the-mind
A readable model of mind, consciousness, self, and AI (based on Joscha Bach’s public work)

What is consciousness?

Consciousness first names the phenomenon to be explained: experience shows up at all.

There is something it is like to be you right now. A present is happening. The world appears, often from a point of view. That is the familiar part.

This page then presents the paper’s working account of that phenomenon: content is present to the system as present, and conscious organization helps make active mental states coherent enough for agency. Primary paper: Machine Consciousness Hypothesis@ p15-18, p20

Bach uses a “bubble of nowness” as a compact analogy for the synchronized present we inhabit. talk: Hacking Conciousness by Joscha Bach at Engineering Conciousness by AGI House@ 00:03:12

That phrase helps because it points at synchronization. A conscious present is not just a lot of information floating around; it is a stretch of time in which multiple contents count as present together for one system. talk: Hacking Conciousness by Joscha Bach at Engineering Conciousness by AGI House@ 00:03:12 Primary paper: Machine Consciousness Hypothesis@ p15-18

The difficult part is explaining how that can be true in a world made of mechanisms. Primary paper: Machine Consciousness Hypothesis@ p1-3

Why the problem feels hard

The paper starts from the classic “hard problem”: why should mechanistic activity give rise to experience at all? It points out that modern philosophy has often treated this gap as nearly impossible to bridge, even while older traditions were often less embarrassed by mind, soul, or spirit language. Primary paper: Machine Consciousness Hypothesis@ p1-3

But the paper does not solve the problem by declaring consciousness magical. It tries to make it tractable by demanding three things from a theory:

  1. capture the phenomenology,
  2. explain the function,
  3. and ideally explain the genesis. Primary paper: Machine Consciousness Hypothesis@ p12

That is a good discipline. It prevents two opposite mistakes:

  • reducing consciousness to vague poetry,
  • or reducing it to behavior without explaining why experience is there.

Consciousness is not the same as mind or self

One of the cleanest lines in the paper is that consciousness is not synonymous with self, mind, or intellect. Primary paper: Machine Consciousness Hypothesis@ p12

That matters because people often mix these together.

Your mind is the larger modeling-and-control matrix. Your self is a model of you as an agent. Your consciousness is the special fact that some of this mental content is present in a distinctive way.

In this framework, that distinction helps explain why some experiences feel conscious without being very self-heavy. You can have awareness with less narrative selfhood than in ordinary waking life.

Phenomenology: what consciousness is like

Phenomenologically, the paper describes consciousness as second-order perception.

That means not only that content is represented, but that the system is aware of the fact that representation is taking place now. There is presentness. There is presence. There may also be the sense of being an observing self in a world. Primary paper: Machine Consciousness Hypothesis@ p15, p20

This is a more precise move than saying consciousness is just “the lights are on” or “what it is like”. Those phrases point in the right direction, but they do not say enough about what is different in the organization.

The paper also makes a subtle but important point: the observer need not always be personal. In meditative or dream states, awareness may still be present even when the observer is not strongly first-personal or even spatial. Minimal consciousness may be just the bare registration of perception as present. Primary paper: Machine Consciousness Hypothesis@ p15-16

Realness and phenomenal reality

The paper introduces two unusually useful terms.

This helps explain why consciousness is not just a list of contents. Conscious contents carry different tags of presentness and actuality. An imagined scene, a memory, a perception, and a hallucination can all appear, but they do not all carry the same realness.

That is also why waking life feels different from daydreaming, and why certain altered states can disturb the sense of what is presently real without making all content disappear.

Function: what consciousness does

The paper rejects the idea that consciousness is just a useless glow over cognition.

Instead, the paper treats consciousness as doing concrete work. Its account associates consciousness with alertness, sustained vigilance, selective response to stimuli, planning, decision-making, and attentional learning. Primary paper: Machine Consciousness Hypothesis@ p16-17

The key functional proposal is that consciousness is a coherence-maximizing operation on mental states. Different subsystems and partial models are active at once. Consciousness helps reduce contradiction among them so the whole mind can act more like one system. Primary paper: Machine Consciousness Hypothesis@ p16-18

That is why consciousness can also be described as the "conductor of a mental orchestra". talk: Synthetic Sentience@ 00:11:15 Primary paper: Machine Consciousness Hypothesis@ p17-18

The point is not that there is a little person inside the brain waving a baton. The proposal is that many specialized processes need a coordination layer that makes their conflicts legible and negotiable at the level of the whole agent. talk: Synthetic Sentience@ 00:11:15 Primary paper: Machine Consciousness Hypothesis@ p16-18

Mechanism: what kind of organization might do this

The paper does not pretend to have the full implementation solved. But it does point toward an architecture:

This matters because it makes consciousness a question about organization, not about adding a mystical ingredient.

In that sense, consciousness is approached as a real feature of causal organization that should, in principle, be modelable.

Genesis: why consciousness may come early

One of the paper’s strongest proposals is that consciousness may not be a late luxury added to an already competent mind.

The Genesis Hypothesis suggests that consciousness appears very early in human development and may be required for building the kind of coherent world-model and self-model that intelligent agency depends on. Primary paper: Machine Consciousness Hypothesis@ p18-20

That changes the tone of the whole topic.

Instead of asking only, “What extra property appears once intelligence is high enough?”, we also ask, “What kind of organization has to be present early for a human mind to form in the first place?”

Why there is no easy test

A lot of public debate wants a shortcut: some question you can ask an AI, or some behavior you can observe, that settles the matter.

The paper explicitly argues that there can be no purely behavioral Turing Test for consciousness.

Why? Because consciousness is not an externally visible performance in the same way intelligence tests often are. A system may perform impressively without implementing the relevant internal organization. Conversely, a conscious system might fail many surface tests. Primary paper: Machine Consciousness Hypothesis@ p21-23

That means the serious question is architectural:

  • what internal structure is present,
  • what role it plays,
  • how it develops,
  • and how it changes the system’s overall organization.

A compact definition

For this site, the working account of consciousness is:

a special mode of mental organization in which content is present to the system as present, often through second-order perception, and in which mental states are made more coherent for agency. Primary paper: Machine Consciousness Hypothesis@ p15-18, p20

Sources