My first post on [what, after reading a comment about it at Catholic and Enjoying It!, I must now call] the Sit on It Model of Toxic Discussions presented the static view, to look at what happens when people with fixed opinions and depth of feeling regarding a particular thing discuss it.
But of course, people don't always have fixed opinions and depth of feeling regarding a particular thing. As a discussion goes on, depth of feeling is especially likely to change -- in fact, generally speaking people simply don't change their opinion during an on-line discussion about something they enter the discussion with a firm opinion on.1,2
I'll propose the following rule of thumb:
Zealots cause an increased depth of feeling in people who do not agree with them.
If, for example, there is a zealot at point Z in the diagram below involved in a discussion, then the effect of his contributions on a person at point A is likely to be a movement to the right. In other words, the zealot will not change the other person's opinion, but he will make that person feel his own opinion more deeply.
That's just good old-fashioned human contrariness. People respond to strong emotion with strong emotion. If the response is strong enough, it can even turn a person into an anti-zealot (e..g., point B).
Typically, though, emotions will calm down, and the depth of feeling will move back to the left. But it may not return to where it was before. Temporary zeal can affect a person, leaving him feeling more deeply about something than he did before (point C). At other times, it can leave him worn out, feeling less deeply than he did (point D).
Thanks to contrariness, then, you don't even need anti-zealots in a population to turn a discussion toxic. Zeal can produce anti-zeal where none existed before.
1. This is not to say on-line discussion doesn't change firmly held opinions. In my experience, though, the change usually happens afterwards, as someone thinks things through.
2. "Firmness" is yet another dimension of a person's opinion that I'm leaving out of the model. It's basically a measure of how hard it is for how highly a person regards a particular thing to change.