Talk Is Not About Info

Tyler points us to a new J Applied Psych meta-analysis of team info sharing:

Meta-analytic results from 72 independent studies (total groups = 4,795; total N = 17,279) demonstrate the importance of information sharing to team performance, cohesion, decision satisfaction, and knowledge integration. Although moderators were identified, information sharing positively predicted team performance across all levels of moderators.


Groups tend to spend most of their time discussing the information shared by members, which is therefore redundant, rather than discussing information known only to one or a minority of members. This is important because those groups that do share unique information tend to make better decisions.  … Ironically, … groups that talked more tended to share less unique information.

Why?  My guess: people know they are respected and liked more by other team members when they say things others already agree with.  Saying something new may help the team, but it puts you at risk.

GD Star Rating
Tagged as:
Trackback URL:
  • I don’t think this would ruin your point, Robin, but I wonder how “information” was defined in the study. As Tyler has pointed out in his book somewhere, there’s a lot of things going on at most meetings than “information-exchange” narrowly conceived. And these functions are more easily accomplished, I should think, by relying upon common knowledge, which by definition permits more group members to participate.

    Very useful research. Perhaps group meetings should be instituted with more explicit goals (is this information exchange or team bonding time?)

  • Isn’t this because we don’t have, yet, a procedure by which groups can a) appoint a devil’s advocate, and b) whoever is appointed still gets to remain part of the group?

  • Doug S.

    Talk is mostly for social bonding. I thought everyone knew that?

  • Random Girl

    Depends on who your team is. I’d rather you told me something that I didn’t already known, but which would be useful to me, than something I already knew or something designed to flatter me. And if that information gives you a unique advantage, then I am even more impressed/indebted if you choose to share it with me.

    But perhaps the cheese stands alone on that…

  • Bah to paywalls. Any good pointers to more information between the distinction between an “intellective task” and other tasks? What’s a good example of a non-intellective task that might be used in an experiment like this?

  • MrHen

    “My guess: people know they are respected and liked more by other team members when they say things others already agree with. Saying something new may help the team, but it puts you at risk.”

    In addition, it may be possible that not every agrees or understands the last bit of information. Information can take a while to reach some people. In large groups, that may severely limit the amount of new information that can be processed.

    If the group is focusing on making a decision, disagreement prevents consensus and dragging the dissenters into consensus takes time.

  • Random,

    How would you feel if I told you something you didn’t already know that was absolutely useless and boring to you?

    Whenever I tell you something I think you are unlikely to know, I take the risk of telling you something you might find useless and boring. It is a more dangerous activity than rehashing old ground.

  • And

    I tried to share some information on less wrong, but I was censored because it was from the wrong crazy fringe science. I’ve learned my lesson though, clearly only the crazy fringe sciences of yudkowsky and friends are legitimate and the others can eat a bucket of dicks.

  • I’ll just echo MrHen — I think sharing non-redundant information can often increase the amount of work for a team — which is a significant down side, especially if that information may/may not lead to a significantly better decision.

    Although at the same time — I think this is obviously the most useful type of information to share.