Translational Psychiatry www.nature.

com/tp

ARTICLE OPEN

Acute neural effects of the mood stabiliser lamotrigine on

emotional processing in healthy volunteers: a randomised
control trial
✉
Marieke A. G. Martens 1,2,3,7 , Tarek Zghoul 1,2,7
, Evelyn Watson1,4,5, Sebastian W. Rieger 1,3
, Liliana P. Capitão6 and
1,2,3
Catherine J. Harmer

© The Author(s) 2024

Lamotrigine is an effective mood stabiliser, largely used for the management and prevention of depression in bipolar disorder. The
neuropsychological mechanisms by which lamotrigine acts to relieve symptoms as well as its neural effects on emotional
processing remain unclear. The primary objective of this current study was to investigate the impact of an acute dose of lamotrigine
on the neural response to a well-characterised fMRI task probing implicit emotional processing relevant to negative bias. 31 healthy
participants were administered either a single dose of lamotrigine (300 mg, n = 14) or placebo (n = 17) in a randomized, double-
blind design. Inside the 3 T MRI scanner, participants completed a covert emotional faces gender discrimination task. Brain
1234567890();,:

activations showing significant group differences were identified using voxel-wise general linear model (GLM) nonparametric
permutation testing, with threshold free cluster enhancement (TFCE) and a family wise error (FWE)-corrected cluster significance
threshold of p < 0.05. Participants receiving lamotrigine were more accurate at identifying the gender of fearful (but not happy or
angry) faces. A network of regions associated with emotional processing, including amygdala, insula, and the anterior cingulate
cortex (ACC), was significantly less activated in the lamotrigine group compared to the placebo group across emotional facial
expressions. A single dose of lamotrigine reduced activation in limbic areas in response to faces with both positive and negative
expressions, suggesting a valence-independent effect. However, at a behavioural level lamotrigine appeared to reduce the
distracting effect of fear on face discrimination. Such effects may be relevant to the mood stabilisation effects of lamotrigine.

Translational Psychiatry (2024)14:211 ; https://doi.org/10.1038/s41398-024-02944-6

INTRODUCTION similar to its antiepileptic properties [8], by inhibiting voltage-

Bipolar disorder (BD) is among the top causes of worldwide gated sodium channels and reducing downstream glutamate
disability, with significant associated morbidity and mortality [1, 2]. release. Whole-cell patch clamp recordings from granule cells in
It usually presents during adolescence or early adulthood and the dentate gyrus have shown that lamotrigine inhibits glutamate
affects 3% of the world’s population [3]. First line treatment for BD release in preclinical models [9]. Functional magnetic resonance
includes monotherapy or adjunctive treatment with mood imaging (fMRI) studies in humans have also shown that
stabilisers such as lithium and various anticonvulsants and lamotrigine pre-treatment attenuates the neural effects of the
antipsychotics [4]. However, treating BD is often complicated by glutamate release promoter ketamine [10, 11]. Further, studies
initial misdiagnosis as unipolar depression and subsequent using the forced swim test animal model of depression have
treatment with conventional antidepressants, which carry a high shown that pre-treatment with the voltage-gated sodium channel
risk of mood destabilisation in this group [5]. The mood stabiliser opener veratrine reverses lamotrigine’s antidepressant effects [12]
lamotrigine is however generally successful in BD therapy at The neuropsychological mechanisms that underpin lamotri-
treating the depressive phase without inducing mania [6]. Initially gine’s mood stabilising effects are however less clear. For example,
synthesized in the 1980s as an antiepileptic drug, lamotrigine is it is not known whether lamotrigine has effects on emotional
also approved by the US Food and Drug Administration(FDA) and information processing that are comparable to those of conven-
European Medicines Agency (EMA) for the long-term maintenance tional antidepressant drugs, like selective serotonin reuptake
treatment of BD [3, 7]. inhibitors (SSRIs). As for unipolar depression, there is evidence that
Despite its widespread use, lamotrigine’s mode of action as a patients with bipolar disorder have altered abnormal emotional
mood stabiliser has yet to be fully elucidated. One hypothesis is information processing which may elicit and maintain a depres-
that lamotrigine exerts its mood stabilising effects in a way that is sive episode [13–15]. According to the cognitive

1
Department of Psychiatry, University of Oxford, Oxford, UK. 2Oxford Health NHS Foundation Trust, Oxford, UK. 3Wellcome Centre for Integrative Neuroimaging, University of
Oxford, Oxford, UK. 4Institute of Sport Exercise and Health, Faculty of Medical Sciences, University College London, London, UK. 5Institute of Cognitive Neuroscience, Faculty of
Brain Sciences, University College London, London, UK. 6Psychology Research Centre (CIPsi), School of Psychology, University of Minho, Braga, Portugal. 7These authors
contributed equally: Marieke A. G. Martens, Tarek Zghoul. ✉email: Marieke.martens@psych.ox.ac.uk

