What about it? What did you object to? Reading through it tomorrow, my concentration is spent for tonight. Also for anyone interested, the paper in question.
It is clearly under-powered. You're never going to find anything that you didn't already know about with n=8/n=10 and n=12/n=10 for boys and girls, respectively. I don't think anyone is suggesting that puberty blockers result in clear organic brain damage, just that there is the
possibility that brain development could be slightly adversely effected. So, you are clearly going to need a proper sample size to gain any information. Their stated conclusion doesn't match their stated results either which is a bit odd. The fMRI stuff is IMO just tacked on to make the study seem cooler. fMRI studies are in general not very reliable and suffer from poor repeatability, this study with its piddly sample sizes is not going to be an exception.
To show just how much of a waste-of-time study this is, we can have a look at the results:
As an aside, this is apparently the dosage for puberty-blockers:
Since someone was curious about that earlier. Also worth noting that some of the suppressed kids had been on PBers for less than 6 months which dilutes any possible effect size further. This works in combination with the fact that the suppressed kids were actually older (0.7 years & 0.8 years for the two groups), now the authors wave this fact away by saying it's "not significant" but that's not how this works. It not being significant just means that the chance of reproducing a given mean by chance by sampling from the same distribution is p<0.05 not that it doesn't have any effect on your results. It very obviously would, and it would do so by attenuating any possible impairment further reducing the non-existent possibility that this study could have ever shown anything.
As for the ToL results, they find one significant effect in which suppressed boys under-perform relative to their IQs which would be
bad news but to be honest it's as likely as not to be a chance finding, a lone p=0.04, particularly when the suppressed reaction times weren't significant and no similar effects were found for FMs. The conclusions the authors give up is basically just them restating what are presumably their priors:
Since they actually did
not find this since since suppressed MFs did have lower accuracy but I guess you're allowed to do this when your study couldn't really be said to be capable of finding anything in the first place.
I don't really think the fMRI stuff matters and it doesn't really look as though they found anything meaningful, again questions over whether they ever could have. Their results seem to be in line with what they expected for their controls and then they found that suppressed kids had very sex-typical activations and unsuppressed kids had less sex-typical activations which just sounds like random nonsense.