Content uploaded by Sunjeev K Kamboj
Author content
All content in this area was uploaded by Sunjeev K Kamboj on Mar 05, 2024
Content may be subject to copyright.
ORIGINAL INVESTIGATION
Cannabidiol enhances consolidation of explicit fear extinction
in humans
Ravi K. Das &Sunjeev K. Kamboj &Mayurun Ramadas &
Kishoj Yogan &Vivek Gupta &Emily Redman &
H. Valerie Curran &Celia J. A. Morgan
Received: 24 September 2012 /Accepted: 6 December 2012 / Published online: 10 January 2013
#Springer-Verlag Berlin Heidelberg 2013
Abstract
Rationale Whilst Cannabidiol (CBD), a non-psychotomimetic
cannabinoid, has been shown to enhance extinction learning in
rats, its effects on fear memory in humans have not previously
been studied.
Objectives We employed a Pavlovian fear-conditioning par-
adigm in order to assess the effects of CBD on extinction
and consolidation.
Method Forty-eight participants were conditioned to a col-
oured box (CS) with electric shocks (UCS) in one context and
were extinguished in a second context. Participants received
32 mg of CBD either following before or after extinction in a
double-blind, placebo-controlled design. At recall, 48 h later,
participants were exposed to CSs and conditioning contexts
before (recall) and after (reinstatement) exposure to the UCS.
Skin conductance and shock expectancy measures of condi-
tioned responding were recorded throughout.
Results Successful conditioning, extinction and recall were
found in all three treatment groups. CBD given post-
extinction enhanced consolidation of extinction learning as
assessed by shock expectancy. CBD administered at either
time produced trend level reduction in reinstatement of
autonomic contextual responding. No acute effects of
CBD were found on extinction.
Conclusions These findings provide the first evidence that
CBD can enhance consolidation of extinction learning in
humans and suggest that CBD may have potential as an
adjunct to extinction-based therapies for anxiety disorders.
Keywords Cannabidiol .Cannabinoids .Conditioning .
Extinction .Consolidation .Pavlovian .Anxiety
Introduction
Pavlovian fear learning mechanisms (Pavlov 1927) are im-
portant in anxiety disorders (Watson and Rayner 1920;
Mineka and Zinbarg 1996). Impaired naturalistic extinction
is thought to be a major contributor to fear persistence in
these disorders. Pharmacological treatments that acutely
enhance extinction learning or potentiate its consolidation
(post-learning stabilisation; Müller and Pilzecker 1900;
Dudai 2004) therefore hold significant promise in improv-
ing the efficacy of exposure-based psychological treatments
for anxiety disorders.
Recently, the endocannabinoid (eCB) system has gained
attention for its role in learning and memory. The eCB
system is comprised of two classes of receptors: CB
1
and
CB
2
, endogenous cannabinoids (eCBs); anandamide and 2-
arachidonylglycerol (2-AG) and enzymes for their hydroly-
sis. Pre-synaptic CB
1
receptors (CB
1
Rs) are expressed most
densely in the hippocampus, amygdala and prefrontal cortex
(Freund et al. 2003) and modulate the firing of both excit-
atory glutamatergic and inhibitory GABAergic neurons in
the amygdala. They thus play an important role in emotional
learning.
However, the role of the eCB system on extinction
learning remains poorly understood. Despite cognitive
impairing effects of Δ9-THC, both CB
1
agonist WIN
55,212-2 (Pamplona et al. 2006; Bitencourt et al. 2008;
but see Chhatwal et al. 2005) and AM404, an anandamide
reuptake inhibitor (Beltramo et al. 1997;Fowleretal.
2004), potentiate extinction of conditioned fear. Converse-
ly, CB
1
R knockout or antagonism generally impairs fear
extinction (Kaplan and Moore 2011; Pamplona et al. 2006)
R. K. Das (*):S. K. Kamboj :E. Redman :H. V. Curran :
C. J. A. Morgan
Clinical Psychopharmacology Unit, University College London,
Gower Street,
London WC1E 7HB, UK
e-mail: ravi.k.das@gmail.com
M. Ramadas :K. Yogan :V. Gupta
UCL Medical School, University College London, London, UK
Psychopharmacology (2013) 226:781–792
DOI 10.1007/s00213-012-2955-y
and increases memory rigidity (as evidenced by failures in
reversal learning; Lutz 2007; Varvel and Lichtman 2002),
leaving fear acquisition intact. These findings suggest spe-
cific and dissociable roles of CB
1
Rs in fear conditioning
and extinction (Marsicano et al. 2002). Thus, CB
1
R blockade
may impair acquisition of extinction learning (Chhatwal et
al. 2005; Marsicano et al. 2002; Suzuki et al. 2004; Niyuhire
et al. 2007). However, glucocorticoid-induced eCB synthesis
and downstream disinhibition of noradrenergic neurons in
the basolateral amygdala (BLA) may also enhance memory
consolidation (Hill and McEwen 2009; de Bitencourt et al.
2013). Consistent with this, WIN 55,212-2 enhances mem-
ory consolidation, and intra-BLA infusions of the CB
1
re-
ceptor antagonist rimonabant prevent the facilitation of fear
memory consolidation by systemic corticosterone in vivo
(Campolongo et al. 2009). ECBs may thus be critical in both
the acquisition and maintenance of conditioned fearful
responding. As such, the eCB system may be an important
target for improving the efficacy of treatment for anxiety
disorders.
Cannabidiol (CBD), a non-psychoactive constituent of
cannabis with effects on the eCB system, is a promising
compound in this respect and has been shown to enhance
extinction in rat models of conditioned fear (Bitencourt et al.
2008). However, the action of CBD is complex and not fully
characterised. It has been shown to have some antagonistic
action at the CB
1
and CB
2
receptor (Thomas et al. 2007;
Pertwee 2008), but binding affinity at these receptors is
generally low (Petitet et al. 1998). Recent evidence suggests
that it exerts its behavioural effects via inhibition of fatty
acid amide hydrolase (FAAH), an enzyme that hydrolyses
endogenous cannabinoids (Leweke et al. 2012). CBD may
thus increase anandamide levels by reducing intracellular
anandamide breakdown. It may therefore be expected to
exert similar effects to AM404 on memory. Indeed, previous
research comparing the two compounds supports this hy-
pothesis, with both CBD and AM404 similarly potentiating
extinction when administered immediately prior to training
(independently of any anxiolytic effects) in animal models
of anxiety (Bitencourt et al. 2008). Given its excellent safety
profile in humans (Nurmikko et al. 2007; Bergamaschi et al.
