The Importance of Common Currency Tasks in Translational Psychiatry

Abstract

Purpose of Review
Common currency tasks are tasks that investigate the same phenomenon in different species. In this review, we discuss how to ensure the translational validity of common currency tasks, summarise their benefits, present recent research in this area and offer future directions and recommendations.

Recent Findings
We discuss the strengths and limitations of three specific examples where common currency tasks have added to our understanding of psychiatric constructs—affective bias, reversal learning and goal-based decision making.

Summary
Overall, common currency tasks offer the potential to improve drug discovery in psychiatry. We recommend that researchers prioritise construct validity above face validity when designing common currency tasks and suggest that the evidence for construct validity is summarised in papers presenting research in this area.

Full text

MOOD AND ANXIETY DISORDERS (E PULCU AND C HARMER, SECTION EDITORS)
The Importance of Common Currency Tasks in Translational
Psychiatry
Alexandra C. Pike
1
&Millie Lowther
1
&Oliver J. Robinson
1,2
Accepted: 6 January 2021
#The Author(s) 2021
Abstract
Purpose of Review Common currency tasks are tasks that investigate the same phenomenon in different species. In this review,
we discuss how to ensure the translational validity of common currency tasks, summarise their benefits, present recent research in
this area and offer future directions and recommendations.
Recent Findings We discuss the strengths and limitations of three specificexamples where common currency tasks have added to
our understanding of psychiatric constructs—affective bias, reversal learning and goal-based decision making.
Summary Overall, common currency tasks offer the potential to improve drug discovery in psychiatry. We recommend that
researchers prioritise construct validity above face validity when designing common currency tasks and suggest that the evidence
for construct validity is summarised in papers presenting research in this area.
Keywords Common currencytasks .Translational tasks .Translational psy chiatry .Validity .Animalmodels .Behavioural assay
Introduction
The rate of drug discovery in psychiatry has not met expecta-
tions for a number of years [1,2], with particular failings in
translating promising pre-clinical findings into humans. This
state of affairs can be attributed, in part at least, to discrepant
findings from human and animal research [3]. Specifically, the
relevance of pre-clinical work to human disease or symptoms
is constrained by the similarities between measures used in
humans and other species [3]. One way to improve this situ-
ation may be to use more ‘common currency’tasks, which
investigate the same phenomenon in different species. In this
review, we will define common currency tasks and discuss
their benefits for translational psychiatric research. Then, we
will summarise some recent research using common currency
tasks, and finally present some promising future avenues and
our recommendations for this area.
Defining Common Currency Tasks
A common currency task is one which has been designed to
measure the same construct across species: key aspects are
maintained when the task is performed by both humans and
animals, although some features of the task may be altered to
account for differences between species (for example, the
range of auditory frequencies that can be perceived differs
substantially between humans and rodents). There may also
be more marked differences between species: a human com-
pleting the CANTAB spatial working memory task must ex-
plore and remember on-screen visuospatial information,
whilst a rat completing a radial arm maze is required to phys-
ically move through space [4]. Some of these differences are
driven by the direction of translation of the specific task.
Common currency tasks may have originated in humans and
been simplified for their translation into animal work (reverse
or back-translation, e.g. the intra-extra dimensional set shift
task from the CANTAB battery [5,6]). Alternatively, they
may have originated in animals, and the context may have
been changed to allow translation into humans (forward
This article is part of the Topical Collection on Mood and Anxiety
Disorders
*Alexandra C. Pike
alex.pike@ucl.ac.uk
1
Anxiety Lab, Neuroscience and Mental Health Group, University
College London Institute of Cognitive Neuroscience, Alexandra
House, 17-19 Queen Square, Bloomsbury, London WC1N 3AR, UK
2
Research Department of Clinical, Educational and Health
Psychology Department, University College London, Gower Street,
London WC1E 6BT, UK
https://doi.org/10.1007/s40473-021-00225-w
/ Published online: 12 February 2021
Current Behavioral Neuroscience Reports (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

