Zajonc, R. B., Social Facilitation, Science, 1965, 149, 269-274
Fundamental forms of interindividual influence are represented by
the oldest experimental paradigm of social psychology -- social facilitation.
Two main experimental effects arose out of early research before WWII.
Audience effects
One study showed that ability of subject to follow rotating object with
stylus improved with an audience. In another study the subjects were told
a supervisor would be watching their progress (and did stop by about four
times during trial). The supervised subjects did 34% better than unsupervised
subjects. At the end, their results were even more different.
However, in other tasks (symbolic memorization, etc.) audience effects had
negative consequences. But later those with audiences remembered better.
So performance is facilitated and learning is impaired by the presence of
spectators. If the dominant responses initially in a new task are wrong,
an audience will impair learning. If they are predominantly right, an audience
will improve learning.
Co-action Effects
A full chicken will continue to eat in the presence of another chicken.
Animals eat more in pairs than alone. Ants accomplish more in groups than
alone.
But learning in animals seems less when accomplished with others (groups
produce inhibition). Proved true with greenfinches and cockroaches in mazes.
Maybe others are distracting in these situations.
Floyd Allport did work on social facilitation in 1920. Subjects did tasks
isolated and seated around a common table. Did tasks such as word association,
vowel cancellation, multiplacation, problem solving, judgements of odors
and weights. Results were that in all but problem solving and judgements
test, performance was better in groups.
Why? Things like word associations have responses that are well established.
The problem solving test (disproving arguments of ancient philosophers)
had less well established responses. In other words, the probability of
wrong answers is dominant.
Avoidance Learning
Rats deprived of water for 48 hours allowed to drink for 5 seconds, then
electicity applied to water. Since drinking was dominant response, and
dominant responses are increase with groups, rats in groups learned to avoid
the water slower than rats alone.
In a study by Ader, students were hooked up to electrodes and forced to
figure out that a red button would delay shocks by 10 seconds. Subjects
in pairs learned much slower than subjects alone in this situation. However,
the performance of the drill once learned was better in groups. Thus learning
is impared by presence of others, but performance of learned task is improved
with others.
Generalization is that the presence of others enhances the emmission of
dominant responses.
Presence of others As a Source of Arousal
Animals together have higher brain chemicals.
Summary is that students should study alone but take exams in presence of
other students and an audience (if they have truly learned the material).