Mirror Neuron Simulation Theory – Giacomo Rizzolatti & Vittorio Gallese

The dawn of contemporary cognitive neuroscience was largely defined by a modular, disembodied conception of the human mind. For decades, classical cognitive science operated under what philosopher Susan Hurley critically termed the “classical sandwich” model: an epistemological architecture wherein perceptual inputs were cleanly segregated from motor outputs by an insulated, amodal central processor responsible for cognition, propositional thought, and mental state attribution. Within this Cartesian-flavored computational framework, the motor system was relegated to a subsidiary status—a mere mechanical executor of dictates computed by higher cortical areas. This disembodied paradigm began to unravel in the early 1990s at the Institute of Human Physiology at the University of Parma, where a research team led by Giacomo Rizzolatti, along with colleagues Vittorio Gallese, Leonardo Fogassi, and Luciano Fadiga, made an empirical discovery that would fundamentally reorganize our understanding of the primate brain: the identification of mirror neurons.

Mirror neurons are a distinct class of visuomotor neurons that fire both when an individual executes a goal-directed motor act and when that individual observes another agent performing a similar or identical act. Initially documented in the ventral premotor cortex (area F5) of the macaque monkey (Macaca nemestrina), these dual-discharge cells demonstrated that the motor system is fundamentally implicated in perception, action interpretation, and intersubjective social cognition. Rather than relying on abstract, symbolic inferences to deduce what another individual is doing, the observing brain deploys its own latent motor repertoire to reconstruct the observed act from within. This neurobiological finding bridged the historic divide between motor physiology and cognitive psychology, providing concrete empirical grounding for the philosophy of embodied cognition.

From this neurophysiological discovery emerged one of the most provocative and influential theoretical frameworks in contemporary cognitive science: the Mirror Neuron Simulation Theory, championed most extensively by Vittorio Gallese. Expanding the initial motor-matching paradigm, Gallese posited that mirror mechanisms are not merely circumscribed to manual actions, but represent an overarching neural architecture of “embodied simulation.” Through this automatic, pre-reflective functional mechanism, actions, emotions, and somatic sensations perceived in others are mapped onto the observer’s internal motor, viscero-motor, and somatosensory circuits. This post provides an exhaustive, multi-disciplinary examination of the mirror neuron simulation theory, tracing its trajectory from serendipitous single-unit recordings in Parma through its profound philosophical, clinical, evolutionary, and computational ramifications.

1. Historical Discovery and Neurophysiological Foundations in Parma

1.1 The Serendipitous Discovery in Area F5 of the Macaque Brain

The identification of mirror neurons in the early 1990s within the laboratory of Giacomo Rizzolatti at the University of Parma stands as one of the most celebrated moments of empirical serendipity in modern neurobiology. The investigative team, comprising Rizzolatti, Vittorio Gallese, Leonardo Fogassi, Luciano Fadiga, and Giuseppe di Pellegrino, was originally engaged in systematically mapping the functional organization of the macaque monkey’s ventral premotor cortex, specifically an agranular frontal region designated as area F5. The experimental paradigm was engineered to delineate how individual neurons within this region code for specific, goal-directed hand and mouth movements, such as precision grips, prehension, manipulation, and tearing of food items.

During these chronic single-unit recording sessions, extracellular microelectrodes were inserted into the cortical mantle while the awake macaque reached for and grasped physical stimuli. The breakthrough occurred during informal interludes: as an experimenter picked up a piece of food to place it within the reach of the monkey or consumed a snack during a laboratory break, the audio monitors connected to the microelectrode amplifier began to crackle with high-frequency neuronal discharges. The cortical neurons in area F5 were firing robustly, despite the macaque remaining completely motionless. Crucially, the cells did not discharge when the monkey was presented with the static sight of food alone, nor did they respond to meaningless visual stimuli, geometrical patterns, or non-goal-directed movements; they fired exclusively when there was an authentic interaction between an animate biological effector and an object.

The initial findings were published cautiously by di Pellegrino et al. in 1992 in Experimental Brain Research, followed by the landmark comprehensive papers by Gallese et al. (1996) and Rizzolatti et al. (1996) in Brain and Cognitive Brain Research. These foundational publications established the term “mirror neurons” to delineate cells that exhibit both motor properties (discharging during the execution of a motor act) and sensory properties (discharging during the observation of the same or a structurally congruent motor act executed by another individual). This marked a radical departure from traditional cortical parcellation, demonstrating that motor circuits possess intrinsic perceptual capacities.

1.2 Anatomical Localization and Functional Architecture

Area F5 forms the posterior bank of the inferior arcuate sulcus and the adjacent cortical convexity in the macaque frontal lobe. Cytoarchitectonically, F5 is an agranular premotor region lacking a distinct granular layer IV, characterized by large pyramidal cells within layer V that project downstream to the primary motor cortex (M1, or area F1) as well as the reticular formation and the spinal cord. Somatotopically, F5 contains two distinct representations: a lateral convexity region primarily representing the mouth and larynx, and a more medial sector situated within the bank of the arcuate sulcus dedicated to hand and forelimb movements.

Area F5 does not operate in isolation; it constitutes the frontal terminus of dedicated, reciprocal parieto-frontal circuits. It receives dense, monosynaptic afferent projections from the rostral inferior parietal lobule (IPL), specifically area PFG and the anterior intraparietal area (AIP). This interconnected parieto-frontal network forms a functionally unified functional loop for sensorimotor transformations. Within this anatomical matrix, neurophysiologists differentiate between two distinct populations of visuomotor neurons: “canonical neurons” and “mirror neurons.” Canonical neurons, predominantly localized within the bank of F5 and AIP, respond visually to the presentation of three-dimensional, graspable physical objects without requiring any observed movement. They extract the pragmatic affordances of an object (its size, shape, and orientation) to automatically compute the appropriate manual posture for interaction.

In contrast, mirror neurons, distributed extensively across the F5 convexity and the PFG area of the parietal lobule, respond visually only to biological actions performed upon objects. At the laminar level, mirror neurons integrate complex visual inputs arriving from the superior temporal sulcus (STS) via parietal relay stations in PFG, processing these sensory inputs within layers II and III, before translating them into motor formats within the pyramidal efferents of layer V. These layer V projections target the hand field of M1, thereby priming the motor system to simulate observed actions at a sub-threshold level without necessarily precipitating overt muscular contraction.

1.3 Methodological Protocols of Single-Unit Extracellular Recordings

The empirical verification of mirror neurons required rigorous electrophysiological paradigms designed to rule out potential artifacts, such as incidental muscular contraction, general attentional shifts, or autonomic arousal. The Parma group utilized microdrive systems positioning varnish-coated tungsten microelectrodes with tip impedances between 0.5 and 2.0 megaohms at 1 kHz. Extracellular action potentials were recorded, amplified, filtered, and continuously routed to an audio monitor and an oscilloscope, enabling researchers to track the spike waveforms of isolated single units with microsecond precision.

To systematically evaluate the functional profiles of these neurons, the researchers implemented a rigorous matrix of behavioral conditions. First, the motor properties were established: monkeys executed varied manual interactions, including precision grips (pinching small food pellets between the thumb and index finger), whole-hand prehension (power grips around large objects), finger-pressing, and wrist rotations. Once the specific motor tuning of a neuron was verified, sensory testing commenced. The experimenter performed the identical action within the monkey’s visual field at various distances and spatial trajectories.

Crucial control experiments were instituted. To confirm that neuronal discharges were not driven by simple spatial attention or anticipation of food, the experimenter moved food toward the monkey using geometric tools (such as pliers or sticks). Intricate mechanical tool use initially failed to elicit responses in native, non-tool-trained macaques; the mirror neurons fired exclusively when a biological effector (a human or monkey hand) interacted with the target. Furthermore, researchers used electromyography (EMG) of the monkey’s peripheral arm, hand, and axial muscles to confirm that the monkey was completely relaxed and executing zero covert motor movements during observation. Kinematic parameters were monitored, validating that mirror discharges were genuine sensorimotor events reflecting a neural translation of observed behavior into the observer’s internal motor architecture.

2. Transition from Action Execution to Action Perception

2.1 The Sensorimotor Direct-Matching Hypothesis

The discovery of mirror neurons led Giacomo Rizzolatti and his colleagues to formulate the sensorimotor direct-matching hypothesis. This theory posits that the fundamental mechanism through which an organism understands the actions of others relies on an unmediated mapping of visual sensory inputs onto the observer’s own motor neural representations. When an individual observes another agent grasping an object, the visual depiction processed by visual cortices and the superior temporal sulcus does not undergo a prolonged, inferential cognitive deduction within higher-order association areas; instead, it accesses the frontoparietal motor system directly.

By mapping observed kinematics directly onto motor circuits that generate identical functional outputs, the observer recognizes the action “from within.” This mechanism circumvents the computational bottleneck of symbolic translation. The brain exploits its existing, evolutionarily sculpted action schemas as an interpretive metric. Rather than constructing a theoretical, detached computational model of the other agent’s physical kinematics, the brain enacts an automatic internal simulation, transforming sensory descriptions of physical motion into pragmatic motor comprehension.

The definitive neurophysiological proof that direct matching represents a predictive simulation rather than a passive visual response came with the landmark visual occlusion experiments conducted by Umiltà et al. in 2001. In this study, macaque mirror neurons in area F5 were recorded under two distinct conditions: full vision of an experimenter grasping an object, and occluded vision, where an opaque screen was placed between the monkey and the object, concealing the actual hand-object interaction. Crucially, the monkey had seen that an object was positioned behind the screen prior to the occlusion. The researchers observed that a substantial subset of mirror neurons discharged equally robustly when the final phase of the action was completely hidden from view. When the identical hand movement was executed behind the screen after the monkey knew no object was present, the neurons remained silent. This demonstrated that mirror neurons do not merely register incoming visual stimuli; they internally represent and extrapolate the goal of the action, reconstructing the unobserved reality via intrinsic motor simulation.