2011), absence of acute subjective effects at low doses and
promising findings on enhancement of extinction learning in
animals, CBD may have a significant clinical potential as an
adjunct to exposure therapies for anxiety disorders. However,
no study to date has examined the effects of CBD on fear
memory in humans.
The present study therefore aimed to assess whether
CBD has potential to help in the treatment of pathological
fear memories by examining its effects on fear memory in
healthy volunteers. A contextual fear-conditioning para-
digm was employed in order to assess the potential effects
of CBD on the acquisition or consolidation of extinction.
We tested the effect of a sub-anxiolytic dose of CBD in
two groups who received the drug either before (so that
CBD would be active during acquisition and consolidation
of extinction) or after (active during consolidation only)
extinction in order to dissociate the contribution of acqui-
sition and consolidation mechanisms. Based on the
hypothesised action of CBD as an FAAH inhibitor, we
predicted that it would potentiate the acquisition (when
given before) and consolidation (when given after) of
extinction. Since conditioned fear responding may reflect
explicit, propositional knowledge as well as relatively
automatic processes (Lovibond and Shanks 2002; Grillon
2009; Bechara et al. 1995), we examined both the explicit
(expectancy ratings) and autonomic (skin conductance
response) aspects of conditioned responding. Such assess-
ment is relevant as our previous findings suggest that
CBD’s ameliorating effects on neuropsychological func-
tioning were evident in measures of explicit, declarative
memory (Morgan et al. 2010).
Methods
Design and participants
In a double-blind, placebo-controlled between-subjects de-
sign, 48 participants were randomised to three groups
(each n=16) to receive either (1) 32 mg of inhaled CBD
prior to extinction (CBD pre-extinction group), (2) 32 mg
of inhaled CBD following extinction (CBD post-extinction
group) or (3) placebo (placebo group). The participants
were recruited via community advertisement and word of
mouth; the inclusion criteria were: 18–35 years old, fluent
in English, no history of serious mental or physical health
problems, no substance misuse problems, normal or
corrected-to-normal colour vision, no learning impair-
ments or neurological history, not pregnant. The study
was approved by the University College London research
ethics committee.
Procedure
After collecting written, witnessed, informed consent, basic
demographic data were collected. The Beck Depression
Inventory (Beck et al. 1996), Spielberger State-Trait Anx-
iety Inventory (Spielberger et al. 1983), Spot-the-Word
(Baddeley et al. 1993) and prose recall task then provided
measures of depressive symptoms, trait anxiety, verbal IQ,
and non-emotional explicit memory, respectively. A drug-
use interview assessed participants’use of psychoactive
substances.
The participants then underwent fear conditioning (de-
scribed below) after which the Mood Rating Scale
782 Psychopharmacology (2013) 226:781–792
(MRS) (Bond and Lader 1974) and Bodily Symptoms
Scale (BSS) were used to assess anxiety, current mood
and physical symptoms. The first drug ‘balloon’was then
administered. The participants in the CBD pre-extinction
group inhaled 32 mg of CBD at this point, while mem-
bers of the other two groups inhaled placebo. Five
minutes following drug administration, the MRS and
BSS were again given to the participants, who then
completed the fear extinction task. Finally, the second
drug balloon was administered. Those in the CBD post-
extinction group then inhaled 32 mg of CBD, the other
groups inhaled placebo, and final mood rating and sub-
jective anxiety measures were given. This concluded the
first day of testing.
Twenty-four hours after the first testing session, the
participants first completed the MRS and BSS and
delayed prose recall. Participants then completed the fear
recall task (described in full under the “Tasks”section).
Following this, the participants were debriefed and testing
was concluded.
Drug administration
CBD and placebo ethanol were provided by STI Pharma-
ceuticals (UK). All drugs were vaporised at 210 °C and
administered via a Volcano Medic vaporiser (Storz &
Bickel, Tuttlingen, Germany). Thirty-two milligrammes of
CBD in 0.08 mg ethanol vehicle was administered to those
in the CBD groups, while 0.08 mg ethanol vehicle only was
used as placebo. This dose of vaporised CBD was chosen
based on extensive pilot work with the vaporiser as an
administration medium and evidence from a previous study
in our lab where 32 mg of CBD was not shown to affect
anxiety levels (Morgan et al. in preparation). The anxiety
subscale of the MRS was used to confirm that this dose of
CBD was sub-anxiolytic.
The participants were told to fast for 4 h prior to testing in
order to facilitate CBD absorption (CBD is lipid soluble)
and maximise peak plasma levels of CBD during the learn-
ing and consolidation windows. In order to maintain the
drug blinding, drugs were prepared and administered by a
separate experimenter to the one collecting behavioural
data. Pharmacological interaction between CBD and ethanol
in the preparation was highly unlikely given the extremely
low dose of ethanol. Previous use of this CBD preparation
in our lab suggested no interaction between the compounds
(Morgan et al. in preparation).
Drugs were administered on a 10-s inhalation cycle.
Participants were instructed to fully exhale, then fully
inhale from the Volcano Medic balloon, hold their breath
for 10 s then fully exhale. This procedure was repeated
until the balloon was empty. All the participants were
familiarised with the inhalation protocol using a placebo
balloon before actual drug administration. In order to main-
tain the drug blinding, drugs were prepared and adminis-
tered by a separate experimenter to the one collecting
behavioural data.
Tasks
Day 1
Conditioning Fear conditioning took place in a virtual
‘context’presented on a 17-inch computer monitor:
‘Room A’, defined by screen background which depicted
a white-walled room (Fig. 1).The unconditioned stimulus
(UCS) was a 250-ms, 4-mA electric shock generated by a
Psychlab generator (Contact Precision Instruments). Con-
ditioned stimuli (CSs) were red and yellow boxes that
appeared on the context background. One of the boxes
(the CS+) co-terminated with the UCS on 50 % of trials.
The other CS (CS−) was never paired with a shock.
Which stimulus (i.e. the red or the yellow box) acted as
the CS+ and CS−was counterbalanced across participants
and drug groups. The participants were first habituated to
both CSs and the context by two presentations of each CS
in the conditioning context. They also received a single
test shock before the start of the task. Each conditioning
trial consisted of context background alone for 10 s, fol-
lowed by CS in the context for between 10 and 14 s
(mean=12 s). Every trial was followed by a plain grey
screen inter-trial interval of 15 s.
