This study by Timothy Bates aimed to answer the question on whether cognitive rationality (CR) and cognitive ability (CA) are the same thing or different. The study suggests that “rationality” can be used to measure one’s cognitive ability.
Researcher conducted a twin study which is used to understand the contributions of genetics and environment to different traits. CR and CA tests were administered to a large twin sample.
Both CA and CR were found to be heritable. And as for the relationship between the two, the study concluded that CR is not a distinct cognitive ability separate from CA and that CR is actually an effective measure of general cognitive ability.
Does this mean that intelligent people tend to be more rational than others?
Looking at the path model, the g factor (general intelligence) shows strong loadings on both traditional CA measures (Vocab, ICAR tests) and the CR scale (.70 loading). The genetic correlation between CR and g appears to be substantial (.93), while unique environmental effects (.37) are modest. This suggests CR and CA share a common genetic architecture. However, the CR scale still has unique variance not captured by g (residual variance from a and e factors), so it’s not completely redundant with traditional IQ measures.
@Marcelo Not exactly. The correlation is strong (.70 loading means about 49% shared variance), but that leaves substantial room for variation. Someone could score high on IQ tests but still make irrational decisions due to cognitive biases, motivated reasoning, or domain-specific knowledge gaps. The relationship is probabilistic, not deterministic.
Absolutely. The high .70 loading of CR onto g and the massive genetic correlation of .93 mean that for all practical purposes, the CR test is functioning as a robust, single-measure proxy for general intelligence (g). The primary takeaway isn’t that CR is a completely separate ability, but rather that it offers an efficient alternative assessment of g, possibly capturing the reflective and effortful side of intelligence that standard CA tests might miss.
The ‘proxy for g’ view is reductionist. If we accept the Dual-Process theory framework, the CR scale measures the implementation of reflective processes. A high loading onto g doesn’t mean CR is g; it means g is a necessary precursor for effective rationality. We should interpret this as a causal constraint: You need high g to score high on CR, but the CR scale is testing the disposition to use that g, a theoretically distinct (though empirically correlated) element. The study simply confirms that the ‘hardware’ (CA) must be robust for the ‘software’ (CR) to run effectively.
@marcelo We can confidently state that a low IQ essentially guarantees low rationality because the cognitive machinery is insufficient. However, moving up the scale, the prediction weakens. While a high IQ is the strongest single predictor of high rationality, the correlation is not 1.0. This means there is a segment of the high-IQ population that falls short on rationality, and it is a key reason why scientists continue to research CR as a separate, albeit heavily linked, construct.
You know, I have concerns about equating rationality with cognitive ability. In my practice, I regularly work with highly intelligent individuals who struggle with catastrophizing or rumination, which are considered “irrational” patterns that their high IQ doesn’t protect against. This study’s conclusion that rationality is just another measure of general intelligence contradicts what I see clinically, that rational thinking in real-world contexts involves emotional regulation and adaptive coping skills that exist somewhat independently of raw cognitive ability.
I’m not sure we can confidently state that low IQ guarantees low rationality. It depends entirely on how we’re defining rationality. If we mean “scoring well on abstract reasoning tests,” sure. But if we mean “making adaptive decisions given one’s circumstances,” that seems like a different question.
This study has some significant limitations. It essentially shows that people who are good at one type of cognitive test are also good at another type of cognitive test. That’s not surprising, but calling this “rationality” seems like conceptual overreach.