2.2 Goal-Directedness and Motor Teleology

A defining characteristic of the mirror neuron system is its profound teleological organization. Mirror neurons are remarkably indifferent to raw spatial kinematics or isolated trajectories of flesh; they are tuned fundamentally to the teleological dimension—the functional goal—of an action. For example, if an experimenter moves their hand through space along an identical arc and velocity profile, but without an object present (an intransitive, non-goal-directed movement), the vast majority of classical F5 mirror neurons will not discharge. The presence of an intentional target object, transforming raw movement into an authentic, transitive “motor act,” is the prerequisite condition for neural recruitment.

This teleological specificity demonstrates that the parieto-frontal network codes actions at a conceptual, motor-semantic level. The system distinguishes between the proximal kinematics (the precise biomechanical movements of individual joints and muscles) and the distal goal (the intended outcome of the action, such as grasping, holding, or manipulating). Neurophysiologists have identified varying degrees of congruence between the observed and executed actions that drive these neurons. “Strictly congruent” mirror neurons, which comprise approximately one-third of the population in F5, fire only when the observed action matches the executed action in both its precise motor mechanics and its ultimate goal (for example, a precision grip using an index-thumb pincer). Conversely, “broadly congruent” mirror neurons, representing roughly two-thirds of the cells, fire when the overarching functional goal is identical, even if the observed effector mechanics differ (for example, discharging during a whole-hand power grasp regardless of whether the experimenter executes a precision grip or a whole-hand grasp, or even uses a mouth effector).

This goal-directed tuning extends beyond the visual modality into multimodal domains. In 2002, Kohler et al. discovered “audiovisual mirror neurons” in macaque area F5. These extraordinary neurons discharge not only when the monkey sees and performs a specific goal-directed action (such as cracking a peanut shell or tearing a sheet of paper), but also when the monkey solely *hears* the acoustic consequence of that action in the total absence of visual input. The auditory signal directly engages the motor program corresponding to the action that produces that sound. The acoustic signature is not processed merely as an abstract auditory waveform, but as an indexical marker of an intentional motor teleology.

2.3 Action Understanding from Within

The direct-matching hypothesis carries profound epistemological implications: it defines action understanding as an intrinsic, first-person bodily capacity. Rizzolatti and his colleagues explicitly differentiate between two distinct cognitive pathways for processing other individuals’ actions: visual categorization and pragmatic motor comprehension. Visual categorization allows an organism to process an external event based on its optical properties, classifying movements via perceptual features in extrastriate, temporal, and prefrontal areas. Pragmatic motor comprehension, by contrast, provides an existential understanding of the action “from within,” grounded in the observer’s own visceral capacity to execute that same act.

This first-person motor understanding is constrained by the biological limitations of the observer’s own physical repertoire. For a mirror mechanism to match an observed action directly onto an internal motor program, the observed act must fall within the motor lexicon of the perceiving species. When an organism witnesses an action that has no motor analogue within its own motor architecture, the direct-matching mechanism cannot engage; the observation must instead be routed through disembodied visual, cognitive channels.

This biological constraint was empirically demonstrated in a seminal fMRI comparative study by Buccino et al. (2004). Human participants were presented with video clips depicting both conspecifics (humans) and heterospecifics (monkeys and dogs) executing two distinct types of actions: ingestive/biting actions and communicative actions. Ingestive biting is a phylogenetically conserved motor act shared across all three species. When human participants observed a human, a monkey, or a dog biting an apple, their own frontoparietal mirror neuron system discharged vigorously in all conditions; the motor act was immediately understood through the shared somatic schema of the oral apparatus. However, when participants observed communicative actions, the neural responses diverged dramatically. Watching a human speak silently activated the left premotor and Broca’s mirror areas, and watching a macaque perform lip-smacking (an ancestral primate communicative gesture) engaged a subset of premotor circuits. Yet, when humans watched a dog barking, there was no activation within the frontoparietal mirror network; the sensory processing was confined strictly to early visual and temporal cortices. Because barking is structurally absent from the human motor repertoire, the action could not be simulated from within; it had to be deciphered as an external, disembodied visual phenomenon.

3. Philosophical Foundations: Embodied Cognition and Phenomenology

3.1 Convergences with Merleau-Ponty’s Corporeal Schema

The neuroscientific model of the mirror neuron system did not emerge into a theoretical vacuum; rather, it provided an empirical realization of insights long articulated within 20th-century continental phenomenology. Chief among these philosophical precursors is the work of French phenomenologist Maurice Merleau-Ponty. In his 1945 magnum opus, Phenomenology of Perception, Merleau-Ponty launched an attack on both classical intellectualism (Cartesian rationalism) and mechanistic empiricism, arguing that consciousness is not an internal, disembodied spectator looking out upon the world, but is fundamentally an embodied, acting entity. Central to this formulation is the conceptual distinction between the objective physical body (le corps objectif) and the lived body (le corps vécu), the latter organized around what he designated as the “corporeal schema” (schéma corporel)—a pre-reflective, dynamic practical awareness of the body’s posture and action possibilities.

Merleau-Ponty anticipated the mirror mechanism through his radical thesis of “intercorporeality” (intercorporéité). He asserted that intersubjective social perception does not proceed through an intellectual calculation, wherein one observes an external body, recognizes it as anatomically analogous to one’s own, and infers that it must possess an internal subjective mind. Rather, Merleau-Ponty claimed that our relation to the other is an immediate, pre-reflective bodily communion:

“I live in the facial expressions of the other, as I feel him living in mine.”

The parieto-frontal mirror circuitry precisely instantiates this intercorporeality at the neurobiological level. When the sight of an other’s hand reaching for an object instantly recruits the observer’s own premotor corporeal schema, the boundary separating the subjective internal self from the external other is transcended at a pre-reflective sensorimotor level. The observer does not look upon the other as an inert mechanical object (Körper); rather, the other is immediately perceived as an intentional, acting, living body (Leib). Through the mirror neuron system, the corporeal schema operates as an epistemic bridge, validating Merleau-Ponty’s claim that perception and bodily action are structurally intertwined facets of being-in-the-world.

3.2 Husserlian Phenomenology of Intersubjectivity

The roots of phenomenological intersubjectivity stretch back to the founder of the phenomenological movement, Edmund Husserl, whose fifth of the Cartesian Meditations (1931) addressed the profound epistemological dilemma of solipsism: how does the transcendental ego constitute an “alter ego”—another subjectivity distinct from itself yet recognized as possessing its own internal experiential life? Husserl introduced the concept of “pairing” (Paarung), a pre-reflective, associative synthesis wherein two phenomena are given together in an intuitive unity. Through pairing, the physical form of the other is associatively linked with one’s own lived, experienced bodily sensations, precipitating primal empathy (Einfühlung).

Phenomenologists such as Husserl, and subsequently his student Edith Stein in her treatise On the Problem of Empathy (1917), maintained that empathy is not an explicit analogical inference. One does not deduce the other’s pain via a logical syllogism; one directly apprehends the other’s intentionality embodied within their kinetic and affective gestures. The discovery of mirror neurons in macaques and the broader mirror neuron system in humans provides the neurobiological mechanism for Husserl’s Paarung. The parieto-frontal mirror network couples the visual depiction of the alter ego with the primordial first-person motor representations of the self.

This neurobiological pairing bridges transcendental phenomenology with cognitive neuroscience. It demonstrates that our recognition of another person as an intentional agent is constitutive of self-consciousness itself. By activating the identical neural patterns that underwrite the observer’s own agency, the brain instantiates a direct, unmediated validation of the other as an alter ego. This biological resonance reveals that the “We” is not a secondary, intellectualized theoretical deduction, but an experiential state woven into the perceptual-motor apparatus of the primate central nervous system.

3.3 Critique of Classical Cartesian Dualism and Disembodied Cognitivism

The mirror neuron paradigm, and its broader integration into Vittorio Gallese’s theory of embodied simulation, represents a systematic dismantling of classical Cartesian dualism and the disembodied computational paradigms that characterized early cognitive psychology. Under the computational paradigm of mind, pioneered by thinkers like Jerry Fodor and Noam Chomsky, cognition was understood as the syntactic manipulation of abstract, amodal symbols. In this framework, sensory perception was conceptualized merely as an input peripheral, motor control as an output peripheral, and cognition as an insulated central operating system—the “classical sandwich” model.

Mirror neurons challenge this functional demarcation. If the identical neuronal populations that execute motor actions are recruited to perceive, interpret, and predict the actions of other agents, the theoretical boundary separating motor output from cognitive perception dissolves. Perception is fundamentally motoric, and motor control is intrinsically perceptual. The motor system, far from being a biological puppet executing abstract central commands, constitutes the neural substrate of social understanding.

This realization forces a total reconceptualization of mental architectures, moving away from disembodied functionalism toward frameworks of embodied, embedded, enacted, and extended cognition (4E Cognitive Science). Action affordances, first articulated by ecological psychologist J.J. Gibson, are not intellectual categorizations applied post-hoc to sensory data; they are the baseline perceptual units of biological agents. The mirror neuron system demonstrates that the brain does not transform the physical world into an amodal, propositional symbolic language before understanding social interaction. Instead, cognition relies on action-oriented representations where the pragmatic motor repertoire of the organism directly grounds social cognition.

4. Vittorio Gallese’s Theory of Embodied Simulation

4.1 Conceptual Architecture of Embodied Simulation

Building on the empirical discoveries in Parma, Vittorio Gallese developed the Theory of Embodied Simulation, elevating the mirror mechanism from a circumscribed motor-matching circuit to a comprehensive, sub-personal neurocomputational architecture that underwrites social cognition across motoric, emotional, and somatosensory domains. Gallese defines embodied simulation as a mandatory, automatic, pre-reflective, and non-propositional functional mechanism. It operates below the threshold of conscious introspective awareness, serving as a biological reuse mechanism: the brain deploys its own sensory, motor, affective, and viscero-motor neural networks as an experiential simulator to model the internal states, actions, and sensations of others.

