The reason I'm asking is this: there's a fundamental difference if we have a rather limited number of loonies, whose tweets then get amplified way beyond their actual proportion among the crowd, or if we have a real avelanche of dozens, hundreds, or thousands of users aggressively dominating the scene.
If it's the latter, we'd be talking about a mob scenario as it's usually claimed, granted. But if it's the former, we'd be talking more about the manufactured impression of a mob. The issue would then be how social media processes help creating mass hysteria (mostly among the opponents of the alleged mob) by unduly amplifying or promoting specific opinions that trigger responses. Or on certain reaction patterns from users that determine which opinions get excessively retweeted and talked about, therefore overrepresented.
(All of this may exist in parallel of course, varying from incident to incident.)
For example, I'm sure that social media related "news" are often deliberately designed this way. Not as part of some masterminded campaign, but through the happy marriage of these sites' economic need for clicks and the ideological need for new scandals among the public.