Sessie 3: Team information
processing; decision-making
and creativity
From cooperative to motivated information sharing in groups: moving
beyond the hidden profile paradigm (Wittenbaum, Hollingshead, &
Botero)
Groups are not able to take advantage of the unique knowledge and expertise of their
members.
This article presents a review and a critique of the literature on collective information
sharing that was initiated by the Stasser and Titus study.
This paper lays out the perspective that information exchange is a motivated
process whereby members deliberately select what information to mention and how
to mention it to particular members in order to satisfy goals.
This paper presents the perspective that the paradigm used for studying collective
information sharing does not capture many features of information exchange that
likely operate in organizational groups.
THE COLLECTIVE INFORMATION-SHARING PARADIGM
In this paradigm, initially unacquainted undergraduate students work in small groups of three
to six members on a decision-making task.
Members read information about the decision alternatives with an understanding that
they may have some information that other members do not have.
Information is distributed such that some information is known by all members (i.e.,
shared information) and other information is known by a single member (i.e.,
unshared information).
Often information is distributed among members as a hidden profile such that
information supporting the best alternative is largely unshared.
Groups rarely discover the hidden profile and discuss proportionally more shared than
unshared information.
Not only is shared information more likely than unshared information to be mentioned
initially, but members are more likely to repeat shared information than unshared
information after it is mentioned.
A BRIEF REVIEW OF THE COLLECTIVE INFORMATION-SHARING LITERATURE
Certain factors influence the relative amounts of shared and unshared information that are
discussed by groups. The literature review is organized into seven types of factors that have
been examined:
1. Information type and distribution
a. Information is more likely to be discussed as the number of members who
know it increases.
b. Attempts to increase the salience of unshared information have been
successful.
c. When at least one member favors the correct alternative prior to discussion,
groups are much more likely to discover the hidden profile.
1
, d. In sum, lessons learned from examining features of the information and its
distribution among members suggest that information pooling and group
decision quality are improved when unshared information is salient and
abundant and when members disagree on the best option.
2. Task features
a. When members collectively recall information during discussion, they mention
more information (albeit, mostly shared) than when they select among the
decision alternatives.
b. The benefit of anticipating collective recall may be due to members attending
to information that they think others will not remember in order to increase the
group’s recall output.
c. Hollingshead (1996b) found that when face-to-face groups rank the decision
alternatives in order of preference, they mention more information and solve a
hidden profile better than groups who choose one best alternative.
d. When members view the hidden profile task as solvable they share
information more thoroughly and choose the best alternative more often than
when members think the group decision is a matter of judgment.
e. In conclusion, the literature examining task features suggests that structuring
the group’s task to aid information exchange is best done by having members
rank order the alternatives and anticipate recalling information during
discussion.
3. Group structure and composition
a. Overall, the findings suggest that larger groups may be better than smaller
ones at pooling information, although the effect is not consistent.
b. The effect of member familiarity on information sharing is mixed.
4. Temporal features
a. Two temporal processes have been examined thus far; the effects of time
pressure and the timing of when shared and unshared information are
introduced into discussion.
b. Members mentioned more unshared information when they had ample time to
learn the information before discussion.
c. Group members mention shared information earlier in discussion than
unshared information.
d. Members mention unshared information later during discussion when they
have run out of shared information to discuss.
e. Longer discussions result in more thorough information exchange.
5. Member characteristics
a. Leaders have been shown to repeat more information than nonleader
members.
b. In particular, leaders initially repeat more shared information than non-leaders
and, over time, repeat more unshared information as well. Directive, as
opposed to participative leaders, were particularly likely to repeat unshared
relative to shared information.
c. Member status and expertise facilitate information pooling.
d. Minority members who hold much shared information (i.e., cognitively central
members) have more influence in the group compared to members who hold
much unshared and little shared knowledge.
e. Together, the findings regarding member characteristics suggest helpful
advice for facilitating the information sharing of members with valuable
unshared information; assign them to a high status position such as group
leader or acknowledge their expertise to others.