The conceptual architecture of embodied simulation rests on the premise that when we observe an external agent, our nervous system does not merely decode the kinematics of their motion; it instantiates a corresponding sub-personal functional state within our own biology. This “neural reuse” occurs without the necessity of explicit motor execution or overt affective contagion. Cortical and subcortical inhibitory mechanisms prevent the simulated motor program from progressing down the pyramidal tract to evoke involuntary muscular mimicry; instead, the activation is maintained as an internal simulation of the observed act or emotion. Through this simulation, the meaning of the observed behavior is grasped directly because the observer experiences a congruent internal functional state.

Crucially, Gallese distinguishes embodied simulation from classical mental simulation. In traditional philosophy of mind, mental simulation was conceived as a high-level, imaginative exercise: an observer consciously projects themselves into the perspective of another, deliberately deliberating “as if” they were in that individual’s circumstances. Embodied simulation, conversely, is non-conscious, pre-reflective, and non-propositional. It is an automatic biological resonance embedded in the sensorimotor architecture itself, providing an immediate somatic translation of the social environment long before conscious deliberation occurs.

4.2 The Shared Manifold Hypothesis

To provide a structured philosophical and neuroscientific foundation for embodied simulation, Gallese formulated the Shared Manifold Hypothesis. This hypothesis addresses the ontological problem of intersubjectivity by proposing that human beings inhabit a shared, relational social space that precedes any formal, cognitive self-other differentiation. Gallese structures the shared manifold into a triad of distinct levels:

• The Phenomenological Level: The subjective experience of identifying others as intentional beings like ourselves, experiencing empathy, shared emotional presence, and mutual intelligibility.
• The Functional Level: The sub-personal simulation routines that convert external visual, acoustic, and contextual signals into internal somatosensory, motor, and affective formats.
• The Sub-Personal (Neural) Level: The discrete neurobiological networks—mirror neurons, the insular cortex, the anterior cingulate cortex, and somatosensory cortices—that simultaneously encode execution and observation, pain experienced and pain witnessed, touch received and touch observed.

The shared manifold gives rise to what Gallese describes as a “we-centric” space. Ontogenetically and phylogenetically, humans do not begin as isolated, solipsistic subjects who slowly learn to construct inferences about the existence of other minds. Rather, our baseline neurobiology is fundamentally relational. The self and the other are instantiated across the identical neural architecture. This shared neural currency allows for interpersonal attunement, enabling humans to operate within a common biological horizon of shared meanings, intentions, and felt somatic realities.

4.3 Reinterpreting Folk Psychology: Simulation vs. Theory of Mind

For decades, cognitive developmental psychology and analytical philosophy of mind were dominated by the “Theory of Mind” (ToM) debate, centered on how human beings understand other agents’ beliefs, desires, and mental states—a capacity known as “folk psychology.” The dispute was polarized between two dominant paradigms: “Theory-Theory” (TT) and “Simulation Theory” (ST). Theory-Theory, advanced by scholars such as Alison Gopnik, Henry Wellman, and Peter Carruthers, argued that children acquire an explicit, theory-like body of folk-psychological laws. According to TT, social cognition is an intellectual, quasi-scientific endeavor wherein an observer uses propositional rules to deduce that an agent possessing belief $P$ and desire $Q$ will perform action $R$.

Conversely, Simulation Theory, defended in its classical, high-level form by philosophers like Alvin Goldman and Robert Gordon, asserted that we do not rely on detached theoretical rules; instead, we project ourselves into the shoes of the target, using our own mind as an off-line cognitive model to simulate their mental states. However, Goldman’s original high-level simulation theory still relied largely on imaginative, propositional perspective-taking.

Gallese, in extensive theoretical collaboration with Alvin Goldman (e.g., Gallese & Goldman, 1998), radically transformed this debate by introducing “low-level embodied simulation” as the foundational scaffold of folk psychology. Gallese demonstrated that before a child can ever process abstract, counterfactual beliefs or engage in propositional reasoning regarding false beliefs, they must possess a low-level, sensorimotor and affective understanding of what other bodies are doing and feeling. The high-level, reflective mindreading network (often identified with the medial prefrontal cortex, the precuneus, and the temporoparietal junction) does not operate in an abstract cognitive vacuum; it is ontogenetically and functionally grounded upon low-level embodied simulation circuits. Embodied simulation provides the immediate, pre-reflective semantic bedrock upon which formal, linguistic, and propositional Theory of Mind is later constructed.

5. The Human Mirror Neuron System (MNS): Evidence and Methodologies

5.1 Non-Invasive Neuroimaging Paradigms

Because microelectrode single-unit recording is an invasive technique reserved primarily for animal models or specific clinical interventions, establishing the existence and functional properties of the human mirror neuron system (MNS) required the deployment of diverse non-invasive neuroimaging and electrophysiological paradigms. Each methodology captures distinct spatiotemporal facets of the mirror mechanism, converging to confirm that an analogous frontoparietal action-observation-execution network operates within the human brain.

Functional Magnetic Resonance Imaging (fMRI) has served as a primary tool, particularly via “fMRI repetition suppression” or “cross-modal adaptation” paradigms. This methodological design exploits the physiological phenomenon wherein neuronal populations exhibit a decreased blood-oxygen-level-dependent (BOLD) signal when an identical cognitive or motor process is engaged sequentially. By demonstrating cross-modal adaptation—where the BOLD response in a specific voxel cluster is attenuated when an executed action follows the observation of that same action (or vice versa)—fMRI studies provide compelling evidence for shared neuronal substrates coding both execution and perception within the identical anatomical coordinates.

Electrophysiologically, the mirror system is investigated via the suppression of the sensorimotor central mu-rhythm. The mu-wave is an oscillatory electroencephalographic (EEG) waveform firing within the 8–13 Hz range over the sensorimotor cortex (electrodes C3, Cz, and C4). In a state of motor rest, these cortical pyramidal neurons fire in synchronized, high-amplitude bursts. When an individual initiates voluntary physical movement, this synchronous activity is disrupted, resulting in amplitude attenuation or “desynchronization.” Crucially, mu-wave desynchronization occurs not only during physical action execution, but also during the passive observation of biological actions performed by others. This attenuation is interpreted as a direct electrophysiological index of premotor mirror resonance. Complementing EEG, Magnetoencephalography (MEG) tracks the millisecond-by-millisecond spatiotemporal cascade of this resonance, demonstrating that visual information processed in the occipital cortex is routed forward to the superior temporal sulcus (STS) within 100 milliseconds, rapidly triggering parietal activation (at ~150–200 ms), and ultimately discharging the primary and premotor motor cortices (at ~250 ms).

Additionally, Transcranial Magnetic Stimulation (TMS) offers causal, physiological verification of mirror motor excitability. In seminal experiments by Fadiga et al. (1995), single-pulse TMS was delivered to the primary motor cortex (M1) while participants passively observed an experimenter executing hand grasping movements. By measuring Motor Evoked Potentials (MEPs) via electromyographic electrodes placed over specific peripheral hand muscles (such as the first dorsal interosseous and abductor pollicis brevis), the researchers demonstrated that corticospinal excitability was selectively elevated in the exact muscles that the observer was passively watching the experimenter use. This proved that action observation automatically primes specific, muscle-congruent motor pathways in the human nervous system.

5.2 Human Anatomical Correlates of the Mirror Mechanism

Through extensive neuroimaging synthesis, the human mirror neuron system has been mapped across a core frontoparietal network, which functions in close coordination with specialized temporal visual processors. The primary nodes of this core network include:

• The Rostral Inferior Parietal Lobule (IPL): Encompassing the angular and supramarginal gyri, as well as the anterior intraparietal area. This region acts as an integrative hub, processing kinesthetic, somatosensory, and visual inputs to code the teleological structure, kinematics, and context of manual actions.
• The Ventral Premotor Cortex (vPMC) and Posterior Inferior Frontal Gyrus (pIFG): Corresponding to Brodmann Area 44 and Brodmann Area 45 (the classic Broca’s area in the left hemisphere) and adjacent premotor Brodmann Area 6. This frontal node translates incoming perceptual schemas into executable motor representations.
• The Superior Temporal Sulcus (STS): While not possessing intrinsic motor properties itself, the STS serves as the essential visual input feed to the core mirror network, parsing biological motion, gaze trajectory, and body orientation, and projecting these rich visual profiles to the parietal lobule.

Beyond this classic tripartite core, the human mirror mechanism features extended components. These include the primary somatosensory cortex (SI and SII), which simulates the tactile, cutaneous feedback associated with observed physical interactions, and the supplementary motor area (SMA) and pre-SMA, which coordinate complex sequence timing, motor readiness, and temporal simulation profiles.

5.3 Direct Intracranial Recordings in Humans

For years, critics contended that indirect methodologies like fMRI, EEG, and TMS, despite their convergence, could not definitively demonstrate the existence of individual mirror neurons in humans, arguing that human activations might merely reflect mixed populations of adjacent, non-mirror sensory and motor neurons blurred together by the coarse spatial resolution of neuroimaging voxels. This methodological objection was directly addressed in 2010 through a breakthrough study published by Mukamel, Ekstrom, Kaplan, Iacoboni, and Fried.

Mukamel and colleagues recorded extracellular single-unit and multi-unit activity directly from 1,177 neurons in the brains of 21 neurosurgical patients undergoing invasive intracranial monitoring for the localization of pharmacoresistant epilepsy. Depth electrodes were implanted into regions dictated entirely by clinical criteria, capturing neuronal activity within the supplementary motor area (SMA), the hippocampus, the parahippocampal gyrus, and the entorhinal cortex while patients executed or observed hand grasping actions and facial expressions.

The recordings provided unequivocal proof: a substantial subset of neurons demonstrated execution-observation matching, firing both when the patient physically executed an action (such as executing a precision grip or smiling) and when the patient observed that same action presented on a video monitor. Astonishingly, Mukamel et al. identified two distinct functional populations: “excitation” mirror neurons, which increased their firing rates during both observation and execution, and “inhibition” mirror neurons, which exhibited robust excitation during physical action execution but marked suppression of firing during action observation. This discovery of inhibition mirror neurons resolved a profound theoretical dilemma: it illuminated the long-sought physiological mechanism that allows an observer to internally simulate an observed action at a sub-personal level while actively suppressing overt, involuntary behavioral mimicry.

