samedi 17 octobre 2015

Brain size and IQ: minor correlation confirmed, causality to be determined

https://en.wikipedia.org/wiki/Elephant_cognition

https://prezi.com/gk49k8qkjnxi/elephant-and-human-brains-compared/

https://neuroscience.stanford.edu/news/ask-neuroscientist-does-bigger-brain-make-you-smarter

"Our findings raise several points of interest. First, brain volume was significantly positively associated with all three investigated intelligence domains (full-scale IQ, performance IQ, verbal IQ). In all, 6%, 4%, and 4% of variance respectively were attributable to these associations, thus yielding a moderate effect (Cohen, 1988). As hypothesized, these associations were stronger for full-scale IQ than for performance IQ and verbal IQ. This result was to be expected, because associations of full-scale IQ should display stronger effects than single components of the general construct due to higher loadings on psychometric g. These differences should be interpreted while keeping in mind that assessment of statistical significance of differences of effect strength was not possible because of data dependency (i.e., most studies provided correlation coefficients for all three intelligence domains based on the same sample)."


"In all, the present study clearly demonstrates a positive moderate association of in vivo brain volume with intelligence. Furthermore, we could show that this effect is observable in healthy individuals as well as (albeit smaller) in clinical samples. Although the association is confounded by reporting bias and therefore smaller than presumed according to previous investigations, it is robust as it generalizes over age, intelligence domain, and sex."
"In addition, the direction of causality between individual differences brain size and intelligence is not completely straightforward. Of course the most intuitive interpretation is that brain size, just as neuroanatomy in general, precedes cognitive development and is thus assumed to cause intelligence differences. Indeed neuroanatomy is highly heritable and strongly genetically correlated with IQ (Posthuma et al., 2002; see alsoDeary et al., 2010, for a review). However, even high heritabilities do not indicate that a trait is innate or genetically determined, and even strong genetic correlations do not necessarily indicate shared underlying genetic variants in any biologically meaningful sense, but might simply indicate a role of one variable in the development of the other (Johnson, Penke & Spinath, 2011; Solovieff et al., 2013). Also, a causal link between brain size and IQ might partly go in the opposite direction, as practice and experience can lead to volume increases in relevant brain areas (Brown et al., 2003 and Steffener and Stern, 2012). Maybe as a consequence, cortical thickness in old age is predicted by childhood IQ, with childhood IQ also fully accounting for the correlation between old-age IQ and cortical thickness (Karama et al., 2014). Similarly, a famous study by Shaw and colleagues (2006) found that IQ was not related to cortical thickness per se, but to the plasticity of cortical thickness during childhood. So even though it is plausible that brain size is at least partially a causal factor for IQ, more research is necessary to fully unravel the interplay between genes, environment, brain anatomy and cognitive development."
http://www.sciencedirect.com/science/article/pii/S014976341500250X

Indeed this correlation is probably linked to neuroanatomy but obviously IQ is more related to the brain wiring than the brain size. These two developments are highly dependent on genomics which explains the relationship between IQ and genomics.

Aucun commentaire:

 
Paperblog