Received: 20 October 2023 Revised: 14 May 2024 Accepted: 17 May 2024

 M.A.G. Martens et al.
2
neuropsychological model of antidepressant drug action, anti- investigators and assessors using sequential numbered containers
depressants target emotional processing rather than mood identical in packaging and appearance.
directly [16–18]. They do this via an ability to (sub-) acutely shift The study flow diagram is presented in Fig. 2. At the start of the research
the processing of emotional information away from a preference visit, female participants completed a pregnancy test. Participants were
for negative relative to positive input [16]. This induced bias in assessed for the following baseline measurements: State-Trait Anxiety
Inventory (STAI), Positive and Negative Affect Schedule (PANAS), Eysenck
favour of positive emotional material is thought to counterbalance Personality Questionnaire (EPQ), Beck Depression Inventory (BDI) and
negatively biased emotional information processing associated visual analogue scales for alertness, calmness, and satisfaction (VAS – [27])
with depression. On the neural level this is reflected in (all secondary outcomes). Following completion of baseline questionnaires,
antidepressant drug effects which have been reported to decrease lamotrigine or placebo was administered. Participants spent 2 h in a quiet
limbic and anterior cingulate cortex (ACC) activity in response to room after which they completed pre-scan questionnaires and mood
negative versus positive information, as well as increasing scales (STAI, PANAS, VAS, secondary outcomes) when the plasma
engagement of prefrontal cortex (PFC) areas such as the concentration of lamotrigine would be expected to be at its peak [28]. A
dorsolateral PFC (dlPFC) [19, 20]. These valence-dependent effects 60-minute MRI scan was then completed, including a structural scan, visual
checkerboard stimulation paradigm (secondary outcome), the covert
have been observed both in patients with depression as well as in
emotional faces gender discrimination task (primary outcome) and a
healthy volunteers, suggesting that they are an inherent effect of resting state scan (reported elsewhere). Following the scan, participants
antidepressant treatment and not simply due to subtle improve- completed the Emotional Task Battery (ETB) (reported in [29]). After the
ments in psychopathology [20]. Importantly, these changes in fMRI (post-scan), participants again completed the STAI, VAS, and PANAS
emotional processing are typically observed long before improve- as endpoint measurements of subjective mood and experience. Side
ments in mood symptoms normally become apparent in patients effects (secondary outcome) were quantified at each of three time points
with depression and these effects are associated with later clinical as well (baseline, pre-scan, post-scan) using a non-validated semi-
response, suggesting that they represent a critical pathway qualitative self-report rating scale (absent=0, mild=1, moderate=2,
through which antidepressants exert their effects [19, 21–25]. severe=3) for the following symptoms: nausea, dry mouth, agitation,
The current study investigated whether lamotrigine affects aggression, headache, drowsiness, dizziness, tremor, back/joint pain, vision
impairment, and rash.
neural response to emotional stimuli in healthy volunteers using a
well validated fMRI paradigm known to reliably elicit amygdala
activity. Healthy volunteers were selected in this study to Behavioural data analysis
investigate the direct effects of lamotrigine unconfounded by Statistical analysis was performed using IBM SPSS Statistics version 26 for
Windows. Questionnaires and scales measuring mood changes throughout
disorder-related factors such as symptom severity and previous
the experiment, side-effects, response accuracy, and response time were
antidepressant exposure, as well as potential symptom improve- analysed using mixed repeated-measures ANOVA. Treatment group (two
ment following lamotrigine administration which would also be levels: lamotrigine and placebo) was the between-subjects factor and
expected to affect emotional processing bias. We hypothesised either time point (three levels: baseline, pre-scan, post-scan) or subjective
that lamotrigine would attenuate amygdala response to negative experience (fear, happy, anger or nausea, dry mouth, agitation, aggression,
emotional stimuli and/ or increase the response to positive headache, drowsiness, dizziness, tremor, back/joint pain, vision, rash) was
emotional stimuli when compared to placebo, similar to the the within-subjects factor. Huynh-Feldt correction was applied where data
profile seen with conventional antidepressant drugs. failed Mauchly’s Test of Sphericity. Mean and SD are reported for
behavioural measures. A p value < 0.05 was used to denote statistical
significance. Adjustments for multiple testing were not made as potential
group differences on most of these behavioural and subjective measures
MATERIALS AND METHODS (like anxiety and side effects) would complicate interpreting the neural
Participants results. Findings are therefore presented uncorrected, with potential
Ethical approval was granted by the University of Oxford Central University confounds included in the analysis as regressors / covariates of no interest.
Research Ethics Committee (CUREC R49749/RE003) and the protocol was
pre-registered with clinicaltrials.gov (NCT04396938). Thirty-six healthy
adult volunteers (24 men, 12 women, mean age 24.11 ± 4.43 years, range fMRI
18 to 32 years) recruited from the Oxfordshire community took part in this fMRI data acquisition. Scanning was performed at the Oxford Centre for
study. The flow of participants is outlined in Fig. 1. Participants were Human Brain Activity (OHBA), University of Oxford, using a 3-Tesla Siemens
recruited through advertising and screened for eligibility. Details of Prisma scanner with a 32-channel head-coil. See supplementary informa-
inclusion and exclusion criteria (pre-established) are included in Supple- tion for the neuroimaging protocol.
mentary Material. In brief, recruited participants were fluent in English,
healthy, not pregnant or breastfeeding and not taking any psychoactive fMRI task designs. The faces task (also called gender discrimination task,
medication. All participants gave written informed consent. No changes to described in full and previously used in similar studies) is a covert task
methods occurred after start of the study. A formal sample size calculation designed to probe emotional processing and has proved sensitive to the
was precluded, because no prior study had determined the acute effect of acute effects of antidepressants on neural processing [30]. The task was a
lamotrigine on brain activity in healthy volunteers. Sample size was block design presenting colour photographs of faces expressing three
therefore modelled on previous successful studies using a similar design. emotions (fear, happy, anger) taken from the NimStim database [31]. For
Acute citalopram was found to reduce amygdala activation with an effect more details, please see supplementary information. Participants were
size of 1.19 (anatomically defined region of interest analysis) [26]. Informed asked to respond by indicating the gender (male or female) of each face as
by these data, an a priori sample size calculation for the current between- quickly and accurately as possible via button press. Reaction time (total
subjects design yielded n = 13 as the minimum sample size required to time between face stimuli presentation and gender classification response)
detect a reduction in amygdala fMRI signal of this magnitude (difference and accuracy (number of faces correctly identified as male/female divided
between two independent means: two tailed, alpha = 0.05, effect size = by total number of faces) were measured and used as a measure of task
1.19, power = 0.8). engagement. Following the faces task, a checkerboard visual paradigm
was presented. This assessed the effect of lamotrigine on the blood
oxygen level-dependent (BOLD) signal in the primary visual cortex, to
Procedures and measures control for a possible confounding effect of global drug-related modula-
The study had a between-subject, double-blind, placebo-controlled design. tion of BOLD signal. For more details, please see supplementary
Eligible participants were randomly allocated to one of two treatment information.
conditions: a single oral dose of lamotrigine (300 mg) or placebo (lactose
capsule). Randomisation was blocked in design (block size = 4), stratified
for sex and undertaken with an online tool (sealed envelope). A researcher fMRI data analysis. Data were analysed using FSL (FMRIB Software Library
not involved in the study was responsible for randomisation and v6.0) tools (https://fsl.fmrib.ox.ac.uk/fsl).
encapsulation. Group allocation was concealed from participants, fMRI data were pre-processed and analysed using FEAT (FMRI Expert
Analysis Tool). For further information on the pre-processing please refer to

Translational Psychiatry (2024)14:211

 M.A.G. Martens et al.
3

Fig. 1 CONSORT diagram to show flow of participants through the study.

