Estimating IQ with a Brain Scan

Comments at Steve Sailer:

* I’m having a hard time penetrating the science jargon, but let me take a guess at what it means. Synaptic networks comprising the human brain make up a fractal pattern. The more space filling that pattern, the smarter you are. In other words, tissue density of the brain is important to IQ just as brain size is important.

Think of your brain as a suitcase you’ve packed for trip. If the shirts are carefully rolled and the pants neatly folded, you’re going to get more things into your suitcase. It’s also going to be easier to find things, which is why highly intelligent people show lower levels of cortical activity to perform any given task.

* Exactly. They use density of connections rather than tissue density but the idea is as you wrote. There have been other MRI studies on Alzheimer’s detection where they literally use the fractal dimension as the estimate of brain health. As people lose the connections the dimension decreases.

In this study, the estimate of 40% IQ variance “explained” by the brain scans is a lower bound. They used a simple model where every additional connection per node in the network gives the same gain in IQ. More realistic models would be used if IQ modeling were the goal. Even with basic methods of using what they had measured, not only did they hit 40%, but their predictions went beyond the total IQ test score to capture some information from the subtests, such as performance on vocabulary vs matrix questions (the visual logic puzzles).

* I’m in the field, and studies like that, involving scans of multiple cohorts of humans and monkeys, behavioral and genetic tests, diffusion tractography etc. are an enormous undertaking. You have to include folks who made substantial contribution to the study as authors, and there is a lot of them in a multi-center study like this. I should say that the senior authors are some of the top-notch people in neuroscience from top imaging centers in the world, and will not be easily shouted down by the SJW crowd.

* Steve, as a longtime reader, I really enjoyed this post which I think cuts to the core of much of your writing: equality is nonsense because people are different.

I wanted to point out why this seems paradoxical to most people. They are accustomed to egalitarian systems where facts are recognized on the basis of consensus, or “most people” thinking they are true.

Once we step outside of the brain-damaged world of egalitarian thinking, it becomes clear that facts are only recognized by those at the top of the IQ spectrum, and only those who are sincere and honest look for those facts.

While I detest democracy in all forms — even a group of people choosing what is for lunch — it would be interesting to imagine future voting booths with brain scans that only allowed entrance for those with high IQ, sincerity and honesty…

* By the way, since brain imaging also provides the full geometry of the skull, science is now in a position to characterize what information is and is not available through phrenology. And to establish how much is predictable from external physical appearance. As Steve likes to say, the glass is partly empty and partly full. The delusional official absolutism of the past 70 years is that it’s discredited pseudoscience, but imaging studies are starting to challenge that.

About Luke Ford

I've written five books (see My work has been noted in the New York Times, the Los Angeles Times, and 60 Minutes. I teach Alexander Technique in Beverly Hills (
This entry was posted in IQ. Bookmark the permalink.