active inference systems are known as multi-agent active inference models. They demonstrate how prediction-based brains might adapt to one another. In these frameworks, two agents do not just predict their environments; they also predict each other’s predictions. This recursive entanglement is similar to the mutual recognition Hegel described. Self-consciousness, in this view, may not only be a property of individual brains but also an emergent feature of coupled systems.
Some might argue that self-consciousness can develop in isolation, such as during extreme solitude or sensory deprivation. However, even solitary reflection depends on internalized social norms and imagined others. The dialogues we have in our minds are often with figures we’ve known, voices we have heard, and expectations we have absorbed. The nature of selfhood may be social, even when no one else is around.
This idea opens up new possibilities across various fields. In psychiatry, phenomena like projection and conditions ranging from dissociation to personality disorders could be understood not only in terms of disordered internal models but also as disruptions in the loops of recognition that sustain selfhood. For example, psychoanalyst Heinz Kohut proposed that individuals with narcissistic personality disorder fail to develop stable self-esteem because early caregivers did not adequately serve as ‘selfobjects’—figures who provide mirroring and empathy necessary for healthy self-development. Without these relational functions, the self becomes fragmented and dependent on external validation.
In artificial intelligence, building ethics into AI applications is an urgent problem. Current approaches include rule-based programming for moral decision-making and reinforcement learning guided by ethical constraints. For instance, some systems are trained to avoid bias in hiring algorithms, while others aim to align chatbot behavior with human norms through supervised learning. However, the kind of empathic understanding required for flexible ethical action may require recursive self-modeling.
Hegel developed his ideas to answer questions that were very different from those of modern neuroscience. However, this major difference in perspective might give us a new way to understand consciousness. The self is not something we are born with, and self-consciousness does not exist in isolation. Both may develop through social interactions. The explanatory gap might not be between brain and mind, but between the individual and others. The next time you feel self-conscious, ask yourself: Through whose eyes are you seeing yourself? The answer could reveal that even your most personal self is deeply connected to those around you.
***
Enjoying the content on 3QD? Help keep us going by donating now.