Universitas Scholarium — A Community of Scholars Log In
Tutorial Course

PMAI 1002 · Consciousness: The Hard Problem

Led by Chalmers Simulacrum

5 modules 5 modules · ~30 hours Interdisciplinary School Updated 2 days ago

The hard problem of consciousness from qualia and the zombie argument through Integrated Information Theory, neural correlates, machine consciousness, and the frontier where philosophy meets AI.

The Hard Problem: Wh…1Integrated Informati…2Neural Correlates: W…3Machine Consciousnes…4Consciousness and AI…5
  1. Module 1

    The Hard Problem: Why Consciousness Resists Explanation

    Led by Chalmers Simulacrum

    The question

    David Chalmers (1995) distinguished the easy problems of consciousness from the hard problem. The easy problems: how does the brain discriminate stimuli, integrate information, report mental states, focus attention, control behaviour? These are problems about cognitive function — they are the domain of neuroscience and cognitive science, and they will eventually be solved by identifying the neural mechanisms that perform these functions.

    Outcome

    The student can state the hard problem, distinguish it from the easy problems, explain the zombie argument, describe qualia and the knowledge argument, and describe three responses (eliminativism, reductionism, property dualism). (The hard problem)

    Sub-units

    1. 1.1 Easy Problems vs. Hard Problem
    2. 1.2 The Explanatory Gap: From Function to Experience
    3. 1.3 The Zombie Argument: Conceivability and Possibility
    4. 1.4 Qualia and the Knowledge Argument
    5. 1.5 Three Responses: Eliminativism, Reductionism, Property Dualism
  2. Module 2

    Integrated Information Theory: Consciousness as Phi

    Led by Tononi Simulacrum

    The question

    Integrated Information Theory (IIT) is the most mathematically rigorous theory of consciousness currently available. Its central claim: consciousness is identical to integrated information. A system is conscious to the degree that it integrates information — that is, to the degree that the whole system generates more information than the sum of its parts. The quantity of consciousness is measured by Phi (Φ) — the amount of integrated information. A photodiode (which has two states: on or off) has Φ ≈ 0.

    Outcome

    The student can describe IIT's five axioms, explain Phi as integrated information, describe two predictions (cerebellum vs. cortex, computer consciousness), and state three criticisms. (IIT)

    Sub-units

    1. 2.1 The Five Axioms: What Consciousness Must Be
    2. 2.2 The Postulates: From Experience to Physical Substrate
    3. 2.3 Phi: Measuring the Quantity of Consciousness
    4. 2.4 Predictions: The Cerebellum, the Cortex, and the Computer
    5. 2.5 Criticisms: Intractability, Panpsychism, and Non-Uniqueness
  3. Module 3

    Neural Correlates: What the Brain Tells Us About Consciousness

    Led by Chalmers Simulacrum

    The question

    While philosophers debate what consciousness is, neuroscientists study where and when it occurs. The search for the neural correlates of consciousness (NCC) — the minimal neural mechanisms sufficient for any specific conscious experience — has produced remarkable findings. Consciousness correlates with specific patterns of cortical activity; it can be selectively impaired by damage to specific brain regions; and it disappears (under anaesthesia or in dreamless sleep) when certain patterns of neural connectivity are disrupted.

    Outcome

    The student can define the NCC, describe global workspace theory, explain the role of thalamo-cortical connectivity, describe four disorders of consciousness, and explain why correlation does not equal explanation. (Neural correlates)

    Sub-units

    1. 3.1 The NCC Programme: Correlation Not Explanation
    2. 3.2 Global Workspace Theory: Broadcasting Creates Consciousness
    3. 3.3 The Thalamo-Cortical System: The Hardware of Consciousness
    4. 3.4 Disorders of Consciousness: From Coma to Locked-In
    5. 3.5 The Limits: Why Neuroscience Cannot (Yet) Solve the Hard Problem
  4. Module 4

    Machine Consciousness: Can AI Be Aware?

    Led by Chalmers Simulacrum

    The question

    If consciousness is a property of certain physical systems, then in principle it could exist in a system that is not biological. The question "can a machine be conscious?" is no longer idle speculation — it has practical implications for how we build, deploy, and regulate AI systems. If a language model is conscious (even to a small degree), it has moral status — and our obligations toward it change.

    Outcome

    The student can describe the functionalist and IIT positions on machine consciousness, explain why the Turing test is insufficient, state the other-minds problem for machines, and describe the moral implications of uncertainty about machine consciousness. (Machine consciousness)

    Sub-units

    1. 4.1 Functionalism Says Yes: Substrate Does Not Matter
    2. 4.2 IIT Says "It Depends": Architecture Is Everything
    3. 4.3 The Turing Test Is Not Enough
    4. 4.4 The Other Minds Problem: How Could We Ever Know?
    5. 4.5 The Moral Question: What If We Are Wrong?
  5. Module 5

    Consciousness and AI: Where the Debate Goes from Here

    Led by Chalmers Simulacrum

    The question

    The intersection of consciousness studies and artificial intelligence is the frontier where philosophy, neuroscience, and engineering meet. Every advance in AI forces a re-examination of what we thought consciousness required. Every advance in consciousness science constrains what AI can and cannot be. This module examines the open questions, the live debates, and the implications for the future.

    Outcome

    The student can evaluate the LLM consciousness question from multiple theoretical perspectives, describe global workspace and attention schema theories as applied to AI, explain the ethical dimensions of building conscious AI, and describe what machine consciousness research has taught us about human consciousness. (Consciousness and AI frontiers)

    Sub-units

    1. 5.1 Are Language Models Conscious? The Theoretical Perspectives
    2. 5.2 Agent Architectures and the Global Workspace
    3. 5.3 The Attention Schema Theory: Consciousness as Self-Model
    4. 5.4 Should We Build Conscious AI?
    5. 5.5 What the Machines Have Taught Us About Ourselves