the Supplementary Material. as test for group differences. Statistics were assessed using the threshold-
A custom three-column format convolved with a gamma hemodynamic free cluster enhancement method with family-wise error correction of 0.05
response function and its temporal derivative was used to model the data. (or 0.95 threshold within randomise) [37]. The general linear model (GLM)
For the covert faces task, the EVs included “fear”, “angry”, and “happy” included 2 groups: placebo and lamotrigine. Contrasts were defined as
faces. Contrasts analysed included means for each condition, mean across placebo greater than lamotrigine, lamotrigine greater than placebo, and
all emotions and directional comparisons. the mean across both groups. To account for the possibility that
The main contrast of interest for the checkerboard task was flashes vs. differences in self-reported anxiety and side effects between the two
baseline. For both task analyses motion parameters estimated by MCFLIRT groups could drive the neural difference, demeaned post-scan state
(Motion Correction by FMRIB’s Linear Image Registration Tool) were added anxiety scores (not pre-scan state anxiety scores due to missing values, no
to the model as nuisance regressors. Absolute and relative motion values significant difference between pre- and post-scan state anxiety scores) and
did not differ significantly between groups and no participant demon- side effect ratings (VAS alertness, calmness, drowsiness, and dizziness),
strated significant movement (all included participants revealed absolute were added as regressors of no interest.
and relative motion <1.5 mm). Significant brain areas were extracted for visualization using the fslmaths
Registration to high-resolution structural images was carried out using and cluster tools, with a threshold of 0.95 (based on the 1-p thresholding
FLIRT (FMRIB’s Linear Image Registration Tool) and further refined using from randomise, described above). To further visualise results, individual
FNIRT (FMRIB’s Non-Linear Image Registration Tool) nonlinear registration. parameter estimate (PE) values were extracted from their custom maps, using
Data were normalized to the Montreal Neurological Institute (MNI) significant clusters as binary masks. In addition, left and right amygdala
template [32–35]. masks were created based on the Harvard-Oxford Atlas (thresholded at 50),
Higher level (group level) analysis was carried out using FSL’s tool for and PE values were again extracted for visualisation. Activations are reported
nonparametric permutation inference Randomise (5000 permutations) [36] using MNI coordinates, and brain regions are reported based on the Harvard-
to assess general effects of task-relevant contrasts on both groups, as well Oxford Cortical and Subcortical Structural Atlas.

Translational Psychiatry (2024)14:211

 M.A.G. Martens et al.
4

Fig. 2 Study flow diagram.

RESULTS participants in the lamotrigine group reported higher scores on

Demographic and baseline clinical characteristics drowsiness (pre-scan: p = 0.031; post-scan: p = 0.033) and dizzi-
Demographic and baseline clinical characteristics are presented in ness (pre-scan: p = 0.003; post-scan: p = 0.013) as reflected in a
Table 1. Participants were recruited between the May 2017 and significant condition by time by side effect interaction
November 2018. The trial ended after the 36th participant was (F(20,580) = 1.97, p = 0.007, η2 = 0.064).
enrolled. The final study sample consisted of 31 participants
(lamotrigine n = 14, placebo n = 17) as 5 participants’ data were Effectiveness of blinding
not able to be analysed due to significant artefacts caused by All participants and the researcher conducting the study visit were
movement during the scan (n = 2, one from lamotrigine and asked to guess whether the participant had been administered
placebo group each), incidental findings identified during pre- lamotrigine or a placebo. The impression of group was not
processing (n = 1 from lamotrigine group), ghosting artefacts completed for one subject who was in the placebo group. Findings
(n = 1 from lamotrigine group), and a software error (n = 1 from reveal that both participants (X2 (1) = 8.62, p = 0.003) and the
lamotrigine group). researcher (X2 (1) = 11.32, p = 0.001) were able to correctly identify
The lamotrigine and placebo groups were well matched on the subject’s condition significantly more often than not (Table 1).
sociodemographic, clinical, and personality parameters. Given the
randomised nature of the study design, statistical analysis on the Behavioural task performance
homogeneity of treatment groups at baseline was not carried out Accuracy in correctly identifying gender during the fMRI faces task
[38]. was overall high ( > 95%, Table 3), confirming that participants
were engaged in the task. However, there was also a significant
Subjective ratings interaction between emotion and treatment (F(2,58) = 4.924,
There were no significant interaction effects between time and p = 0.011, η2 = 0.145). Post-hoc tests revealed that this effect
treatment on self-reported state anxiety and mood ratings as was driven by fearful faces, with the lamotrigine group being
measured by the STAI and PANAS (all F’s < 0.75, p’s > 0.1, more accurate than the placebo group (p = 0.045) at classifying
η2’s < 0.080) (Table 1). However, there was a significant difference gender for faces showing this emotion, but not anger (p = 0.441)
between groups on state anxiety across all time points nor happiness (p = 0.281). In addition, participants in the placebo
(F(1,28) = 6.54, p = 0.016, η2 = 0.189), where the lamotrigine group group performed worst for fearful faces overall, reflected in trend
had a significantly greater mean state anxiety score (M = 31.1, significant differences between fear vs happy (p = 0.081) and fear
SD = 7.5) than the placebo group (M = 26.0, SD = 5.7). vs anger (p = 0.052), but not happy vs anger p = 0.851). There was
There was a significant main effect of group on VAS ratings of no significant difference between the lamotrigine-treated and
calmness as well (F(1,28) = 6.24, p = 0.019, η2 = 0.182) with the placebo-treated groups in reaction time (F(1,29) = 0.06, p = 0.806,
lamotrigine group (mean, SD = 66.3, 27.9) reporting feeling η2 = 0.002), nor an interaction between emotion and treatment
significantly less calm (e.g., more excited and tense) across all (F(2,58) = 1.33, p = 0.272, η2 = 0.044).
timepoints compared to the placebo group (mean, SD = 41.8,
29.9), even prior to drug treatment (Table 2). fMRI results
There was also significant group by timepoint interaction for Main effect of task—faces task. To determine if the task engaged
VAS ratings of alertness (F(2,56) = 6.67, p = 0.003, η2 = 0.192) (see brain regions previously associated with emotional and facial
also Table 2). Post-hoc comparisons showed that the lamotrigine stimuli, neural activation in response to fear vs. baseline, happy vs.
group (mean = 420, SD = 156) reported feeling less alert (e.g., baseline, anger vs. baseline and mean faces vs. baseline was
more drowsy, clumsy, and lethargic) than the placebo group compared across groups. A whole-brain analysis revealed
(mean = 242, SD = 127) pre-scan (after drug administration) significant activation in response to emotional stimuli vs. baseline
(p = 0.002). There were no significant group differences at across groups in a network of areas (Supplementary Fig. 1).
baseline or post-scan (p’s > 0.093). The groups did not differ on Significant activity was observed in clusters that include the
VAS ratings of satisfaction (F’s < 2.27, p’s > 0.143, η2’s < 0.139) occipital fusiform gyrus, bilateral amygdala, angular gyrus and
(Table 2). ACC. These results are consistent with previous reports using the
There was a significant time by condition interaction for side same task, suggesting that the task was successful in probing
effects (F(2,58) = 5.95, p = 0.004, η2 = 0.170) with the lamotrigine emotional processing [26, 30, 39].
group presenting significantly more side effects than the placebo
group after treatment (both pre-scan (p = 0.004) and post-scan Effect of lamotrigine administration. A whole-brain analysis
(p = 0.007)) but not at baseline (p = 0.277) (Table 2). Specifically, revealed a range of areas with reduced BOLD activation in the

