Behavioural Processes 86 (2011) 178–183
Contents lists available at ScienceDirect
Behavioural Processes
journal homepage: www.elsevier.com/locate/behavproc
Learned helplessness in the rat: Effect of response topography in a within-subject
design
Cristiano Valerio dos Santosa,∗
, Tauane Gehmb
, Maria Helena Leite Hunzikerb
a
Universidad de Guadalajara, Mexico
b
Universidade de São Paulo, Brazil
a r t i c l e i n f o
Article history:
Received 2 March 2010
Received in revised form 22 October 2010
Accepted 24 November 2010
Keywords:
Gender
Learned helplessness
Rat
Response topography
Time course
a b s t r a c t
Three experiments investigated learned helplessness in rats manipulating response topography withinsubject
and different intervals between treatment and tests among groups. In Experiment 1, rats
previously exposed to inescapable shocks were tested under an escape contingency where either jumping
or nose poking was required to terminate shocks; tests were run either 1, 14 or 28 days after
treatment. Most rats failed to jump, as expected, but learned to nose poke, regardless of the interval
between treatment and tests and order of testing. The same results were observed in male and female
rats from a different laboratory (Experiment 2) and despite increased exposure to the escape contingencies
using a within-subject design (Experiment 3). Furthermore, no evidence of helplessness reversal
was observed, since animals failed to jump even after having learned to nose-poke in a previous test
session. These results are not consistent with a learned helplessness hypothesis, which claims that shock
(un)controllability is the key variable responsible for the effect. They are nonetheless consistent with
the view that inescapable shocks enhance control by irrelevant features of the relationship between the
environment and behavior.
© 2010 Elsevier B.V.
1. Introduction
Exposure to inescapable shocks has been shown to impair subsequent
performance on an escape contingency, whereas escapable
shocks produce no such effect (Seligman and Maier, 1967). This
behavioral effect, named learned helplessness, has been reported
many times, by different laboratories, and with different species,
which attests to the generality and robustness of the phenomenon
(Peterson et al., 1993). However, despite being consistently reproduced,
the precise process that gives rise to it is not yet
clear.
A handful of hypotheses, with varying degrees of success, have
been put forward to explain this effect. Some of them claimed
that the inescapable shocks produce motor inactivity that interferes
with escape learning (Glazer and Weiss, 1976). However,
the most accepted explanation, also named the learned helplessness
hypothesis, maintains that organisms exposed to inescapable
shocks somehow learn that their responses bear no relationship to
shock onset/offset. So, because this ﬁrst experience is opposite to
the second one, when a relationship does exist (under the escape
∗ Corresponding author at: Centro de Estudios e Investigaciones en Comportamiento,
Francisco de Quevedo 180, C.P. 44130 Arcos Vallarta, Guadalajara, Mexico.
Tel.: +52 33 38 18 07 30x5804; fax: +52 33 38 18 07 30x4.
E-mail address: behaca@yahoo.com (C.V.d. Santos).
contingency), an impairment of performance is produced (Maier
and Seligman, 1976).
Despite the considerable amount of empirical support this
hypothesis has received, many questions remain unanswered and
need to be addressed before a general account of the phenomenon
is offered. For example, impairment in the test is observed when
nose poking is required to terminate shocks only if the animal is
allowed to move freely inside the experimental chamber (Yano and
Hunziker, 2000); when its movements are restricted (e.g., by conﬁning
subjects into a plastic tube and delivering shocks through the
tail), previously inescapable shocks facilitate responding (Glazer
and Weiss, 1976). Also, it is not yet clear whether organisms learn
that there is no relationship between their responses and shocks
or whether they learn that there is no relationship between their
responses and environmental events in general. For instance, while
some have found that inescapable shocks also impair learning
maintained by positive reinforcement (Rosellini et al., 1982), others
found no effect (Capelari and Hunziker, 2009; Mauk and Pavur,
1979).
