A response to Bird’s vituperative attack

Why Bird did not test the strongest hereditarian hypothesis as he claims

Bird “has merely demonstrated that population differences in cognitive abilities are incompatible with neutral evolution—a finding that would actually support, not contradict, hereditarian expectations”.

Introduction: A Response to Unscientific Discourse

Bird's response to my legitimate scientific criticism represents a troubling departure from academic standards. Rather than addressing methodological concerns through reasoned discourse, Bird chose to launch a vituperative attack on me, describing me as "ignominious".

This ad hominem approach is particularly concerning given the serious methodological issues raised about Bird's work. When researchers resort to personal attacks rather than addressing technical critiques, it often indicates that the underlying criticism has merit. Bird's hostile response, combined with his demonstrably flawed methodology, reveals a pattern of intellectual dishonesty that undermines confidence in his conclusions.

The following analysis examines Bird's methodological choices and responses to demonstrate how his approach systematically biases results toward predetermined conclusions while dismissing contradictory evidence.

The Core Issue with Neutrality Assumptions

Bird's methodology relies on a fundamental assumption that IQ-related genetic variants evolved under neutral selection. This assumption becomes problematic when examined against his own empirical findings. His Qx test analysis produced contradictory results. In his own words: “the less biased sibling GWAS polygenic scores did not show signs of selection while the regular GWAS polygenic scores showed did”.

Bird interpreted this discrepancy as evidence that educational polygenic scores were biased by stratification, similar to height polygenic scores, and concluded that neutral evolution was plausible. However, this interpretation contains significant logical inconsistencies that undermine his subsequent analyses. More concerning is that Bird's choice to proceed with neutrality-based models despite contradictory evidence appears to be methodologically misleading, as it allows him to reach predetermined conclusions about minimal genetic contributions while ignoring evidence that suggests otherwise.

Bird's Misrepresentation of the Hereditarian Position

Bird has defended his methodological choices by claiming he was "exploring what values would look like if the strongest hereditarian case were true and then comparing that to what the molecular data show. It's a simple way of showing that the existing data is wildly incompatible with hereditarianism." This defense is fundamentally dishonest and reveals a deep misunderstanding—or deliberate misrepresentation—of the hereditarian position.

The hereditarian hypothesis explicitly rests on the assumption that cognitive abilities have been subject to divergent selection pressures across populations. This is not an ancillary component but the central theoretical foundation of hereditarian claims about population differences in cognitive abilities. By using a model that assumes neutral evolution, Bird has not tested the "strongest hereditarian case" but rather tested a straw man that no serious hereditarian researcher would endorse.

Bird's claim to have refuted hereditarianism is therefore false. He has merely demonstrated that population differences in cognitive abilities are incompatible with neutral evolution—a finding that would actually support, not contradict, hereditarian expectations. The molecular data being "wildly incompatible" with his neutral model would be precisely what hereditarians would predict if their selection-based hypotheses were correct.

This misrepresentation appears to be either a fundamental misunderstanding of the theoretical framework he claims to be testing, or a deliberate attempt to mislead readers about the nature of his analysis. Either interpretation raises serious questions about the integrity of his research approach.

The Logical Flaws in Bird's Evolutionary Framework

Bird's response to criticism reveals several fundamental logical errors that further undermine his analysis:

Dismissal of Established Evolutionary Theory

Bird dismisses the classical interpretation of Qst > Fst as evidence for selection driving trait divergence beyond what would be expected from drift alone. This interpretation, established by Whitlock (2008) and widely accepted in evolutionary biology, holds that when quantitative trait differentiation (Qst) exceeds neutral genetic differentiation (Fst), this indicates selection pressure. Bird redefines this well-established discrepancy as solely environmental without providing evidence to disprove the selection hypothesis. This represents a rejection of foundational evolutionary theory without adequate justification.

Circular Reasoning in Model Application

Bird's analytical approach contains a circular logical structure that precludes the detection of selection even when it exists:

First, he assumes neutrality to justify his model choice. Second, he interprets the Qst/Fst gap as environmental based on this assumption. Third, he rejects evidence of selection because it violates his initial neutrality assumption. This circular reasoning ensures that his methodology cannot detect selection regardless of the underlying evolutionary reality.

Ignoring Contradictory Polygenic Score Evidence

Bird's analysis ignores crucial evidence from the same data he uses elsewhere. My Qst estimate of 0.6 was derived from educational attainment polygenic scores—the identical molecular data that Bird employs to calculate his molecular Fst of 0.11. When these scores demonstrate Qst >> Fst, this constitutes evidence for selection on cognition-associated alleles. Bird's model systematically ignores this evidence, instead treating it as purely environmental variance, only because a much less powered within-family GWAS did not pass the Qx selection test.

