Comprehensive Study Notes on the Neurobiology and Philosophy of Consciousness

Historical Evolution of Philosophy of Mind and Cognitive Science

  • Mind versus Consciousness:

    • The Mind (Mente): The overall complex of cognitive operations, processes, and faculties (ranging from somatosensory experiences to self-awareness) that an individual performs to gain knowledge of the world and react to it through behavior.

    • Consciousness (Coscienza): The subjective awareness of cognitive operations; it represents the phenomenological inner experience, or "what it feels like" (ciò che si prova).

  • Ancient Philosophical Foundations:

    • Socrates: Shifted philosophy from the search for a single universal generating principle (Arché) to anthropology, interiority, and the rigorous search for oneself and one's own consciousness.

    • Plato: Formulated classical Dualism, postulating a fundamental separation between the physical world and the world of immaterial forms/soul.

    • Aristotle: Formulated classical Monism, viewing the soul as the formal organization or hylomorphic structure of the biological body.

  • Emergence of Philosophy of Mind and Cognitive Science in the 20th Century:

    • Philosophy of mind matured into an autonomous discipline in the 20th century following the cognitive turn (svolta cognitiva), logic and mathematical breakthroughs, linguistics, computer science, and neuroscience.

    • Cognitive Hexagon (Esagone Cognitivo): The interdisciplinary domain defining Cognitive Science, comprising six key fields:

      1. Philosophy (Filosofia)

      2. Linguistics (Linguistica)

      3. Anthropology (Antropologia)

      4. Neuroscience (Neuroscienze)

      5. Artificial Intelligence (Intelligenza Artificiale)

      6. Psychology (Psicologia)

    • Core Epistemological Postulates of Cognitive Science:

      • Cognitive activities are the primary object of study and must be analyzed at a level separate from biological and socio-cultural levels.

      • The electronic computer serves as the mandatory reference model for human mental operations.

  • René Descartes and Modern Mind-Body Thought:

    • Regarded as the father of modern philosophy because he criticized traditional knowledge and grounded truth in human reason rather than in Being (Greek philosophy) or God (Medieval philosophy).

    • Methodological and Hyperbolic Doubt (Dubbio Iperbolico):

      • Sensory knowledge is unreliable because senses deceive us and dream experiences are indistinguishable from waking states.

      • Mathematical knowledge seems universally true in both dreams and waking states, but could be an illusion engineered by an Evil Demon (Genio del Male).

    • The Cogito Principle: The act of doubting proves the existence of the thinking subject (Cogito, ergo sum).

    • Descartes' Tripartite Classification of Ideas:

      1. Innate Ideas (Idee innate): Present from birth; concern the inherent capacity to think and form ideas.

      2. Adventitious Ideas (Idee avventizie): Originate from external physical things and are captured via sensory perception.

      3. Factitious Ideas (Idee fittizie): Created or invented by the subject (e.g., chimeras or mythical creatures).

    • Properties of Matter:

      • Objective / Quantitative Properties: Real, measurable, extended properties satisfying the criterion of evidence: size, figure, movement, position, duration, and number.

      • Subjective / Qualitative Properties: Unreal properties not belonging to bodies themselves, but representing internal responses to external stimuli: color, taste, smell, sound.

    • Cartesian Substance Dualism:

      • Res Cogitans (Thinking Substance): Immaterial, unextended, intense, conscious, and free.

      • Res Extensa (Extended Substance): Material, spatial, unthinking, determined, and governed by mechanical laws.

      • Pineal Gland (Ghiandola Pineale): The specific anatomical locus identified by Descartes where res cogitans and res extensa interact.

    • Historical Merits of Cartesian Dualism:

      • Eliminated anthropomorphic, teleological, magical, and astrological explanations of nature, giving rise to modern deterministic physical science.

      • Safeguarded human spiritual freedom from mechanical physical laws.

      • Allowed biology and medicine to study the body as a machine while paving the way for psychology and cognitive science (comparing mind to software and brain to hardware).

  • 20th-Century Dualist and Non-Reductive Frameworks:

    • John C. Eccles:

      • Applied quantum field probability hypotheses to synaptic neurophysiology.

      • Proposed that the immaterial mind acts as an abstract probability field without physical or energetic mass.

      • Introduced Psychons (Psiconi): Immaterial mental units that interact with cortical dendritic units (Dendrons) to select synaptic vesicles for exocytosis via the quantum uncertainty principle.

    • Karl Popper:

      • Proposed a causal dualism between mind and brain organized into Three Worlds:

        • World 1: The physical world of objects, states, and the biological brain.