Shock predictability seems to matter as well: if shocks are predictable,
a learning impairment was observed in male, but not in
female, rats; unpredictable shocks, on the other hand, produced
impairment regardless of gender (Castelli, 2004; Kirk and Blampied,
1985).
Finally, the escape response used to evaluate the effect also
may be important. Hunziker and Santos (2007) have shown that
0376-6357 © 2010 Elsevier B.V.
doi:10.1016/j.beproc.2010.11.005
Open access under the Elsevier OA license.
Open access under the Elsevier OA license.
C.V.d. Santos et al. / Behavioural Processes 86 (2011) 178–183 179
the learning impairment following inescapable shocks depended
on the topography of the response required to escape. When running
from one side of a shuttle box to the other was required, a
difference between experimental and naïve groups was observed
when rats had to go forth and back (replicating the results of Maier
et al., 1973); however, in spite of this difference, a more detailed
analysis of the pattern of responding in the test revealed that not
even naïve subjects learned to escape (i.e., response latencies did
not show a downward trend), which casts doubt on the adequacy of
this response requirement to assess learning deﬁcits. When a single
response of jumping a barrier was required, naïve animals clearly
learned to escape, but impairment was observed with inescapable
animals. Moreover, this impairment was still present 28 days after
exposure to inescapable shocks, which is contrary to previous ﬁndings
that showed a transient effect after 72 h or more (Maier, 2001).
In the following three experiments we further explore the importance
of the topography of the escape response to the learned
helplessness effect.
2. Experiment 1
The effect of inescapable shocks is reduced with the passage
of time when the response of running (Maier, 2001), but jumping
(Hunziker and Santos, 2007) is used in the test. Nose-poking
has already been shown to be equally impaired by inescapable
shocks (Yano and Hunziker, 2000), but no data has been obtained
on how durable this effect is. This experiment evaluated this question
by exposing male rats to inescapable shocks and later testing
them under two conditions: jumping and nose-poking, which
were conducted 1, 14 or 28 days after exposure to inescapable
shocks.
2.1. Method
2.1.1. Subjects
Forty-two three-month-old naïve male albino Wistar rats
served as subjects. They came from Butantan Institute in São Paulo
and were housed individually, with water and food freely available.
The experiment was carried out during the light phase of a
12-h light/12-h dark cycle.
2.1.2. Equipment
Three experimental chambers were used: two square chambers
and one rectangular shuttle box. The square chambers, made
of aluminum and plexiglass, were 21.5 cm long, 21.5 cm wide
and 21 cm high. The right wall contained a round opening, 3 cm
in diameter, through which the rat could insert any of its body
parts (usually the nose). On the other side of the wall, a photoelectric
cell was mounted. If the rat inserted any body part at
least 1.5 cm deep into the opening, a light beam was interrupted
and a nose-poking response was registered. The grid ﬂoor of each
of the experimental chambers was constructed of stainless steel
rods 0.3 cm in diameter and spaced 1.3 cm apart. These chambers
were connected to two Lehing Valey 113-33 electric shock
generators, which delivered scrambled shocks through the grid
ﬂoor.
The shuttle box was 50 cm long, 15.5 cm wide and 20 cm high.
It consisted of two compartments of equal size, separated by an
acrylic wall. In this centre wall, there was a 7.5-cm-high and
6-cm-wide rectangular opening 8 cm above the grid ﬂoor. This
opening allowed the rat to jump from one side to another. Each
compartment had an independent grid ﬂoor, constructed of stainless
steel rods 0.3 cm in diameter and spaced 1.3 cm apart, which
was depressed by the animal’s weight. When this happened, a
microswitch was activated, registering the animal’s presence in
that compartment. Two cylinder metal rods (similar to those on
the ﬂoor) were located at the base of the opening that separated
the compartments. A BRS Foringer 901 electric shock generator and
a scrambler delivered shocks to the grid ﬂoor and the metal rods at
the base of the opening.
The three experimental chambers were housed in light- and
sound-attenuating boxes equipped with a fan for ventilation and
masking noise. Sessions were run by a PC, with a piece of software
especially developed for this experiment.
