The maximum group size for a proper conversation is four.
This discovery, revealed in the four studies summarised below, has profound implications for those who design and facilitate workshops, unconferences, Open Space meetings and large-scale gatherings.
The four studies
Krems and Wilks (academic)
Fay, Garrod and Carletta (academic)
Dunbar, Duncan and Nettle (academic)
Study 1 (academic): Four people maximum, according to Jaimie Krems and Jason Wilkes
Research conducted by Jaimie Krems, an assistant professor of psychology at Oklahoma State University, and Jason Wilkes, a graduate student in evolutionary cognitive psychology at the University of California, reveals that the maximum number for a conversation is four people.
Here is an abstract of their paper, Why are conversations limited to about four people? A theoretical exploration of the conversation size constraint, published in Evolution and Human Behavior:
It is genuinely difficult to sustain a casual conversation that includes more than four speakers. Add a fifth speaker, and the conversation often quickly fissions into smaller groups.
Termed ‘the dinner party problem,’ this four-person conversation size limit is believed to be caused by evolved cognitive constraints on human mentalizing capacities. In this view, people can mentally manage three other minds at any one time, leading to four-person conversations. But whereas existing work has posited and empirically tested alternative accounts of what drives the conversation size constraint, to our knowledge, no work has explored the question of why this capacity is specifically four.
In this theoretical paper, we (a) review research demonstrating this cognitive constraint in sociality, (b) review the relevant working memory literature, which has explored the “why four” question at some length, and (c) we begin to pose possible answers to our specific social “why four” question. Using simple mathematical models of small-scale sociality, which we imbue with evolutionarily-relevant content, we present one novel possible explanation for the four-person conversation size constraint.
Pairs (or “dyads,” in psychology research parlance) are the essential building blocks of a society. Let’s imagine a conversation between four hypothetical humans: you, Chris, Pat, and Taylor. In a four-person conversation, there are six possible pairs of people who can be talking to one another at once. you and Chris, you and Pat, you and Taylor, Chris and Pat, Chris and Taylor, and Pat and Taylor. That’s three pairs you’re part of, and three pairs you’re not. Essentially, you have a role in influencing half of the possible conversations that could be happening in that group.
If there are three people in the conversation, there are three possible pairs, only one of which excludes you. If there are five people, there are 10 possible pairs, and the majority—six—don’t include you, which makes it harder to get your point across.
Source: Why can’t more than four people have a conversation at once? by Corinne Purtill, on Quartz, citing Krems’ and Wilkes’ research
Study 2 (academic): Five people maximum, according to Nicolas Fay, Simon Garrod, and Jean Carletta
Current communication models draw a broad distinction between communication as dialogue and communication as monologue. The two kinds of models have different implications for who influences whom in a group discussion.
The experiments reported in this paper show that in small, 5-person groups, the communication is like dialogue and members are influenced most by those with whom they interact in the discussion.
However, in large, 10-person groups, the communication is like monologue and members are influenced most by the dominant speaker.
The difference in mode of communication is explained in terms of how speakers in the two sizes of groups design their utterances for different audiences.
Source: Group Discussion as Interactive Dialogue or as Serial Monologue: The Influence of Group Size, by Nicolas Fay (University of Glasgow), Simon Garrod (University of Glasgow) and Jean Carletta (University of Edinburgh).
Study 3 (academic): Four people maximum, according to Robin Dunbar, Neil Duncan and Daniel Nettle
The researchers “evaluated J. E. Cohen’s (1971) hypothesis of an upper limit to conversation group sizes by sampling conversations, and examined structural aspects of human conversational groups. Observation of 802 college cliques during lunchtime and other groups from various settings suggests that no more than four individuals can interact in spontaneous conversation.” – Abstract fragment | view source
Download the paper: Size and Structure of Freely Forming Conversational Groups (pdf), by R. I. M. Dunbar, N. D. C. Duncan and D. Nettle
Study 4 (empirical): Five people maximum, according to David Gurteen
David Gurteen (pictured) is a professional speaker and conversational facilitator. He is the founder of the Gurteen Knowledge Community, a global learning community of more than 22,000 people in 160 countries; the author of Conversational Leadership, an online book; and the originator of the Knowledge Café, in which people come together for unfacilitated conversations on a topic of mutual interest. Over the course of 15 years, David Gurteen has facilitated Knowledge Cafés in more than 30 countries.
In his article Interactive Dialogue or Serial Monologue: The Influence of Group Size on conversation, David Gurteen reveals that the ideal group size for unfacilitated interactive conversation is four people. If not four, then five is OK but three is better.
Best: four people.
Second best: three people.
Maximum: five people.
David Gurteen made this discovery by experimenting with different configurations of participants. He found that conversation does not work so well when the group consists of more than five people. One or two people are likely to dominate; the conversation will probably break into two, even three; someone may be excluded from the conversation; and group energy will be low.
“Schisming is the observation that in a conversation with at least four participants, the conversation may systematically split into two or more smaller conversations.” – Abstract fragment | view source
Work by James, Hart and others in the early 1950s showed that:
Interaction patterns change — becoming more concentrated on the talkative few — as group size rises.
Naturally occurring interactive groups are not observed with more than six people.
If you know more about the work of James and Hart as mentioned here, please send me a note as I’ve been unable to find anything on the web.
At Vagas, Brazil’s market leader in e-recruitment, the average group consists of 3 to 5 employees.
Source: Self-Management At Vagas: Zero Functional Hierarchy, No Targets, And Powerful Consensus, on Corporate Rebels.
Three’s company at Basecamp
At Basecamp, three is a magic number.
Nearly all product work is done by teams of three people. A team of three is usually composed of two programmers and one designer. And if it’s not three, it’s two or one — not four or five. We don’t throw more people at problems, we chisel problems down until they can be tackled by three people, at most.
We rarely have meetings at Basecamp, but when we do, you’ll hardly ever find more than three people around a table. Same with conference calls or video chats. Any conversation with more than three people is typically a conversation with too many people.
Source: Three’s company, by Jason Fried, founder & CEO at Basecamp.
How might meetings, workshops and large-scale gatherings be designed and facilitated in accordance with the max4 principle?
I offer some practical suggestions here.
Where do you stand on this continuum?
Conversational Leadership, an online book by David Gurteen
Group Discussion as Interactive Dialogue or as Serial Monologue: The Influence of Group Size, by Nicolas Fay (University of Glasgow), Simon Garrod (University of Glasgow) and Jean Carletta (University of Edinburgh)
The Knowledge Café meeting format (on this website)
Knowledge Café (David Gurteen’s dedicated website)
Knowledge Café tipsheet, by David Gurteen and Steve O’Hagan
Real conversations, by David Gurteen, on LinkedIn
Size and Structure of Freely Forming Conversational Groups (pdf), by R. I. M. Dunbar, N. D. C. Duncan and D. Nettle
Why are conversations limited to about four people? A theoretical exploration of the conversation size constraint, by Jaimie Krems, an assistant professor of psychology at Oklahoma State University, and Jason Wilkes, a graduate student in evolutionary cognitive psychology at the University of California, in Evolution and Human Behavior
Why can’t more than four people have a conversation at once? by Corinne Purtill, on Quartz
Why size matters: groups, dialog and high quality participation, a set of slides by Stephen Mugford