        • World 2: The subjective world of mental states, thought processes, and consciousness.

        • World 3: The objective world of cultural products, scientific theories, language, and art.

      • Argued that classical materialism died with the concept of physical forces (e.g., gravity, electromagnetic fields), which are immaterial yet causally effective.

    • Thomas Nagel:

      • Contested physicalist reductions of consciousness, emphasizing the necessity of the subjective "point of view."

      • Bat Thought Experiment: Asking "What is it like to be a bat?" highlights that bats experience the world through echolocation/sonar—a perceptual system radically different from human senses. To know what it feels like to be a bat, one must share its specific sensory system.

      • Formulated Property Dualism (Dualismo delle proprietà): There is only one material entity (the brain), but it possesses both physical properties and non-physical/phenomenological properties.

    • Donald Davidson:

      • Formulated Anomalous Monism: Mental events are identical to physical events, but mental types cannot be captured by physical nomological laws.

      • Introduced the Supervenience (Sopravvenienza) of the mental on the physical: Two events cannot differ in any mental aspect without also differing in some physical aspect.

  • Historical Transition to Naturalistic Models of Mind:

    • Thomas Hobbes: Viewed thought as mechanical, arithmetic calculation over physical brain elements; considered a conceptual progenitor of Artificial Intelligence.

    • David Hume: Formulated a naturalistic Bundle Theory (Teoria del fascio) of the mind. Denied a persistent, unified Self, asserting that humans are merely bundles of constantly changing perception, comparing the mind to a theater.

    • Immanuel Kant: Solved the problem of mental unity by asserting that experience requires a subject. Introduced the Empirical Self and the transcendental unity of apperception, where the mind operates as an active organ structuring raw sensations into unified conscious thought.

    • Theodor Fechner: Advanced psychophysics with the Weber-Fechner Law, mathematically quantifying mental processes by establishing that subjective sensation intensity SS is proportional to the logarithm of physical stimulus intensity II:         S=Kln⁡(I)S = K \ln(I)

Foundations of Artificial Intelligence and Formal Logic

  • Mathematization of Logic and Symbolic Formalism:

    • George Boole:

      • Unified logic and mathematics through Boolean Algebra, transforming logical reasoning into algebraic calculation using binary states (00 and 11).

      • Distinguished between primary propositions (classes/concepts) and secondary propositions (judgments/statements).

      • Calculated primitive recursive functions, such as the successor function:             S(x)=x+1S(x) = x + 1

    • Gottlob Frege:

      • Overcame Boolean dualism by inventing Ideography (Begriffsschrift), an autonomous formal language independent of arithmetic notation.

      • Introduced the universal quantifier (∀\forall) and conditional link (→\rightarrow). For example, transformed the primary proposition "All men are mortal" into the formal conditional statement:             ∀x(U(x)→M(x))\forall x (U(x) \rightarrow M(x))

      • Replaced traditional subject-predicate grammar with the mathematical concept of Function and Argument, laying the formal foundation for the Computational Theory of Mind.

  • Mechanical Calculation Milestones:

    • Wilhelm Schickard (1623): Invented the first automatic mechanical calculating clock.

    • Blaise Pascal (1642): Developed the Pascalina, utilizing interlocking gears and an automatic carry mechanism to perform additions and subtractions.

    • Gottfried Wilhelm Leibniz (1671): Extended the Pascalina to execute multiplication and division mechanically.

    • Charles Babbage and the Analytical Society (1812–1833):

      • Designed the Difference Engine for computing mathematical tables.

      • Designed the Analytical Engine, introducing total programming, store/memory, mill/factory processing, conditional jumps, and punched card instructions.

  • Alan Turing and the Algorithmic Paradigm:

    • Definition of Algorithm: An effective, mechanical procedure satisfying completeness, non-ambiguity, and internal coherence.

    • Turing Machine Model: A theoretical computational device consisting of:

      1. An infinite paper tape divided into discrete cells.

      2. A reading/writing head ("eye-hand").

      3. An alphabet of basic symbols (S1,S2,…,SnS_1, S_2, \dots, S_n).

      4. A set of internal states (q1,q2,…,qnq_1, q_2, \dots, q_n).

      5. Local directional instructions: Right (DD), Left (SS), or Stay (FF).

      • Example quintuple rule describing operation:             q1S1S2Dq2q_1 S_1 S_2 D q_2             (If in state q1q_1 and reading symbol S1S_1, replace S1S_1 with S2S_2, move Right (DD), and transition to state q2q_2).