2.1.3. Procedure
The subjects were randomly divided into three experimental
groups (n = 14). Each experimental group was initially exposed to
one session of inescapable electric shocks (treatment). During this
session, subjects received 60 electric shocks in the square chambers,
each with a ﬁxed duration of 10 s and with an intensity of
1.0 mA, delivered on a variable time schedule 60 s (range 10–110 s).
The oriﬁce on the right wall of these chambers was covered with
a round metal plate, so that subjects were not allowed to emit the
nose-poking response. In this session, the rat had no control over
any aspect of the shocks.
These three groups were then exposed to two different escape
test sessions, 24 h apart. In both tests the animals received 30
shocks with an intensity of 1.0 mA and maximum duration of 10 s,
delivered through the grid ﬂoor on a variable time 60 s schedule
(range 10–110 s). The tests differed on the escape contingency in
effect. The nose-poking test was conducted in the same square
chamber used in the ﬁrst session, except that the oriﬁce on the
right wall was uncovered. In this test session, the shock was interrupted
immediately if the rat emitted a nose-poking response. If
the animal did not emit this response during the shock, it terminated
automatically after 10 s. The jumping test was conducted in
the shuttle box where shock was interrupted if the rat jumped
from one compartment to the other. If the animal did not jump
during the shock, it was terminated automatically after 10 s. In
both tests, each shock started one trial, and the time between the
shock onset and termination was recorded as the latency of that
trial.
The ﬁrst experimental group was exposed to these tests 1 and 2
days (24 and 48 h, respectively) after treatment; the second experimental
group was exposed to these tests 14 and 15 days after
treatment; the third experimental group was exposed to these tests
28 and 29 days after treatment. Half of the subjects were initially
tested for nose-poking and then for jumping; the other half was
tested in the reversed order.
2.1.4. Data analysis
A repeated-measures analysis of variance, with order of exposure
and interval between treatment and tests as between-subject
factors and trials as the within-subject factor, was conducted. The
dependent variable (latency) was tested for normality (ShapiroWilk),
homogeneity of variance (Levene), and sphericity (Mauchly)
and all tests corroborated the adequacy of the analysis. After an
effect of the independent variables was found, univariate tests, corrected
for multiple comparison using the Bonferroni method, were
conducted to ﬁnd which conditions differed. To do so, the SPSS 13
software was used and a level of signiﬁcance of 0.05 was adopted
in all experiments.
2.2. Results and discussion
Because the manipulation of the interval between treatment
and test produced no signiﬁcant results, data from subjects with
different intervals were grouped in the following analysis. Fig. 1
shows escape latencies on both tests. The left panel presents data
from groups initially required to jump and then to nose-poke;
the right panel presents data from groups initially required to
180 C.V.d. Santos et al. / Behavioural Processes 86 (2011) 178–183
Fig. 1. Response latencies, grouped in blocks of ﬁve trials, for subjects previously exposed to inescapable shocks with different intervals between treatment and test (see
legend). On the x-axis, the number before the underscore stands for the day of testing (on the ﬁrst or second day) and the number after the underscore stands for trial block.
The left panel shows data from groups initially tested under the jumping response criterion (ﬁlled symbols) and later under the nose-poking response criterion (unﬁlled
symbols). The right panel shows data from groups initially tested under the nose-poking response criterion and later under the jumping response criterion. These tests were
conducted 24 h apart, regardless of the order.
nose-poke and then to jump. When the required response was
nose-poking, animals learned to escape shocks, showing higher
latencies in the ﬁrst block of trials and a gradual decrease as
session progressed. Conversely, when the required response was
jumping, animals did not learn to escape: response latencies were
either stable or increased as session progressed. The analysis of
variance for repeated measures showed no signiﬁcant effect of
trials [F(5,128) = 2.46, p = 0.06] or order of exposure to the tests
[F(1,40) = 1.58, p = 0.21], but a signiﬁcant interaction between trials
and order of exposure [F(5,144) = 9.85, p < 0.001]. Bonferroni tests
with the signiﬁcance corrected for multiple comparisons revealed
both within- and between-subjects differences: nose-poking latencies
were lower than jumping latencies in most blocks of trials
(p < 0.05).
