We find a marked developmental shift from a predominantly negative correlation between intelligence and cortical thickness in early childhood to a positive correlation in late childhood and beyond. Additionally, level of intelligence is associated with the trajectory of cortical development, primarily in frontal regions implicated in the maturation of intelligent activity.
“Intellectual ability and cortical development in children and adolescents,” by Philip Shaw et al. (PDF)
So: IQ and cortical thickness are NEGATIVELY correlated in kids age 4-8. The correlation flips at around age 8 and then attenuates, but remains positive, until adulthood. What’s really weird about this is its implication that frontal cortex is BAD when you’re young for the same things it’s GOOD for when you’re old. Which seems slightly too weird to be right. Isn’t the real question more like: How are kids and adults solving IQ test problems differently, such that more frontal cortex hurts kids and helps adults?
I wrote some version of the preceding paragraph yesterday, and I think it’s a good observation, so I’m leaving it. But things are a little bit more complicated. First of all, other papers by the same group observe that “cortical thinning” may actually be artifactual — better myelinated axons may produce the appearance of thinner cortex in this kind of morphometry procedure. This is not viewed as a big deal for some reason, possibly a very good reason. The other complication is that people with “superior” intelligence (IQ 121+) show a very different developmental trajectory than do people with normal or high IQ.
Look at that image: At early ages, superior IQ (the blue line) accompanies thinner cortex than does average IQ (the red), but merely high IQ (the brown line) accompanies thicker cortex than does average. That makes you wonder whether this is all due to some ebb and flow in the effect of myelination on measures of cortical thickness — more cell bodies is good, which accounts for the difference between average and high IQ, but you need more myelin to be really good, and that artificially lowers the thickness measurements. Maybe really smart kids have both more cell bodies and heavier myelination, but the latter fact deflates estimates of cortical thickness.
It’s also interesting to do think a bit about why myelin would be so important to younger kids, but not adolescents or adults. Heavily myelinated axons mean that action potentials are transmitted faster, but a corollary of that is that they’re transmitted with greater precision — they ought to be less smeared, better separated. (This is, admittedly, impressionistic thinking.) But you quit seeing differences in adulthood, mostly. This makes me wonder whether the importance of myelination isn’t really related, not to online computation, but to patterning — not to maintaining the fidelity of the neural signal, but to creating the right set of attractors.
I’m reading and digesting a bunch of papers on this and related topics in the hopes of sparking a good research proposal for my NRSA (note that big IQ-related differences in developmental timing seem to show up in dorsomedial prefrontal cortex), so I may have more thoughts on this in due course.