    • Universal Turing Machine (UTM): A Turing machine capable of simulating any other specific Turing machine by reading its instructions from the tape, providing the theoretical blueprint for modern programmable computers.

    • The Turing Test: An operationalist test for machine intelligence stating that if an interrogator cannot distinguish a machine's responses from a human's, the machine exhibits human-equivalent intelligence.

    • John Searle's Chinese Room Paradox: A counter-argument to the Turing Test. A person inside a closed room manipulating Chinese symbols using syntactical rulebooks can pass as a native speaker without understanding any Chinese. Demonstrates that pure formal syntax (symbol manipulation) does not produce semantical understanding or intrinsic Intentionality.

  • Birth of AI and Cognitive Computer Architectures:

    • Dartmouth Conference (1956): Organized by Marvin Minsky and John McCarthy (who coined the term "Artificial Intelligence"), alongside Allen Newell and Herbert Simon.

    • John von Neumann: Invented the Stored-Program Architecture (implemented in machines like JOHNNIAC), integrating data and instructions in the same internal memory space using binary code.

    • Physical Symbol System Hypothesis (Newell & Simon): Asserts that a physical symbol system has the necessary and sufficient means for general intelligent action. Intelligence is functional and independent of its physical substrate (whether biological brain tissue or silicon hardware).

Cognitive Psychology, Linguistics, and Theoretical Paradigms

  • Evolution of Experimental Psychology:

    • Wilhelm Wundt: Founded scientific psychology using controlled experimental introspection to measure mental processes (sensation, perception, memory), treating individual laboratory variance as measurement error.

    • John B. Watson and Behaviorism: Rejected internal mental states as unscientific, restricting psychology to objectively observable Stimulus-Response (S−RS-R) behaviors.

    • Gilbert Ryle and Philosophical Behaviorism: Criticized mentalistic language as a source of Category Errors (Errori Categoriali) and rejected the concept of private mental language in favor of public, intersubjective language games.

  • Noam Chomsky's Cognitive Revolution:

    • Refuted the Behaviorist S−RS-R formula, demonstrating that an internal cognitive mediator (the Mind) must process stimuli and generate responses.

    • Introduced the Language Acquisition Device (LAD) and Universal Grammar (Grammatica Universale): Innate, genetically inherited, non-conscious syntactic parameters allowing children to rapidly acquire native languages despite sparse input (poverty of the stimulus).

    • Formulated Generative Grammar: Language is produced by rule-governed formal mental structures.

    • George Lakoff: Coined the term Cognitive Unconscious (Inconscio cognitivo) to denote innate, automatic mental structures operating below awareness to organize human symbolic activity.

  • Computational-Representational Theory of Mind (CRTM / TCRM):

    • Conceives mental processes as computations over symbolic representations.

    • Hierarchical breakdown of cognitive processes: Complex intelligent behavior emerges from a hierarchy of simple, non-intelligent sub-processes down to mechanical electronic logic gates (flip-flops).

    • John Haugeland's Motor Analogy: A complex cognitive system functions like an engine; its macro-level capacity emerges from functional interactions among non-intelligent micro-components.

    • Hilary Putnam's Functionalism:

      • Identifies mental states with the functional states of a computational machine.

      • Extended machine functionalism through:

        1. Complexity: Replicating complex human state transitions.

        2. Extended Turing Machines: Interacting with external environments ("oracles").

        3. Probabilistic Automata: Assigning statistical probability distributions to state-action outputs to mimic human uncertainty.

  • Propositional Attitudes and Intentionality:

    • Intentionality (Franz Brentano): The intrinsic property of mental states to be directed about or towards objects/events in the world.

    • Propositional Attitudes: Mental states expressed through psychological verbs taking propositional clauses (e.g., "Enzo believes that the sun is a star").

    • Jerry Fodor's Language of Thought (LOT / Mentalese):

      • States that computational operations occur on an innate, sub-personal symbolic language (Mentalese).

      • Mental states operate as syntactic manipulations of formulas in Mentalese.

    • Three Levels of Mental Representations in Cognitive Psychology:

      1. Neurophysiological: Neural spiking and brain tissue activity.

      2. Informatic / Algorithmic: Procedural execution code.

      3. Linguistic: Propositional attitudes and semantically interpretable symbols.

Cognitive Architectures, Memory, and Representational Formats

  • Shift to Content-Based Cognitive Models:

    • George A. Miller (1956): Published The Magical Number Seven, Plus or Minus Two, establishing an absolute informational bottleneck in human short-term memory capacity:         7±2 items7 \pm 2\ \text{items}

    • Atkinson and Shiffrin Modal Model: Structured memory into three sequential stores: Sensory Register, Short-Term / Working Memory, and Long-Term Memory.