As reported by Hunziker and Santos (2007), jumping latencies
decrease as the escape session progresses for naïve subjects, but
not for rats previously exposed to inescapable shocks, regardless
of the interval between treatment and test. The present results
replicate and extend these ﬁndings, which suggest that requiring
a single response of jumping a barrier provides a sensitive measure
to the manipulation of inescapable shocks. When nose-poking
was required in the present experiment, however, response latencies
also decreased. This result is at odds with previous ﬁndings
that showed that nose-poking latencies remain high and relatively
stable after exposure to inescapable shocks (Yano and Hunziker,
2000).
Before trying to explain these results in relation to any theoretical
position, two additional experiments were conducted to
ensure the reliability of these ﬁndings and to evaluate whether
some subject variables may have been responsible for the failure
to replicate. The ﬁrst variable is gender: Yano and Hunziker’s
rats were female, while ours were male, and there is evidence
that, in tests of learned helplessness with the jumping
response, gender may interact with some variables (like drugs
or shock unpredictability – Gouveia, 2001) in determining the
learned helplessness effect. Second, their rats and ours originated
from different laboratories. Even though both laboratories follow
strict procedures for breeding animals, the unintended selection
of a biological characteristic, such as different pain thresholds
(or other), may have inadvertently occurred. To permit a more
direct comparison of the results, the second experiment partially
replicated the ﬁrst, using male and female rats coming
from the same laboratory as those used by Yano and Hunziker
(2000).
3. Experiment 2
3.1. Method
3.1.1. Subjects
Eight male and eight female rats, similar to the ones used in
Experiment 1, served as subjects. These rats came from Adolfo Lutz
Institute and were housed in the same conditions as in Experiment
1.
3.1.2. Equipment
Same as Experiment 1.
3.1.3. Procedure
Subjects were exposed to one treatment session (inescapable
shocks) and two test sessions (nose-poking and jumping, in
this order), 24 h apart each. Sessions were conducted exactly as
described in Experiment 1.
3.1.4. Data analysis
A repeated-measures analysis of variance, with sex as the
between-subject factor and trials as the within-subject factor,
was conducted. The dependent variable (latency) was tested for
normality (Shapiro-Wilk), homogeneity of variance (Levene), and
sphericity (Mauchly) and all tests corroborated the adequacy of the
analysis. After an effect of the independent variables was found,
univariate tests, corrected for multiple comparisons using the Bonferroni
method, were conducted to ﬁnd which conditions differed.
3.2. Results and discussion
Fig. 2 shows escape latencies, grouped in blocks of ﬁve trials, in
both tests. As in Experiment 1, nose-poking latencies were higher
in the ﬁrst block of trials and gradually decreased as session progressed.
Conversely, jumping latencies remained relatively stable.
An analysis of variance for repeated measures showed a signiﬁcant
effect of trials [F(5,46) = 8.438, p < 0.001], which indicates a difference
between the response topographies, but no signiﬁcant effect
of gender [F(1,14) = 0.084, p = 0.77] or interaction gender × trials
[F(5,46) = 2.282, p = 0.08].
