And yet not definite enough that a single sporting organization in the world uses them as proof of steroid use.
Its an interesting article, but its about the combination of training and self-administration of steroids, and its conclusion is that on average there is a higher ratio of androgen receptors in the neck than in the limbs. That means that if someone is using, it'll show up faster in the neck muscles. However, no where does it suggest that the same results can't be achieved with training alone, or that the same ratios hold for every individual (ie its likely but not always true). And that, btw, is why no sporting organization uses well developed traps (or other neck muscles) as proof of steroid use -- the same results can be gained by training alone, and genetic variation among humans means different individuals will have different results.
That and the fact that their total sample size was 23 people, which statistically is too small a number to apply to the population as a whole.