Lost an article: internet behaviour, negative reactions cause increase in hostility

[endoperez]

Hi all,

I recently (in the last week or so, I think) read an article, a summary or some other form of text about a study. I forgot where, I can't find it, and I'd love to find it again.

The study was about how internet communities differ in certain ways from real-life communities. The scientists had studied how people react to negative reactions to a person's comments. They found that in internet communities, negative reactions caused the maker of that comment to act in an even more hostile manner in the future, with the rest of the community finding his comments even less worthy as the feedback loop continued.

I believe they had done a study on Facebook, but I have a hazy, perhaps mistaken memory about a "thumbs down/thumbs up" sort of thing.

I'm interesting in this because when I read it, I thought it's basically a group of scientists figuring out that internet causes trolling, and that you shouldn't feed the trolls. Does anyone remember reading anything like this?

[NichG]

Hi all,

I recently (in the last week or so, I think) read an article, a summary or some other form of text about a study. I forgot where, I can't find it, and I'd love to find it again.

The study was about how internet communities differ in certain ways from real-life communities. The scientists had studied how people react to negative reactions to a person's comments. They found that in internet communities, negative reactions caused the maker of that comment to act in an even more hostile manner in the future, with the rest of the community finding his comments even less worthy as the feedback loop continued.

I believe they had done a study on Facebook, but I have a hazy, perhaps mistaken memory about a "thumbs down/thumbs up" sort of thing.

I'm interesting in this because when I read it, I thought it's basically a group of scientists figuring out that internet causes trolling, and that you shouldn't feed the trolls. Does anyone remember reading anything like this?

If its a feedback loop, it actually does suggest that there's some utility in feeding the trolls, at least trolls which are on the line. If that pattern holds, then that basically encourages the nascent troll to ramp things up, which leads to them discrediting themselves. So basically, the first person to get visibly emotional about the discussion loses the debate - which makes sense, since responding with emotion and frustration is often a visible cue that the person can't actually back up what they're saying through reason.

The problem with the pattern is more that the troll doesn't generally disengage after they discredit themselves, and the community doesn't generally disengage even after they have a consensus that the troll is in fact a troll/is posting worthless things. So I'd be curious if there are results about the tendency of people to respond to something and how that correlates with their perception of the worthiness of the post they're responding to (hard to measure objectively of course).

I'd expect an inverse correlation, that the less value someone puts in someone's post the more likely they are to try to respond to it - which would basically correspond to the tendency of trolls to successfully garner attention despite being reviled by everyone who is feeding them.

[Frozen_Feet]

It's called negative reciprocity (http://en.wikipedia.org/wiki/Reciprocity_%28social_psychology%29). It's not a thing limited solely to internet either, anonymity and lack of consequences simply exacerbates it.