Brian Mullen

Who they are

The vault holds a note on a Brian Mullen described as a Syracuse University psychologist who led a 1984 study of newscasters' facial expressions and repeated it in 1988. That study, reported in 1986 in the Journal of Personality and Social Psychology with Futrell, Stairs, Tice, Baumeister and others as co-authors, asked whether a presenter's smile could shift viewers' votes. It is the work for which the vault first recorded his name, through its notes on Malcolm Gladwell's The Tipping Point.

The two are taken to be the same person. A research profile places a Brian Mullen at Syracuse University and lists the 1994 groupthink meta-analysis among his publications, and a co-author on that paper, Tara Anthony, is listed at Syracuse. The abstract of the 1994 paper, as read for this page, gives the author list, the journal, the volume, the issue and the page range, and does not state an affiliation for him, so the Syracuse link rests on the profile and the co-author's affiliation. The two bodies of work sit comfortably in one career, since both concern how people behave in the presence of other people.

The 1994 paper itself is a meta-analysis. It does not run a new experiment on groups. It collects the existing tests of the groupthink hypothesis and combines them, which turns the question from whether one study found an effect into what the pooled studies show on average and under what conditions the effect changes.

Key papers

• Mullen, B., Anthony, T., Salas, E., & Driskell, J. E. (1994). Group cohesiveness and quality of decision making: An integration of tests of the groupthink hypothesis. Small Group Research, 25(2), 189-204. doi:10.1177/1046496494252003.
• Mullen, B., Futrell, D., Stairs, D., Tice, D. M., Baumeister, R. F., and others (1986). Newscasters' facial expressions and voting behavior of viewers: can a smile elect a president? Journal of Personality and Social Psychology, 51(2), 291-295. doi:10.1037/0022-3514.51.2.291.

The 1994 paper was read for this page through its abstract on the SAGE Journals record, and not in full.

Key ideas

No overall effect of cohesiveness. The headline result of the meta-analysis is that cohesiveness had no significant effect on the quality of group decisions. Janis's theory made cohesion the central antecedent, the condition from which the other elements followed, so a null average effect is awkward for the theory in its plainest form.

Directive leadership as the amplifier. The abstract reports that more cohesive groups made poorer decisions when another groupthink antecedent, directive leadership, was present. That keeps a version of Janis's account alive. A close-knit group under a leader who signals the answer he wants can decide badly, while the same cohesion without that leadership carried no average penalty.

Which cohesiveness. The harm was greater when cohesiveness was measured as interpersonal attraction than when it was measured as commitment to the task or as group pride. In plain terms, a group that simply likes one another posed more risk than a group committed to its work.

Group size. More cohesive groups produced poorer decisions as group size increased. The abstract does not say why, and this page does not guess.

How it fits the later record. James Esser's 1998 review judged it too early to pass judgment on groupthink theory, Roderick Kramer's 1998 reappraisal found cohesion pressures played a smaller causal role in the Bay of Pigs than Janis argued, and Robert Baron's 2005 chapter proposed that the processes are common to groups in general. The Mullen analysis is the earliest of these in the sequence and the most quantitative, and its pattern of exceptions is a reason the later work focused on conditions and not on a single cause.

---