Thirty-two conditioning trials (16 of each CS) were rand-
omised with the stipulation that no more than two of one CS
type could appear in succession. At the beginning of the
task, the participants were instructed that they would see red
and yellow boxes in ‘Room A’and that they would occa-
sionally receive shocks. They were instructed that they
should try to discern the relationship between the boxes
and the shocks. Throughout the conditioning trials, skin
conductance response (SCR) was measured via silver/silver
chloride electrodes attached to the medial phalanges of the
index and middle fingers of participants’non-dominant
hands. SCR was amplified and recorded by a Psychlab
modular skin conductance amplifier unit (Contact Precision
Instruments) with a 0.02 µS low-pass filter. In order to
assess explicit learning of the contingency between the
CS+ and UCS, shock expectancy ratings were collected
on every trial. When the CS appeared, the participants
rated (with their free hand) how much they expected to
receive a shock from 0 (certain there will not be a
shock) to 5 (certain there will be a shock).
Extinction Fear extinction was carried out in ’Room B’,
defined by screen background (Fig. 1). Prior to the extinc-
tion task, participants were told that they would be seeing
Psychopharmacology (2013) 226:781–792 783
the same boxes that they had seen in the conditioning task,
but that they would now be appearing in ‘Room B’. The
extinction task was otherwise identical to conditioning,
except no shocks were presented.
Recall The fear recall task, 24 h later, consisted of two
phases: recall and reinstatement. In the recall phase, CSs
were presented in the conditioning and extinction con-
texts in an alternating manner. The reinstatement phase
was identical to recall, except prior to the reinstatement
trials, the UCS was presented in each context once in the
absence of any CSs. At the beginning of the task, par-
ticipants were told that they would be seeing the same
boxes as the first day in the same rooms to maximise
reliance on memory from conditioning and extinction and
to avoid occasion-setting effects. SCR was continuously
recorded through the task and shock expectancy ratings
were made twice per trial, one when the context alone
was visible and one when the CS appeared. This rating
protocol was used in order to examine utilisation of cue
and context information (Fig. 2).
Data analyses
Skin conductance
All skin conductance records we collected at a sampling rate
of 100 Hz and down-sampled to 50 Hz prior to analysis.
Records were smoothed and visually inspected for move-
ment artefacts. If artefacts were found, the trial in which
they occurred was excluded from the analysis. SCRs were
calculated for the last 10 s of each CS presentation during
fear conditioning. Analysis was performed in this way to
best capture anticipatory autonomic responses. SCRs were
calculated by subtracting the skin conductance level at the
start of the trial from the maximum level reached within the.
If skin conductance level flat-lined or reduced throughout
the block, responses were marked as 0.
50 %
Conditioning: Room A
X
DAY 1
10 s
10 - 14 s
15 s
10 s
10 - 14 s
15 s 15 s
Extinction: Room B
10 - 14 s
15 s
10 s
10 s
10 - 14 s
Fig. 1 Conditioning and
extinction tasks. This figure
represents two trials of
conditioning and extinction to
demonstrate trial timing and
shock contingencies. For both
conditioning and extinction,
this cycle is repeated, in a
pseudo-randomised order, until
each CS had been repeated 16
times
784 Psychopharmacology (2013) 226:781–792
Other analyses
Data were analysed with IBM SPSS version 19 for Windows.
Mixed analyses of variance (ANOVAs) were used to analyse
SCR, subjective and computer-based response data. Signifi-
cant and trend-level (p<.1) between-subjects main effects and
interactions were followed up with, Dunnett’stpost hoc tests
(as we were interested in comparing effects of CBD to placebo
control), ttests or planned Helmert contrasts, where appropri-
ate. Where sphericity was violated, a Greenhouse-Geisser
correction was applied. Adjusted dfs and pvalues are reported
in this instance. Identical ANOVAs were used for shock
expectancy and SCR data and for the recall and reinstatement
phases. For conditioning, these were 2×4×3 RMANOVAs
with within-subjects factors of CS (CS+/CS−) and block (four
blocks of four trials each) and a between-subject factor of
group (placebo, CBD pre-extinction, CBD post-extinction).
For extinction, an extra level of block was added to compare
the last block of conditioning to extinction blocks, yielding 2 ×
5×3 RMANOVAs. For the recall and reinstatement tasks,
responding to contexts alone was assessed by 2 (conditioning
context/extinction context)×6 (trial 1–6)× 3 (group) RMA-
NOVAs, and responding to CSs within contexts by 2(CS+/
CS−)× 2(conditioning context/extinction context)×3 (trial 1, 2
or 3)× 3 (group) RMANOVAs.
Results
Demographics The groups did not differ on any standar-
dised questionnaire measures of mental health (anxiety,
depression or verbal IQ). Some participants had used
15s
15s
15s
15s
10s
10s
10s
10s
10s
10s
10s
10s
Recall Reinstatement
15s
15s
15s
15s
10s
10s
10s
10s
10s
10s
10s
10s
Reinstating UCS
Presentations
Fig. 2 In the recall task,
contexts were presented in an
alternating fashion, with each
CS appearing in each context
before context change. Two
trials are presented in this figure
to demonstrate this trial
ordering. The reinstatement
phase was identical to the recall
phase but was preceded by a
single UCS presentation in each
context in order to reactivate
contextual fear memory. In both
phases, each CS appeared in
each context three times
Psychopharmacology (2013) 226:781–792 785
tobacco and cannabis. There were no differences between
drug groups in use, with the exception of recency of tobacco
use [F(2,20)=3.882, p=.038], due to more recent tobacco
use in the CBD pre-extinction group than the CBD post-
extinction group [t(12)=2.617, p=.023] see Table 1for
descriptive statistics.
Associative fear learning
Excluded and missing data
A proportion of skin conductance datafor extinction trials was
lost due to equipment failure and participant non-response for
each task. Final Ns at each stage are as follows (all given as
placebo N, CBD pre-extinction N, CBD post-extinction N);
conditioning n=13, n=15, n=14; extinction n=8, n=13, n=
12; recall n=12, n=13, n=13, reinstatement n=12, n=12, n=
12. For shock expectancy, five participants failed to make
responses during the recall task, leaving n=15, n=15, n=13.
Habituation
No differences in SCRs to the CS+ or CS−or between groups
were found during habituation (all p>.2). Stimuli and groups
were thus well matched for responding at baseline.
