This data was compiled from a battery of dog-tailored cognitive tasks such as self-control, communication, and memory.
Let’s take a look at the genetic and evolutionary basis of dog intelligence 🧵
Smaller dogs have smaller brains; weight and brain size correlate very well.
Brain size and intelligence correlate at r=0.45. Correcting for measurement error and averaging among breeds we can infer a correlation of nearly 0.65.
Genes go beyond morphology; four measures of cognition were found to be highly heritable.
Inhibitory Control - 0.5-0.7
Communication - 0.39
Physical Reasoning - 0.21
Memory - 0.17
Among breed metrics also return heritability estimates from 0.56 to 0.73.
Like in humans, genes associated with intelligence are often expressed in the brain.
It may also be that dogs have lost some brain volume during the selection pressures that diverged them from wolves. Similarly-sized wolves have larger brains than matched dogs!
For reference here is the phylogenetic tree (There isn't an association between genetic distance from wolves and intelligence)
I also looked into the relationship between intelligence and violence. I found a -0.08 correlation between intelligence and dog bite counts but it was not significant.
However for some reason, If I only test for below-average breeds the correlation is -0.33. p=0.1
This should be expected as most dog violence is from dogs specifically bred for this behavior rather than from other associations.
Pit bulls are bred to be violent; this has inevitable costs on society.
What you might not expect is that activists and owners are working hard to socialize these costs.
A thread on a microcosm of fairness, "justice", and discrimination. 🧵
Pit bulls are not like other dogs: they are bred to fight.
In the pit, dogs are expected to clash round after round. The first round begins with the dogs both released at once, they must race and attack the other—the round ends when the dog who is overpowered "Turns" away.
You would be forgiven to think that this is like boxing, where, after a number of rounds, a winner is decided, emphasizing strength and skill. This is NOT how dog fights are decided: the winner is selected by attrition or death.
Above is the average genetic score of high school students, but I'll get to that later.
Below are the results of a meta-analysis on education. Here, education, using different methods, points to a 1-3 IQ point increase per additional year!
So there is certainly an effect! But let's put that into context: a 3 IQ point increase equals a Cohen's d of 0.2, or 1% of the variance.
1% isn't nothing, but look at what the rest shows us: the smartest students are considerably better, 3 IQ points is small in comparison.
McNamara’s Folly and The Denial of Individual Differences 🧵
The utmost importance of Intelligence in war, and the grim reality of what happened when the Military drafted over 300,000 low IQ men.
Robert McNamara, the eighth Secretary of Defense, was a genius.
At different points in his life was an Eagle Scout, the youngest and most highly paid assistant professor at Harvard Business School, and a president at Ford Motor Company.
He had mastered quantitative analysis by running the B-29 Bomber schedules and statistics in WWII and then later at Ford. In the 1960s, as Sec. of Defense he attempted to apply a similar process to the military.
Taleb / Carr have an erroneous 'insight' over the nonlinearity of IQ along with a conceptual misunderstanding.
On linear and non-linear IQ relationships 🧵
Here's the interaction, the premise is that :
a) There exists some (unspecified) degree of nonlinearity
b) This is somehow a 'devasting' critique of IQ
Note that the above is a simulation. This can motivate a point but you need to back it up with data, as we'll see this only happens in certain situations.
Critically, the point that IQ isn't efficacious at the high end fails here.
About the paper behind this figure: The moral circle
🧵
The paper is from Nature, entitled "Ideological differences in the expanse of the moral circle" by Waytz et al. They combined a few different surveys to produce this study.
The first takes a broad look at the moral differences between conservatives and liberals:
Their 'love' scale produced statistically significant differences for all measures, although the absolute differences themselves were small.