The data from Experiments 1 and 2, taken together, suggest
that neither subject’s gender not its origin was responsible for the
differences between Yano and Hunziker’s results and ours related
to the reduction of nose-poking latencies, but not jumping laten-
C.V.d. Santos et al. / Behavioural Processes 86 (2011) 178–183 181
Fig. 2. Response latencies, grouped in blocks of ﬁve trials, for female (triangles) and
male (circles) subjects previously exposed to inescapable shocks. Curves on the left
(unﬁlled symbols) refer to the nose poking test and the curves on the right refer to
the jumping test (ﬁlled symbols). On the x-axis, the number before the underscore
stands for the day of testing (on the ﬁrst or second day) and the number after the
underscore stands for trial block.
cies, after exposure to inescapable shocks. The third experiment
was designed to further explore the phenomenon using a reversal
design with increased exposure to the response requirement
in the shuttle box (jumping). Given more opportunity to respond,
will jumping latencies eventually decrease? In case subjects fail to
jump consistently, will the added experience with shocks that were
not escaped impair the subsequent performance of the nose poking
response? In case subjects learn to nose poke, will jumping be
affected when re-exposed to the escape contingency in the shuttle
box?
4. Experiment 3
4.1. Method
4.1.1. Subjects
Eight male rats, similar to those used in Experiment 2 and kept
under the same housing conditions, served as subjects.
4.1.2. Equipment
Same as Experiment 1.
4.1.3. Procedure
Subjects were initially exposed to one treatment session
(inescapable shocks), conducted exactly as in Experiment 1, and
four tests sessions, conducted 1, 8, 9, and 10 days after treatment.
The ﬁrst, second and fourth test sessions were conducted in the
shuttle box, where jumping was required to terminate shocks. The
third session was conducted in the square chambers where nosepoking
was required to terminate shocks. Test sessions were run
exactly as in Experiment 1, except that 60 shocks were delivered
in the ﬁrst and second sessions, while 30 shocks were delivered in
the third and fourth sessions.
4.1.4. Data analysis
A repeated-measures analysis of variance, with trials blocks as
a within-subject factor, was conducted. The dependent variable
(latency) was tested for normality (Shapiro-Wilk), homogeneity
of variance (Levene), and sphericity (Mauchly) and all tests corroborated
the adequacy of the analysis. After an effect of the
independent variables was found, univariate tests, corrected for
multiple comparison using the Bonferroni method, were conducted
to ﬁnd which conditions differed.
4.2. Results and discussion
Fig. 3 shows escape latencies, grouped in blocks of ﬁve trials,
in both types of test. In the ﬁrst jumping test, subjects presented
an erratic response pattern, without any evidence of a
systematic reduction in response latencies. In the second jumping
test, conducted seven days after the ﬁrst one, the same erratic
response pattern was observed. In the nose-poking test session,
subjects emitted the required response consistently, with latency
reduction throughout the session. When jumping was once again
required for shock termination in the last test session, latencies
remained at high levels comparable to the second jumping test
with the same erratic pattern of responding. These results were
corroborated by an analysis of variance for repeated measures,
which showed a main effect of trials [F(35,245) = 8.826, p = 0.001].
Multiple comparisons among blocks of trials revealed that nosepoking
latencies were signiﬁcantly lower than jumping latencies
(p < 0.05).
The third experiment was an attempt to answer three questions,
namely: (1) would the subjects previously exposed to inescapable
shocks learn to jump eventually when given more trials than in
previous experiments?; (2) would the added shocks received in
case of failure to jump affect nose poking?; and (3) would jumpFig.
3. Response latencies, grouped in blocks of ﬁve trials, for subjects previously exposed to inescapable shocks. The ﬁrst, second, and fourth graphs refer to the jumping
test and the third graph (unﬁlled circles) refers to the nose poking test.
182 C.V.d. Santos et al. / Behavioural Processes 86 (2011) 178–183
ing be affected when re-exposed to the shuttle box in case subjects
learned to nose-poke (as in the previous two experiments)? The
answer to all three question seems to be no. Arranging 60 trials
in two days had no effect on jumping acquisition for all subjects
except one. It is possible that arranging even more trials would
facilitate responding, but this seems unlikely given each shock
that an animal fails to escape is a new “inescapable” shock that
might add to the inescapable shocks it had received previously.
Nose poking, on the other hand, was not affected by the added
exposure to shocks and learning to nose poke did not affect subsequent
performance in the shuttle box either, which suggests that
these response topographies might differ in some critical feature
that makes them particularly sensitive or insensitive to the effects
of inescapable shocks. Four possible candidates are discussed
below.