6. Neural Substrates of Empathy: Affective and Somatosensory Simulation

6.1 The Insular Cortex and the Vicarious Experience of Disgust

Embodied simulation is not confined to the domain of voluntary skeletal motor actions. In an essential theoretical and empirical expansion, Vittorio Gallese, in collaboration with Christian Keysers, Bruno Wicker, and colleagues, demonstrated that the mirror architecture extends into the viscero-motor and interoceptive affective structures of the primate brain. The most definitive empirical proof of this emotional resonance emerged from studies investigating the neural substrate of disgust.

Disgust is a biologically primitive, viscero-motor emotion engineered to protect an organism from the ingestion of toxic, contaminated, or pathogen-rich substances, accompanied by distinct physiological reflexes including nausea, salivation, gagging, and orofacial expressions. In an influential fMRI experiment, Wicker et al. (2003) exposed human participants to disgusting olfactory tastants (such as the smell of rotten eggs and sewage) and recorded their BOLD responses. In an alternate condition, the same participants passively watched video clips of actors drinking from a glass and displaying spontaneous, authentic facial expressions of intense disgust. The results revealed striking neuroanatomical convergence: both the direct, first-person visceral experience of disgust and the passive observation of disgust in another individual activated the identical anatomical subregion of the anterior insula and the adjacent anterior cingulate cortex (ACC).

The anterior insula acts as a central clearinghouse for interoceptive awareness, monitoring internal bodily sensations and regulating homeostatic, viscero-motor autonomic states. When we witness another person retching or recoiling in disgust, our own anterior insula does not calculate an abstract semantic proposition (“That person is experiencing disgust”); it instantiates a congruent viscero-motor simulation. We understand their visceral state by activating the very cortical coordinates that govern our own somatic disgust. Clinical lesion studies corroborate this finding: neurological patients with localized damage to the anterior insula and basal ganglia lose not only the capacity to experience physiological disgust themselves, but also the specific ability to recognize or categorize facial and vocal expressions of disgust in other human beings. The lived internal capacity for an emotional state is the physiological prerequisite for its intersubjective perception.

6.2 The Pain Matrix and Empathic Resonance

A second foundational axis of affective simulation resides within the neural architecture of pain. Pain is an intrinsically multidimensional phenomenon comprising two dissociable neuroanatomical systems: the “sensory-discriminative” circuit (mediated by the primary and secondary somatosensory cortices SI and SII, and the posterior insula), which encodes the physical location, intensity, and mechanical quality of a noxious stimulus; and the “affective-motivational” circuit (mediated by the dorsal anterior cingulate cortex [dACC] and the anterior insula [AI]), which codes the intrinsic aversiveness, suffering, and subjective distress associated with pain.

In a groundbreaking fMRI investigation of empathic pain resonance, Tania Singer and colleagues (2004) examined female participants while a painful cutaneous electrical shock was delivered directly to their own hand, or to the hand of their romantic partner seated adjacent to the scanner. Singer found that while direct cutaneous nociception activated the full pain matrix (both the sensory-discriminative and affective-motivational networks), observing pain delivered to the loved one recruited the affective-motivational circuit—the anterior insula and the anterior cingulate cortex. The observers internally simulated the emotional suffering and aversive distress of the other person, transforming an observed event into an authentic, first-person affective resonance.

Subsequent TMS and fMRI investigations by Avenanti et al. (2005) demonstrated that under conditions of high spatial and visual resolution—such as watching a video of a hypodermic needle physically puncturing another person’s specific index finger muscle—the simulation cascades into the sensory-discriminative system as well, causing an immediate, highly localized reduction in corticospinal excitability within the observer’s congruent hand muscle. Furthermore, this empathic pain resonance is dynamically modulated by high-level contextual variables. In studies evaluating the influence of social appraisal (Singer et al., 2006), when participants observed a player who had previously acted uncooperatively or unfairly receive a painful shock, the empathic response in the anterior insula and dACC was significantly suppressed (and in male participants, replaced by activation in the nucleus accumbens, a key reward center). The mirror architecture for pain is not a mechanistic automaton; it is an integrated simulation system dynamically tuned by social, contextual, and interpersonal valences.

6.3 Tactile Mirroring and the Secondary Somatosensory Cortex

The reach of embodied simulation encompasses the sense of touch. When we observe another person being touched—whether it is the gentle caress of an arm, a sudden slap, or an insect crawling across skin—our nervous system does not merely catalog the optical coordinates of the collision; it activates our own somatosensory cortices. In a series of pioneering fMRI paradigms led by Christian Keysers and colleagues (2004), healthy human participants were scanned while their legs were tactilely stimulated with a physical brush, and subsequently while they passively watched video clips of another human being’s leg (or an inanimate object) being touched by the same brush. The results revealed that the secondary somatosensory cortex (SII) discharged bilaterally both during direct physical tactile stimulation and during the visual observation of touch applied to another individual.

This somatosensory resonance reveals that tactile perception is an intersubjective domain. However, under typical neurobiological conditions, the observer does not physically feel the observed touch on their own skin; the simulation operates as a sub-threshold, non-conscious mapping. The physiological differentiation between the self and the other is maintained by precise inhibitory controls, largely governed by the right temporoparietal junction (rTPJ) and the anterior insular-prefrontal networks, which attenuate the simulated sensory signal and register that the primary sensory afferents from the skin are currently silent.

When this inhibitory gating mechanism breaks down or exhibits hyper-excitability, the phenomenon of “mirror-touch synesthesia” manifests. Described extensively by Banissy and Ward (2007), individuals with mirror-touch synesthesia literally experience an authentic, physical tactile sensation upon their own bodies when they observe someone else being touched. If they see someone touched on the left cheek, they immediately experience a simultaneous, localized physical sensation on their own right (or left) cheek. Mirror-touch synesthesia provides a dramatic phenomenological demonstration of the shared manifold: it shows that the neural architecture for feeling one’s own bodily sensations and the architecture for perceiving another’s sensations are identical, held apart only by delicate inhibitory thresholds that preserve the boundaries of the corporeal ego.

7. Motor Intentionality and Contextual Action Understanding

7.1 Encoding ‘Why’ Versus ‘What’: The Parieto-Frontal Chain

A central criticism leveled against early formulations of mirror neuron theory was that simple kinematic matching could at best code *what* an action was (e.g., “a hand is closing around a piece of food”), but could not deduce *why* the action was executed (e.g., “the hand is grasping the food to eat it” versus “to discard it”). This fundamental distinction between immediate action identification and teleological intention attribution was definitively elucidated in 2005 through a transformative neurophysiological study conducted by Leonardo Fogassi, Giacomo Rizzolatti, and colleagues.

Fogassi et al. inserted microelectrodes into 165 neurons within the rostral inferior parietal lobule (area PFG) of macaque monkeys. The experimental paradigm was meticulously structured into two primary behavioral conditions. In the first condition, the monkey observed an experimenter reach for a piece of food, grasp it, and bring it to their mouth to eat it (“grasping-to-eat”). In the second condition, the monkey observed the experimenter reach for the identical piece of food using the exact same kinematic trajectory, but grasp it and place it into a small container positioned next to the food (“grasping-to-place”). Crucially, during the initial reaching and grasping phases of the motor act, the raw biomechanical movements, velocity profiles, and muscular kinematics were physically identical across both conditions.

The empirical findings revealed that the vast majority of parietal mirror neurons discharged differentially during the identical grasping phase depending entirely on the ultimate *distal intention* of the action. A neuron that discharged intensely during grasping-to-eat exhibited little to no activity during grasping-to-place, and vice versa. The researchers discovered that parietal mirror neurons are organized into chained, sequential circuits. An individual neuron does not fire in isolation; it belongs to a dedicated functional pathway of “action chains.” The activation of the first link in the chain (reaching/grasping) automatically and predictively activates subsequent links (eating or placing). Consequently, when an observer perceives the initial phase of an action, the downstream teleological chain is pre-reflectively activated. The parieto-frontal mirror network does not merely code kinematic mechanics; it directly computes the distal behavioral intention—the “why”—via predictive motor chaining.

7.2 Contextual Modulation of Motor Resonance

In humans, this capacity to predictively compute distal intentionality through motor resonance is deeply intertwined with environmental and situational context. This contextual modulation was demonstrated in an influential fMRI experiment led by Marco Iacoboni and colleagues in 2005, colloquially designated the “Tea Party” study. Human participants were presented with three conditions of video stimuli: “Action,” “Context,” and “Intention.” The Context clips depicted a table set with teacups, teapots, and pastries in two states: pristine, ready for high tea; or disheveled, covered in crumbs, dirty napkins, and empty cups. The Action clips depicted a hand executing a precision grasp around a teacup against a plain, neutral background, devoid of context. The Intention clips depicted the identical hand executing the exact same grasping action embedded within the two distinct contexts: either the clean table (contextual cues indicating an intention of “drinking”) or the disheveled table (cues indicating an intention of “cleaning up”).

The results revealed that while the Action condition activated the core premotor mirror system, the Intention conditions elicited a massive, highly significant increase in premotor activation, specifically within the posterior inferior frontal gyrus (Brodmann Area 44/45). Furthermore, the premotor activation was markedly differentiated between the drinking intention and the cleaning-up intention. This demonstrated that the human mirror neuron system does not process motor acts in isolated vacuum spaces. Visual contextual cues are immediately synthesized with the observed motor kinematics to dynamically select the appropriate predictive forward model within the motor hierarchy.

This process operates as a computational forward model: the motor system uses context to select the most probable distal intention, anticipating the subsequent behavioral trajectory before it unfolds mechanically. While complex social contexts involving ambiguous motivations or deceptive behaviors require additional engagement from the “mentalizing network” (mPFC, rTPJ, precuneus), the immediate parsing of everyday pragmatic intentions is handled by this bottom-up, context-sensitive parieto-frontal mirror architecture.

