Corrections — What We Actually Know About Consciousness

Corrections to What We Actually Know About Consciousness, one of five essays checked in The first fact-check — 17–19 August 2026.

Nine factual changes. The most instructive on the site, and the reason this entry leads with it: one sentence was rewritten four times in thirty-one hours, and the error rate rose with the confidence.

One sentence, four versions

1 — 17 August, 07:28

A serious academic effort has catalogued over 350 distinct proposed explanations for consciousness, spanning physicalist to non-physicalist positions.

2 — 18 August, 22:34

Philosopher David Chalmers has separately catalogued roughly 200 to 225 distinct proposed explanations for consciousness in the peer-reviewed academic literature […] a public consciousness-theories database maintains a looser, more inclusive count above 350.

3 — 18 August, 22:54

neuroscientist Robert Lawrence Kuhn’s 2024 peer-reviewed taxonomy, “A Landscape of Consciousness,” catalogues more than 325 distinct proposed explanations across ten categories

4 — 18 August, 23:09 — current

science broadcaster Robert Lawrence Kuhn — creator and host of the long-running PBS series Closer to Truth, with a doctorate in brain research though not an active research career — published a peer-reviewed taxonomy in 2024, “A Landscape of Consciousness,” cataloguing roughly 225 distinct proposed explanations across ten categories

Version	Specificity	Status
1	vague	asserts nothing checkably false, but the number came from an informal database it did not name
2	specific	false — the taxonomy is not Chalmers’s
3	specific	wrong descriptor, wrong count
4	specific	correct

Each rewrite sounded more authoritative than the last, and each new piece of precision introduced a new wrong thing, until verification caught up at version four. Version one is the least impressive sentence on the page and the only early one that asserts nothing checkably false.

Specificity is a claim about how much checking you did. Adding it without checking is a promise you haven’t kept.

The number was right in version 2, and version 3 threw it away

v2   "roughly 200 to 225"   attributed to Chalmers    right number, wrong person
v3   "more than 325"        attributed to Kuhn        right person, wrong number
v4   "roughly 225"          attributed to Kuhn        both right

Version 3 did not merely fail to reach the truth. It discarded a correct figure that was already on the page while fixing a false attribution — and version 4 then recovered the same number version 2 had carried. The count did not improve steadily. It was right, lost, and found again.

That is a harder version of this entry’s thesis than the four-row table shows. “Corrections introduce fresh errors” is one claim. “Corrections destroy true things while fixing false ones” is a different and more uncomfortable one, and this is a clean demonstration of the second. The correction was concentrating on the attribution and dropped an accurate detail going past.

One qualification, because the clean version of that story is slightly too clean. Version one is not exactly “true.” “A serious academic effort has catalogued over 350” takes its figure from the informal companion database while the phrasing implies the peer-reviewed taxonomy. It is unfalsifiable rather than true — nothing in it can be checked and found false, which is a weaker defense than being right. The four-stage pattern holds. The first rung is just less solid than it looks.

The same misattribution, one sentence later

Before: A separate, informal companion database built on the same taxonomy extends the count further, past 350.

After: Kuhn’s own companion website, built on the same taxonomy, has since grown further, past 500 entries across eleven categories.

The database described as “separate, informal” is Kuhn’s. The headline error and this one are the same error twice in adjacent sentences, caught in the same pass.

The conceded bet

Before: No confirmed neural correlate of consciousness was found — a result significant enough that a leading neuroscientist publicly conceded a long-standing bet that one would be identified by a set deadline.

After: The preliminary results, presented at a 2023 consciousness science conference, were significant enough that neuroscientist Christof Koch publicly conceded a long-standing bet on the spot — handing over a case of wine to philosopher David Chalmers, whom he’d wagered against 25 years earlier that a clear neural correlate of consciousness would be found by then.

Two changes in sequence: an unsourced claim about what the study found became a dated account of what actually happened, and then the anonymous participants were named.

What the peer-reviewed results showed

Before: The formal, peer-reviewed results followed nearly two years later, in 2025, confirming no confirmed neural correlate had been identified.

After: The formal, peer-reviewed results followed nearly two years later, in 2025, published in Nature.

The paper did not confirm that. The claim was removed rather than softened.

Two smaller ones

An unsupported date went, and the study gained its actual design: “In 2025, researchers ran a structured collaboration…” became “Researchers ran a structured, preregistered collaboration…”.

And “That number alone tells you how unsettled the field is” became “Either number tells you the same thing”, following the count sentence from one figure to two.

A correction inside this entry, found while writing it

The revisions were extracted on 1 September 2026 to write this file. The extraction came back flagging that the internal fact-check record had the sequence wrong — specifically that it misplaced where the figure of 325 appeared.

Checked against the record: it did not. The record describes the chain as “a serious academic effort” (vague but true) → “David Chalmers” (false) → “neuroscientist Kuhn … 325” (right person, wrong descriptor, wrong count) → correct. That is the order the revisions show. The flag was wrong, and had it been trusted, this entry would have opened by correcting a document that was already right.

It is recorded here because it is this entry’s own subject arriving on schedule. The correction is where the error hides — and it turned up during the drafting of a log about that exact failure mode.