5. General discussion
Jumping and nose poking (and their respective experimental
preparations) differ in some features that might be – together or
in isolation – responsible for these results. The ﬁrst one could
be response difﬁculty. It could be argued that jumping might
be more difﬁcult than nose poking (i.e., involving a longer chain
of responses), and inescapable shocks might affect only difﬁcult
responses. In previous studies, a learning impairment was observed
only when the response required after exposure to inescapable
shocks was made more difﬁcult, such as when rats were required
to run back and forth in a shuttle box (Maier et al., 1973) or press
a lever three times while receiving shocks (Seligman and Beagley,
1975). However, we have no evidence that jumping was actually
more difﬁcult than nose poking. When naïve subjects are exposed
to the jumping contingency, their response pattern is virtually identical
to naïve subjects exposed to the nose-poking contingency:
higher latencies in the ﬁrst trials and a gradual decrease throughout
the session (Yano and Hunziker, 2000). Thus, it seems unlikely
that response difﬁculty played a role.
The second feature is context similarity. Nose poking tests were
conducted in the same experimental chambers used to deliver
inescapable shocks, except that the opening in the wall was covered
by a metal plate, making nose poking impossible, whereas
jumping tests were conducted in a different chamber (shuttle box).
It might be argued that context similarity facilitated responding
in some way. However, there is evidence that context similarity,
rather than mitigating the effect of inescapable shocks, maximizes
it. For example, in an experiment conducted by Maier et al. (1995),
rats were exposed to inescapable shocks and later tested for escape
responding in an environment that could be either the same as
or different from where they had received inescapable shocks.
When tests occurred in the same environment, learned helplessness
was observed seven days after exposure to inescapable shocks;
when the test environment was different, helplessness was not
observed after two days. Thus, it also seems unlikely that being
tested in the same chamber where inescapable shocks were delivered
should facilitate subsequent responding that results in shock
termination.
A third possible feature is the amount of motor activity required
to emit both responses. If the exposure to inescapable shocks
reduces motor activity, as has been suggested by some (Glazer
and Weiss, 1976), escape learning should be facilitated when nose
poking is used in the test – because nose poking entails lower
activity – and impaired when jumping is used, since a high level
of activity is needed to emit this response. However, systematic
observations (not presented) of the behavior of the rats during electric
shocks suggested a high level of motor activity in both tests,
which should have caused the opposite of what was observed.
Nevertheless, a careful and precise measurement of the level of
motor activity during shocks is needed to rule out this possibility
altogether.
Finally, in Yano and Hunziker’s study, electromechanical equipment
was used to control events in the session, while the events
in our sessions were controlled by a computer. We hypothesize
that the use of electromechanical equipment may have introduced
a short delay between response emission and shock offset and this
short delay may have made the detection of the contingency more
difﬁcult. Supportive evidence for this hypothesis was provided by
Minor et al. (1984), who exposed rats to inescapable shocks and
later to an escape task where shocks could be terminated by choosing
one arm of a Y-shaped maze; choosing the other arm had no
effect over shocks. When there was a short delay (350 ms) between
response emission and shock offset, performance on the task was
poorer compared to a no-delay group. This effect was greatly magniﬁed
when, in addition to the delay, there was an experimenter
in the room. The authors concluded that inescapable shocks might
render animals more sensitive to irrelevant features of the environment
(for example, the small delay between response and shock
offset) and less sensitive to proprioceptive cues that might lead to
the correct response. So, if inescapable shocks make the subject
more sensitive to different characteristics of the contingency, the
delay between response emission and shock offset inherent to Yano
and Hunziker’s procedure may have been sufﬁcient to prevent animals
previously exposed to inescapable shocks from detecting a
new contingency between what they did and what happened as a
result. This view that inescapable shocks cause some, but not all, of
the characteristics of the contingency to gain disproportionate control
over the animal’s behavior is also supported by Lee and Maier’s
(1988) work, in which, compared with naïve rats, rats previously
exposed to inescapable shocks were more likely than naïve rats to
ﬁnd the correct arm of a water maze during an escape test when it
was signaled by an accurate cue, which suggests more control by a
particular feature of the external environment.
