3 ms·
There are so many effects here that no advice this unguarded has any chance of being correct. 1) there is an economic consideration. 100% of the conversation t
by ouid 4y ago
There are so many effects here that no advice this unguarded has any chance of being correct.
1) there is an economic consideration. 100% of the conversation time must be used. By the pigeonhole principle, someone is going to have to speak at least proportionately, or theres going to be silence. How appreciated overtalkers are is proportional to the average utilization of the conversation time for the people being conversed with.
2) Observational research of the form, "people who do X have better lives", is immediately invalidated as soon as it becomes popular. For a similar example, impostor syndrome became a huge buzzword, and all of a sudden everyone who had any reason at all to think they were underperforming was just a victim of impostor syndrome. This happened basically overnight. People who have been advised to shut up by their parents or researchers or therapists or the media vastly outnumber those who have been encouraged to talk more. Aside from creating an influx to the quiet part of the population, this interacts with point 1, and the next point which is...
3) Whether you are a good conversationalist depends on your audience, and how interesting you are, far far more than it depends on your ratio. People who are interesting to listen to are rewarded with interest when they talk. They are much less likely to be swayed by "research" that suggests that they should do something different. They already probably have what they want.
4) conversations are not independent. Whatever your ratio of talking to listening is, it should tend towards the central value the longer you spend with the conversational group. Of course this research is explicitly limited in scope, but the "false belief" people have, may actually be a long term heuristic. "This person will like me more in the long run if we figure out something to talk about that we both essentially contribute equally to.
5) that last arrow runs in reverse as well. The rule that people like you more when you talk more is exactly reversed. You talk more when you detect that people like you. But both rules are valid, because as a human, you only care about situations in which you like each other, and in that case it is strategic to share the conversation, so you talk less in order to show them your interest in what they have to say because of the golden rule.