Translational Psychiatry (2024)14:211

 M.A.G. Martens et al.
5
superior parietal lobule towards angular gyrus, supramarginal
Table 1. Sociodemographic, clinical, and personality characteristics.
gyrus, right precentral gyrus, right middle frontal gyrus and right
Placebo Lamotrigine inferior frontal gyrus (Supplementary Fig. 2). No effects of
(n = 17) (n = 14) lamotrigine were found suggesting that the observed effects on
Gender emotional processing did not reflect global haemodynamic
changes.
Male n = 11 n=9
Female n=6 n=5 Sensitivity analysis. Adding demeaned subjective ratings of
Age 24.65 (4.3) 23.21 (4.4) alertness, calmness, dizziness, or drowsiness as a nuisance
Verbal IQ (NART) 118.59 (5.4) 118.36 (3.9) regressor together with state anxiety in the fMRI analysis resulted
in similar results. When correcting just for alertness or drowsiness
Beck Depression Inventory (BDI)
the main effect of group (mean all emotions > baseline) remains in
Screening visit 1.82 (1.9) 3.36 (3.3) amygdala and ACC (Supplementary Fig. 3).
Testing visit baseline 1.41 (1.9) 3.43 (2.6) When adding state anxiety as a covariate (as per imaging
PANAS analysis) and/or drowsiness, the accuracy and reaction time results
of the gender identification task remain similar, accuracy: F’s >
Positive (baseline) 34.29 (7.0) 32.14 (6.2)
4.21, p’s < 0.020; reaction time: F’s < 0.91, p’s > 0.410.
Negative (baseline) 11 (1.1) 11.36 (1.4)
Positive (pre-scan) 30.38 (8.7) 26.93 (7.7)
Negative (pre-scan) 10.75 (1.1) 11.07 (1.6) DISCUSSION
The main aim of the current study was to investigate the effects of a
Positive (post-scan) 33.81 (8.3) 28.71 (7.9)
single dose of lamotrigine on subjective mood and neural measures
Negative (post-scan) 10.5 (1.1) 12.14 (5.8) of emotional processing in healthy volunteers. Participants in the
Spielberger State and Trait Anxiety Inventories lamotrigine group were more accurate at identifying the gender of
Trait-Anxiety (screening 30.00 (5.0) 34.36 (4.6) fearful faces compared to the placebo group. In addition to this
visit) behavioural effect, we saw a reduction in BOLD activity in the
State-Anxiety (baseline 25.06 (5.9) 30.0 (5.9)* lamotrigine group relative to placebo in response to fearful, happy,
and angry faces (versus baseline) in a network of regions associated
State-Anxiety (pre- 26.94 (6.0) 31.29 (6.8)* with emotional processing, including the amygdala, insula and ACC.
scan)
Therefore, not in line with our hypothesis, lamotrigine had valence-
State-Anxiety (post- 25.29 (5.4) 32.00 (9.6)* independent rather than emotion-specific effects on neural
scan) measures of emotional processing. There were no effects of
Eyseneck Personality Questionnaires (EPQ) lamotrigine in response to visual stimulation, supporting the
Neuroticism 4.81 (3.0) 7.71 (4.9) conclusion that the effects on emotional processing were not
Psychoticism 2.63 (2.5) 3.00 (1.8)
driven by global drug-related modulation of the BOLD signal.
Undesired side effects related to lamotrigine included reduced
Extraversion 15.88 (3.8) 14.00 (4.9) alertness and increased drowsiness. However, neural results were
Lie/Social Desirability 8.81 (3.4) 9.50 (4.9) not affected by adding subjective ratings as nuisance regressors.
Perceived lamotrigine treatment (%) Together these findings indicate that lamotrigine has broad ranging
Participant impression 11.76 64.29
effects on neural response to emotional stimuli. This would suggest
an effect of lamotrigine that is not specific to negative emotional
Researcher impression 5.88 64.29 stimuli, at least at a neural level, which may be relevant for models
Values are means (SD). of mood stabilising action to consider.
Time points: baseline = prior to drug administration. The amygdala and ACC play an important role in emotional
Pre-scan = 2.5 h post drug/placebo administration. processing, and are postulated as a key site of traditional
Post-scan = 3.5 h post drug/placebo administration.
antidepressant action [19]. Specifically, acute doses of conven-
NART National Adult Reading Test, EPQ Eysenck Personality Questionnaire,
BDI Beck Depression Inventory, PANAS Positive and Negative Affective tional antidepressants have been shown to reproducibly and
Schedule. significantly reduce amygdala response to fearful faces and/or
*p < 0.05. increase activation to happy faces in this region [20, 26, 40]
perhaps reflecting early normalisation of negative affective bias in
depression. These changes in emotional processing have been
associated with later improvements in mood, suggesting that they
lamotrigine group relative to placebo, as a main effect of group in represent a critical pathway through which antidepressants exert
response to the mean of all faces versus baseline (78236 voxels, their effects [23, 25]. In our current study lamotrigine reduced
peak voxel location: x = 12, y = −12, z = 16 right thalamus, t-max BOLD activity in the amygdala and ACC in response to both
= 5.53, p = 0.001)). These brain areas include bilateral amygdala, positive and negative faces. The amygdala plays a key role in
hippocampus, ACC, insula, superior temporal gyrus, anterior PFC, detecting salient information in the environment [41, 42]. It is
frontal medial cortex, paracingulate gyrus, nucleus accumbens, therefore possible that this effect of reducing activity in the
posterior cingulate cortex (PCC), precuneous cortex and pre-and amygdala in response to both positive and negative valences
post-central gyrus (Fig. 3 and Supplementary Table 1). Similar could contribute to the mood-stabilising effect of lamotrigine,
results were found for each individual emotion (vs. baseline) given that patients with BD suffer from mood instability and their
(Supplementary Table 1). No group differences were seen for the interpretation of positive and negative events can oscillate greatly
contrasts comparing the different emotions with each other (i.e., a between mood episodes. This hypothesis remains speculative
group x emotion interaction). however and further research is needed to further explore this.
Moreover, group differences between lamotrigine and placebo
Visual checkerboard. In the checkerboard task, visual stimulation were also seen in other areas, including areas related to motor
was associated with a number of significant activation clusters, control and movement, attention, and cognition. In addition,
across groups. These included the bilateral occipital cortex, activity in areas related to emotional processing and attention

Translational Psychiatry (2024)14:211

 M.A.G. Martens et al.
6
Table 2. Reported side effects at baseline, pre-scan, and post-scan.
Baseline Pre-scan Post-scan

