remember people mainly run a full node to see when their transactions arrive and fully validate the transactions they hold as the main selfish/personal concern..
[...]so thats why my mindset is that the psychology of full node users is predominantly about personal use and network protect is the after thought. not the other way round
I still don't think that much nodes would get lost. There is other effect that has to be taken into account: Even if it's true that a drivechain is a "pegged altcoin", people that use it would like to run "Bitcoin" and not an "Europecoin" for example. So they would look for a Bitcoin client. And Bitcoin.org probably would offer to download clients with the main chain and the main drivechains (as opt-in).
also the amount of merchants/businesses is not 7000 nodes. but more like a couple hundred nodes.
That would change if the drivechains would made Bitcoin more attractive for merchants because of the lower transaction fees. (I would like to see a serious estimation, though
)
All that makes me believe that there could be a small drop of full nodes initially, but they would soon be outnumbered by the new nodes due to new merchants.
im personally less concerned of the node count drop due to 'drivechain'. and more concerned with a nodecount drop due to the cesspit of all the stripped, prunned nodes leaching(torrent analogy) the network vs the amount of true seed(torrent analogy) full archival nodes
But I estimate that would be worse if we switch to significantly larger blocks (8MB+). Then the cost to run a full archival node would be much higher. In the drivechain scenario, a typical full node would archive the main blockchain and the preferred drivechain. As the main blockchain archive is the most important one, this would lead to higher redundancy of the most important transaction data.
I consider this point not that important, though - as the main "sense" of full nodes for the security of the ecosystem is the assumption that they are a economic counter-power to miners, but nodes with small holdings wouldn't add much security here. So the couple of hundreds or thousands 24/7 online business nodes would be, in fact, more important than the (larger) number of standard full nodes.
Drivechains could even maybe be designed in a way that they expire after a certain date and a new drivechain is created, and so they would virtually not add any significant "storing/archival cost". In this case a full node including the main chain could be cheaper than today, with 10x or more capacity of the system.