    • Roger Shepard and Stephen Metzler: Conducted seminal 3D Mental Rotation Experiments. Demonstrated that mental rotation latency increases linearly with the angular rotation difference between target shapes, proving that mental images preserve analog spatial-temporal properties.

  • Formats of Mental Representation:

    1. Propositional Representations: Abstract, predicate-argument formal structures (P(x,y)P(x, y)) tied to logic and language truth-values.

    2. Pictorial / Imagery Representations: Spatial-visual matrix representations.

    3. Sensorimotor Representations: Grounded directly in motor execution and sensory feedback.

  • Complex Structural Representations:

    • Semantic Networks (Collins & Quillian): Hierarchical node-link structures where specific traits are stored at subordinate nodes and shared traits are inherited from superordinate nodes (e.g., "bird" inherits "breathes" from "animal").

    • Frames (Marvin Minsky): Data structures representing stereotypical situations or visual scenes.

    • Minsky's Society of Mind: Model of mind composed of many small, specialized, non-conscious agents interacting cooperatively without a centralized processor (contrasting Fodorian isolated modules).

  • David Marr's Computational Theory of Vision:

    • Asserted that neurobiology alone is insufficient to explain cognitive states, requiring three distinct analytical levels:

      1. Computational Level: Defines what problem the system solves and why (goal and logical strategy).

      2. Algorithmic / Representational Level: Defines how the problem is solved (input/output representations and transformation algorithms).

      3. Physical / Implementational Level: Defines the physical/biological substrate realizing the algorithm (neurons or silicon hardware).

  • Stephen Kosslyn's Visual Imagery Model:

    • Visual Buffer: A spatial matrix in short-term memory acting as a transient display medium.

    • Long-Term Memory Archives: Stored structural data used to generate spatial scenes in the buffer.

  • Critique of Classic Cognitivist Dualism:

    • Classic Cognitivism established a new functional dualism separating software (mind) from hardware (brain).

    • Limitations of classic AI (e.g., IBM's Deep Blue defeating Garry Kasparov): Performs brute-force calculations without semantic comprehension, emotion, embodiment, evolutionary history, or subjective qualia.

Neuroscientific Methodology and Neural Correlates of Consciousness (NCC)

  • Definition of Neural Correlates of Consciousness (NCC):

    • The minimal neural mechanisms jointly sufficient for any one specific conscious percept or experience (quale).

    • Distinct from enabling background conditions (e.g., heart rate, brainstem reticular formation, metabolic support).

  • Empirical Research Methodologies:

    • In Vivo Recording: Direct insertion of microelectrodes for single-unit extracellular recordings of individual neurons or local populations (conducted in animal models or surgical epilepsy patients).

    • In Vitro Methods: Useful for single-channel ion kinetics, but incapable of investigating conscious states.

  • Non-Invasive Human Neuroimaging:

    • High Temporal / Low Spatial Resolution: Electroencephalography (EEG) and Magnetoencephalography (MEG).

    • High Spatial / Low Temporal Resolution: Functional Magnetic Resonance Imaging (fMRI) and Positron Emission Tomography (PET) (which measure delayed hemodynamic/blood-oxygen-level-dependent responses).

    • Hybrid Approaches: Simultaneous EEG-fMRI recordings.

  • Key Experimental Paradigms in Consciousness Research:

    • Benjamin Libet's Experiments:

      • Voluntarily Initiated Actions: Found that neural Readiness Potential (Bereitschaftspotential) in motor cortex precedes conscious intent (W-time) by approximately 350−400 ms350-400\,\text{ms}, and precedes action execution by approximately 500 ms500\,\text{ms}.

      • Direct Somatosensory Stimulation: Cortical stimulation required a continuous pulse train duration of at least 0.5 s0.5\,\text{s} (500 ms500\,\text{ms}) to evoke conscious sensation.

      • Backward Referral Hypothesis (Riferimento retrorso): The brain back-dates conscious awareness by 0.5 s0.5\,\text{s} to the onset of the peripheral sensory impulse.

    • Cutaneous Rabbit Illusion (Coniglietto cutaneo): Rapid sequential mechanical taps to the forearm cause perceived taps at equidistant un-stimulated intermediate locations, proving that tactile consciousness is retroactively synthesized across time.