7.3 Motor Expertise and Motor Repertoire Constraints

If the direct-matching hypothesis and embodied simulation theory are correct in asserting that action understanding is grounded in an internal motor simulation “from within,” a critical empirical prediction emerges: the degree of mirror neuron activation should correlate directly with the observer’s own motor expertise. An individual who has mastered a physical motor skill should exhibit markedly deeper, more nuanced sensorimotor resonance when observing that specific skill than an individual who lacks the corresponding motor repertoire.

This empirical prediction was tested in a classic fMRI experiment by Calvo-Merino, Glaser, Grèzes, Passingham, and Haggard (2005). The researchers scanned two groups of elite physical experts: classical ballet dancers and expert capoeira dancers (an Afro-Brazilian martial art combining acrobatic kinematics with dance). While inside the scanner, all participants watched short, silent video clips of classical ballet sequences and capoeira sequences. The visual stimuli were completely matched for visual complexity, athletic dynamism, and biological motion. The crucial difference was the somatic motor repertoire of the observers.

The fMRI data revealed a profound double dissociation. When classical ballet dancers watched ballet, their frontoparietal mirror network (vPMC, IPL, and STS) discharged with intense, robust BOLD signals; when they watched capoeira, this activation was significantly attenuated. Conversely, when capoeira dancers watched capoeira, their mirror circuits fired vigorously, but showed marked reduction when observing classical ballet. In a critical follow-up study (Calvo-Merino et al., 2006), the researchers scanned male and female classical ballet dancers viewing male-specific and female-specific dance steps. Because male and female dancers train together every day, their visual familiarity with both types of steps was identical; yet, mirror neuron activation was significantly higher when dancers observed movements belonging to their own gender-specific motor repertoire. This confirmed that the parieto-frontal resonance is driven primarily by internal motor training rather than passive visual familiarity.

Analogous findings have been demonstrated in the auditory-motor domain with expert musicians. Neuroimaging studies by Haueisen and Knösche (2001) and Haslinger et al. (2005) demonstrated that when classically trained pianists passively listen to a piano composition without moving, their primary motor cortex, premotor area, and supplementary motor area display immediate somatotopic activation corresponding to the precise finger movements required to play those musical notes. Their brains play the piano silently in motor formats. The mirror architecture is plastic, continuously sculpted and refined by an individual’s personal sensorimotor biography.

8. The Evolutionary Trajectory: From Gestural Communication to Language

8.1 Broca’s Area as the Evolutionary Homologue to Macaque Area F5

One of the most consequential evolutionary implications of the Parma discoveries concerns the phylogenetic origin of human language. Comparative neuroanatomists, including Giacomo Rizzolatti and Michael Arbib, noted that the monkey ventral premotor area F5 is not an arbitrary motor strip; it represents the precise phylogenetic and cytoarchitectonic homologue of human Broca’s area, specifically Brodmann Area 44 (the pars opercularis) and Brodmann Area 45 (the pars triangularis) in the inferior frontal gyrus.

For over a century, Broca’s area was classified primarily as an executive motor center for linguistic speech production, dedicated to coordinating the vocal apparatus for verbal articulation. However, the identification of mirror neurons within area F5 revealed that this ancestral cortical territory was originally dedicated to executing and parsing manual and orofacial actions: reaching, grasping, holding, tearing, and lip-smacking. In their seminal 1998 paper, “Language within our grasp,” Rizzolatti and Arbib formulated a detailed evolutionary model proposing that human language did not emerge abruptly *de novo* out of primitive vocalizations or primate alarm calls (which are mediated primarily by subcortical, limbic circuits such as the periaqueductal gray and amygdala). Rather, human language evolved out of a flexible, voluntarily controlled gestural communication system grounded in the parieto-frontal mirror architecture.

Under this evolutionary framework, the mirror mechanism solved the “parity problem” in communication: the necessity that an emitted signal must convey the identical meaning to both the sender and the receiver. Because mirror neurons automatically match the execution of a motor gesture with its perceptual observation, the gesture becomes an inherently intersubjective, shared communicative currency. This manual-orofacial mirror substrate provided the biological scaffolding through which primates moved from pragmatic object manipulation, to communicative pantomime, to formal symbolic signing, and ultimately to acoustic speech.

8.2 The Motor Theory of Speech Perception Revisited

In the mid-20th century, Alvin Liberman and his colleagues at Haskins Laboratories formulated the provocative Motor Theory of Speech Perception. Liberman posited that the basic objects of speech perception are not the acoustic sounds of the phonemes themselves, but the intended phonetic gestures of the speaker’s vocal tract (such as lip closure, tongue elevation, and velopharyngeal opening). According to Liberman, listeners do not decipher speech through general-purpose auditory decoding; rather, they decode speech by accessing their own internal motor representations of speech production.

While Liberman’s theory was initially dismissed as computationally unfeasible and biologically ungrounded, the discovery of the mirror neuron system ignited a massive resurgence of the paradigm. Neuroscientists recognized that the mirror mechanism provided the exact neural substrate Liberman had predicted: a biological engine converting acoustic speech signals directly into articulatory motor programs. In an essential TMS experiment, Fadiga, Craighero, Buccino, and Rizzolatti (2002) stimulated the primary motor representation of the tongue while human participants passively listened to spoken words containing either a double “r” sound (which requires intense mechanical tongue-tip vibration in Italian, such as terra) or a double “f” sound (which requires labiodental contact without tongue movement, such as baffo).

The results were unequivocal: passive auditory perception of phonemes requiring tongue movement produced an immediate, highly significant elevation of Motor Evoked Potentials in the listener’s tongue muscles, whereas listening to labiodental phonemes produced no such enhancement. Subsequent fMRI and MEG studies confirmed that listening to speech phonemes selectively engages the premotor and primary motor cortices responsible for producing those precise articulatory gestures (Pulvermüller et al., 2006). The auditory perception of speech is fundamentally an embodied, sensorimotor simulation.

8.3 The Gestural Protolanguage Hypothesis

The convergence of comparative neuroanatomy and speech perception gave rise to the Gestural Protolanguage Hypothesis, advanced by scholars such as Michael Corballis. The central thesis states that the communicative medium of early hominins was not acoustic speech, but a rich system of deictic (pointing) and iconic (pantomimic) manual and facial gestures. Hominin hands, freed from terrestrial locomotion by the advent of bipedalism, possessed high degrees of mechanical freedom, controlled by an expanding parieto-frontal mirror network capable of creating transparent, iconic representations of the physical world.

Pantomimic gestures possess immediate semantic intelligibility: imitating the movement of cracking an egg, stalking prey, or throwing a spear relies on the observer’s parieto-frontal mirror system to unpack the intention instantly without needing a formalized abstract syntax. Over evolutionary time, these iconic manual gestures underwent conventionalization and symbolic compression, gradually shifting toward the mouth and vocal tract. Anthropologist Peter MacNeilage proposed that the rhythmic jaw, lip, and tongue movements of vocal speech evolved directly out of the motor mechanisms of primate chewing, sucking, and lip-smacking—communicative orofacial gestures mediated by the ventral convexity of area F5. Through this evolutionary sequence, the mirror neuron system served as the phylogenetic bridge transforming bodily motor action into modern linguistic syntax.

9. Neurodevelopmental Perspectives and Clinical Implications

9.1 Ontogeny of the Mirror Mechanism: Innate Tuning vs. Hebbian Learning

The developmental emergence of the mirror neuron system has ignited one of the most intense theoretical debates within cognitive neuroscience. The debate centers on whether mirror neurons represent an innate, evolutionarily hardwired adaptation specifically selected for social cognition, or whether they emerge ontogenetically through general-purpose, sensorimotor Hebbian associative learning.

The evolutionary adaptationist model, championed by Rizzolatti and Gallese, points to evidence from infant research. In landmark studies by Meltzoff and Moore (1977), human neonates mere hours or days old were shown to successfully imitate facial gestures, such as tongue protrusion and mouth opening, executed by an adult experimenter. Because a neonate cannot physically see their own face, translating the visually perceived gesture of an adult onto their own unseen motor effectors requires an active cross-modal mapping mechanism—an innate somatic core that Gallese aligns with the early ontogeny of the shared manifold.

Conversely, cognitive scientist Celia Heyes formulated the Associative Sequence Learning (ASL) model. Heyes argues that mirror neurons are not an innate evolutionary adaptation for mindreading, but are the byproducts of associative, domain-general synaptic plasticity. According to the ASL model, an individual is not born with mirror neurons; rather, mirror neurons develop through continuous, contingent Hebbian pairing (“cells that fire together, wire together”). Throughout early development, an infant constantly looks at its own hands while reaching, grasping, and manipulating objects, creating thousands of paired instances where the sensory visual input of a moving hand is paired with the simultaneous motor firing of the premotor cortex. Furthermore, social interactions—such as parents imitating an infant’s facial expressions, playing mirror games, or culturally scaffolded rituals (like dancing or clapping)—provide the environmental contingencies necessary to forge bidirectional sensorimotor connections. Contemporary perspectives increasingly favor an epigenetic synthesis: an initial, coarse phylogenetic biological pre-wiring is continuously calibrated and shaped by rich, Hebbian sensorimotor interactions throughout childhood development.

9.2 The ‘Broken Mirror’ Hypothesis of Autism Spectrum Conditions

In the late 1990s and early 2000s, researchers sought to determine whether dysfunctions within the mirror neuron system might account for the core social, communicative, and empathic difficulties that characterize Autism Spectrum Conditions (ASC). This paradigm, formulated prominently by V.S. Ramachandran and Lindsay Oberman (2006), became known as the “Broken Mirror Hypothesis” of autism.

The empirical foundation of the hypothesis initially derived from EEG investigations of the sensorimotor mu-rhythm. Early studies by Oberman et al. (2005) demonstrated that while neurotypical participants exhibited robust mu-wave suppression both when executing an action and when observing an action, children and adolescents with autism exhibited normal mu suppression during physical action execution, but displayed an absence of mu suppression during the observation of hand movements. Subsequent structural and functional MRI studies reported atypical cortical thickness, reduced BOLD activation, and aberrant functional connectivity within the core pars opercularis of the inferior frontal gyrus and the rostral inferior parietal lobule in individuals with autism during social observation tasks.

