1. Modes of expressing uncertain beliefs

Uncertain beliefs can be expressed in an "internal mode" (e.g., "I am pretty sure that I am going away on the weekend," "I am 80% confident that your stock will rise") or in an "external mode" (e.g., "There is a pretty good chance that I am going away on the weekend," "I think here is an 80% probability that your stock will rise"). These two modes differ in the following way:

1. The internal mode is used more often for beliefs about yourself, your future actions, whereas the external mode is used for objective events (e.g., whether Clinton will be reelected).

2. People who express their predictions in the internal mode are seen as having more expertise and credibility (bookie, real-estate expert, doctor).

3. People who express their predictions in the internal mode get more credit when they are right but also more blame when they are wrong.

2. Negotiations

Research on the Prisoners' Dilemma (see Plous, p. 245) and on negotiations in general makes clear that communication is the central issue in dyadic decision processes. We find, for example, that people conceal some of their knowledge as well as their true interests (in order to deceive or out of fear of being taken advantage of), thereby failing to realize mutually attractive bargaining zones, and we find that people are suspicious of communicated offers from the opponent's side (whereas the same offer from a third party may be acceptable).

3. Group decision making

A. Task analysis

Additive tasks: Outcome is defined and is the sum of individual efforts (e.g., pulling ropes, some team sports). Main danger: social loafing. Outcome is better than any individual contribution.
Conjunctive tasks:Outcome is one correct solution (e.g., cognitive problems, fact knowledge). Group performs as well as (but no better than) the best individual; still, other individuals can share a solution they might not have found themselves or find it faster than they would have otherwise.
Disjunctive tasks:Outcome is undefined and emerges from interaction of individual members; resources are pooled and combined; the more variance the better. Ideal for creative tasks. Group outcome is better than any individual's and often better than the additive sum of individual efforts.

B. Non-sharing of information

Stasser & Titus (1985): Hiring decision with one condition in which group members had identical information (83% picked best candidate); another condition in which individual group members had some shared and some unique information. Group members were unlikely to share unique information and therefore failed to recognize the quality candidate (only 24% picked best candidate).

C. Groupthink

Examples: Challenger disaster; Vietnam escalation; Korea invasion; Bay of Pigs invasion; Pearl Harbor.

Antecedents: High cohesion, group isolation, directive leader, high stress, no agreement on decision-making procedures (research shows that decision quality is predicted by whether or not the group spends time discussing discussion and decision procedures ). Symptoms: Illusion of invulnerability, belief in moral correctness, conformity pressures, illusion of unanimity. Consequences: Incomplete survey of alternatives, failure to criticize favored option, poor information search and processing.

D. Group polarization

First discovered by Stoner (1961) as the "risky shift" phenomenon. Further experiments showed that groups can shift into either extreme position (whether more risky or more conservative) depending on the initial average leaning of the individuals. Why does it occur? Other people's unique arguments strengthen one's own intuition; extreme opinions are more likely to be voiced and thus "pull" others along; some people are unsure, don't disagree, either conform or give the appearance of conforming.