    • Rodney Cotterill's Vital Triangulation: Consciousness depends on continuous re-entrant processing loops connecting sensory cortex, frontal cortex, and thalamic nuclei.

  • Perceptual Instability Paradigms:

    • Ambiguous / Bistable Figures: Physical stimulus remains constant while conscious experience (qualia) toggles between two interpretations (e.g., Rubin's Vase, Necker Cube, Duck-Rabbit figure).

    • Binocular Rivalry: Presenting incompatible images to each eye causes the brain to alternate perceptually between the images rather than fusing them.

    • Frequency Tagging: Oscillating rival stimuli at distinct flickering frequencies to trace their unique neural signatures in EEG/MEG signals.

    • Suppression Paradigms: Flash Suppression and Continuous Flash Suppression (CFS) (using high-contrast flickering noise in one eye to unconsciously suppress target images in the other eye for extended durations).

  • Neural Mechanisms of Qualia Integration:

    • High Neural Activation: Consciously perceived stimuli evoke neural activation rates 5−10 times5-10\ \text{times} higher than unperceived stimuli.

    • Gamma Synchronization: High-frequency bioelectromagnetic oscillatory activity in the gamma band (40−150 Hz40-150\,\text{Hz}) mediates Binding—the integration of isolated features (color, motion, shape) into unified conscious percepts.

    • Winner-Take-All Dynamics: Competitive inhibitory neural circuits select a single winning representation for conscious access.

    • Francis Crick and Christof Koch's NCC Postulates:

      • NCC must form an explicit representation projected directly to frontal/effector motor areas.

      • Primary Visual Cortex (V1) is not a direct substrate for visual consciousness because V1 neurons have simple receptive fields and lack direct projections to frontal motor execution centers.

Functional Anatomy of the Visual System and Brain Rhythms

  • Electroencephalographic (EEG) Brain Rhythms:

    • Delta Waves: Frequency around 2 Hz2\,\text{Hz}; high amplitude. Typical of deep sleep (NREM Stage 3/4).

    • Theta Waves: Frequency around 4 Hz4\,\text{Hz}; medium amplitude. Characteristic of light sleep and deep relaxation.

    • Alpha Waves: Frequency around 10 Hz10\,\text{Hz}; low amplitude. Associated with relaxed wakefulness (eyes closed) and REM sleep.

    • Beta Waves: Frequency around 20 Hz20\,\text{Hz}; low amplitude. Present during active wakefulness, problem-solving, and cognitive load.

    • Gamma Waves: Frequency range 40−150 Hz40-150\,\text{Hz} (minimal amplitude). Crucial for feature binding, perceptual integration, and conscious awareness.

    • Systematic Relationship: An inverse relationship exists between wave amplitude and consciousness level, while a direct relationship exists between wave frequency and conscious content.

  • Retinal Microarchitecture:

    • Retina thickness: Approximately 0.25 mm0.25\,\text{mm}.

    • Five Layers of Retinal Cells:

      1. Photoreceptors (Cones for daylight color vision; Rods for low-light monochromatic vision).

      2. Bipolar Cells.

      3. Ganglion Cells.

      4. Horizontal Cells.

      5. Amacrine Cells.

    • Ganglion Receptive Fields:

      • Center-On: Excited by central light, inhibited by surround darkness.

      • Center-Off: Excited by central darkness, inhibited by surround light.

      • Functions to highlight local contrast boundaries rather than uniform illumination.

  • Subcortical and Cortical Visual Pathways:

    • Superior Colliculus Subcortical Pathway: Mediates Blindsight (Visione cieca). Patients with V1 destruction consciously experience total blindness, yet accurately localize visual targets unconsciously via midbrain superior colliculus projections.

    • Lateral Geniculate Nucleus (LGN): Thalamic relay preserve retinotopic spatial maps.

    • Primary Visual Cortex (V1 / Striate Cortex):

      • Contains Simple Cells (detecting oriented lines/edges) and Complex Cells (detecting oriented motion).

      • Organized into functional orientation columns.

    • Extrastriate Visual Cortices:

      • Area V4: Specialized for color processing; bilateral damage causes Achromatopsia (loss of color experience).

      • Area MT / V5: Specialized for visual motion processing; bilateral damage causes Akinetopsia (motion blindness).

    • Dual Stream Hypothesis:

      • Dorsal Stream ("Where / How Pathway"): Projects from occipital to posterior parietal cortex. Operates automatically and unconsciously to guide real-time motor action (online control).

