Average response time in a conversation?

AI Thread Summary
The discussion centers on the variability of response times in spoken conversations, emphasizing that numerous factors influence these times, making it difficult to establish a meaningful average. Participants note that scripted conversations tend to have shorter response times compared to unscripted ones, and complex questions often receive immediate answers. Factors such as language proficiency, emotional considerations, and the nature of the topic significantly affect how quickly individuals respond. Cultural differences also play a role, with specific examples provided regarding Native American languages and the nuances of conversational pauses. The conversation touches on the application of these insights in game design, suggesting that realistic response times could be tailored based on the context of interactions between characters. Overall, the consensus is that while averages may be elusive, understanding the distribution and context of response times is more valuable.
kolleamm
Messages
476
Reaction score
44
Is there any average time between responses in a simple conversation using speech (not text)?

For example
Bob : Hello
1sec gap
Mike : Hey!
 
Physics news on Phys.org
kolleamm said:
Is there any average time between responses in a simple conversation using speech (not text)?

For example
Bob : Hello
1sec gap
Mike : Hey!
I think there are too many factors, making up quite a spread in response times, for an "average" to be meaningful.
 
  • Like
Likes kolleamm
I have noticed that when people have scripted conversations on television shows, the average time of responses is shorter than in an unscripted conversation and complex questions are answered immediately. The delay time needed to formulate an answer in one's head is a variable that depends on a lot of parameters as @phinds noted. It may be longer if one is not a native speaker of the language, if one tries to cover up the truth or remember the lies he/she has already told, if one needs time to think how to spare someone else's feelings, etc. etc.
 
  • Like
Likes Klystron and kolleamm
kuruman said:
I have noticed that when people have scripted conversations on television shows, the average time of responses is shorter than in an unscripted conversation and complex questions are answered immediately. The delay time needed to formulate an answer in one's head is a variable that depends on a lot of parameters as @phinds noted. It may be longer if one is not a native speaker of the language, if one tries to cover up the truth or remember the lies he/she has already told, if one needs time to think how to spare someone else's feelings, etc. etc.
There is also the fact that some people are just naturally slower in conversation. And some topics require much more thought than others, and on and on.

By the way, @kolleamm, why do you care? What would you do with the information even if it existed?
 
kolleamm said:
Is there any average time between responses in a simple conversation using speech (not text)?

For example
Bob : Hello
1sec gap
Mike : Hey!
I find this topic very interesting, because I always find it difficult to judge when it is safe to say something without interrupting - ie has the pause been long enough that the other has finished speaking? I even wrote an essay about it, expressing my perplexity.

I would find the shape of the distribution of response times more interesting than the mean. I suspect it would have a fat, long tail. I wonder if anybody has ever done a research project on it in psychology. I'd love to read it if there were one.
 
  • Like
Likes Auto-Didact
It has a cultural component. My kids speak some Native American languages, Navajo & Keres - this is one of the complaints of Native speakers using English with non-Natives. Exactly the idea of being impolite or interrupting by boring ahead too soon. The reason my kids know about it is they started learning with English "implicit rules", and some of the older ladies used to correct them. Not by lecturing, but by making a non-offensive joke about it. IMO that works better than lecturing. One of the "jokes" was a coyote story, all of which are hilarious.

http://navajopeople.org/blog/coyote-tales-coyote-and-rabbit/

(Tell the stories only when frost is on the ground...)
 
phinds said:
There is also the fact that some people are just naturally slower in conversation. And some topics require much more thought than others, and on and on.

By the way, @kolleamm, why do you care? What would you do with the information even if it existed?
I would like my game characters to give more realistic response times.
 
Ah, so you are designing a game. What you are trying to do for a realistic feel is commendable. Perhaps you could identify "kinds" of conversations then adjust the average and the variance according to where the conversations are headed. For example, if two mortal enemies are facing each other, the response times should be fast:
Good guy: I'm going to shoot you right now!
(Three second pause)
Bad guy: You do that and the hostages are dead.
(Two and a half second pause)
Good guy: You're bluffing.
(Four second pause)
Bad guy: Oh yeah? Then pull the trigger. You got nothing to lose.

If people have to process the information that is conveyed then the response times should be slow:
Wizard: If you drink this elixir, you will become invincible but its side effects are that your wife will leave you, your kids will hate you, your dog will bite you and you will be an outcast for the rest of your days.
(A tenth of a second pause)
Hero: Give it to me.

See what I mean? It's a lot of work if you want the dialog to be believable.
 
phinds said:
I think there are too many factors, making up quite a spread in response times, for an "average" to be meaningful.
An important factor being, I think, whether it is face-face or remote.
 
  • Like
Likes Auto-Didact
  • #10
My best students are much faster with their sarcasm than I am.
 
  • Like
Likes Auto-Didact
Back
Top