Conditioning
SCRs Rapid conditioning was shown by a main effect of CS
[CS+>CS−;F(1,39)=18.50, p<.001, η
p
2
=.322]. Significant
conditioned responding between CSs was evident from the
second block onwards (all p<0.005) with a trend for
conditioned responding in the first block [t(40)= 2.005,
p=.052]. As it is clearly shown in Fig. 3,successfulcondi-
tioning of SCRs was achieved within the first two blocks. A
main effect of block [F(2.40,93.2)=3.85, p<.03, η
p
2
=.084],
reflected a general decrease in SCRs over the course of the
trials indicating habituation to the UCS. This was driven by a
significant decrease in SCRs from the first block to later
blocks only [F(1,37)=10.266, p=.003, η
p
2
=.217]. Together
these data suggest ceiling-level conditioning by the end of the
second block.
Expectancy Conditioning was also evidenced by expectancy
ratings which showed a main effect of CS (CS+ >CS−);
[F(1,44)= 131.845, p< .001, η
p
2
=.75], qualified by a CS×
block interaction [F(2.34,102.94)= 5.323, p=.005,
η
p
2
=.106], indicating a significant reduction in ratings for
the CS−between blocks 1 and 2 [t(46)=4.668, p<.001] but
no reduction in ratings for the CS+ (all p>.05). Concordant
with SCR data, this shows rapid conditioning.
Extinction
SCRs Extinction was evidenced by a main effect of block [F
(2.838,79.367)= 4.977, p=.004, η
p
2
=.142], showing a signif-
icant decrease in responding between the last block of condi-
tioning and all extinction blocks [F(1,30) = 6.092, p=.02,
η
p
2
=.169] and between the first and later blocks of extinction
trials [F(1,30) = 12.983, p= .001]. A borderline significant
main effect of CS was also found [F(1,28)=4.23, p=.05,
η
p
2
=.131]. Helmert contrasts showed that this was driven by
greater conditioned responding (CS+>CS−)inthelast
conditioning block compared to all extinction blocks
[F(1,28)=5.228, p=.03, η
p
2
=.157]. Confirmatory ttests
Table 1 Descriptive statistics
for baseline demographic data
BDI Beck depression inventory
CBD pre-extinction (n=16) CBD post-extinction (n=16) Placebo (n=16)
Gender (M/F) 8:8 10:6 12:4
No. of alcohol users 13 12 13
Alcohol days per month 8.69± 5.57 7.96± 5.23 8.23± 5.76
Alcohol units per session 4.73± 2.2 6± 3.93 6.54± 4.29
Days since last use 3.38± 3.93 19.42± 14.5 6.85± 8.67
No. of cigarette smokers 8 6 9
Tobacco days per month 7.31±10.25 5.67± 11.98 1.67± 1.41
Cigarettes per day 1.56± 1.31 1± 0.89 1.11± 0.93
Days since last use 9.25± 9.95 330.25±325.65 171.75±234.12
No. of cannabis users 4 6 4
Cannabis days per month 3.17±3.82 1.75±2.22 0.75±0.5
Cannabis time to smoke eighth 4.5 ± 3.54 12.33 ± 6.81 5.75± 1.5
Days since last use 55.7± 59.25 223.5 ± 320.62 67.5± 95.39
Spielberger Trait Anxiety score 35.88± 6.4 36± 10.06 36.29± 5.65
BDI score 5.63±4.3 6.27 ±6.2 5.88± 4.33
Spot-the-Word score 47.33± 5.3 48.6± 3.64 50.06± 3.05
786 Psychopharmacology (2013) 226:781–792
showed a significant effect of CS (CS+>CS−)inthelast
block of conditioning only [t(41)= 3.355, p= .002] with
no differences between CSs during any extinction
blocks. Thus successful extinction occurred within the
first two blocks of extinction trials. No group effects
were found.
Expectancy
A CS×block interaction was found for ratings during ex-
tinction [F(2.289,89.268)=11.256, p=<0.001, η
p
2
=.224].
The interaction was driven by a significant increase in shock
expectancy between the last block of conditioning and first
block of extinction for the CS−[t(41)=4.224, p<0.001] but
no change between these time points for CS+ ratings (p>
0.5). This suggests an occasion-setting effect in the early
block of extinction, with participants expecting a contingen-
cy reversal to occur. However, a main effect of block
[F(2.757,107.537)=20.981, p<0.001, η
p
2
=.35] showed sig-
nificant reduction in ratings to both CSs across the first three
blocks of extinction [all p<.001], but not between the third
and last block (p>.5), suggesting that ceiling-level extinc-
tion of expectancies had occurred by the final extinction
block. No group effects were found.
Recall
SCRs
Contexts An interaction was found between group, context
(conditioning or extinction) and trial (1–6) [F(6.68,110.26)=
2.231, p=.039, η
p
2
=.119]. One-way ANOVAs revealed a
trend for a group difference only on the sixth trial in the
extinction context [F(2,33)=3.625, p=.038]. However, post
hoc tests did not reach post-correction significance.
Conditioned stimuli Recall in all drug groups was evidenced
by a context×CS interaction [F(1,35)=6.828, p=.013,
η
p
2
=.163; see Fig. 4], driven by significantly greater SCRs
to the CS+ than the CS−in the conditioning context
[t(37)= 2.796, p=.008] but no differences in the extinction
context. No effects of group or context were found.
Expectancy
Contexts A context×trial interaction [F(3.06,119.34)=4.014,
p=.009, η
p
2
=.093] showed no difference between contexts in
ratings on trials 1 and 2 but greater ratings for the conditioning
context for all trials thereafter (all p<.002).
Conditioned stimuli Explicit recall was shown by a con-
text×CS interaction [F(1,40) = 4.509, p= .04, η
p
2
=.101],
whereby expectancies for the CS+ were higher in the con-
ditioning context than the extinction context [t(42) = 2.738,
p=.009], but no contextual discrimination was found for the
CS−(p>.4) showed. A trend for a main effect of group was
found [F(2,40)=2.991, p=.062, η
p
2
=.13], reflecting lower
post correction ratings in the CBD post-extinction group
than placebo group [p= .047] but no differences between
placebo and CBD pre-extinction.
Reinstatement
SCRs
Contexts A trend for a context ×group interaction was found
[F(2,35)=2.545, p=.097, η
p
2
=.132], suggesting lower
responding to the extinction context in the CBD post-
extinction group than placebo (see Fig. 5). However, no
significant differences emerged following post hoc tests. A
main effect of trial [F(2.696,94.348)= 27.289, p<.001,
η
p
2
=.438] indicated a decrease in overall SCR, driven by a
large decrease in SCRs between the first trial and all later trials
[F(1,35)=50.348, p<0.001].