      • Ventral Stream ("What Pathway"): Projects from occipital to inferior temporal cortex. Mediates conscious object identification, form recognition, and semantic assignment (e.g., Fusiform Face Area, where lesions cause Prosopagnosia—the inability to consciously recognize faces).

    • Posterior Hot Zone: Posterior cortical regions (parietal, temporal, occipital) constitute the core neural substrate for the quality and content of conscious experience.

Higher Cognitive Processes, Split-Brain, and Consciousness Disorders

  • Memory, Attention, and Consciousness:

    • Gerald Edelman's Theory of Neuronal Group Selection (Neural Darwinism):

      • Re-entry (Rientro): Continuous, reciprocal, dynamic signaling across cortical maps.

      • Primary Consciousness: Present-bound experience tied to short-term memory.

      • Secondary Consciousness: Higher-order consciousness enabled by language, linking memory to past reflection and future planning.

    • Dissociation of Attention and Consciousness: Attention (exogenous/automatic vs. endogenous/voluntary) filters sensory input to prevent informational overload, but consciousness and attention can occur independently.

  • Split-Brain Research (Roger Sperry and Michael Gazzaniga):

    • Studied patients subjected to surgical callosotomy (severing the corpus callosum) to control intractable epilepsy.

    • Functional Hemispheric Asymmetries:

      • Left Hemisphere: Language production, verbal articulation, analytical logic. Controls right visual field and right hand.

      • Right Hemisphere: Spatial processing, face recognition, emotional evaluation, non-verbal synthesis. Controls left visual field and left hand.

    • Experimental Findings: Presenting a visual word exclusively to the right hemisphere (left visual field) prevents verbal naming, yet the patient can select or draw the object using their left hand.

    • Sperry's Chemoaffinity Hypothesis: Axons track target connections using specific chemical cell-surface signals.

  • Reticular Formation and Global Arousal:

    • Reticular Activating System (RAS) (Giuseppe Moruzzi and Horace Magoun): Neural network in the brainstem controlling wakefulness, vigilance, and global states of consciousness ("the container").

    • RAS project to intralaminar thalamic nuclei, which diffusely activate the cerebral cortex to enable specific conscious contents.

  • Spectrum of Severe Consciousness Disorders:

    • Coma:

      • Complete loss of both wakefulness (arousal) and content (awareness).

      • Unarousable, unresponsive, eyes continuously closed, absent sleep-wake cycles, non-reactive pupils.

      • Assessed via the Glasgow Coma Scale. Typical maximum duration: less than 1 month.

    • Vegetative State (VS) / Unresponsive Wakefulness Syndrome (UWS):

      • Preservation of wakefulness (arousal) without conscious content (awareness).

      • Irregular, unsynchronized sleep-wake cycles; intact autonomous brainstem functions (spontaneous breathing, cardiovascular regulation, swallowing, grimacing, primitive reflexes).

      • Severe cortical metabolic reduction. Termed Permanent Vegetative State if it persists beyond 1 to 3 months (etiology dependent).

    • Locked-In Syndrome:

      • Full preservation of consciousness, cognition, self-awareness, and arousal, coupled with complete paralysis of voluntary muscles.

      • Caused by bilateral ventral pontine lesions in the brainstem, interrupting descending corticospinal motor tracts while sparing higher cortical structures.

      • Communication is restricted to vertical eye movements and eyelid blinking.

    • Minimally Conscious State (MCS):

      • Fluctuating, intermittent, but reproducible non-reflexive behavioral evidence of self or environmental awareness (e.g., visual tracking, reaching for objects, responding to simple verbal commands).

      • Akinetic Mutism: A rare subtype marked by severe apathy, reduced motor initiative, and delayed responsiveness despite intact wakefulness.

  • Anatomical Hypotheses and Comparative Consciousness:

    • The Claustrum Hypothesis (Francis Crick and Christof Koch): Proposed the claustrum as a thin subcortical structure acting as a central conductor integrating multi-modal sensory information across the cortex.

    • Semir Zeki's Micro-Consciousnesses: Consciousness is spatially and temporally distributed into asynchronous processing units operating in specialized cortical areas (e.g., color and motion are consciously perceived at slightly different millisecond intervals before binding).

    • Development and Species Distribution:

      • Fetal vs. Neonatal: Fetuses remain mostly asleep in utero; human consciousness matures weeks after birth as thalamocortical pathways complete functional myelination.

      • Animal Consciousness: Observed in mammals, birds, cephalopods, and arthropods. Example: Scrub jays (ghiandola della california) show episodic-like memory by adjusting food caching strategies based on food degradability and social observation.