However, the broken mirror hypothesis has faced substantial empirical and theoretical challenges. Extensive meta-analyses and rigorous replication efforts (most notably by Antonia Hamilton, 2013) have demonstrated that the mirror neuron system in autistic individuals is often structurally intact and functionally responsive under ecologically valid, engaging conditions. When autistic participants are explicitly attended to the stimuli, or when the observed actions are emotionally meaningful, mirror resonance operates normally. Contemporary consensus has largely shifted away from a simplistic “broken mirror” deficit model toward a more nuanced framework: social-perceptual differences in autism are more likely driven by upstream alterations in social attention, top-down perceptual filtering, and altered sensory processing, which modulate how the mirror system is recruited rather than representing an intrinsic structural absence of the mirror neurons themselves.

9.3 Neurorehabilitation Paradigms: Action Observation Therapy

The translational application of mirror neuron theory has driven transformative innovations in clinical neurorehabilitation, particularly for patients recovering from ischemic stroke, cerebral palsy, and limb amputations. The premier therapeutic manifestation of this research is Action Observation Therapy (AOT).

Developed on the principle of direct matching, AOT is engineered to stimulate neuroplastic reorganization within damaged motor cortices by exploiting the mirror system’s capacity to activate motor pathways without requiring overt physical execution. In standard AOT clinical protocols for chronic stroke patients suffering from hemiparesis, patients systematically observe daily video sequences of mundane, everyday manual actions (e.g., reaching for a cup, turning a key, brushing hair) filmed from a first-person perspective. Immediately following the observational phase, patients are guided to execute the identical physical actions using their paretic limb. Extensive randomized controlled trials have demonstrated that patients receiving AOT exhibit statistically significant, sustained recoveries in motor function, manual dexterity, and daily independence compared to control cohorts receiving standard physical therapy alone.

Simultaneously, the principles of embodied simulation underwrite the success of Mirror Visual Feedback (MVF) Therapy, originally pioneered by V.S. Ramachandran for the alleviation of phantom limb pain. In MVF therapy, an amputee positions a vertical mirror between their intact limb and their phantom limb. By looking into the mirror and executing bilateral, symmetrical movements, the visual reflection of the intact moving arm creates a vivid visual illusion that the phantom limb is moving normally and painlessly. This congruent visuomotor feedback resolves the sensorimotor mismatch in the brain’s parietal corporeal schema, dampening the pathological, agonizing cramping signals of the phantom limb. Today, AOT and MVF are being merged with immersive Virtual Reality (VR) and robotic exoskeleton interfaces, allowing clinicians to build custom-tailored simulation environments that accelerate neural rewiring through targeted, embodied feedback loops.

10. Neuroaesthetics: Embodying Art, Literature, and Cinema

10.1 Bodily Resonance and Empathy in Visual Art

In 2007, Vittorio Gallese formed a collaboration with Columbia University art historian David Freedberg, publishing the landmark treatise “Motion, emotion and empathy in esthetic experience” in Trends in Cognitive Sciences. Freedberg and Gallese sought to rescue the field of aesthetics from purely formalist, intellectual, or semiotic paradigms, arguing that the profound emotional and psychological power of visual art is rooted in pre-reflective bodily resonance mediated by the mirror neuron system.

Their framework operates across multiple artistic dimensions. First, when an observer views a figurative artwork depicting intense physical movement, bodily tension, or profound suffering—such as the struggling musculature of Michelangelo’s Slaves or the twisting, agonizing torment of the Hellenistic statue Laocoön and His Sons—the observer’s frontoparietal mirror circuits and somatosensory cortices simulate the physical forces, strains, and somatic states displayed in the sculpture. The aesthetic experience is not a detached, disembodied intellectual appraisal; it is an authentic, somatic empathic engagement with the depicted body.

Second, and more radically, Freedberg and Gallese extended this principle to abstract and non-figurative art. They demonstrated that observers experience intense sensorimotor resonance not only with depicted human bodies, but also with the *physical traces of the artist’s creative acts*. When viewing the violent, slashed canvases of Lucio Fontana or the explosive, dynamic paint drips of Jackson Pollock, the human brain automatically reconstructs the physical gestures required to produce those marks. Neuroimaging studies utilizing EEG and fMRI have confirmed that looking at a slashed canvas or a forceful brushstroke activates the observer’s own motor and premotor cortices corresponding to the cutting or splashing action. The viewer internally simulates the hand of the artist moving across the canvas. Art appreciation becomes an embodied, intersubjective dialogue between the kinetic agency of the creator and the resonant motor architecture of the spectator.

10.2 Cinematic Immersion and Felt Spatiality

The moving image represents the absolute technological exploitation of the human mirror neuron system. Film theorist Michele Guerra, working in close collaboration with Vittorio Gallese (culminating in their 2015 book The Empathic Screen), demonstrated that cinema is fundamentally an art of embodied simulation. Rather than approaching film through semiotics, Lacanian psychoanalysis, or linguistic structuralism, Guerra and Gallese analyzed cinematic immersion through the lens of sensorimotor resonance and phenomenological film theory.

A crucial neuroscientific insight in cinematic theory concerns the phenomenological status of the camera itself. The invention of dynamic camera mobility—specifically the Steadicam by Garrett Brown in the 1970s—fundamentally altered cinematic cognition. A static camera framing mimics the view of an external spectator looking through a proscenium window. Conversely, a Steadicam shot, which floats through space matching the biological kinematics, acceleration curves, and rhythmic sway of a walking human body, directly engages the viewer’s vestibular and premotor mirror systems. The viewer does not merely look at the cinematic space; they somaticize it, simulating the physical trajectory of the camera as if navigating the environment themselves.

Furthermore, cinematic editing techniques capitalize on the predictive chaining of the mirror network. Fast-paced action editing, sudden match cuts, and kinetic tracking shots exploit the viewer’s forward motor models, triggering rapid premotor anticipations of movement that register as physiological suspense or kinetic exhilaration. Extreme close-ups of human faces—a hallmark of directors like Ingmar Bergman or Sergio Leone—evoke instant, automatic viscero-motor simulations via the anterior insula and cingulate cortices. Cinema does not communicate primarily with our intellectual, propositional reasoning; it speaks directly to our flesh, utilizing the mirror neuron system to orchestrate shared emotional and kinesthetic realities in the dark of the theater.

10.3 Narrative Comprehension and Linguistic Motor Simulation

Does the mirror neuron system disengage when we step away from direct visual and acoustic perception to immerse ourselves in the abstract, symbolic realm of written literature? Neuroscientific investigations into narrative comprehension reveal that the comprehension of written language is itself grounded in embodied sensorimotor simulation. Reading does not consist of decoding arbitrary symbols into an amodal semantic database; it involves the internal sensory and motor reenactment of the described narrative reality.

In groundbreaking fMRI studies led by cognitive neuroscientist Friedemann Pulvermüller and colleagues (2005), human participants read action verbs referring to different physical effectors: face/mouth verbs (e.g., “lick,” “chew”), arm/hand verbs (e.g., “grasp,” “write”), and leg/foot verbs (e.g., “kick,” “stomp”). The results demonstrated clear somatotopic organization within the premotor and primary motor cortices during reading. Reading the word “lick” activated the lateral premotor regions governing the tongue; reading “grasp” activated the dorsal premotor hand representations; and reading “kick” recruited the medial dorsal areas dedicated to the leg. Comprehending the linguistic meaning of an action verb requires the brain to run a rapid, sub-threshold motor simulation of that action.

This linguistic simulation extends to metaphorical, narrative, and poetic prose. Work by Lacey, Stilla, and Sathian (2012) revealed that reading sensory metaphors, such as “he had a rough day” or “she had a velvet voice,” activates the primary and secondary somatosensory cortices, whereas non-metaphorical literal equivalents (“he had a bad day”) engage only standard linguistic circuits. When an author constructs a visceral narrative—describing the chill of freezing wind, the physical exhaustion of a grueling climb, or the acute sensation of a physical wound—the reader’s frontoparietal, somatosensory, and insular mirror circuits simulate those states. Literary empathy is an authentic neurobiological phenomenon: the page becomes an embodied portal, enabling the reader to inhabit the lived bodily experiences of fictional characters through the shared manifold of simulation.

11. Epistemological and Methodological Critiques

11.1 The High-Level vs. Low-Level Mindreading Controversy

Despite its immense explanatory power, the mirror neuron simulation theory has generated profound theoretical debates within analytical philosophy of mind and cognitive science. One of the primary epistemological critiques concerns the profound “explanatory gap” between low-level sensorimotor matching and high-level, propositional mental state attribution—an objection articulated forcefully by philosophers such as Peter Carruthers, Alvin Goldman, and Shannon Spalding.

These critics argue that while the parieto-frontal mirror system can effortlessly explain how an observer recognizes *what* another person is doing (e.g., grasping a glass) or even predict *how* they will interact with immediate objects, it is structurally insufficient to infer complex, high-level propositional attitudes. Social life is dominated by the interpretation of non-perceptual, internal mental states, such as counterfactual beliefs, deceptive intentions, unexpressed desires, and abstract political allegiances. For instance, if an observer watches an executive sign a corporate contract, no degree of motor simulation of the grasping hand can reveal whether the executive believes the contract is legally binding, intends to commit fraud, or is signing under duress.

This critique has led to the development of “dual-system architectures” in social cognitive neuroscience. Scholars increasingly accept that human sociality relies on two complementary, interacting cognitive engines:

1. The Low-Level Sensorimotor System (The Mirror System): A rapid, automatic, pre-reflective heuristic engine that provides immediate, real-time action parsing, emotional resonance, and motor intentionality directly from perceptual cues.
2. The High-Level Mentalizing System (The Theory of Mind Network): A slower, reflective, inferential network mediated by the medial prefrontal cortex (mPFC), the precuneus, and the temporoparietal junction (TPJ), which computes abstract beliefs, counterfactual scenarios, and long-term socio-cultural motives.

