The reality is that inadequate research on homosexuality, which might have taken hundreds of years to provide conclusive results (all we have to this day that is fairly well established is the birth-order effect, which is an odd and scarcely explained finding), was eventually laid to rest by people pointing out that the research was highly inadequate and based on prejudicial assumptions.

The problem with the control setup of Cantor's study wasn't that he used teleiphilic heterosexual non-sex offenders, though there are obviously problems there as I pointed out. The problem is that the needed pedophile non-offender control group was completely lacking.

If you can't design an experiment so that it can be meaningfully interpreted, then at the very least you need to be extremely modest and cautious about your findings.

The fact is that he cannot have any idea whether his statements about pedophiles apply to pedophiles as a group or not. If the primary topic of your research is the area where the most troublesome question marks exist, you are in an embarrassing spot.

Such research is by no means in its infancy. It is quite middle aged. The availability of scans is just a technical upgrade. Before that there were hormones, before that, anatomy. The thought processes need to upgrade as well.