Conditioned stimuli SCRs in the conditioning context were
higher than the extinction context [F(1,33)=6.206, p=.018,
η
p
2
=.158], indicating reinstatement of contextual responding.
A CS× context interaction was found [F(1,33)= 4.312, p=.046,
η
p
2
=.116], driven by greater responses to the CS+ in the
conditioning context than extinction context [t(37)= 2.707,
p=0.01], but no contextual difference in responding to the
CS−(ps>0.5). A trend for a group×context interaction was
also found [F(2,35)=2.836, p=.073, η
p
2
=.147], driven by a
trend for greater differential responding between contexts in
the placebo group than CBD groups (p= .045), although
this effect did not reach significance post-correction.
Expectancy
Contexts Expectancy ratings were higher overall in the con-
ditioning context than extinction context [F(1,38)=5.032,
***
Fig. 3 Conditioned SCRs to the CS+ and CS−across blocks of
conditioning (one block=four trials). Bars represent mean skin con-
ductance responses±SEM. The asterisks indicate CS+ > CS−(p< 0.05)
Psychopharmacology (2013) 226:781–792 787
p=.031, η
p
2
=.117]. A trial× context interaction was found
[F(5,190)=3.274, p=.007, η
p
2
=.079], driven by a significant
increase between the first and second trials of the conditioning
context only [t(41)=4.471, p<.001].Thisisevidenceofrein-
statement of contextual contingency with the UCS following
the reinstating shock. A trial×group interaction was also
found [F(10,190)=2.274, p=.016, η
p
2
=.107]. An effect of
trial was only found in the CBD post-extinction group
[F(5,65)=3.5, p=.007, η
p
2
=.212] driven by a significant in-
crease in expectancy between the first and second trials only
[F(1,13)= 12.444, p=.003], showing a greater reactivity to
contingency violation by reinstating UCS shocks paired with
context only (and not CS+) in this group only.
Conditioned stimuli A main effect of group was found
[F(1,40)=4.76, p=.014, η
p
2
=.192], driven by lower overall
ratings in the CBD post-extinction group compared to the
placebo [t(26)=2.215, p=.036] and CBD pre-extinction group
[t(26)=3.046, p=.005; see Fig. 6]. Overall recall was shown
by an interaction between CS and context [F(1,40)=15.915,
p<.001, η
p
2
=.285], with greater responses to the CS+ in
the conditioning context than the extinction context
[t(42)=3.789, p<0.001], but no difference for the CS−.
Subjective effects of CBD
Descriptive statistics for all MRS data are given in Table 2.
No effect of group was found at any time on any subscales
of MRS score. A significant main effect of measurement
time was found on anxiety scores [F(3,132) =2.902, p=.037,
η
p
2
=.062]. This was driven by a reduction in anxiety in all
drug groups from before to after inhaling the first balloon,
irrespective of whether the balloon contained drug or place-
bo [t(46)=2.347, p=0.023].
Prose recall
Amaineffectofgroup[F(2,44)=3.305, p=0.046, η
p
2
=.131;
see Fig. 6] reflected greater immediate [t(30)=2.456, p=.02]
and delayed [t(30)= 2.2946, p=.029] recall in the placebo
group than the CBD pre-extinction group. Delayed recall
was poorer than immediate recall time in all groups
0.015
0.065
0.115
0.165
0.215
0.265
0.315
0.365
0.415
0.465
0.515
Placebo CBD Post-conditioning CBD post-ex tinction
SCR (µSieme ns)
Drug Group
Conditioning Context
Extinction Context
Fig. 4 SCRs to the extinction
and conditioning contexts by
drug group during
reinstatement. Bars show mean
±SEM
0
0.05
0.1
0.15
0.2
0.25
0.3
Placebo CBD pre-extinction CBD post-extinction
SCR (µSiemens)
Drug Group
Conditioning Context
Extinction Context
p =0.045
Fig. 5 SCR across groups by context during reinstatement. Bars show
mean± SEM
Fig. 6 Main effect of drug on overall ratings during fear reinstatement
task. Bars show means+SEM. Significant at *p< .05; **p< .01
788 Psychopharmacology (2013) 226:781–792
[F(1,44)=35.402, p<.001, η
p
2
=.446]. To assess whether this
difference in explicit memory capacity could explain memory
differences in the renewal and reinstatement tasks, prose recall
scores were correlated with trial-by-trial scores on skin con-
ductance and shock expectancy in these tasks, but no correla-
tions were found (all ps>.2), suggesting the observed recall
effects are independent of baseline differences in explicit
memory capacity.
Post hoc power calculation Owing to the loss of SCR data,
post hoc power calculations were performed to assess the
impact of the final SCR Ns on power to detect a CBD effect
during the extinction, recall and reinstatement tasks. The η
p
2
value of .192 (the observed effect size of the between-
groups CBD effect on expectancies) was used as the best
available estimate of a possible CBD effect size. This found
that power was not overly compromised for SCR with
achieved power of 0.77, 0.82 and 0.81 for extinction, recall
and reinstatement tasks, respectively.
Discussion
The current study examined the effects of CBD on extinc-
tion and consolidation of contextual conditioned fear mem-
ory. CBD administered after extinction learning led to a
generalised attenuation of explicit fearful responding during
recall and reinstatement, accompanied by greater sensitivity
to violations in learned context/CS/UCS contingencies.
CBD administered either pre- or post-extinction also pro-
duced a trend-level reduction in reinstatement of contextual
responding on SCRs. This suggests that CBD can potentiate
consolidation of extinction memory in humans, leading to
extinction trace dominance at recall.
This is the first study to show extinction-enhancing
effects of CBD in human aversive conditioned memory.
There is currently no information on how CBD may affect
emotional memory in humans when administered after dis-
crete phases of learning (e.g. conditioning and extinction).
These results suggest that, consistent with the known role of
eCBs in memory consolidation in rats (Campolongo et al.
2009), giving CBD after extinction preferentially enhanced
consolidation of extinction memory relative to acquisition
memory, leading to extinction trace dominance and the
global attenuation of explicit fearful responding at test seen
here. A preferential enhancement of extinction consolida-
tion may also partly explain the dissociation between the
effect of CBD given pre- and post-extinction, with peak
plasma levels falling more in the consolidation window in
the latter.
