>I am not so sure how universal this is if it varies between 7ms and 470ms, the article says "there is variation but it's minuscule", to me a two-order of magnitude variation is all but small.

Only if we are comparing it with itself.

If we compare it with the times involved in the general context of speaking (the time to say even a small phrase) the variation is insignificant.

> I am not so sure how universal this is if it varies between 7ms and 470ms, the article says "there is variation but it's minuscule", to me a two-order of magnitude variation is all but small.

Agreed completely. Given the scale, and the mention that the average conversational segment lasts 2 seconds, 7-470ms seems like a major difference.

It also doesn't seem like "the minimum human response time to anything", as the article quotes, if it has that much variation. Humans across all cultures can react far faster than 200ms.

> And growing up in Europe and moving to North America I can definitely feel that the "polite" amount to pause in a conversation here feels way longer, it might be small in terms of milliseconds, but if you are used to, say, 50ms, having to pause for 200ms is very, very noticeable

I've had the same experience in reverse: if you've come to expect longer conversational pauses, short ones give the conversation a very different tone.

And on the flip side, the necessary pause can grow noticeably shorter between people who know each other very well.

Looking at the data, it is not that people are reacting in 7ms, but that people sometimes start talking before their partner has finished, and the mean ends up at 7ms. It seems that the Japanese are more prone to interrupting others: http://www.pnas.org/content/106/26/10587/F2.expansion.html

It could be that the minimum is 7, the maximum is 470, but almost all values cluster around 200. Then it would be fair to say that variation is minuscule.

That's what I thought they meant, too, but it is not. From the research article (http://www.pnas.org/content/106/26/10587.full):

"The means display somewhat more variation, as shown in Fig. 2. Danish has the slowest response time on average (+469 ms) and Japanese has the fastest (+7 ms). The mean response offset for the full dataset is +208 ms, and the language-specific means fall within ≈250 ms either side of this cross-language mean, approximately the length of time it takes to produce a single English syllable (37)."

The point, I think, is that these means are close compared to the tails: http://www.pnas.org/content/106/26/10587/F1.large.jpg

I think the Atlantic article conveyed the wrong conclusion. The full text of the article is here: http://www.pnas.org/content/106/26/10587.full

The abstract has their conclusions:

"Informal verbal interaction is the core matrix for human social life. A mechanism for coordinating this basic mode of interaction is a system of turn-taking that regulates who is to speak and when. Yet relatively little is known about how this system varies across cultures. The anthropological literature reports significant cultural differences in the timing of turn-taking in ordinary conversation. We test these claims and show that in fact there are striking universals in the underlying pattern of response latency in conversation. Using a worldwide sample of 10 languages drawn from traditional indigenous communities to major world languages, we show that all of the languages tested provide clear evidence for a general avoidance of overlapping talk and a minimization of silence between conversational turns. In addition, all of the languages show the same factors explaining within-language variation in speed of response. We do, however, find differences across the languages in the average gap between turns, within a range of 250 ms from the cross-language mean. We believe that a natural sensitivity to these tempo differences leads to a subjective perception of dramatic or even fundamental differences as offered in ethnographic reports of conversational style. Our empirical evidence suggests robust human universals in this domain, where local variations are quantitative only, pointing to a single shared infrastructure for language use with likely ethological foundations."

The research article claims that all languages have this gap, and that within a culture, the gap is quite consistent. Across cultures and languages, however, the gap has a decent spread, as you observed. In the discussion, the research article authors say,

"Amid a strong universal pattern, we do see measurable cultural differences. However, the range that we show, mean offset of next turn in each language departing no more than a quarter-second from the overall mean, is not of the kind that would imply fundamentally different types of turn-taking systems in the different languages, as the cultural variability hypothesis would suggest."

Honestly, I'm still trying to reason through how their data and conclusions line up, but their conclusions are more subtle than what the Atlantic article explains. (Which is not surprising to me.)

edit: Looking at the distribution graphs, it is true that these means are close compared to the tails of the distributions: http://www.pnas.org/content/106/26/10587/F1.large.jpg

Thanks for this -- the "single shared infrastructure for language use with likely ethological foundations" seems to be the interesting bit; looks like the Atlantic missed the point altogether.