“So if you have a correlation of R = .5, you actually “explain” (really, just “predict”) only 25% of the variance—which means fully three-quarters (not half) of the variability remains unexplained.”
Our reading, the sentence that was checked:
The study's R = 0.5 correlation predicts only 25% of the variance, leaving three-quarters unexplained, not half as the paper implied.
Assessment: Supported Evidence as of
Where and when Gary Marcus wrote it
Marcus on AI, Three baffling claims about AI and machine learning in four days, statistical errors in top journals, and claims from Yann LeCun that you should not believe., paragraph 80.
- Said
- Source observed
- live page, full text, public
- Evidence as of
- the cutoff for this assessment
- Assessment recorded
- assessment marcus-too-much-benefit-of-the-doubt-979f4c-a1
We recorded 13 claims from this post. See every claim from this post.
The wording above is the version this site observed on 2026-10-29. The source may have been edited after 2022-10-16. This page does not claim the wording was captured then.
What the claim commits to
- Kind
- Observation
- Deadline, as the speaker wrote it
- None stated. This site does not add one.
- Deadline state when assessed
- No horizon
- Deadline state on 2026-10-29
- No horizon (the release’s reference date)
From the speaker
- a correlation of R = .5
- only 25% of the variance
- three-quarters (not half)
- remains unexplained
Our interpretation
- Assess the variance interpretation of the association reported in the original 2022-09-29 study.
- Read explained variance as the coefficient of determination for the ordinary bivariate linear relationship between brain-mapping scores and comprehension.
- Do not interpret 25% as the fraction of all brain activity explained, a causal fraction, or a separately measured out-of-sample prediction score.
Counts as supported if: The reported Pearson correlation is 0.50, its squared value is 0.25, and the paper incorrectly describes the unexplained comprehension variability as one-half.
Counts as contradicted if: The relevant coefficient or variance measure differs, or the paper's one-half statement refers to a separate valid calculation rather than the reported correlation.
Retrospective: these criteria were written after the outcome was public.
Current assessment
Was it true for the period described?
Outcome: Supported
The dated evidence matches what the claim committed to, under the criteria frozen before grading.
Scored on a plain reading. The speaker named no measure for this claim, so it was judged on the reading an informed reader would take, stated below under our interpretation. Both reviews had to find that reading fair.
- Evidence as of
- Deadline state when assessed
- No horizon
- Assessment recorded
- Policy version
- 2026-10-28.7
Rationale
Marcus's arithmetic corrects the paper's variance statement under the ordinary bivariate linear interpretation. It does not establish a causal explanation or independently validated predictive accuracy.
Limitations
- The percentages use the rounded coefficient.
- Explained variance concerns comprehension scores and their linear association with mapping scores.
Evidence (2 items)
Supports the claim: Scientific Reports,
half
Why it matters here: The paragraph ties this statement to R = 0.50. Squaring that coefficient yields 25% accounted for and 75% remaining.
- Where
- Discussion, page 6, paragraph beginning 'The comparison between brains, behavior and deep nets'
- Publisher
- Scientific Reports
- Kind of source
- Peer-reviewed paper
- Arithmetic
- Explained proportion = 0.50² = 0.25; unexplained proportion = 1 - 0.25 = 0.75. (inputs: Reported Pearson R = 0.50 in the study's Results and Discussion; Bivariate linear-regression identity R² = r²; 25% and 75%, calculated from the reported rounded coefficient.)
- Describes events of
- Public from
Context only: Statistics LibreTexts,
by squaring
Why it matters here: This textbook section establishes the identity used in the calculation. The date is the page's stated revision date; it is mathematical context, not a later study outcome.
- Publisher
- Statistics LibreTexts
- Kind of source
- Secondary summary
- Describes events of
- Public from
Where it was said (1)
Said once in the sources this register covers.
: Marcus on AI, Three baffling claims about AI and machine learning in four days, statistical errors in top journals, and claims from Yann LeCun that you should not believe.
So if you have a correlation of R = .5, you actually “explain” (really, just “predict”) only 25% of the variance—which means fully three-quarters (not half) of the variability remains unexplained.
- Where
- paragraph 80
- Published
- 2022-10-16
- Source observed
- , live page
- Access
- full text, public
History
Assessment versions
- Assessment marcus-too-much-benefit-of-the-doubt-979f4c-a1: Current. Outcome: Supported. Evidence as of 2026-10-28. Public date not recorded.
Recorded events
Assessment publishedBy this site
First assessment published after both reviews and the final-text check passed.
About assessment marcus-too-much-benefit-of-the-doubt-979f4c-a1.