It is interesting that this effect should only be seen in
shock expectancy ratings during recall, although a trend-
level reduction in SCR responses to the extinction context
alone was found during reinstatement in the CBD post-
extinction group, consistent with a potentiation of extinction
consolidation. Such dissociations of psychophysiological
and expectancy metrics in conditioning are well docu-
mented (Mineka 1979; Bechara et al. 1995; Sevenster et
al. 2012) and there is currently debate as to the role explicit
and implicit associative processes play, with authors vari-
ously regarding conditioning as an automatic, low-level
process, a dual-level process (Ohman and Minkea 2001),
or a more propositional process, where UCS expectancy
plays a central role (DeHouwer 2009). The current findings
highlight the importance of collecting multiple measures of
conditioning in human studies (Boddez et al. 2012), espe-
cially if conditioning and extinction models are to be ap-
plied successfully to anxiety disorders. The finding that
CBD can attenuate fearful expectancy processes is therefore
striking and highly encouraging for its use in anxiety
patients where expectancy of fearful outcomes plays a cen-
tral role in disease symptomatology (Reiss 1991).
A trend for a reduction in reinstatement of contextually
discriminant responding in the presence of conditioned
Table 2 Mean MRS subscale
scores pre- and post-drug ad-
ministration across drug groups
MRS subscale Time CBD pre-extinction CBD post-extinction Placebo
Anxiety Pre-inhalation 1 31.17±19.54 35.67±23.52 34.15 ± 31.82
Post-inhalation 1 21.93 ±23.98 30.57 ±25.67 22.47± 21.98
Pre-inhalation 2 24.3± 17.51 32.53±24.55 15.91 ± 12.93
Post-inhalation 2 21.97 ±16.33 38.9± 26.8 18.91± 19.24
Discontentedness Pre-inhalation 1 23.61± 9.12 28.95 ± 15.42 24.45±21.49
Post-inhalation 1 22.79 ±11.27 25.52 ±16.18 22.62± 19.12
Pre-inhalation 2 22.53±12.82 22.91±14.83 19.17 ± 12.99
Post-inhalation 2 22.01 ±10.58 29.56 ±18.28 16.87± 13.98
Sedation Pre-inhalation 1 26.5 ± 13.64 28.4 ±16.26 24.94± 13.67
Post-inhalation 1 34.34 ±24.71 33.84 ±20.73 30.22± 20.68
Pre-inhalation 2 34.59±19.81 27.64±18.64 28.36 ± 21.97
Post-inhalation 2 32± 18 33.94± 20.65 29.12 ± 18.97
Psychopharmacology (2013) 226:781–792 789
stimuli was seen in both groups administered CBD, suggest-
ing that CBD administered either before or after extinction
may provide some protection against reinstatement of con-
textual memory. This may be due to some lingering action
of CBD during consolidation (albeit at a lower level than
CBD post-extinction) in the pre-extinction CBD group. In
contrast to previous preclinical findings (Pamplona et al.
2008), no acute effects of CBD on extinction were observed,
although very rapid ceiling-level extinction made seeing a
subtle drug effect on extinction rate unlikely. As this rapid
extinction learning is typical in humans, acute potentiation
of extinction may not be a feasible or necessary mechanism
for enhancing extinction-based learning in anxiety disor-
ders, with endurance and dominance of the extinction trace
during memory retrieval more likely important for long-
term outcome improvement.
As noted above, these reinstatement effects may reflect
enhanced memory flexibility following CBD. Activation of
the eCB system is critical in behavioural flexibility and
emotional re-learning, processes which underlie successful
extinction learning following conditioning (Marsicano et al.
2002; Pamplona et al. 2008; Hill et al. 2006), with CB
1
R
knockout mice showing deficits in reversal learning (Varvel
and Lichtman 2002). Consistent with the hypothesised role
of CBD as an inhibitor of the transport and hydrolysis of
anandamide, CBD may increase the flexibility of acquired
memory traces (as evidenced by potentiated extinction learn-
ing) via increasing extracellular anandamide. This flexibility
may then persist at test via glucocorticoid/eCB-mediated
consolidation enhancement (Campolongo et al. 2009).
Importantly, the reinstating shocks (one administered in
each context) violated learned CS/context/UCS contingen-
cies. Before these UCS presentations, the UCS had never
been presented in the absence of the CS+ and never associ-
ated with purely contextual information. That these viola-
tions reinforced differential contextual responding during
reinstatement in the placebo group, demonstrates a rigidity
of conditioned memory in this group that was resistant to
contingency violation. That is, novel context/UCS pairings
reinstated learned contextual fear contingencies, rather than
shifting responding towards new learning (as was the case in
the CBD groups). The lack of contextual differentiation
following this contingency violation in both CBD groups
may be indicative of a rapid shift in responding based on
new information (i.e. greater memory flexibility), with less
reactivity to stimuli that no longer predicted aversive out-
comes. This resistance to reinstatement effects is consistent
with a more dominant extinction trace and increased flexi-
bility at test. The increase in responding during presentation
of contexts only (when reinstating UCSs were administered)
within the first two trials in the CBD post-extinction group
further supports this interpretation, indicating a shift in UCS
expectancy from old learned associations to new learning.
Future studies should pay close attention to the point at
which violations of expected and actual outcomes occurs,
as this mismatch is thought to be critical for triggering
extinction or reconsolidation of memory traces (Osan et al.
2011) and CBD has recently been found to have effects on
the latter process (Stern et al. 2012). While disrupted recon-
solidation or enhanced extinction may present the same
effect at test, complicating interpretation of results, the pres-
ent findings cannot be attributed to reconsolidation blockade
due to the lack of sufficient time between conditioning and
extinction for memory consolidation to occur.
These findings have important implications for the use of
CBD as an adjunct to extinction-based pharmacotherapies
for anxiety disorders, particularly PTSD, where aversive
memory persistence and reinstatement are major contribu-
tors to maintenance of pathological fearful responding. En-
hancing consolidation of extinction learning and increasing
flexibility of memory with CBD could therefore be an
excellent strategy for improving efficacy of behavioural
therapy in these patients. Furthermore, reducing reinstate-
ment of contextually fearful responding following
extinction-based therapy is critical for long-term relapse
prevention in anxiety disorders. Thus, while CBD has dem-
onstrated potential for use in anxiety disorders due to its
anxiolytic properties and safety profile alone, its optimal use
in the treatment of anxiety disorders may be via combined
use for judicious management of anxiety during exposure
therapy (Rachman et al. 2008) and via enhancing consoli-
dation of extinction learning.
In the current study, the attenuation of fearful responding in
the CBD post-extinction group was unlikely to be due to
anxiolysis as there was no evidence of reduced anxiety fol-
lowing CBD. However, as eCBs are known to interact with
glucocorticoids, which vary with stress levels (Atsak et al.
