Since my I descend from one of those communities (Trevi nel Lazio) from father's side, I decided to analyze the yDna of the village, and the yDna of the surrounding villages, to later compare the results to the average italian results. The sample is composed of 170 people.
Haplogroup - Number of people - Percentage of people - (National percentage)
E1b1b - 35 - 20,58% - (13,50%)
G2a - 31 - 18,23% - ( 9,00%)
I1 - 14 - 8,23% - ( 4,50%)
I2 - 4 - 2,35% - ( 5,50%)
J1 - 6 - 3,52% - ( 3,00%)
J2 - 30 - 17,64% - (15,50%)
R1a - 3 - 1,76% - ( 4,00%)
R1b - 34 - 20,00% - (39,00%)
T - 7 - 4,11% - ( 2,50%)
Q - 6 - 3,52% - ( 0,00%)
The interesting part is that the neolithic migrants all are present in double (or nearly double) percentages in comparisions to the rest of Italy, while the indoeuropean (mainly roman, in this case) R1b is half the national percentage. This means that the romans just put those communities under roman political control, without ever really migrating in great quantities to those places (which was already the most likely possibility, but I still expected some more migrations from R1b during those millennia). J2 is still quite present, but I relate (probably wrongly) this haplogroup more to merchant communities, in comparision to the belligerent and power driven romans.
It makes a lot of sense for farmers and herders to prosper so well in such territories, since farming and herding was still the main activity of the population since not so long ago. And it's also important to note that neolithic farmers present in more plain territories probably had the find more isolated and secure place with the indoeuropean arrival.
The fact that ALL the neolithic farmers are present in great numbers doesn't only mean that the R1b didn't have much interest in those territories. It also means that neolithic migrants were able to form STRONG AND COMPACT farming and herding based communities, who let no space for the older native I2 groups.
That's kind of the reason why in Sardinia I2 is so strongly present, while both neolithic and bronze age migrants ain't: the place was difficult too reach and the oldest community was too solid.
The other interesting part of this analysis is the high I1 percentages, which are even higher if we consider the central italian percentages in place than the national ones. 8,23% vs 1,5% is quite a big difference. The germanic invasions ain't considered as a main factor for modern italian genetic composition, but it's quite likely that those peoples had a bigger impact at least on some territories. And in some cases they probably had to retreat in more isolated communities.
P.S. I'm not an expert. I'm new to haplogroups, so I accept criticism in order to learn how things really are, but don't insult.