Persuasive Communication - How Audiences Decide. 2nd Edition

Heuristics and Biases in Audience Decision Making 235

misinformed voters are about a policy, the more confi dent they are in their opinions about it,

and the more resistant they are to changing their opinions, even when presented with factual

information to the contrary.^334

Experts as well as novices are susceptible to the confirmation bias. Like the mock recruiters

mentioned previously, seasoned managers’ preinterview evaluations of job applicants influ-

ence their postinterview evaluations of the applicants’ qualifications.^335 Moreover, managers

with favorable preinterview impressions are more likely to attribute good interview perfor-

mances to the applicants’ qualifications for the job and to attribute poor performances to

external factors. Political scientists rarely change their theories even when events prove their

predictions wrong. Instead, they tend to keep their theories and discount the evidence.^336

Rather than change their initial diagnoses, medical doctors tend to distort important diag-

nostic cues.^337 And when police detectives uncover evidence that disconfirms their initial

suspicion against a suspect, they often rate it as less reliable and generate more arguments to

question it.^338

The Confi rmation Bias in Groups

The confi rmation bias not only affects the information acquisition process of individuals, it also

affects information acquisition in group discussions. Group members tend to discuss and repeat

information, or schema slot values, that confi rms their prior preferences and decisions.^339 In addi-

tion, they perceive information that supports the alternative they prefer as being more valuable,

more credible, and more relevant than information that opposes it.^340

Even if group members are not motivated to advocate the alternative they personally prefer,

they are more likely to mention information that is consistent with it.^341 Once the group makes

a decision, group members begin to prefer information that supports their choice.^342 Groups that

cannot agree on the best alternative are less confi dent about the correctness of the group’s decision

and less committed to implementing it.

The Common Knowledge Effect in Groups

One reason people meet in groups is to allow group members to share information relevant to

making a decision, in other words, to share and acquire information that fi lls their schema slots.

Yet groups typically fail to make better decisions than individual decision makers. In a seminal

study of the common knowledge effect, social psychologists Daniel Gigone and Reid Hastie

asked 40, three-person groups to determine the grades received by 32 students in an introduc-

tory psychology class.^343 Each group of three was given information about six attributes or

decision criteria relevant to the students’ grades—their aptitude test scores, high-school grade

point averages, attendance, enjoyment of the course, overall workload, and self-rated anxiety. For

each student a group evaluated, all three group members always received information about the

same two attributes, information about two other attributes was always shared by two of the

group members, and information about the remaining two attributes was always given to only

one group member, with the specifi c group members receiving a particular unshared cue vary-

ing across cases. Thus, each group member always received four items of information about each

case. If the three group members pooled their information, information about all six attributes

would be available to them.

Surprisingly, the groups consistently failed to discuss all the information that was unshared or to

use it to make their decisions. Even though the groups were in possession of more information than

any of the individuals, placing people in groups did not result in better judgments than would have