Rather than claiming that mirror neurons single-handedly solve the entirety of social cognition, contemporary models position embodied simulation as an indispensable, primary scaffold that feeds rich, bodily grounded inputs upward into the reflective mentalizing network.

11.2 Gregory Hickok’s The Myth of Mirror Neurons

The most sustained, comprehensive neuroscientific critique of the mirror neuron paradigm was mounted by cognitive neuroscientist Gregory Hickok, synthesized definitively in his 2014 book The Myth of Mirror Neurons: The Real Neuroscience of Communication and Cognition. Hickok challenged the foundational thesis that motor system activation is necessary for action perception and speech comprehension, arguing that the Parma group conflated correlation with causation.

Hickok’s critique rests on a broad range of neuropsychological and clinical evidence demonstrating a double dissociation between action execution and action comprehension. First, in the domain of speech perception, Hickok highlighted classic aphasiology: patients with severe Broca’s aphasia, who suffer catastrophic damage to the inferior frontal gyrus and exhibit profound motor speech production deficits (apraxia of speech), typically retain fluent, nuanced auditory speech comprehension. If the motor execution system were strictly necessary to understand speech, damage to the motor articulatory centers should abolish speech perception; empirically, it does not.

Second, Hickok examined motor-perceptual dissociations in limb apraxia. Neurological patients suffering from extensive frontal premotor lesions who have completely lost the motor capacity to execute complex, goal-directed manual actions can still easily categorize, understand, and discriminate those exact same actions when performed by others. Conversely, patients with ventral visual stream damage (such as visual agnosia) often lose the capacity to visually recognize an action while remaining completely capable of physically performing it. Hickok argued that mirror neurons are not the causal engines of action comprehension; rather, they are the computational consequences of motor control. According to Hickok, mirror firing represents motor prediction and preparation generated via sensory-motor integration circuits, functioning to refine the organism’s *own* motor outputs rather than serving as the foundational perceptual engine for social understanding.

11.3 Methodological Limits: Spatial Resolution and Cross-Species Extrapolation

Beyond theoretical disputes, mirror neuron research has confronted substantial methodological critiques regarding spatial resolution, imaging confounds, and the hazards of cross-species extrapolation. A primary methodological critique focuses on the massive leap from macaque single-unit electrophysiology to human functional neuroimaging. In the macaque monkey, mirror neurons were identified as discrete single cells discharging within circumscribed cytoarchitectonic regions. In contrast, the vast majority of human evidence for over a decade relied on fMRI, which measures blood-oxygenation changes across voxels that typically measure 2 to 3 millimeters on each side.

A single standard fMRI voxel contains approximately 5.5 million neurons, along with millions of synapses, glia, and complex microvascular beds. Therefore, observing that an fMRI voxel exhibits BOLD activation during both action execution and action observation does not prove that the *identical single neurons* are discharging across both tasks. It is entirely possible that the voxel contains two intermingled, distinct populations of cells: 2.7 million purely motor neurons and 2.7 million purely visual sensory neurons. This “spatial averaging confound” meant that early human neuroimaging could not definitively claim the existence of mirror neurons at the cellular level (a limitation finally resolved by Mukamel et al.’s intracranial study in 2010, though in a different anatomical region—the SMA and MTL).

Furthermore, critics such as Dinstein et al. (2008) raised serious concerns regarding publication bias and non-replicability in early human mirror neuron studies. Many early fMRI and EEG experiments suffered from small sample sizes, flexible analytical pipelines, and inadequate control conditions for spatial attention, visual salience, and cognitive task demands. Additionally, evolutionary skeptics caution against over-generalizing from non-human primates to humans: macaque monkeys possess robust populations of area F5 mirror neurons, yet macaques are completely incapable of imitation, possess no language, and demonstrate exceptionally limited, rudimentary Theory of Mind capabilities. This profound evolutionary divergence underscores that the mere presence of mirror neurons is insufficient to generate human-level sociality; the human brain must possess unique architectural, connectivity, and cultural scaffolds that fundamentally transform basic motor resonance into high-level social cognition.

12. Future Horizons and the Synthesis of Intersubjective Neuroscience

12.1 Second-Person Neuroscience and Hyperscanning Paradigms

As the mirror neuron simulation paradigm enters its fourth decade, social cognitive neuroscience is undergoing a paradigm shift: the transition from a detached “third-person” neuroscience to an interactive “second-person neuroscience,” championed by Leonhard Schilbach and colleagues. For decades, mirror neuron experiments forced participants to remain entirely passive, isolated observers, lying immobilized in an MRI scanner or seated before a computer monitor watching pre-recorded video clips of hands. Yet, real-world human sociality is not an isolated spectator sport; it is an active, reciprocal, real-time dynamic coupling between interacting agents.

To overcome this ecological limitation, researchers are deploying dual-brain “hyperscanning” methodologies. Using dual-EEG, dual-fNIRS (functional near-infrared spectroscopy), and dual-fMRI systems, hyperscanning allows the continuous, simultaneous recording of two or more human brains engaged in spontaneous, face-to-face interaction, joint physical action, or open-ended dialogue. These studies reveal the emergence of “inter-brain synchronization”—the dynamic phase-locking of neural oscillations across the frontoparietal mirror networks of interacting partners.

Framed through the lens of dynamical systems theory, inter-brain coupling demonstrates that the mirror neuron system does not merely run an internal simulation of an external event; it operates as a continuous, closed-loop sensorimotor interface. During live human cooperation, the predictive motor representations of Agent A immediately shape the behavioral outputs of Agent B, which simultaneously feed back to recalibrate Agent A. The boundary between observer and actor dissolves into a shared, reciprocal system of mutual embodiment, providing empirical realization for Gallese’s vision of a truly intersubjective shared manifold.

12.2 Integrating Computational Neuroscience and Predictive Processing

The contemporary evolution of mirror neuron theory has found its most powerful mathematical and computational formulation within the framework of predictive processing and the Free Energy Principle, pioneered by Karl Friston and adapted to the mirror system by James Kilner (Kilner, Friston, & Frith, 2007).

Under the predictive coding paradigm, the brain is not a passive, reactive stimulus-detector; it is an active, hierarchical Bayesian inference engine. The brain continuously minimizes sensory “prediction error” (variational free energy) by projecting top-down predictions down the cortical hierarchy to meet bottom-up sensory streams. When applied to the mirror neuron system, this framework resolves the historic tension between motor execution and sensory perception through a unified computational architecture:

At the highest level of the cortical hierarchy (the frontal premotor cortex and Broca’s area), the brain generates a prior prediction regarding the distal goal or intention of an observed action (e.g., “grasping the cup to drink”). This teleological prediction is translated down the hierarchy into the inferior parietal lobule (IPL), which translates the intention into a predicted sequence of kinematic movements (e.g., “arm extension, wrist pronation, precision grip”). This kinematic prediction is then projected down to the superior temporal sulcus (STS) and visual cortices, which compare the predicted visual motion against the actual incoming sensory visual stream.

Any discrepancy between what the brain expects to see based on its internal motor simulation and what is actually observed produces a sensory “prediction error.” This prediction error is propagated back up the hierarchy to update and recalibrate the higher-order premotor hypotheses. The mirror neuron system is thus reframed as the biological engine of hierarchical Bayesian active inference for social cognition. Intentionality is not an elusive, mysterious ghost in the machine; it is the mathematically optimal, top-down motor hypothesis that best minimizes prediction error across the parieto-frontal sensorimotor hierarchy.

12.3 Toward an Integrated Model of Human Sociality

The journey of the mirror neuron simulation theory—from its unexpected discovery during single-unit recordings in the macaque arcuate sulcus to its contemporary synthesis within predictive computational neuroscience and phenomenological philosophy—has permanently transformed the landscape of cognitive science. Giacomo Rizzolatti and Vittorio Gallese dismantled the Cartesian dogma that isolated perception, action, and thought into autonomous silos. They proved that the body is not an accidental biological vessel for a disembodied computational mind; rather, our flesh, our visceral systems, and our motor schemas are the fundamental epistemological bedrock through which we understand the social world.

The future of social cognitive neuroscience lies in a grand synthesis: an integrated, multi-tiered architecture of human sociality. In this mature model, low-level embodied simulation and high-level inferential mentalizing are recognized not as warring, mutually exclusive paradigms, but as deeply collaborative, functionally integrated layers of a unified cognitive whole. The mirror neuron system provides the rapid, pre-reflective, somatic attunement that grounds us instantly within the physical actions, emotional distress, and tactile sensations of others. Upon this foundation, language, culture, developmental scaffolding, and the higher-order Theory of Mind networks construct the magnificent, nuanced edifice of human intersubjectivity.

As cognitive science advances into the frontiers of artificial intelligence, social robotics, and brain-computer interfaces, the principles of embodied simulation formulated in Parma remain profoundly vital. To build an artificial agent that truly understands human beings, or to engineer technologies that seamlessly interface with our neural architecture, we cannot rely solely on the manipulation of abstract, disembodied algorithms. We must reckon with the foundational lesson of Rizzolatti and Gallese: to understand another mind is to run a simulation within the living flesh of our own shared, embodied reality.

Conclusion

The mirror neuron simulation theory, pioneered by Giacomo Rizzolatti and Vittorio Gallese, represents an intellectual paradigm shift in our understanding of the primate mind. By demonstrating that the motor system is fundamentally implicated in perception, action interpretation, and empathic attunement, their work bridged the historical divide between neurophysiology, the philosophy of embodied cognition, and phenomenology. The discovery that the same neural substrates code both the first-person execution and the third-person observation of actions, emotions, and sensations revealed that social understanding is not merely an abstract, inferential computation, but an immediate, experiential capacity grounded in the body.