Interestingly, the hypothesis that inescapable shocks cause
some characteristics of the contingency to gain more rather than
less control over the animal’s behavior also may help explain why
learning to jump was impaired even when subjects had learnt to
escape shocks by nose-poking the day before. This datum is inconsistent
with a learned helplessness hypothesis if we consider that
experiencing an actual contingency between behavior and shock
offset should have counteracted the previous experience where
such a contingency was absent. However, a difference in the delays
between responses and shock offset also may help account for these
results. When nose poking was required, subjects only had to insert
the nose (or any body part) within the oriﬁce in the wall and this
caused an immediate termination of the shock. Jumping, on the
other hand, required the subject to position itself facing the opening
in wall that separated the compartments and propel its body
through it, receiving electric shocks in the meantime. We suggest
that this small difference in delay between response initiation and
shock termination may have made the detection of a contingency
more or less difﬁcult and this small delay may be very important
for rats previously exposed to inescapable shocks but not naïve
subjects.
Moreover, another procedural detail might have made the
detection of a contingency with nose-poking easier: when the animal
emitted the correct response, the electric shock that was being
administered through the ﬂoor of the same chamber terminated.
When an animal jumped from one compartment to the other in the
shuttle box, the relationship between response and consequence
is less clear: to escape shocks, the animal removes itself from the
situation where shocks are happening and it does not have experience
with shock elimination per se in the same compartment (i.e.,
shocks could have ended on the other side long after the animal
C.V.d. Santos et al. / Behavioural Processes 86 (2011) 178–183 183
jumped the barrier). So, it seems that the control over shock termination
was made much less explicit when jumping was used in the
test.
The explanation that inescapable shocks cause some characteristics
of the contingency to gain more control over the animal’s
behavior seems particularly promising for two reasons. First, it
provides a more molecular account of the phenomenon, instead
of relying on more molar variables, such as learning that there is
no relationship between what one does and what happens in the
environment. It is not yet clear how an organism might come to
detect this lack of relationship and until this question is answered,
an explanation based on such a process might not be as parsimonious
as one wishes. Second, this explanation may be easily put
to test. For example, if it is correct, then one would expect that
animals exposed to inescapable shocks should not learn to nose
poke if a delay is imposed between the response and shock offset.
These empirical tests might tell us which explanation better suits
the results.
It is important to note that our data do not refute the learned
helplessness hypothesis, given that control groups not exposed
to inescapable shocks were not run in the present study. Control
subjects, however, do learn to jump barriers to escape shocks,
as reported elsewhere (Hunziker and Santos, 2007; Sanavio and
Savardi, 1980). The reduction in response latencies when nosepoking
was required that we observed here also is evidence of
learning, and this response pattern is in clear contrast to the pattern
observed with jumping. Therefore, our data might be in a way
comparable to previous work that observed impairment in performance
after exposure to inescapable shocks. However, given that
the hypothesis of learned helplessness emphasizes the lack of control
over shocks as the critical variable to produce to effect, it is
difﬁcult to see how it might explain why an experience of control
over shocks does not override the effects of inescapable shocks.
Acknowledgements
The present research was supported by FAPESP (process number
01/13097-5 and 08/51294-6) and CNPq (process number
306007/2006-1). Correspondence and requests for reprints may
be sent to the ﬁrst author at the Centro de Estudios e Investigaciones
en Comportamiento, Universidad de Guadalajara. Calle
Francisco de Quevedo 180, CP 44130 Guadalajara, Mexico (e-mail:
behaca@yahoo.com).
References
Capelari, A., Hunziker, M.H.L., 2009. Aprendizagem discriminativa após choques
incontroláveis (Discriminative learning after uncontrollable shocks). Interac¸ ão
em Psicologia 13 (1), 81–90.
