THE ROUGH GUIDE TO PSYCHOLOGYwas more accurate than the competition winner and more accurate than
cattle experts at the exhibition.
There’s no magic or mystery to the “wisdom of the crowd” phenom-
enon. Individuals are rarely one hundred percent accurate, even if they
are well-informed and expert in the question at hand. By averaging the
estimates made by a crowd, the random error introduced in people’s
verdicts – sometimes in one direction, sometimes in the other – tends to
cancel out, thereby homing in on the actual answer. The phenomenon
only works, however, when certain conditions are met. Thus the group
must be diverse, with individuals having unique insights into the problem
at hand – obviously, if everyone brings the same information to the table,
the averaged estimate will be no better than any individual’s best guess.
Group members must also be independent, in the sense that their own
judgement is free from contamination or influence by the other group
members. Again, this makes sense – if one individual in a crowd persuades
all the others of the superiority of his insight, then this will spoil the
balancing influence of the other estimates.
The wisdom of the crowd in one person
Amazingly, we can use the principles of the “wisdom of the crowd”
phenomenon to improve the accuracy of our own judgements, even
when we’re on our own. Stefan Herzog and Ralph Hertwig demon-
strated this in a study published in 2009 in which they asked partici-
pants to estimate historical dates, such as the year that electricity
was discovered. Crucially, half the participants were coached to make
an initial estimate, then pause to consider how it might be wrong,
before using this new perspective to make a new estimate. Herzog
and Hertwig call this solo technique “dialectical boot-strapping”. The
control participants simply made two best guesses for each question.
It turned out that the average of each dialectical boot-strapper’s
two guesses was, on average, 4.1 percent more accurate than their
initial estimate. By contrast, the average of each control participant’s
two estimates was, on average, just 0.3 percent more accurate than
their initial estimate. “Part of the wisdom of the many resides in an
individual mind”, was Herzog and Hertwig’s conclusion. “Dialectical
boot-strapping is a simple mental tool that fosters accuracy by lever-
aging people’s capacity to construct conflicting realities.” However, the
pair also pointed out that their dialectical boot-strapping technique
is no match for harnessing the wisdom of others. If a boot-strapping
participant’s initial estimate was averaged with the estimate of another
participant, this improved accuracy by about twice as much as
averaging that first participant’s own initial and second guesses.