      • Plants: Plant neurobiology demonstrates action potentials, but plants lack neural complexity, central processing, pain experience, or subjective conscious content.

Connectionism and Artificial Neural Networks

  • Principles of Parallel Distributed Processing (PDP):

    • Connectionism models cognition using Artificial Neural Networks (ANNs) inspired by biological brain structures.

    • Replaces serial, deterministic symbolic programming with parallel processing, high interconnectivity, distributed representations, and self-training through algorithmic weight updates.

  • Biological vs. Artificial Neurons:

    • Biological Neuron Components: Dendrites (signal input), Soma (summation), Axon (signal output wrapped in myelin with Nodes of Ranvier for saltatory conduction), Synaptic Cleft (neurotransmitter release).

    • Neurotransmission: Glutamate (excitatory) and GABA (inhibitory) bind to ionotropic or metabotropic receptors.

    • Action Potential Mechanics (Hodgkin Cycle): Depolarization triggers voltage-gated Na+Na^+ channels to open, causing a rapid potential spike. Subsequent K+K^+ channel opening restores resting potential, followed by a refractory period.

    • Artificial Neuron Mathematical Architecture:

      • Inputs (xix_i) are multiplied by adjustable synaptic weights (wiw_i), summed, and processed through a activation function (ff) to yield output yy.

      • Activation Functions: Step function (binary thresholding) and Sigmoid / Logistic function (continuous non-linear output):             f(x)=11+e−xf(x) = \frac{1}{1 + e^{-x}}

  • Historical Models of ANNs:

    • McCulloch-Pitts Model (1943): The first computational model of a neuron. Uses binary inputs/outputs and linear threshold logic gates (AND, OR, NOT). Based on Boolean logic; incapable of self-learning due to fixed weights.

    • Donald Hebb's Principle (1949): Formulated synaptic plasticity: "Neurons that fire together, wire together." Synaptic weight (wijw_{ij}) increases when pre- and post-synaptic units are activated simultaneously.

    • Frank Rosenblatt's Perceptron (1958): A two-layer feed-forward network with variable, learnable weights updated via supervised learning. Limited to solving linearly separable problems (unable to compute XOR logic).

    • Error Backpropagation Algorithm (Rumelhart, Hinton, Williams): Multi-layer perceptrons (MLPs) use hidden layers to resolve non-linear problems. Propagates calculated output error backwards through hidden units using gradient descent to adjust weights across the network.

  • Connectionist Simulations:

    • Predator-Prey Simulations:

      • Simulates predators (PP) seeking static prey (a,b,ca, b, c) in 2D environments. Inputs register prey direction, hidden layers process spatial data, and output layers drive motor actions.

      • Demonstrates that goal-directed, intelligent behaviors emerge naturally from non-conscious, mechanical interactions within neural networks.

    • Genetic Algorithms and Biological Evolution:

      • Simulates natural selection by encoding neural network weights into binary DNA bitstrings.

      • Agents consuming prey gain energy and reproduce via cloning with random mutations. Over generations, natural selection optimizes weight configurations without explicit supervision.

    • Ontogenetic Learning and Reward Systems:

      • The brain's mesolimbic-cortical reward system spans the Ventral Tegmental Area (VTA), Nucleus Accumbens, Prefrontal Cortex, and Amygdala.

      • Mediated by dopamine, glutamate, and GABA. Encodes Motivational Valence (positive vs. negative reward) and Motivational Salience (action drive intensity).

  • High-Level Cognitive Simulations:

    • Motivation and Attention: Internal energy states (hunger vs. satiety) modulate network weights, shifting predator attention dynamically between seeking food and seeking mates.

    • Categorization: Networks adjust internal hidden unit weights to group distinct stimuli, learning to differentiate edible from poisonous mushrooms based on feedback.

    • Language Acquisition and Abstraction: Interactive adult-child agent models show that linguistic label inputs accelerate categorization and direct attention, enabling agents to isolate abstract properties (e.g., extracting "redness" across different objects like cherries and cars).

    • Constraint Satisfaction Networks:

      • Units represent micro-features and connections act as relational constraints.

      • Networks relax iteratively until reaching a stable global energy minimum, termed the State of Maximum Goodness (Stato di massima bontà).

      • Models high-level psychological schemas and frames as dynamic emergent activation states rather than static symbolic structures.

Integrated Information Theory (IIT)

  • Fundamental Postulate (Giulio Tononi):

    • Consciousness is an intrinsic, fundamental physical property of matter—analogous to mass, electric charge, or gravity.