Placebo Mean ± Lamotrigine Placebo Mean Lamotrigine Placebo Mean Lamotrigine

SD Mean ± SD ± SD Mean ± SD ± SD Mean ± SD
VAS
Alertness 207.1 124.5 242.6 116.7 242.4 59.6 419.9* 31.3 248.8 27.3 343.7 33.0
Satisfaction 88.9 127.2 101.1 156.4 84.4 62.8 125.0 38.5 87.3 29.3 119.2 39.9
Calmness 42.5 147.8 65.1* 150.6 41.3 65.7 58.9* 79.5 41.5 33.0 75.0* 44.9
Side effects
Nausea 0 0 0 0 0.06 0.24 0.14 0.36 0 0 0.14 0.36
Dry mouth 0.12 0.33 0.07 0.27 0.12 0.33 0.29 0.61 0.12 0.33 0.21 0.43
Agitation 0 0 0.07 0.27 0 0 0 0 0 0 0.14 0.53
Aggression 0 0 0 0 0 0 0 0 0 0 0.07 0.27
Headache 0 0 0.14 0.36 0.12 0.33 0.29 0.61 0 0 0.29 0.47
Drowsiness 0.12 0.33 0.14 0.36 0.24 0.44 0.71* 0.73 0.24 0.56 0.79* 0.80
Dizziness 0 0 0.07 0.27 0 0 0.71* 0.91 0.06 0.24 0.57* 0.76
Tremor 0 0 0 0 0.06 0.24 0.14 0.36 0 0 0.07 0.27
Back/joint pain 0 0 0.07 0.27 0 0 0 0 0 0 0 0
Vision 0 0 0 0 0 0 0.07 0.27 0 0 0 0
Rash 0 0 0 0 0 0 0.14 0.53 0 0 0.07 0.27
*p < 0.05.

healthy volunteers are less accurate and slower at identifying the

Table 3. Accuracy and reaction times on the faces task. gender of fearful faces than that of happy faces in this task (e.g.,
Placebo Mean Lamotrigine Total Mean ± [39]), suggested to reflect a distraction effect from the threat
± SD Mean ± SD SD relevant content of the fearful faces interfering with the unrelated
Accuracy decision regarding facial gender. The improved accuracy in
classifying the fearful faces in those receiving lamotrigine could
Fear 95.29 (4.3) 97.86 (1.7)* 96.45 (3.6) represent reduced threat distraction by the fearful face content,
Happy 97.06 (3.9) 95.71 (2.7) 96.45 (3.4) even in the absence of differences in neural response to fear vs
Anger 97.21 (2.9) 96.43 (2.5) 96.85 (2.7) happy faces. Difficulties in reducing activation within the DMN has
Reaction times been linked to increased rumination and impaired control of
action [45, 52, 53]. Thus, the pattern of decreased DMN activation
Fear 497.13 (84.5) 498.78 (85.4) 497.87 (83.5) and reduced fear distraction following lamotrigine appears
Happy 496.03 (94.9) 483.27 (85.1) 490.27 (89.3) consistent and highlights a potential mechanism of action. Future
Anger 494.55 (89.7) 482.83 (76.8) 489.26 (82.9) resting state studies may clarify our findings further.
*p < 0.05. To our knowledge this is the first study that has investigated the
neural effects of lamotrigine in healthy volunteers. As part of the
same trial as this current study, we also investigated the effects of
lamotrigine on behavioural measures of emotional processing in
other than the amygdala and ACC, such as insula, anterior PFC, different cognitive domains, using the Emotional Test Battery
PCC, precuneous, supramarginal gyrus and paracingulate gyrus, (ETB) [29]. The ETB is a validated tool also proven to be sensitive to
was also significantly reduced in the lamotrigine group. These the early effects of antidepressant medication [17, 18]. There were
findings indicate that lamotrigine may reduce brain activity in no effects on emotional face recognition, in contrast to the effects
several areas related to emotional processing in a different way seen here in neural response to face stimuli. Such a difference may
than other studies using SSRI antidepressants have reported. As occur because of increased sensitivity of fMRI measures, which are
lamotrigine regulates glutaminergic release by inhibiting voltage- closer to mechanistic sites of action, compared to behavioural
gated channels associated with glutamate, it is perhaps not readouts. However, the behavioural results from the ETB also
surprising that widespread changes in activation patterns are suggested a decrease in negative vs positive memory recall
seen. Interestingly the areas identified overlap with regions that following lamotrigine [54]. The effects of lamotrigine on emotional
form the default mode network (DMN), the salience network (SAN) processing are therefore likely complex and may reflect dissoci-
and the affective network (AN). These resting state networks able effects on neural circuity underpinning these different
integrate cognitive control, affective and reward-systems of the perceptual vs memory functions. Alternatively, it is possible that
brain and have been implied in the pathophysiology of bipolar reducing the impact of emotional salient stimuli (shown here with
disorder [43–45]. The glutamatergic system also plays an reduced neural reactivity) could have the behavioural conse-
important role in regulating these networks [46–50]. For example, quence of releasing more memory capacity to recall the relatively
it has been shown that high glutamate concentration in the PCC more positive items i.e., reducing the priority often given to
and precuneous area is associated with reduced deactivation of negative stimuli.
the DMN [51]. The effect of lamotrigine on brain activity has been somewhat
Behaviourally lamotrigine improved the performance on a more widely explored in the context of emotional and cognitive
simple gender discrimination task for fearful faces only. Typically tasks in patients with bipolar disorder. In terms of resting state

Translational Psychiatry (2024)14:211

 M.A.G. Martens et al.
7

Fig. 3 fMRI Faces Task results. A Sagittal, coronal, and axial images depicting significantly reduced BOLD activation (deactivation) in the
lamotrigine group relative to placebo in response to mean of all faces versus baseline, in a whole range of areas including bilateral amygdala,
insula and ACC. Cursor in the left amygdala MNI coordinates: x = −18, y = −4, z = −16. Results are shown TFCE-corrected with a family-wise
error cluster significance of 1 – p > 0.95. B Parameter estimates extracted from the whole cluster, and C left and right amygdala anatomical
mask thesholded at 50. Error bars represent the standard error of the mean.

