Choice as a disrupter of performance in paired-associate learning

Abstract

Conducted 2 experiments in paired-associate learning in which 20 undergraduates were permitted to choose their responses from alternatives following the procedure of L. C. Perlmuter, R. A. Monty, and G. A. Kimble . The other 20 Ss were forced to learn the responses chosen by their yoked masters. Both groups of Ss were forced to learn an interposed A-C list following the choice-force procedure but prior to learning the A-B list. The performance of force Ss was reliably superior to that of choice Ss on A-C; however, when both groups learned A-B, the typical facilitative effect of choice was disturbed. In Exp II with 40 undergraduates, interposing a C-D list produced equivalent performance in both groups on C-D. Similarly, the facilitative effect of choice was not observed on the subsequent A-B trials. The degraded performance of choice Ss is discussed in terms of frustration as an additional source of motivation.

Full text

Journal
of
Experimental
Psychology
1974,
Vol. 102,
No. 1,
170-172
CHOICE
AS A
DISRUPTER
OF
PERFORMANCE
IN
PAIRED-ASSOCIATE
LEARNING1
LAWRENCE
C.
PERLMUTER
»
Bowdoin
College
RICHARD
A.
MONTY
Human
Engineering
Laboratory,
Aberdeen Proving Ground,
Aberdeen,
Maryland
PETER
M.
CROSS
Bowdoin
College
In 2
experiments,
half
of the 5s
were permitted
to
choose their responses
from
alternatives
following
the
1971 procedure
of
Perlmuter,
Monty,
and
Kimble.
The
other
5s
were forced
to
learn
the
responses chosen
by
their yoked
masters.
Both groups
of &
were forced
to
learn
an
interposed
A-C
list
following
the
choice-force procedure
but
prior
to
learning
the A-B
list.
The
performance
of
force
5s was
reliably superior
to
that
of
choice
5s on
A-C;
however, when both
groups learned A-B,
the
typical facilitative
effect
of
choice
was
disturbed.
In
a
second experiment, interposing
a
C-D
list produced equivalent perform-
ance
in
both groups
on
C-D. Similarly,
the
facilitative
effect
of
choice
was
not
observed
on the
subsequent
A-B
trials.
The
degraded performance
of
choice
5s was
discussed
in
terms
of
frustration
as an
additional source
of
motivation.
Allowing
5s the
opportunity
to
choose
a
portion
of
the
response materials
to be
learned
on a
subse-
quent paired-associate list
was
examined with
a
paradigm
that
manipulated both
the
amount
of
choice
allowed during
the
selection procedure
and
the
locus
or
point
at
which choice took place (Monty,
Rosenberger
&
Perlmuter, 1973).
The
amount
of
material
chosen
is not as
important
as the
point
at
which
choice
is
exercised,
as
indicated
by the
finding
that
even when only
3 out of 12
items were chosen,
performance
on the
entire list
was
nearly
as
good
as
that
when
all 12
items were chosen, provided
that
choice
was
allowed
early
in the
experiment. Such
an
observation ruled
out the
associative interpre-
tation
of the
choice
vs.
force
phenomenon
as
sug-
gested
by
Perlmuter, Monty,
and
Kimble (1971)
and
seemed
to
favor
a
motivational explanation.
Subsequently, Perlmuter
and
Monty (1973)
found
some preliminary evidence
that
motivational factors
may
even work
in a
detrimental
fashion
under
certain conditions,
A
rather extensive literature exists
on the
relation-
ship
between motivational level
and
performance.
It is
generally asserted that performance increases
as
motivation increases. However, excessively high
levels
of
motivation have also been implicated
in a
deterioration
of
performance (Brown, 1961). Thus,
the
purpose
of the
present
experiments
is to
study
further
the
role
of
choice
in
learning
and to
examine
the
possibility
that
choice
may
have
an
inhibitory
as
well
as a
facilitatory
effect
on
performance.
'This
research
was
supported
in
part
by
Grant
MH-15315
from
the
National
Institute
of
Mental
Health
to the first
author.
Experiment
I was
presented
at the
meeting
of the
Southeastern Psychological Association,
New
Orleans,
May
1973.
2
Requests
for
reprints
should
be
sent
to
Lawrence
C.
Perl-
muter,
who
is
now at the
Department
of
Psychology,
Virginia
Polytechnic
Institute
and
State
University,
Blacksburg,
Virginia
24,061.
EXPERIMENT
I
Employing
the
basic design
of
Perlmuter
et
al.
(1971),
half
of the 5s
chose their
A-B
responses
(choice
group),
while
the
other
5s
were
forced
to
learn
those responses chosen
by
their yoked masters
(force
group).
In
contrast
to
earlier experiments,
an
A-C
list
was
forced upon both groups prior
to
the
learning
of
A-B, with
the
expectation
that
inhibitory
effects
would cause
the
choice group
to be
inferior
to the
force
group
on A-C
trials,
while
motivation
produced
by
prior choosing would cause
them
to be
superior
to the
force group
on the
subse-
quent
A-B
trials. Thus,
the
analysis
of the A-B
trials should serve
to
determine
the
durability
of
the
beneficial
effect
of
choice.
Method.
The 5s
were
40
male
and
female
Uni-
versity
of
Delaware students, aged 19-25 yr.,
who
were
paid $2.00 each
for
participating
in the
experi-
ment.
They
were alternately assigned
to 2
groups
of
20 5s
each
in
order
of
their appearance.
The
apparatus
and
general procedures were
identical
to
those
employed
in
previous studies.
Briefly,
the
last
84
5-letter
words were taken
from
a
list generated
by
Taylor
and
Kimble (1967)
and
were
considered
to be of
relatively high meaning.
These
items were assembled into
12
groups
of 7
each. Within each group,
one
item
was
designated
arbitrarily
as the
stimulus word
and one
item
as the A-C
response word.
The
remaining
5
items
served
as the 5
potential
response
words
on the
A—B
list.
Each
5 was first
presented with
3
paired-
associate (PA) practice trials utilizing common
words.
The
choice group
was
then presented with
12
slides,
one at a
time. Each slide contained
a
stimulus
word centered
on the
left
and 5
potential
response words listed vertically
on the
right.
The
5 was
instructed
to
read
the
stimulus word
and the 5
potential response words aloud
and to
then select
170