There just isn't that much else in the border zone that would make for an attractive target. East of Seoul is Gangwon province, which is relatively mountainous and sparsely populated. Mountains are hard for an army to move through, especially an army with tanks, and the sparse population (around 1.5 million people, compared to 10 million in Seoul itself and 15 million more elsewhere in the capital region) means that there aren't a lot of targets of high economic value there.
About the only military value of a move towards Gangwon instead of Seoul would be strategic surprise -- an attempt to shock the allied forces by appearing from a direction where they aren't expected, the way the Germans shocked the French and British in 1940 by driving into France through the (generally considered impassable to tanks) Ardennes Forest instead of via the open plains of Belgium (see https://en.wikipedia.org/wiki/Battle_of_Sedan_(1940)). But that kind of surprise would be much harder to pull off in an age of high-resolution satellite reconnaissance and instant electronic communication, and it would mean forgoing the hammer blow on South Korean morale that the loss or destruction of Seoul would cause. So it would be quixotic at best.
Actually, most of them weren't special forces. It was a North Korean submarine that was doing its routine(?) spying activity when the submarine was stuck in the shore. They had to abandon the ship, and most of them were navy soldiers with (I assume) no training in spying activity. Besides, you cannot exactly hide a submarine stuck in the shore, so of course a national emergency was declared and the South Korean military went full alert mode.
Reference: https://en.wikipedia.org/wiki/1996_Gangneung_submarine_infil...
In an ordinary situation, a North Korean spy agent would act much less conspicuously and have much more time to prepare for life in South Korea. (Personally, though, I don't believe they would make much difference in a full-scale war.)
I would not plan on North Korean special forces being so witless.
The eastern side of Korea has lots of issues for an attacking army to overcome, especially an army which would be depending so heavily on a quick operational breakthrough.
Such as what? Taking control of Seoul under the element of surprise is the only conceivable endgame for a DPRK invasion. Anything else would end in a protracted fight where US air superiority would absolutely demolish them.
>In the 1990s I remember reading an account of North Korean special forces landing by submarine and trying to infiltrate ... and immediately being tripped up by unfamiliarity with modern technology. One of the captured agents had no idea every farmhouse had a phone, and was therefore surprised that the alarm was raised as soon as they were spotted.
This sounds like propaganda. North Koreans are not stupid.
http://www.newsweek.com/reds-rocks-178000
Other accounts did mention the agent was surprised that farmers would have a telephone.
He also did not or could not hide his North Korean accent, as the article states.
Playing devil's advocate: How do we know Seoul is the operational objective? It seems obvious (capital city located near the border) but what if DPRK went for something not so obvious? i.e. "Attack him where he is unprepared, appear where you are not expected" (classic Sun/Mao approach)
Also, what if Pyongyang tries to draw in a sympathetic third party that would also give the US/ROK pause? China is the obvious choice, but there is also Russia ...
An interesting feature of this war is that since both sides look and speak more or less alike, covert insertion and operation is easier for each side—but especially so for North Korean agents who may move freely within South Korea’s open society.
In the 1990s I remember reading an account of North Korean special forces landing by submarine and trying to infiltrate ... and immediately being tripped up by unfamiliarity with modern technology. One of the captured agents had no idea every farmhouse had a phone, and was therefore surprised that the alarm was raised as soon as they were spotted.
There is still a huge technology divide, not to mention social giveaways relating to accents and other behavior. Sure, they can be trained, but it won't make them impervious to detection.