2
processing; decision-making
and creativity
From cooperative to motivated information sharing in groups: moving
beyond the hidden profile paradigm (Wittenbaum, Hollingshead, &
Botero)
Groups are not able to take advantage of the unique knowledge and expertise of their
members.
This article presents a review and a critique of the literature on collective information
sharing that was initiated by the Stasser and Titus study.
This paper lays out the perspective that information exchange is a motivated
process whereby members deliberately select what information to mention and how
to mention it to particular members in order to satisfy goals.
This paper presents the perspective that the paradigm used for studying collective
information sharing does not capture many features of information exchange that
likely operate in organizational groups.
THE COLLECTIVE INFORMATION-SHARING PARADIGM
In this paradigm, initially unacquainted undergraduate students work in small groups of three
to six members on a decision-making task.
Members read information about the decision alternatives with an understanding that
they may have some information that other members do not have.
Information is distributed such that some information is known by all members (i.e.,
shared information) and other information is known by a single member (i.e.,
unshared information).
Often information is distributed among members as a hidden profile such that
information supporting the best alternative is largely unshared.
Groups rarely discover the hidden profile and discuss proportionally more shared than
unshared information.
Not only is shared information more likely than unshared information to be mentioned
initially, but members are more likely to repeat shared information than unshared
information after it is mentioned.
A BRIEF REVIEW OF THE COLLECTIVE INFORMATION-SHARING LITERATURE
Certain factors influence the relative amounts of shared and unshared information that are
discussed by groups. The literature review is organized into seven types of factors that have
been examined:
1. Information type and distribution
a. Information is more likely to be discussed as the number of members who
know it increases.
b. Attempts to increase the salience of unshared information have been
successful.
c. When at least one member favors the correct alternative prior to discussion,
groups are much more likely to discover the hidden profile.
1
, d. In sum, lessons learned from examining features of the information and its
distribution among members suggest that information pooling and group
decision quality are improved when unshared information is salient and
abundant and when members disagree on the best option.
2. Task features
a. When members collectively recall information during discussion, they mention
more information (albeit, mostly shared) than when they select among the
decision alternatives.
b. The benefit of anticipating collective recall may be due to members attending
to information that they think others will not remember in order to increase the
group’s recall output.
c. Hollingshead (1996b) found that when face-to-face groups rank the decision
alternatives in order of preference, they mention more information and solve a
hidden profile better than groups who choose one best alternative.
d. When members view the hidden profile task as solvable they share
information more thoroughly and choose the best alternative more often than
when members think the group decision is a matter of judgment.
e. In conclusion, the literature examining task features suggests that structuring
the group’s task to aid information exchange is best done by having members
rank order the alternatives and anticipate recalling information during
discussion.
3. Group structure and composition
a. Overall, the findings suggest that larger groups may be better than smaller
ones at pooling information, although the effect is not consistent.
b. The effect of member familiarity on information sharing is mixed.
4. Temporal features
a. Two temporal processes have been examined thus far; the effects of time
pressure and the timing of when shared and unshared information are
introduced into discussion.
b. Members mentioned more unshared information when they had ample time to
learn the information before discussion.
c. Group members mention shared information earlier in discussion than
unshared information.
d. Members mention unshared information later during discussion when they
have run out of shared information to discuss.
e. Longer discussions result in more thorough information exchange.
5. Member characteristics
a. Leaders have been shown to repeat more information than nonleader
members.
b. In particular, leaders initially repeat more shared information than non-leaders
and, over time, repeat more unshared information as well. Directive, as
opposed to participative leaders, were particularly likely to repeat unshared
relative to shared information.
c. Member status and expertise facilitate information pooling.
d. Minority members who hold much shared information (i.e., cognitively central
members) have more influence in the group compared to members who hold
much unshared and little shared knowledge.
e. Together, the findings regarding member characteristics suggest helpful
advice for facilitating the information sharing of members with valuable
unshared information; assign them to a high status position such as group
leader or acknowledge their expertise to others.
2