functional connectivity (rsFC) a recent study by [55] suggested clinical data, as well as significant differences in behavioural task
that preserved rsFC between the frontoparietal network (FPN) and performance between groups, and unsuccessful participant and
the dorsal attention network (DAN) (the networks involved in researcher blinding. The lamotrigine and placebo groups in our
cognitive control), and the hub of the posterior DMN (the study differed significantly in several important assessments,
precuneus), was critical for good response to lamotrigine as an including subjective state anxiety and side-effect profile. This may
add-on treatment in patients with bipolar depression. With have served as a confounding factor that interacted with
regards to task-based fMRI, some studies have suggested a link emotional processing and neural response. We accounted for this
between symptomatic improvement and normalization and by adding these subjective measures as nuisance regressors. In
therefore increased BOLD activity in a number of brain regions, addition, further unknown between-group differences could also
including the PFC and ACC. However, sample sizes were small and have adversely impacted the results and it is therefore not feasible
most of these studies included children and adolescents [56–58]. to attribute group differences to solely the effect of lamotrigine. It
Turning to other pharmacological agents that influence down- could be hypothesized that the lamotrigine group’s significantly
stream levels of glutamate, results have been mixed. For example, reduced neural response to emotional stimuli resulted from the
ebselen is a bioavailable antioxidant shown to lower glutamate participants not processing the stimuli as much. This could be due
levels in the ACC in healthy volunteers [59, 60]. Ebselen has been to the lamotrigine group being significantly more anxious than
found to differentially influence the recognition of positive vs. the placebo group, or the lamotrigine group experiencing
negative facial expressions in the Facial Expression Recognition significantly greater negative side effects, including greater
Task (FERT), a behavioural measure of emotional processing in one drowsiness and reduced alertness. It would be possible that
study [61], but not in another [62]. Experimental medicine studies lamotrigine group’s anxiety and reduced arousal caused them to
with neural outcome measures of emotional processing are avoid the emotional stimuli compared to the placebo group and
currently underway [60]. Ketamine on the other hand is a non- therefore explain the lamotrigine group’s reduced neural response
competitive N-methyl-D-aspartate (NMDA) receptor antagonist to presented emotional faces compared to baseline. However, the
causing increased presynaptic glutamate release and extracellular lamotrigine group was significantly better at identifying gender of
glutamate concentrations. Similar to the results from the current fearful faces which argues against this interpretation. A within-
study, ketamine has been shown to reduce neural reactivity in the subjects design would have been able to overcome some of the
amygdala after emotional stimulation with both positive and difficulties with group matching via randomisation, however a
negative pictures [63, 64]. However, ketamine and lamotrigine parallel-group design was utilised to avoid the possibility of
would be expected to have a different profile of effect on the practice effects, and habituation to the emotional stimuli used in
glutamate system suggesting that the relationship between this task. Finally, the relatively small sample size (n = 31) may have
emotional processing and glutamate is likely to be complex. impacted the power of the study to detect broader effects of
Several factors that may have influenced the current study must lamotrigine on emotional pressing, in addition to type 2 errors.
be taken into consideration when interpreting results. This The generalisability of the current results to clinical populations is
includes significant differences between groups in self-report limited. Using healthy participants for early mechanistic work in

Translational Psychiatry (2024)14:211

 M.A.G. Martens et al.
8
humans has a number of advantages; in particular, it allows the exhibiting depressed symptoms, but not when exhibiting manic symptoms.
characterisation of neurocognitive effects without confounding Cogn Neuropsychiatry. 2006;11:505–20.
due to symptom change. However, prospective conclusions about 15. Lemaire M, Aguillon-Hernandez N, Bonnet-Brilhault F, Martineau J, El-Hage W.
the potential effect of lamotrigine on emotional processing and Subjective and physiological emotional response in euthymic bipolar patients: a
pilot study. Psychiatry Res. 2014;220:294–301.
their relevance for the treatment of bipolar disorder remain to be
16. Harmer, Goodwin GM, Cowen PJ. Why do antidepressants take so long to work?
validated in future work that uses a well-matched, high-powered A cognitive neuropsychological model of antidepressant drug action. Br J Psy-
sample of patients across a broader age range. chiatry. 2009;195:102–8.
17. Harmer, O’Sullivan U, Favaron E, Massey-Chase R, Ayres R, Reinecke A, et al. Effect
of Acute Antidepressant Administration on Negative Affective Bias in Depressed
CONCLUSION Patients. AJP. 2009;166:1178–84.
A single dose of lamotrigine significantly reduced activation in a 18. Harmer CJ, Duman RS, Cowen PJ. How do antidepressants work? New perspec-
range of brain areas important for emotional processing, including tives for refining future treatment approaches. Lancet Psychiatry. 2017;4:409–18.
bilateral amygdala and ACC, in response to both positive and 19. Godlewska BR, Harmer CJ. Cognitive neuropsychological theory of antidepressant
action: a modern-day approach to depression and its treatment. Psycho-
negative emotions. Therefore, lamotrigine has an effect that
pharmacology. 2021;238:1265–78.
seems to be valence-independent rather than valence-specific. 20. Ma Y. Neuropsychological mechanism underlying antidepressant effect: a sys-
Together, while also considering the limitations of the current tematic meta-analysis. Mol Psychiatry. 2015;20:311–9.
study, this suggests lamotrigine has a broad-acting mechanism of 21. Browning M, Kingslake J, Dourish CT, Goodwin GM, Harmer CJ, Dawson GR.
action that is not specific to negative emotional stimuli. This data Predicting treatment response to antidepressant medication using early changes
is important as it suggests a mechanism of action underlying in emotional processing. Eur Neuropsychopharmacol. 2019;29:66–75.
