HTML Preview Simultaneous Conditioning page number 1.
Journal
of
Comparative Psychology
1992,
Vol.
106,
No. 2,
114-119
Copyright
1992
by the
American Psychological Association, Inc.
0735-7036/92/S3.00
Simultaneous
Conditioning
in
Honeybees
(Apis
mellifera)
John
D.
Batson,
James
S.
Hoban,
and M. E.
Bitterman
University
of
Hawaii
Honeybees
(Apis
mellifera)
were classically conditioned
with
odor
as
conditioned stimulus (CS),
sucrose
as
unconditioned stimulus (US),
and
proboscis extension
as
response.
The
purpose
of
Experiment
1 (Ns = 26 and 27) was to
look
for
facilitation
of
forward
conditioning
by
CS-US
overlap,
but
rapid conditioning without overlap
left
little room
for
improvement.
In 2
further
experiments,
CS and US
were simultaneous,
and
response
to
odor alone
was
measured
in
subsequent
tests.
In
Experiment
2, a
Simultaneous group
(N = 25)
responded more
to the
training
odor than
did an
Unpaired control group
(N =
25).
In
Experiment
3, a
differentially
conditioned Simultaneous group
(N = 29)
responded more
to an
odor paired
with
sucrose
in
training
(S+) than
to an
odor presented alone (S-).
The
implications
of the
results
for the
problem
of the
role
of
amount
of
reward
in
honeybee learning
are
considered.
Foraging honeybees trained with
two
targets
different
in
odor,
one of
which always contains
a
20-jd
drop
of 50%
sucrose solution
and the
other
a
5-/nl
drop
of the
same
solution, quickly develop
a
preference
for the
20-/xl
odor,
a
preference
that
is
most simply explained
in
terms
of a
stronger
association
with sucrose
(Couvillon,
Lee,
&
Bitterman,
1991).
Because
the
flying
animals seem
to
detect
20-jtl
drops more
readily
than
5-/J
drops (Walker, Lee,
&
Bitterman,
1990),
it
might
be
thought that
the
stronger association
is due to
closer
contiguity between
the
perception
of the
20-jul
odor
and the
initial taste
of
sucrose (the delay hypothesis),
but
that possi-
bility
can be
discounted
on the
basis that
the
preference
develops even when
the
locations
of the
drops
are
clearly
marked (Lee
&
Bitterman,
1990b).
A
second possibility, more
difficult
to
test,
is
that
the
drops
are
differentially
reinforcing
by
virtue
of
their visual appearance
or of
some physical
property that
is
detected—as
the
concentration
of
sucrose
is
detected—on
initial contact
of the
proboscis (the immediate-
evaluation
hypothesis).
A
third possibility,
the
focus
of
interest
here,
is
that strength
of
association
increases with
the
duration
of
concurrent
odor-taste
stimulation, which
is
greater
for the
larger
drop than
for the
smaller because
the
time required
for
ingestion
is
greater (the concurrent-stimulation hypothesis).
In
a
suggestive experiment
on
simultaneous conditioning
in
rats
(Burkhardt
&
Ayres,
1978),
suppression
of
drinking
by a
noise previously paired with shock
was
found
to
increase with
the
duration
of
concurrent noise-shock stimulation.
The
assumption
of
concurrent
odor-taste
association
in
honeybees
might seem insupportable
in the
light
of
some
This research
was
supported
by
Grant
BNS
9010609
from
the
National
Science Foundation.
The
participation
of
John
D.
Batson,
on
sabbatical leave
from
Furman
University,
was
made
possible
in
part
by a
supplementary Research Opportunity Award
from
the
National Science Foundation
and in
part
by
Research Centers
for
Minority
Institutions Grant
RR03061
from
the
National
Institutes
of
Health.
The
participation
of
James
S.
Hoban
was
made
possible
by
Minority
Biomedical
Research Support Grant
RRO8125
from
the
National Institutes
of
Health.
Correspondence concerning this article should
be
addressed
to M.
E.
Bitterman,
Bekesy
Laboratory
of
Neurobiology,
1993
East-West
Road, Honolulu, Hawaii 96822. Electronic
mail
may be
sent
to
experiments
by
Opfinger
(1949)
that were designed
to
deter-
mine when
in the
course
of a
visit
to a
feeding place
foraging
honeybees learn about
its
odor.
Opfinger's
method
was to
wait
until
her
animals
had
landed
at an
odor
source
and
begun
to
feed
then
lift
them
briefly,
and
substitute
a
second
odor
source
for the
first.
Tested subsequently,
the
animals
preferred
the
landing odor
to the
feeding odor, which
led
Opfinger
to the
questionable conclusion that they
had
learned
nothing
about
the
feeding
odor.
In
many instances,
in
fact,
the
feeding
odor
was
preferred
to no
odor,
although that
preference
in
itself
is not
sufficient
evidence
of an
association
between
the
feeding
odor
and
sucrose.
One
of the
many techniques used
to
demonstrate simulta-
neous conditioning
in
vertebrates
is the
within-compound
conditioning technique (e.g.,
Rescorla
&
Cunningham,
1978),
which
has
been used also
to
look
for
evidence
of
association
between
concurrent odors
and
colors
in
free-flying
honeybees
(Couvillon
&
Bitterman,
1982).
The
procedure was,
first,
to
reinforce
each
of
two
color-odor
compounds (orange-jasmine
and
yellow-lemon targets
with
sucrose solution were pre-
sented
on
successive visits
to the
laboratory), then
to
differ-
entially
reinforce either
the
odors
or the
colors alone (e.g.,
jasmine
reinforced
vs.
lemon unreinforced),
and finally to
look
for
differential
response
to the
remaining components
(e.g.,
orange
vs.
yellow)
in a
choice
test.
Because
the
colors
of
the
targets probably were detected
before
their odors
in the
first
stage
of
training (see
von
Frisch,
1920),
color-odor
associations might have been expected
on the
basis
of
sequen-
tial stimulation alone.
The
effectiveness
of
concurrent stim-
ulation
can be
inferred,
however,
from
the
symmetry
of the
results
for
color
and
odor;
in the
example given, orange
was
preferred
in the
third stage
after
jasmine
had
been
differen-
tially reinforced
in the
second,
and
jasmine
was
preferred
in
the
third stage
after
orange
had
been differentially reinforced
in
the
second.
In
these
experiments
we
look
for
evidence
of
concurrent
odor-taste
association
with
a
classical
conditioning technique
that
affords
somewhat better control
of
stimulation than
is
possible
in
work with
free-flying
foragers
(Frings,
1944;
Ku-
wabara,
1957; Takeda,
1961).
The
animals
are not
depended
on
to
expose themselves
to the
training stimuli
but are
trained
while
they
are
harnessed
in
small tubes.
The
conditioned
114