2011;Akirav2011; Campolongo et al. 2012), eCB-
glucocorticoid interactions may thus be critical in the effects
of CBD on memory (de Bitencourt et al. 2013). As stress can
have bi-directional effects on memory consolidation (depend-
ing on severity), the mediating role of stress on CBD’smne-
monic effects should be assessed in future studies.
The CBD pre-extinction group performed worse on the
prose recall task at both time points than the placebo group.
This is indicative of pre-existing group differences in mem-
ory capacity. It is possible that the relative lack of effects of
pre-extinction CBD is related to pre-existing memory ca-
pacity in this group. However, no correlations were found
between prose recall performance and any measures during
conditioning, extinction, renewal or reinstatement.
Limitations
Due to the continuous learning protocol employed in the
current study (extinction followed acquisition by ~15 min),
790 Psychopharmacology (2013) 226:781–792
it was not possible to examine the effects of CBD on
consolidation of fear acquisition, without potential interfer-
ence from new learning during extinction. This may have
contributed to the lack of drug effects on consolidation of
conditioning. An optimal protocol for assessing these po-
tential effects would be a delayed extinction paradigm,
where acquisition memory is allowed to consolidate before
extinction learning. However, due to the recently discovered
effectofCBDonmemoryreconsolidation(Sternetal.
2012), insufficiently long extinction sessions may make
any effects within this protocol difficult to interpret.
Due to equipment failure, a significant proportion of
psychophysiological data was lost. This may have negative-
ly affected test power and may explain some of the observed
dissociation between explicit and autonomic memory meas-
ures. Nevertheless, the findings of extinction consolidation
enhancement by low-dose CBD on expectancy are compel-
ling and warrant further investigation of the effects of CBD
on multiple measures of conditioning and extinction in a
larger cohort. The current study, however, should not have
been taken as evidence against effects of CBD on autonomic
measure of conditioned responding.
An issue in all conditioning studies in humans is the
effect of propositional mechanisms concerning task
demands during novel tasks. There was evidence of these
‘occasion setting’effects in the current study on initial trial
of new tasks. In extinction, for example, there was an initial
increase in CS−ratings, reflecting expectancy of contingen-
cy reversal. However, following these initial orienting
responses, appropriate responding was achieved and main-
tained, suggesting that occasion setting effects did not affect
learning of contingencies. These effects are somewhat un-
avoidable in human studies, although the collection of mul-
tiple measures of conditioning allows appraisal of the role
they may play.
Conclusion
The current study found that a sub-anxiolytic dose of CBD
given post-extinction potentiated the consolidation of ex-
tinction learning as evidenced by a globally reduced UCS
expectancy at test. This was accompanied by an increased
sensitivity to learned contingency violation in this group
(i.e. increased memory flexibility), consistent with the
hypothesised action of CBD as an anandamide hydrolysis
and reuptake inhibitor. These preliminary findings suggest
CBD may be a potentially excellent adjunct to extinction-
based therapies for anxiety disorders and warrant further
investigation into its mnemonic effects.
Acknowledgments This study was supported by a grant awarded to
SKK, CJAM and HVC by the Medical Research Council (UK). The
authors report no conflict of interest, financial or otherwise, in this
research. All data were collected in compliance with UK law.
References
Akirav I (2011) The role of cannabinoids in modulating emotional and
non-emotional memory processes in the hippocampus. Front
Behav Neurosci 5:34
Atsak P, Roozendaal B, Campolongo P (2011) Role of the endocanna-
binoid system in regulating glucocorticoid effects on memory for
emotional experiences. Neuroscience 204:104–116
Baddeley A, Emslie H, Nimmo-Smith I (1993) The spot-the word test:
a robust estimate of verbal intelligence based on lexical decision.
Br J Clin Psychol 32:55–65
Bechara A, Tranel D, Damasio H, Adophs R, Rockland C, Damasio
AR (1995) Double dissociation of conditioning and declarative
knowledge relative to the amygdala and hippocampus in humans.
Science 269:1115–1118
Beck AT, Steer RA, Brown GK (1996) Manual for the Beck Depres-
sion Inventory-II. Psychological Corporation, San Antonio
Beltramo M, Stella N, Calignano A, Lin SY, Makriyannis A, Piomelli
D (1997) Functional role of high-affinity anandamide transport, as
revealed by selective inhibition. Science 277:1094–1097
Bergamaschi MM, Quieroz RHC, Crippa JAS, Zuardi AW (2011)
Safety and side effects of cannabidiol, a Cannabis sativa constituent.
Current Drug Safety 6(4):237–49
Bitencourt RM, Pamplona FA, Takahashi RN (2008) Facilitation of
contextual fear memory extinction and anti-anxiogenic effects of
AM404 and cannabidiol in conditioned rats. Eur Neuropsycho-
pharmacol 18:849–859
Boddez Y, Baeyens F, Luyten L, Vanteenwegen D, Hermans D, Beckers
T (2012) Rating data are underrated: validity of US expectancy in
human fear conditioning. J Behav Ther Exp Psychiatry 44:201–206
Bond A, Lader M (1974) The use of analogue scales in rating subjec-
tive feelings. Br J Med Psychol 47:211–218
Campolongo P, Roozendaal B, Trezza V, Hauer D, Schelling G,
McGaugh JL, Cuomo V (2009) Endocannabinoids in the rat
basolateral amygdala enhance memory consolidation and enable
glucocorticoid modulation of memory. PNAS 106:4888–4893
Campolongo P, Ratano P, Manduca A, Scattoni ML, Palmery M,
Trezza V, Cuomo V (2012) The endocannabinoid transport inhib-
itor AM404 differentially modulates recognition memory in rats
depending on environmental aversiveness. Front Behav Neurosci
6:11
Chhatwal JP, Davis M, Maguschak KA, Ressler KJ (2005) Role of
endogenous cannabinoids in cognition and emotionality. Enhancing
cannabinoid neurotransmission augments the extinction of condi-
tioned fear. Neuropsychopharmacology 30(3):516–524
de Bitencourt RM, Pamplona FA, Takahashi RN (2013) A current
overview of cannabinoids and glucocorticoids in facilitating ex-
tinction of aversive memories: potential extinction enhancers.