lamotrigine that may be relevant to its mood stabilisation effects. 22. Browning M, Bilderbeck AC, Dias R, Dourish CT, Kingslake J, Deckert J, et al. The
Future research in a larger sample is needed to further clarify the clinical effectiveness of using a predictive algorithm to guide antidepressant
neural mechanisms underlying lamotrigine administration, their treatment in primary care (PReDicT): an open-label, randomised controlled trial.
Neuropsychopharmacol. 2021;46:1307–14.
relevance for the treatment of bipolar disorder and the translation
23. Godlewska BR, Browning M, Norbury R, Cowen PJ, Harmer CJ. Early changes in
to unipolar depression. emotional processing as a marker of clinical response to SSRI treatment in
depression. Transl Psychiatry. 2016;6:e957.
24. Shiroma PR, Thuras P, Johns B, Lim KO. Emotion recognition processing as early
DATA AVAILABILITY predictor of response to 8-week citalopram treatment in late-life depression. Int J
Anonymised subjective and behavioural data, as well as normalised MRI data (such Geriatr Psychiatry. 2014;29:1132–9.
that unique brain structure is not shared) that support the findings of this study, are 25. Tranter R, Bell D, Gutting P, Harmer C, Healy D, Anderson IM. The effect of
available from the corresponding author MM, upon reasonable request. The data are serotonergic and noradrenergic antidepressants on face emotion processing in
not publicly available due to restrictions on consent given by participants. depressed patients. J Affect Disord. 2009;118:87–93.
26. Murphy SE, Norbury R, O’Sullivan U, Cowen PJ, Harmer CJ. Effect of a single dose
of citalopram on amygdala response to emotional faces. Br J Psychiatry.
REFERENCES 2009;194:535–40.
1. Belmaker RH. Bipolar disorder. N. Engl J Med. 2004;351:476–86. 27. Bond A, Lader M. The use of analogue scales in rating subjective feelings. Br J
2. Kupfer DJ. The increasing medical burden in bipolar disorder. JAMA. Med Psychol. 1974;47:211–8.
2005;293:2528–30. 28. Goldsmith DR, Wagstaff AJ, Ibbotson T, Perry CM. Lamotrigine. Drugs.
3. Miranda ASde, Miranda ASde, Teixeira AL. Lamotrigine as a mood stabilizer: 2003;63:2029–50.
insights from the pre-clinical evidence. Expert Opin Drug Discov. 2019;14:179–90. 29. Zghoul T, Hortelano PA, Kaltenboeck A, Wright L, Goodwin G, Capitão L, et al. A
4. Connolly KR, Thase ME. The clinical management of bipolar disorder: a review of single dose of lamotrigine induces a positive memory bias in healthy volunteers.
evidence-based guidelines. Prim Care Companion CNS Disord. 2011;13:26102. 2023:2023.12.06.23299597.
5. Godlewska BR, Emir UE, Masaki C, Bargiotas T, Cowen PJ. Changes in brain Glx in 30. Capitão LP, Chapman R, Murphy SE, Harvey C-J, James A, Cowen PJ, et al. A single
depressed bipolar patients treated with lamotrigine: A proton MRS study. J Affect dose of fluoxetine reduces neural limbic responses to anger in depressed ado-
Disord. 2019;246:418–21. lescents. Transl Psychiatry. 2019;9:1–9.
6. Solmi M, Veronese M, Zaninotto N, Loos L, van der MLM, Gao K, et al. Lamotrigine 31. Tottenham N, Tanaka JW, Leon AC, McCarry T, Nurse M, Hare TA, et al. The
compared to placebo and other agents with antidepressant activity in patients NimStim set of facial expressions: Judgments from untrained research partici-
with unipolar and bipolar depression: a comprehensive meta-analysis of efficacy pants. Psychiatry Res. 2009;168:242–9.
and safety outcomes in short-term trials. CNS Spectr. 2016;21:403–18. 32. Andersson JL, Jenkinson M, Smith S. Non-linear registration, aka Spatial normal-
7. Weisler RH, Calabrese JR, Bowden CL, Ascher JA, DeVeaugh-Geiss J, Evoniuk G. isation FMRIB technical report TR07JA2. FMRIB Anal Group Univ Oxf. 2007;2:e21.
Discovery and development of lamotrigine for bipolar disorder: A story of ser- 33. Andersson JL, Jenkinson M, Smith S. Non-linear optimisation FMRIB technical
endipity, clinical observations, risk taking, and persistence. J Affect Disord. report TR07JA1. Practice. 2007.
2008;108:1–9. 34. Jenkinson M, Bannister PR, Brady M, Smith SAC. Improved optimization for the
8. Rogawski MA, Löscher W. The neurobiology of antiepileptic drugs. Nat Rev robust and accurate linear registration and motion correction of brain images.
Neurosci. 2004;5:553–64. NeuroImage. 2002;17:825–41.
9. Lee C-Y, Fu W-M, Chen C-C, Su M-J, Liou H-H. Lamotrigine inhibits postsynaptic 35. Jenkinson M, Smith S. A global optimisation method for robust affine registration
AMPA receptor and glutamate release in the dentate gyrus. Epilepsia. of brain images. Med Image Anal. 2001;5:143–56.
2008;49:888–97. 36. Winkler AM, Ridgway GR, Webster MA, Smith SM, Nichols TE. Permutation
10. Deakin JFW, Lees J, McKie S, Hallak JEC, Williams SR, Dursun SM. Glutamate and inference for the general linear model. Neuroimage. 2014;92:381–97.
the neural basis of the subjective effects of ketamine: a pharmaco–magnetic 37. Smith SM, Nichols TE. Threshold-free cluster enhancement: addressing problems
resonance imaging study. Arch Gen Psychiatry. 2008;65:154–64. of smoothing, threshold dependence and localisation in cluster inference. Neu-
11. Doyle OM, Simoni SD, Schwarz AJ, Brittain C, O’Daly OG, Williams SCR, et al. roimage. 2009;44:83–98.
Quantifying the attenuation of the ketamine pharmacological magnetic reso- 38. Senn S. Testing for baseline balance in clinical trials. Stat Med. 1994;13:1715–26.
nance imaging response in humans: a validation using antipsychotic and glu- 39. Martens MAG, Kaltenboeck A, Halahakoon DC, Browning M, Cowen PJ, Harmer CJ.
tamatergic agents. J Pharm Exp Ther. 2013;345:151–60. An experimental medicine investigation of the effects of subacute pramipexole
12. Prica C, Hascoet M, Bourin M. Antidepressant-like effect of lamotrigine is reversed treatment on emotional information processing in healthy volunteers. Pharma-
by veratrine: A possible role of sodium channels in bipolar depression. Beha- ceuticals. 2021;14:800.
vioural Brain Res. 2008;191:49–54. 40. Rawlings NB, Norbury R, Cowen PJ, Harmer CJ. A single dose of mirtazapine
13. Bilderbeck AC, Reed ZE, McMahon HC, Atkinson LZ, Price J, Geddes JR, et al. modulates neural responses to emotional faces in healthy people. Psycho-
Associations between mood instability and emotional processing in a large pharmacology. 2010;212:625–34.
cohort of bipolar patients. Psychological Med. 2016;46:3151–60. 41. Zheng J, Anderson KL, Leal SL, Shestyuk A, Gulsen G, Mnatsakanyan L, et al.
14. Gray J, Venn H, Montagne B, Murray L, Burt M, Frigerio E, et al. Bipolar patients Amygdala-hippocampal dynamics during salient information processing. Nat
show mood‐congruent biases in sensitivity to facial expressions of emotion when Commun. 2017;8:14413.

