Source
Barsade (2002), LIT-706, read in full in NOTE-551 from the August 2001 manuscript. Her abstract and General Discussion state contagion as a property of work groups and its effect on cooperation and conflict as an effect, not an association.
What is asserted
That in work groups generally, not only in the ad hoc student groups studied, one member’s displayed mood spreads to the others; and that pleasant contagion causes more cooperation and less conflict in the group’s process, rather than merely going with them.
What the Active entry establishes toward it
THEORY-123 holds the measured part. In Study 1 a confederate randomly assigned to show pleasant or unpleasant mood moved members’ mood with its valence, by self-report and by blind coders, so the spread is causal in that setting. The process link is a correlation between measured contagion and coder-rated cooperation (r = .44) and conflict (r = −.42); the manipulated conditions were not compared on process. Study 2 found members’ moods converging without a confederate (ICC .00 to .24) in MBA groups.
What would confirm or reject it
The promote_when, inherited unchanged from THEORY-123: a second experiment with a manipulated source in ongoing teams, several confederates and mood measured during the interaction, and process compared across the manipulated conditions. It is refuted if the manipulated conditions spread mood but leave cooperation and conflict unchanged.
What this does not say
- It does not say how contagion works. Mimicry with afferent feedback and conscious social comparison both fit Barsade’s data.
- It does not say a group has a mood of its own. That is the question of the convergence account, THEORY-126, which this does not decide.