Neuropharmacology 64:389–395
DeHouwer J (2009) The propositional approach to associative learning
as an alternative for association formation models. Learn Behav
37:1–20
Dudai Y (2004) The neurobiology of consolidation, or, how stable is
the engram? Annu Rev Psychol 55:51–86
Fowler CJ, Tiger G, Ligresti A, López-Rodriguez ML, Di Marzo V
(2004) Selective inhibition of anandamide cellular uptake versus
enzymatic hydrolysis –a difficult issue to handle. European
Journal of Pharmacology 492:1–11
Freund TF, Katona I, Piomelli D (2003) Role of endogenous cannabi-
noids in synaptic signalling. Physiol Rev 83:1017–1066
Psychopharmacology (2013) 226:781–792 791
Grillon C (2009) D-Cycloserine facilitation of fear extinction and
exposure-based therapy might rely on lower-level, automatic
mechanisms. Biol Psychiatry 66:636–641
Hill MN, McEwen BS (2009) Endocannabinoids: the silent partner of
glucocorticoids in the synapse. PNAS 106:4579–4580
Hill MN, Froese LM, Morrish AC, Sun JC, Floresco SB (2006) Alterations
in behavioral flexibility by cannabinoid CB1 receptor agonists and
antagonists. Psychopharmacology 187:245–259
Kaplan GB, Moore KA (2011) The use of cognitive enhancers in
animal models of fear extinction. Pharmacol Biochem Behav 99
(2):217–228
Leweke FM, Piomelli D, Pahlisch F, Muhl D, Gerth CW, Hoyer C,
Klosterkötter J, Hellmich M, Koethe D (2012) Cannabidiol
enhances anandamide signaling and alleviates psychotic symp-
toms of schizophrenia. Transl Psychiatry 2:e94
Lovibond PF, Shanks DR (2002) The role of awareness in Pavlovian
conditioning: empirical evidence and theoretical implications. J
Exp Psychol 28:3–26
Lutz B (2007) The endocannabinoid system and extinction learning.
Mol Neurobiol 36:92–101
Marsicano G, Wotjak CT, Azad SC, Bisogno T, Rammes G, Cascio
MG (2002) The endogenous cannabinoid system controls extinc-
tion of aversive memories. Nature 418:530–534
Mineka S (1979) The role of fear in theories of avoidance learning,
flooding and extinction. Psychol Bull 86:985–1010
Mineka S, Zinbarg R (1996) Conditioning and ethological models of
anxiety disorders: stress-in-dynamic-context anxiety models. In: Hope
DA (ed) Nebraska symposium on motivation, vol 43. Perspectives
on anxiety, panic, and fear. University of Nebraska Press, Lincoln
Morgan CJA, Schafer G, Freeman TP, Curran HV (2010) Impact of
cannabidiol on the acute memory and psychotomimetic effects of
smoked cannabis: naturalistic study. Br J Psychiatry 197:285–290
Müller GE, Pilzecker A (1900) Experimentelle Beiträge zur Lehre vom
Gedächtnis. Z Psychol Ergänzungsband 1:1–300
Niyuhire F, Varvel SA, Thorpe AJ, Stokes RJ, Wiley JL, Lichtman AH
(2007) The disruptive effects of the CB(1) receptor antagonist
rimonabant on extinction learning in mice are task-specific. Psy-
chopharmacology 191:223–231
Nurmikko TJ, Serpell MG, Hoggart B, Toomey PJ, Morlion BJ, Haines D
(2007) Sativex successfully treats neuropathic pain characterised by
allodynia: a randomised, double-blind, placebo controlled clinical
trial. Pain 133:210–220
Ohman A, Minkea S (2001) Fears, phobias and preparedness: toward
an evolved module of fear and fear learning. Psychol Rev
108:483–522
Osan R, Tort ABL, Amaral OB (2011) A mismatch-based model for
memory reconsolidation and extinction in attractor networks.
PLoS One 6(8):e23113
Pamplona FA, Prediger RD, Pandolfo P, Takahashi RN (2006) The
cannabinoid receptor agonist WIN-55,212-2 facilitates the extinc-
tion of contextual fear memory and spatial memory in rats. Psy-
chopharmacology 188:641–649
Pamplona FA, Bitencourt RM, Takahashi RN (2008) Short- and long-
term effects of cannabinoids on the extinction of contextual fear
memory in rats. Neurobiol Learn Mem 90:290–293
Pavlov IP (1927) Conditioned reflexes. Oxford, London
Pertwee RG (2008) The diverse CB1 and CB2 receptor pharmacology of
three plant cannabinoids: Δ9-tetrahydrocannabinol, cannabidiol and
Δ9-tetrahydrocannabivarin. Br J Pharmacol 153(2):199–215
Petitet F, Jeantaud B, Reibaud M, Imperato A, Dubroeucq MC (1998)
Complex pharmacology of natural cannabinoids: evidence for
partial agonist activity of delta9-tetrahydrocannabinol and antago-
nist activity of cannabidiol on rat brain cannabinoid receptors. Life
Sci 63:PL1–PL6
Rachman S, Radomsky AS, Shafran R (2008) Safety behaviour: a
reconsideration. Behav Res Ther 46:163–173
Reiss S (1991) Expectancy model of fear, anxiety and panic. Clin
Psychol Rev 11(92):141–153
Sevenster D, Beckers T, Kindt M (2012) Retrieval per se is not
sufficient to trigger reconsolidation of human fear memory. Neu-
robiol Learn Mem 97:338–345
Spielberger CD, Gorssuch RL, Lushene PR, Vagg PR, Jacobs GA
(1983) Manual for the State-Trait Anxiety Inventory. Consulting
Psychologists Press, Inc., Palo Alto
Stern CAJ, Gazarini L, Takahashi RN, Guimaraes FS, Bertogliuo LJ
(2012) On disruption of fear memory by reconsolidation block-
ade: evidence from cannabidiol treatment. Neuropsychopharma-
cology 37:2132–2142
Suzuki A, Josselyn SA, Frankland PW, Massuhige S, Silva AJ, Kida S
(2004) Memory reconsolidation and extinction have distinct tem-
poral and biochemical signatures. J Neurosci 24:4787–4795
Thomas A, Baillie GL, Phillips AM, Razdan RK, Ross RA, Pertwee
RG (2007) Cannabidiol displays unexpectedly high potency as an
antagonist of CB1 and CB2 receptor agonists in vitro. Br J
Pharmacol 150(5):613–623
Varvel SA, Lichtman AH (2002) Evaluation of CB1 receptor knockout
mice in the Morris water maze. J Pharmacol Exp Ther 301:915–
924
Watson JB, Rayner R (1920) Conditioned emotional reactions. J Exp
Psychol 3:1–14
792 Psychopharmacology (2013) 226:781–792
Copyright of Psychopharmacology is the property of Springer Science & Business Media B.V. and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.
