I haven’t been able to access the article yet, but in the Catalyst story, von Hippel proposed that it could be explained through the role risk taking plays as a signal of fitness. It demonstrates skill or (in case of failure) robustness.

Another evolutionary explanation, and one that applies particularly to young males, was put forward in 1979 by Rubin & Paul (An Evolutionary Model of Taste for Risk, Economic Inquiry, 17:4).They noted that adolescents, having attracted zero mates, have little to lose from risk seeking activity. By taking the risk, they have a chance of increasing their number of mates from zero. Failure to take the risk leaves them with zero mates with a probability of one. The ‘risky’ activity is not risky from the perspective of the desired result.

An extension of the skateboarding experiment to test this other hypothesis could involve using older males or males with long-term partners. It would be interesting to see their testosterone response compared to the young, single cohort.

If this hypothesis were true, you would expect to see more risk taking where there were, say, an excess of males or some males monopolising the females. Some cross-society analysis could be interesting.