The Fundamental Absurdity of Bird's Resolution

Bird's logic follows this pattern: "If I assume neutrality, then Qst (0.6) must equal my phenotypic Fst (0.6). Since molecular Fst is only 0.11, the gap proves environment explains approximately 90% of variance." This reasoning contains a critical flaw: if neutrality is false—as the Qst >> Fst pattern suggests—then his "phenotypic Fst" of 0.6 is not a valid neutral expectation but rather a value likely inflated by selection effects.

The observed gap between phenotypic and molecular Fst could reflect both selection and environmental factors, but Bird's model attributes it entirely to environment. This represents a fundamental misunderstanding of how selection affects quantitative trait evolution and leads to systematic underestimation of genetic contributions to population differences.

Methodological Problems with the Neutrality Inference

Misinterpretation of Contradictory Results

The presence of conflicting Qx test outcomes does not constitute evidence for neutrality. When standard GWAS methods detect selection signals, this finding suggests either environmental confounding or genuine selection pressure. When sibling GWAS methods fail to detect selection, this indicates only that selection was not detectable using that particular approach—not that selection is absent.

These contradictory results should be interpreted as inconclusive rather than supportive of neutrality. Bird's conclusion that neutrality is "plausible" based on mixed evidence represents an error where the absence of evidence is treated as evidence of absence. This selective interpretation raises questions about methodological integrity, as Bird appears to have chosen the interpretation that supports his preferred conclusion while dismissing evidence that contradicts it.

Problems with the Height Analogy

Bird draws an analogy between IQ and height polygenic scores, noting that both show reduced selection signals when corrected for stratification. However, this analogy is scientifically inappropriate and potentially misleading. Height has undergone documented strong selection pressures in human populations, with clear evidence of directional selection for increased stature in certain populations (particularly northern Europeans). The fact that height polygenic scores are biased by stratification does not negate the reality of selection on height itself.

Similarly, correcting for stratification bias in IQ polygenic scores does not eliminate the possibility that cognitive abilities have been subject to selection pressures. Bird's use of this analogy appears deliberately misleading, as it conflates methodological bias correction with biological reality to support his neutrality assumption.

Insufficient Evidence for Neutrality Claims

The Relethford-Blangero model employed by Bird requires demonstration of neutral evolution as a prerequisite for valid application. Neutrality is not a default assumption but rather a testable hypothesis requiring positive evidence. Bird's approach of using contradictory Qx test results to infer neutrality fails to meet this evidentiary standard.

The choice to proceed with neutrality-based models despite inconclusive evidence appears methodologically dishonest. Bird had access to data showing selection signals in standard GWAS approaches, yet chose to dismiss this evidence in favor of an assumption that would support his environmental conclusions. This selective use of evidence suggests a predetermined outcome rather than objective scientific inquiry.

Implications for Study Conclusions

The neutrality assumption directly affects the study's key quantitative findings. If the assumption is incorrect, several major conclusions become questionable:

The reported "phenotypic Fst" value of 0.6 may be artificially inflated, leading to overestimation of environmental contributions to population differences. The claimed upper bound of 10% genetic contribution to IQ differences may significantly underestimate the true genetic component. The comparison between phenotypic Fst (0.6) and genetic Fst (0.11) becomes biased toward environmental explanations.

Alternative Interpretations

Rather than concluding neutrality from ambiguous results, the contradictory Qx test outcomes should be interpreted as indicating scientific uncertainty about the selection history of IQ-related variants. This uncertainty has important implications for model choice and interpretation of results.

When empirical tests for neutrality produce conflicting results, the appropriate scientific response is to acknowledge this uncertainty rather than proceed with analyses that assume neutrality. The authors' approach of treating uncertain evidence as confirmation of their preferred assumption represents a significant methodological weakness.

Recommendations for Future Research

Future studies should address the neutrality question more rigorously before applying models that depend on this assumption. This might involve developing more sophisticated methods for detecting selection that account for the complexities of cognitive trait evolution, or employing analytical approaches that are robust to uncertainty about selection history.

Additionally, researchers should consider the broader implications of their methodological choices and avoid treating model assumptions as established facts when the empirical evidence remains ambiguous.

Conclusion

The study's environmental conclusions rest on an unvalidated assumption of neutral evolution for IQ-related genetic variants. The authors' own data provide contradictory evidence regarding this assumption, yet they proceed to treat neutrality as established fact. This represents a significant methodological flaw that undermines confidence in their quantitative conclusions about the relative contributions of genetic and environmental factors to population differences in cognitive abilities.

Until the question of selection history is resolved with more definitive evidence, claims about the minimal genetic contribution to IQ differences remain scientifically unsupported. The field would benefit from more rigorous approaches to testing evolutionary assumptions before drawing strong conclusions about the sources of human cognitive variation.

References

Whitlock M. C. (2008). Evolutionary inference from QST. Molecular ecology, 17(8), 1885–1896. https://doi.org/10.1111/j.1365-294X.2008.03712.x