Translational Psychiatry (2024)14:211

 M.A.G. Martens et al.
9
42. Liberzon I, Phan KL, Decker LR, Taylor SF. Extended amygdala and emotional performance neuroimaging analyses would not have been possible. Last but least our
salience: a PET activation study of positive and negative affect. Neuropsycho- gratitude goes to all other support staff in our Centre whose professional expertise
pharmacology. 2003;28:726–33. and skillset supports our research on a daily basis. This research was funded in whole,
43. Perry A, Roberts G, Mitchell PB, Breakspear M. Connectomics of bipolar disorder: a or in part, by the Wellcome Trust. For the purpose of open access, the author has
critical review, and evidence for dynamic instabilities within interoceptive net- applied a CC BY public copyright licence to any Author Accepted Manuscript version
works. Mol Psychiatry. 2019;24:1296–318. arising from this submission.
44. Wang Y, Gao Y, Tang S, Lu L, Zhang L, Bu X, et al. Large-scale network dysfunction
in the acute state compared to the remitted state of bipolar disorder: A meta-
analysis of resting-state functional connectivity. EBioMedicine. 2020;54:102742. AUTHOR CONTRIBUTIONS
45. Zovetti N, Rossetti MG, Perlini C, Maggioni E, Bontempi P, Bellani M, et al. Default Substantial contributions to the conception or design of the work; or the acquisition,
mode network activity in bipolar disorder. Epidemiol Psychiatr Sci. 2020;29:e166. analysis, or interpretation of data for the work: MM, TZ, EW, SR, LC, CH. Drafting the work
46. Duncan NW, Wiebking C, Tiret B, Marjańska M, Hayes DJ, Lyttleton O, et al. or revising it critically for important intellectual content: MM, TZ, EW, SR, LC, CH. Final
Glutamate concentration in the medial prefrontal cortex predicts resting-state approval of the version to be published: CH. Agreement to be accountable for all
cortical-subcortical functional connectivity in humans. PLoS ONE. 2013;8:e60312. aspects of the work in ensuring that questions related to the accuracy or integrity of any
47. Duncan NW, Wiebking C, Northoff G. Associations of regional GABA and gluta- part of the work are appropriately investigated and resolved: MM, TZ, EW, SR, LC, CH.
mate with intrinsic and extrinsic neural activity in humans—A review of multi-
modal imaging studies. Neurosci Biobehav Rev. 2014;47:36–52.
48. Falkenberg LE, Westerhausen R, Specht K, Hugdahl K. Resting-state glutamate
level in the anterior cingulate predicts blood-oxygen level-dependent response FUNDING
to cognitive control. Proc Natl Acad Sci USA. 2012;109:5069–73. This research study was funded by the NIHR Oxford Health Biomedical Research Centre
49. Levar N, Van Doesum TJ, Denys D, Van Wingen GA. Anterior cingulate GABA and (OH BRC). The views expressed here are those of the authors and not necessarily those
glutamate concentrations are associated with resting-state network connectivity. of the NHS, the NIHR or the Department of Health. The Wellcome Centre for Integrative
Sci Rep. 2019;9:2116. Neuroimaging is supported by core funding from the Wellcome Trust (203139/Z/16/Z
50. McCutcheon RA, Pillinger T, Rogdaki M, Bustillo J, Howes OD. Glutamate con- and 203139/A/16/Z). Liliana Capitão currently works at the Psychology Research Centre
nectivity associations converge upon the salience network in schizophrenia and (PSI/01662), School of Psychology, University of Minho, supported by the Foundation for
healthy controls. Transl Psychiatry. 2021;11:1–10. Science and Technology (FCT) through the Portuguese State Budget (Ref.: UID/PSI/
51. Hu Y, Chen X, Gu H, Yang Y. Resting-state glutamate and GABA concentrations 01662/2020). She is individually funded by a Research Fellowship also from the
predict task-induced deactivation in the default mode network. J Neurosci. Foundation for Science and Technology (FCT) (Ref: 2021.00415.CEECIND).
2013;33:18566–73.
52. Doucet GE, Janiri D, Howard R, O’Brien M, Andrews-Hanna JR, Frangou S.
Transdiagnostic and disease-specific abnormalities in the default-mode network COMPETING INTERESTS
hubs in psychiatric disorders: a meta-analysis of resting-state functional imaging CH has received consultancy fees from P1vital Ltd., Janssen Pharmaceuticals, Sage
studies. Eur Psychiatry. 2020;63:e57. Therapeutics, Pfizer, Zogenix, Compass Pathways, and Lundbeck. CH holds grant
53. Uddin LQ, Clare Kelly AM, Biswal BB, Xavier Castellanos F, Milham MP. Functional income from Zogenix, UCB Pharma, Janssen Pharmaceuticals. The remaining authors
connectivity of default mode network components: correlation, anticorrelation, declare no competing interests.
and causality. Hum Brain Mapp. 2009;30:625–37.
54. Murrough JW, Abdallah CG, Mathew SJ. Targeting glutamate signalling in
depression: progress and prospects. Nat Rev Drug Discov. 2017;16:472–86. ETHICAL STANDARDS
55. Martens M, Filippini N, Masaki C, Godlewska BR. Functional connectivity between The authors assert that all procedures contributing to this work comply with the
task-positive networks and the left precuneus as a biomarker of response to ethical standards of the relevant national and institutional committees on human
lamotrigine in bipolar depression: a pilot study. Pharmaceuticals. 2021;14:534. experimentation and with the Helsinki Declaration of 1975, as revised in 2008.
56. Haldane M, Jogia J, Cobb A, Kozuch E, Kumari V, Frangou S. Changes in brain
activation during working memory and facial recognition tasks in patients with
bipolar disorder with Lamotrigine monotherapy. Eur Neuropsychopharmacol.
ADDITIONAL INFORMATION
2008;18:48–54.
57. Jogia J, Haldane M, Cobb A, Kumari V, Frangou S. Pilot investigation of the Supplementary information The online version contains supplementary material
changes in cortical activation during facial affect recognition with lamotrigine available at https://doi.org/10.1038/s41398-024-02944-6.
monotherapy in bipolar disorder. Br J Psychiatry. 2008;192:197–201.
58. Passarotti AM, Sweeney JA, Pavuluri MN. Fronto-limbic dysfunction in mania pre- Correspondence and requests for materials should be addressed to
treatment and persistent amygdala over-activity post-treatment in pediatric Marieke A. G. Martens.
bipolar disorder. Psychopharmacology. 2011;216:485–99.
59. Masaki C, Sharpley AL, Godlewska BR, Berrington A, Hashimoto T, Singh N, et al. Reprints and permission information is available at http://www.nature.com/
Effects of the potential lithium-mimetic, ebselen, on brain neurochemistry: a magnetic reprints
resonance spectroscopy study at 7 tesla. Psychopharmacology. 2016;233:1097–104.
60. Ramli FF, Cowen PJ, Godlewska BR. The potential use of ebselen in treatment- Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims
resistant depression. Pharmaceuticals. 2022;15:485. in published maps and institutional affiliations.
61. Masaki C, Sharpley AL, Cooper CM, Godlewska BR, Singh N, Vasudevan SR, et al.
Effects of the potential lithium-mimetic, ebselen, on impulsivity and emotional
processing. Psychopharmacology. 2016;233:2655–61.
62. Singh N, Sharpley AL, Emir UE, Masaki C, Herzallah MM, Gluck MA, et al. Effect of
the putative lithium mimetic ebselen on brain myo-inositol, sleep, and emotional Open Access This article is licensed under a Creative Commons
processing in humans. Neuropsychopharmacology 2016;41:1768–78. Attribution 4.0 International License, which permits use, sharing,
63. Loureiro JRA, Leaver A, Vasavada M, Sahib AK, Kubicki A, Joshi S, et al. Modulation adaptation, distribution and reproduction in any medium or format, as long as you give
of amygdala reactivity following rapidly acting interventions for major depres- appropriate credit to the original author(s) and the source, provide a link to the Creative
sion. Hum Brain Mapp. 2020;41:1699–710. Commons licence, and indicate if changes were made. The images or other third party
64. Scheidegger M, Henning A, Walter M, Lehmann M, Kraehenmann R, Boeker H, material in this article are included in the article’s Creative Commons licence, unless
et al. Ketamine administration reduces amygdalo-hippocampal reactivity to indicated otherwise in a credit line to the material. If material is not included in the
emotional stimulation. Hum Brain Mapp. 2016;37:1941–52. article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly
from the copyright holder. To view a copy of this licence, visit http://
creativecommons.org/licenses/by/4.0/.
ACKNOWLEDGEMENTS
First of all, we would like to thank all volunteers who participated in this study. We
would furthermore like to thank the radiographers who kindly helped acquiring the © The Author(s) 2024
data. Great thanks also goes to our computing team as without them our high-

Translational Psychiatry (2024)14:211