Through Vittorio Gallese’s formulation of the Shared Manifold Hypothesis and embodied simulation, this neurobiological discovery was developed into an overarching framework for human sociality. Despite rigorous debates, computational reformulations within predictive coding, and critiques regarding the explanatory limits of motor resonance, the legacy of the Parma group remains foundational. They showed that the mind is inherently embodied, relational, and attuned to others from the inside out—transforming our conception of what it means to perceive, to feel, and to inhabit a shared human world.

References

• Avenanti, A., Bueti, D., Galati, G., & Aglioti, S. M. (2005). Transcranial magnetic stimulation highlights the sensorimotor mapping of thin needle penetration into hands. Neuron, 46(4), 655–666. https://doi.org/10.1016/j.neuron.2005.03.011
• Banissy, M. J., & Ward, J. (2007). Mirror-touch synesthesia is linked with empathy. Nature Neuroscience, 10(7), 815–816. https://doi.org/10.1038/nn1926
• Buccino, G., Lui, F., Canessa, N., Patteri, I., Lagravinese, G., Benuzzi, F., Porro, C. A., & Rizzolatti, G. (2004). Neural circuits involved in the recognition of actions performed by nonconspecifics: An FMRI study. Journal of Cognitive Neuroscience, 16(1), 114–126. https://doi.org/10.1162/089892904322755601
• Calvo-Merino, B., Glaser, D. E., Grèzes, J., Passingham, R. E., & Haggard, P. (2005). Action observation and acquired motor skills: An FMRI study with expert dancers. Cerebral Cortex, 15(8), 1243–1249. https://doi.org/10.1093/cercor/bhi007
• Calvo-Merino, B., Grèzes, J., Glaser, D. E., Passingham, R. E., & Haggard, P. (2006). Seeing or doing? Influence of visual and motor familiarity in action observation. Current Biology, 16(19), 1905–1910. https://doi.org/10.1016/j.cub.2006.07.036
• di Pellegrino, G., Fadiga, L., Fogassi, L., Gallese, V., & Rizzolatti, G. (1992). Understanding motor events: A neurophysiological study. Experimental Brain Research, 91(1), 176–180. https://doi.org/10.1007/BF00230027
• Dinstein, I., Thomas, C., Behrmann, M., & Heeger, D. J. (2008). A mirror up to nature. Current Biology, 18(1), R13–R18. https://doi.org/10.1016/j.cub.2007.11.004
• Fadiga, L., Craighero, L., Buccino, G., & Rizzolatti, G. (2002). Speech listening specifically modulates the excitability of tongue muscles: A TMS study. European Journal of Neuroscience, 15(2), 399–402. https://doi.org/10.1046/j.1460-9568.2002.01960.x
• Fadiga, L., Fogassi, L., Pavesi, G., & Rizzolatti, G. (1995). Motor facilitation during action observation: A magnetic stimulation study. Journal of Neurophysiology, 73(6), 2608–2611. https://doi.org/10.1152/jn.1995.73.6.2608
• Fogassi, L., Ferrari, P. F., Gesierich, B., Rozzi, S., Chersi, F., & Rizzolatti, G. (2005). Parietal lobe: From action organization to intention understanding. Science, 308(5722), 662–667. https://doi.org/10.1126/science.1106097
• Freedberg, D., & Gallese, V. (2007). Motion, emotion and empathy in esthetic experience. Trends in Cognitive Sciences, 11(5), 197–203. https://doi.org/10.1016/j.tics.2007.02.003
• Gallese, V. (2001). The ‘shared manifold’ hypothesis: From mirror neurons to empathy. Journal of Consciousness Studies, 8(5–7), 33–50. https://doi.org/10.1080/09515080110056801
• Gallese, V. (2005). Embodied simulation: From neurons to phenomenal experience. Phenomenology and the Cognitive Sciences, 4(1), 23–48. https://doi.org/10.1007/s11097-005-4737-z
• Gallese, V., Fadiga, L., Fogassi, L., & Rizzolatti, G. (1996). Action recognition in the premotor cortex. Brain, 119(2), 593–609. https://doi.org/10.1093/brain/119.2.593
• Gallese, V., & Goldman, A. (1998). Mirror neurons and the simulation theory of mind-reading. Trends in Cognitive Sciences, 2(12), 493–501. https://doi.org/10.1016/S1364-6613(98)01262-5
• Gallese, V., & Guerra, M. (2015). The empathic screen: Cinema and neuroscience. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198758037.001.0001
• Hamilton, A. F. de C. (2013). Reflecting on the mirror neuron system in autism: A systematic review of current theories and evidence. Neuropsychologia, 51(1), 91–108. https://doi.org/10.1016/j.neuropsychologia.2012.11.021
• Haslinger, B., Erhard, P., Altenmüller, E., Schroeder, U., Boecker, H., & Ceballos-Baumann, A. O. (2005). Transmodal sensorimotor networks during action observation in professional pianists. Journal of Cognitive Neuroscience, 17(2), 282–293. https://doi.org/10.1162/0898929053124893
• Haueisen, J., & Knösche, T. R. (2001). Involuntary motor activity in pianists evoked by music perception. Journal of Cognitive Neuroscience, 13(6), 786–792. https://doi.org/10.1162/08989290152541449
• Heyes, C. (2010). Where do mirror neurons come from? Neuroscience & Biobehavioral Reviews, 34(4), 575–583. https://doi.org/10.1016/j.neubiorev.2009.11.007
• Hickok, G. (2014). The myth of mirror neurons: The real neuroscience of communication and cognition. W. W. Norton & Company.
• Iacoboni, M., Molnar-Szakacs, I., Gallese, V., Buccino, G., Mazziotta, J. C., & Rizzolatti, G. (2005). Grasping the intentions of others with one’s own mirror neuron system. PLoS Biology, 3(3), e79. https://doi.org/10.1371/journal.pbio.0030079
• Keysers, C., Wicker, B., Gallese, V., & Rizolatti, G. (2004). A touching sight: SII/PV activation during the observation and experience of touch. Neuron, 42(2), 335–346. https://doi.org/10.1016/S0896-6273(04)00156-4
• Kilner, J. M., Friston, K. J., & Frith, C. D. (2007). Predictive coding: An account of the mirror neuron system. Cognitive Processing, 8(3), 159–166. https://doi.org/10.1007/s10339-007-0170-2
• Kohler, E., Keysers, C., Umiltà, M. A., Fogassi, L., Gallese, V., & Rizzolatti, G. (2002). Hearing sounds, understanding actions: Action representation in mirror neurons. Science, 297(5582), 846–848. https://doi.org/10.1126/science.1070311
• Lacey, S., Stilla, R., & Sathian, K. (2012). Metaphorically feeling: Comprehending textural metaphors activates somatosensory cortex. Brain and Language, 120(3), 416–421. https://doi.org/10.1016/j.bandl.2011.05.002
• Meltzoff, A. N., & Moore, M. K. (1977). Imitation of facial and manual gestures by human neonates. Science, 198(4312), 75–78. https://doi.org/10.1126/science.198.4312.75
• Mukamel, R., Ekstrom, A. D., Kaplan, J., Iacoboni, M., & Fried, I. (2010). Single-neuron responses in humans during execution and observation of actions. Current Biology, 20(8), 750–756. https://doi.org/10.1016/j.cub.2010.02.045
• Oberman, L. M., Hubbard, E. M., McCleery, J. P., Altschuler, E. L., Ramachandran, V. S., & Pineda, J. A. (2005). EEG evidence for mirror neuron dysfunction in autism spectrum disorders. Cognitive Brain Research, 24(2), 190–198. https://doi.org/10.1016/j.cogbrainres.2005.01.014
• Pulvermüller, F., Hauk, O., Nikulin, V. V., & Ilmoniemi, R. J. (2005). Functional links between motor and language systems. European Journal of Neuroscience, 21(3), 793–797. https://doi.org/10.1111/j.1460-9568.2005.03900.x
• Ramachandran, V. S., & Oberman, L. M. (2006). Broken mirrors: A theory of autism. Scientific American, 295(5), 62–69. https://doi.org/10.1038/scientificamerican1106-62
• Rizzolatti, G., & Arbib, M. A. (1998). Language within our grasp. Trends in Neurosciences, 21(5), 188–194. https://doi.org/10.1016/S0166-2236(98)01260-0
• Rizzolatti, G., & Craighero, L. (2004). The mirror-neuron system. Annual Review of Neuroscience, 27(1), 169–192. https://doi.org/10.1146/annurev.neuro.27.070203.144230
• Rizzolatti, G., Fadiga, L., Gallese, V., & Fogassi, L. (1996). Premotor cortex and the recognition of motor actions. Cognitive Brain Research, 3(2), 131–141. https://doi.org/10.1016/0926-6410(95)00038-0
• Rizzolatti, G., & Sinigaglia, C. (2008). Mirrors in the brain: How our minds share actions and emotions. Oxford University Press.
• Schilbach, L., Timmermans, B., Reddy, V., Costall, A., Bente, G., Schlicht, T., & Vogeley, K. (2013). Toward a second-person neuroscience. Behavioral and Brain Sciences, 36(4), 393–414. https://doi.org/10.1017/S0140525X12000660
• Singer, T., Seymour, B., O’Doherty, J., Kaube, H., Dolan, R. J., & Frith, C. D. (2004). Empathy for pain involves the affective but not sensory components of pain. Science, 303(5661), 1157–1162. https://doi.org/10.1126/science.1093535
• 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(7075), 466–469. https://doi.org/10.1038/nature04271
• Umiltà, M. A., Kohler, E., Gallese, V., Fogassi, L., Fadiga, L., Keysers, C., & Rizzolatti, G. (2001). I know what you are doing: A neurophysiological study. Neuron, 31(1), 155–165. https://doi.org/10.1016/S0896-6273(01)00369-2
• Wicker, B., Keysers, C., Plailly, J., Royet, J. P., Gallese, V., & Rizzolatti, G. (2003). Both of us disgusted in My insula: The common neural basis of seeing and feeling disgust. Neuron, 40(3), 655–664. https://doi.org/10.1016/S0896-6273(03)00679-2