Differential effects of reinforcement of an inhibitory feature after serial and simultaneous feature negative discrimination training

Abstract

Two experiments using a conditioned suppression procedure with 64 Sprague-Dawley rats found that postconditioning reinforcement of an inhibitory feature stimulus (X) had substantially different effects depending on whether a serial or a simultaneous feature negative discrimination procedure was used to establish the inhibition. With the simultaneous procedure, acquisition of excitation to the previously inhibitory X was retarded when X was paired with shock. Subsequent summation tests showed no evidence of inhibition to X after reinforced X presentations. However, acquisition of excitation to X was unaffected by prior serial feature negative training, and X-shock pairings had relatively little effect on X's inhibitory power in summation tests. Data suggest that the nature of inhibition established in feature negative discriminations differs substantially depending on the temporal arrangement of stimuli. One possibility is that inhibitors established using simultaneous stimulus arrangements modulate behavior by acting on a representation of the UCS, but inhibitors established with serial procedures act on particular CS–UCS associations. (24 ref)

Full text

Journal
of
Experimental Psychology:
Animal
Behavior
Processes
1984,
Vol.
10, No. 4,
461-475
Copyright 1984
by the
American
Psychological
Association,
Inc.
Differential
Effects
of
Reinforcement
of an
Inhibitory Feature
After
Serial
and
Simultaneous Feature Negative
Discrimination Training
Peter
C.
Holland
University
of
Pittsburgh
Two
experiments
using
a
conditioned suppression procedure with
rat
subjects
found
that postconditioning reinforcement
of an
inhibitory
feature
stimulus
(X)
had
substantially
different
effects
depending
on
whether
a
serial (A+,
X
—»
A-)
or
a
simultaneous
(A+,
AX-)
feature
negative discrimination procedure
was
used
to
establish
the
inhibition. With
the
simultaneous procedure, acquisition
of
excitation
to the
previously
inhibitory
X was
retarded
when
X was
paired
with
shock. Subsequent summation tests showed
no
evidence
of
inhibition
to X
after
reinforced
X
presentations. However, acquisition
of
excitation
to X was
unaffected
by
prior serial feature negative training,
and
X-shock pairings
had
relatively little
effect
on X's
inhibitory power
in
summation
tests.
These data suggest that
the
nature
of
inhibition established
in
feature
negative discriminations
differs
sub-
stantially
depending
on the
temporal arrangement
of
stimuli.
One
possibility
is
that inhibitors established using simultaneous stimulus arrangements modulate
behavior
by
acting
on a
representation
of the
unconditioned stimulus,
but
inhibitors established
with
serial procedures
act on
particular conditioned
stimulus-
unconditioned
stimulus associations.
Conditioned inhibition
is
often
described
as
the
learning that
two
stimuli, usually
an
initially
neutral stimulus
and an
uncondi-
tioned
stimulus (US),
occur
apart
in
time
or
space.
Typically,
we
assume this information
to be
represented
as an
inhibitory association
between
representations
of
those
two
stimuli.
However,
there
is
very little evidence concern-
ing
the
nature
or
content
of
inhibitory learn-
ing.
This lack
of
evidence
is
surprising given
the
recent interest
in the
content
of
excitatory
learning
and the
extensive investigation
of
the
impressive variety
of
procedures
that
generate inhibitory phenomena.
The
experi-
ments reported here extend
a
recent series
of
experiments
from
my
laboratory, which sug-
Portions
of
these data were presented
at a
conference
held
in
Binghamton,
New
York,
in
June, 1983.
This
research
was
supported
in
part
by a
grant
from
the
National Institute
of
Mental Health.
I
wish
to
thank
Jennifer
Lamarre,
Anthony
Dickinson,
Ruth
Colwill,
and an
anonymous reviewer
for
their
helpful
comments,
and
John Gory, Marie
Buggey,
and
Wendy
Pekich
for
their technical assistance.
Reprints
are not
available. Correspondence should
be
addressed
to
Peter Holland, Department
of
Psychology,
University
of
Pittsburgh, Pittsburgh, Pennsylvania
15260.
gest
that
the
content
of
inhibitory learning
may
be
substantially influenced
by
variations
in
training procedures.
Holland
and
Lamarre
(in
press;
see
also
Holland, 1984) used transfer
tests
to
investi-
gate
the
content
of
inhibitory learning
in one
of
the
more popular procedures
for
generating
conditioned inhibition,
the
Pavlovian condi-
tioned inhibition
(Pavlov,
1927)
or
feature
negative
(Jenkins
&
Sainsbury,
1969) proce-
dure.
In
that procedure,
one
stimulus,
A, is
reinforced
when presented alone,
but not
reinforced
when presented
in
compound with
another
(feature)
stimulus,
X.
Using
a
con-
ditioned suppression procedure with rats,
we
examined
X's
ability
to
inhibit conditioned
responding
to a
conditioned stimulus (CS)
other
than
A
after
feature negative training
in
which
(a) A and X
were presented simul-
taneously
on
nonreinforced compound
trials,
or (b) X
preceded
A on
nonreinforced com-
pound trials.
We
found
that
although
X
readily
inhibited conditioned responding
to
a
different
excitor when
X had
been estab-
lished
using
the
simultaneous procedure,
X
did not
inhibit responding
to a
different
excitor when
X had
been
trained
with
the
461