    • It measures a system's capacity for integrated information, starting from phenomenological experience rather than physical brain structures.

  • The Five Phenomenological Axioms of IIT:

    1. Intrinsic Existence: Consciousness exists intrinsically from its own internal point of view, independent of external observers.

    2. Composition: Experience is structured and composed of multiple phenomenal distinctions (e.g., shapes, colors, spatial locations).

    3. Information: Experience is highly differentiated; each conscious state is unique and distinct from all other possible conscious experiences.

    4. Integration: Experience is unified and indivisible into independent sub-components.

    5. Exclusion: Experience has precise spatial and temporal boundaries, excluding other contents or time scales.

  • The Five Mechanistic Postulates of Physical Systems:

    1. Intrinsic Cause-Effect Power: System elements must have intrinsic cause-effect power on themselves (past cause, future effect).

    2. Composition: System must be structured into sub-mechanisms composed of multiple elements.

    3. Information: Mechanism specifies a cause-effect repertoire outlining possible past and future states.

    4. Integration: The cause-effect structure must be irreducible to independent parts, quantified by integrated information (Φ\Phi).

    5. Exclusion: The physical complex must be specified by the exact set of elements that maximizes integrated information (Φmax\Phi_{\text{max}}).

  • Structural Comparisons: Cerebellum vs. Cerebral Cortex:

    • The Cerebellum: Contains roughly 80%80\% of the brain's total neurons, but its damage does not eliminate consciousness. Its feed-forward, modular architecture operates in parallel without reciprocal cross-connections, resulting in low integrated information (Φ≈0\Phi \approx 0).

    • The Thalamocortical System: Highly interconnected with re-entrant loops, generating high integrated information (Φmax\Phi_{\text{max}}) essential for consciousness.

  • Sleep, Dreams, and TMS-EEG Neurophysiology:

    • NREM Deep Sleep: Cortical neurons display bistability (oscillating between active UP states and hyperpolarized DOWN states), disrupting long-range connectivity. Transcranial Magnetic Stimulation (TMS) combined with EEG reveals local, non-integrated responses, causing a drop in Φmax\Phi_{\text{max}}.

    • REM Sleep and Dreaming: Re-entrant cortical connectivity is restored. TMS-EEG responses show complex, widespread propagation similar to waking states.

  • Differentiation and Integration Comparison:

    • Galileo vs. Photodiode: A photodiode discriminates light from dark (11 bit of information), but lacks integration or sub-components. A camera sensor has millions of photodiodes that function independently without integration. Human conscious experience requires both high differentiation (vast state repertoire) and tight integration.

  • Qualia Space and Silent Neurons:

    • Qualia Space (Spazio dei qualia): A high-dimensional space where concepts are points with dimensions proportional to their irreducibility.

    • Role of Silent Neurons: Inactive or silent neurons contribute to conscious experience by constraining the system's cause-effect repertoire. Only destroyed or physically disconnected neurons cease contributing to Φ\Phi.

  • Quantitative Metrics of Consciousness:

    • Big Phi (Φ\Phi): Quantifies the integrated cause-effect information of an entire system. If Φ=0\Phi = 0, the system is a mere aggregate; if Φ>0\Phi > 0, it exists as an irreducible entity.

    • Small Phi (ϕ\phi): Measures integrated information at the level of individual internal mechanisms.

    • Perturbation Complexity Index (PCI):

      • A clinical approximation of Φ\Phi used to assess unconscious or non-responsive patients.

      • TMS pulses perturb the cortex, and the resulting EEG spatiotemporal responses are compressed using the Lempel-Ziv algorithm.

      • Conscious States (Wakefulness / REM / MCS): Produce complex, uncompressible EEG patterns resulting in a high PCI.

      • Unconscious States (NREM / Coma / General Anesthesia): Produce uniform, easily compressed responses resulting in a low PCI.

  • System Topology and Simulation Limits:

    • Aggregates: Groupings of individuals (e.g., 10 people hearing individual words of a sentence) do not create a unified collective consciousness because PhimaxPhi_{\text{max}} remains localized within individual brains.

    • Feed-Forward Networks: Purely feed-forward networks (such as deep convolutional neural networks) have Φ=0\Phi = 0 due to a lack of recurrent feedback loops, rendering them non-conscious regardless of computational task performance.

    • Software Computer Simulations: Running a brain simulation on a digital computer models virtual function, but lacks the physical cause-effect architecture required to generate high intrinsic Φ\Phi. Consequently, software simulations cannot possess subjective conscious experience, even if they pass the Turing Test.