Castelli, M.C.Z., 2004. Efeitos da (im)previsibilidade de estímulos no desamparo
aprendido: uma comparac¸ ão entre ratos machos e fêmeas [Effects of stimulus
(un)predictability on learned helplessness: male and female rats compared].
Unpublished doctoral dissertation, Universidade de São Paulo, Brasil. Available
from: http://www.usp.br/labc/downloads/castelli2004.pdf (retrieved 26.02.10).
Glazer, H.I., Weiss, J.M., 1976. Long-term interference effect: an alternative to
learned helplessness. J. Exp. Psychol.: Anim. Behav. Proc. 2, 202–213.
Gouveia Jr., A., 2001. Efeitos da administrac¸ ão aguda de imipramina sobre o desamparo
aprendido em ratos machos e fêmeas (Effects of acute administration of
imipramine on learned helplessness of male and female rats). Acta Comport. 9,
19–29.
Hunziker, M.H.L., Santos, C.V., 2007. Learned helplessness: effects of response
requirement and the interval between treatment and testing. Behav. Proc. 76,
183–191.
Kirk, R.C., Blampied, N.M., 1985. Activity during inescapable shock and subsequent
escape avoidance learning: female and male rats compared. J. Psychol. 14, 9–13.
Lee, R.K.K., Maier, S.F., 1988. Inescapable shock and attention to internal versus
external cues in a water escape discrimination task. J. Exp. Psychol.: Anim. Behav.
Proc. 14, 302–311.
Maier, S.F., 2001. Exposure to the stressor environment prevents the temporal
dissipation of behavioral depression/learned helplessness. Biol. Psychiatry 49,
763–773.
Maier, S.F., Albin, R.W., Testa, T.J., 1973. Failure to learn to escape in rats previously
exposed to inescapable shock depends on nature of escape response. J. Comp.
Physiol. Psychol. 85, 581–592.
Maier, S.F., Grahn, R.E., Watkins, L.R., 1995. 8-OH-DPAT microinjected in the region
of the dorsal raphe nucleus blocks and reverses the enhancement of fear conditioning
and the interference with escape produced by exposure to inescapable
shock. Behav. Neurosci. 109, 404–413.
Maier, S.F., Seligman, M.E.P., 1976. Learned helplessness: theory and evidence. J. Exp.
Psychol. Gen. 105, 3–46.
Mauk, M.D., Pavur, E.J., 1979. Interconsequence generality of learned helplessness.
Bull. Psychon. Soc. 14, 421–423.
Minor, T.R., Jackson, R.L., Maier, S.F., 1984. Effects of task-irrelevant cues and
reinforcement delay on choice-escape learning following inescapable shock:
evidence for a deﬁcit in selective attention. J. Exp. Psychol.: Anim. Behav. Proc.
10, 543–556.
Peterson, C., Maier, S.F., Seligman, M.E.P., 1993. Learned Helplessness: A Theory for
the Age of Personal Control. Oxford University Press, New Cork, 376 pp.
Rosellini, R.A., DeCola, J.P., Shapiro, N.R., 1982. Cross-motivational effects of
inescapable shock are associative in nature. J. Exp. Psychol.: Anim. Behav. Proc.
8 (4), 376–388.
Sanavio, E., Savardi, U., 1980. Observational learning of a discriminative shuttlebox
avoidance by rats. Psychol. Rep. 44, 1151–1154.
Seligman, M.E.P., Beagley, G., 1975. Learned helplessness in the rat. J. Comp. Physiol.
Psychol. 88, 534–541.
Seligman, M.E.P., Maier, S.F., 1967. Failure to escape traumatic shock. J. Exp. Psychol.
74, 1–9.
Yano, Y., Hunziker, M.H.L., 2000. Desamparo aprendido e imunizac¸ ão com diferentes
respostas de fuga (Learned helplessness and immunization with different escape
responses in rats). Acta Comport. 8, 143–166.