translation [7]). Some tasks have been translated multiple
times between species (e.g. the ‘ambiguous-cue interpreta-
tion’task [8–10]), leading to multiple species-specific
alterations.
Regardless of apparent similarities or differences be-
tween tasks, it is a non-trivial problem to ensure that all
versions of a task are measuring the intended construct.
Existing recommendations for the development of com-
mon currency tasks include minimal verbal instructions,
non-verbal stimuli, similarity of task parameters (such as
number of trials and stimulus timing) and required re-
sponses, as well as consistent statistical analyses follow-
ing data collection [11], all of which influence the
translationalvalidityofcommoncurrencytasks.
Translational Validity
The extent to which tasks are truly ‘common’is known
as ‘translational validity’[12,13]. Translational validity
is composed of multiple types of validity, including face
validity, predictive validity and construct validity
(Fig. 1). Face validity is the degree of phenomenologi-
cal similarity between test contents and a construct (e.g.
whether the task appears, at face value, to be assessing
workingmemory[14]). Predictive validity is the ability
of a measure derived from a task to predict a subse-
quent score or outcome, such as response to treatment
(e.g. whether the number of items recalled in a
working-memory task increases after administration of
a pro-cognitive drug [14]). Construct validity is the ex-
tent to which a task actually probes the intended under-
lyingvariable(e.g.workingmemory[15]).
Face Validity
Face validity is arguably the easiest component of translational
validity to assess. If two tasks appear similar, then it increases
confidence that they are measuring the same thing and that suc-
cess on the task is not driven by different strategies between
species. However, high face validity does not guarantee construct
validity—for example, humans and rodents may use spatial strat-
egies to different extents when completing the Morris water
maze, despite the use of virtual reality in humans to promote face
validity [16]. Face validity may therefore be a red herring—
researchers may focus too closely on making sure the task ap-
pears similar in both species, and less time focusing on whether it
requires the same strategies and processes and is implemented
using similar neural circuitry [12].
In particular, it is common for tasks in humans and animals
to use different reinforcers—primary reinforcers such as food
or water are often used in animal work, and secondary rein-
forcers such as money or points are often used in human work.
Whilst this reduces face validity, it can be argued to increase
construct validity. On the one hand, using money or points
across species would likely evoke little reward-seeking behav-
iour in animals, as they do not have the same learnt associa-
tions between money and primary reinforcers as humans do.
On the other hand, using food is unlikely to elicit strong re-
sponses from humans who generally have sufficient access to
food, and food-restricting humans to ensure that they are sen-
sitive to primary reinforcers (which is routine in animal exper-
iments) would be difficult due to ethical considerations.
However, it is worth noting that the brain circuitry involved
in responding to primary and secondary reinforcers is not
identical [17,18], and different individuals within a species
may have different associations with secondary reinforcers
Fig. 1 Graphic displaying the
components of translational
validity, which may be defined as
the extent to which tasks designed
to capture the same phenomenon
in different species achieve this
goal, along with their descriptions
2 Curr Behav Neurosci Rep (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

[11]. Ultimately, the ability to increase face validity by direct-
ly matching reinforcers across humans—who voluntarily con-
sent to participate for research—and animals—whose entire
existence falls within the research context—is always going to
be a challenge.
Predictive Validity
Predictive validity is also important, because if performance
on a common currency task is not sensitive to the same inter-
ventions across species, it cannot be used for drug discovery.
However, a focus on predictive validity has frequently result-
ed in the development of tasks which are only sensitive to
‘me-too’pharmacological compounds (a term which refers
to compounds that are chemically similar to an original, pro-
totypical compound and therefore have the same mechanism
of action as drugs already known to change behaviour in that
task) [19–21]. Even tasks with high face validity, such as
approach-avoidance conflict tasks used to investigate anxiety,
may show predictive validity for one class of drug more con-
sistently than another. For example, punishment-induced con-
flict tasks are sensitive to benzodiazepine administration, but
SSRI administration does not consistently change responding,
even though both classes of drug are effective in treating anx-
iety symptoms in humans [22,23]. Ultimately, predictive va-
lidity depends on what is being predicted, which may be dif-
ficult to standardise because many of the things that matter to
human patients (e.g. reduced ‘feelings’of anxiety or more
enjoyment of everyday activities) are not measurable in
animals.
Construct Validity
Finally, construct validity may be the most important contrib-
utor to translational validity, as tasks that measure the same
construct should rely on common, evolutionarily preserved,
cognitive, neural and biological mechanisms, thus ensuring
that pharmacological agents should have similar effects on
these tasks even when performed by different species [12].
However, it is difficult to prove that a given task demonstrates
construct validity, as we have no access to the ground truth of
which psychological phenomenon causes patterns of
responding on any given task. Furthermore, by default, many
tasks recruit several different psychological capabilities (e.g.
working memory and reward learning in many reversal learn-
ing paradigms [24]), resulting in difficulty disentangling the
respective contributions of each construct to performance.
The Benefits of Common Currency Tasks
Despite the challenges, the key promise of common currency
tasks is that by allowing the same endpoint to be measured in
both preclinical and clinical drug development trials, the drug
discovery pipeline in psychiatry will become more efficient
[3,19,25,26]. The failure of drugs in clinical trials for anxiety
and depression can be partially attributed to lack of common
endpoints. For instance, the promising pre-clinical trials of
neurokinin-1 used forced-swim, tail suspension and stress par-
adigms [27,28], whereas the human clinical trials, which
failed, used symptom questionnaires [27].
Additionally, it is useful to have directly corresponding
measures, rather than just two separate tasks for separate spe-
cies that are purported to measure the same underlying con-
struct. The primary benefit of this is that the data from differ-
ent species can then be easily and directly compared, allowing
any discrepancies on how species are performing the task to
be detected (e.g. different patterns of accuracy in different
conditions) and resolved. This direct comparison of results
also allows changes in performance due to interventions or
manipulations to be compared across species. For example,
using a common currency task, Ironside et al. were able to
show corresponding side-by-side plots of the probability of
approach vs. avoidance responses for humans and non-
human primates, allowing direct visual comparison of the be-
haviour of two different species on this task [29].
Another key advantage of valid common currency tasks is
that they can allow us to obtain causal evidence for mechanisms,
which is not possible using cross-sectional correlational designs:
it is possible to directly manipulate genes, brain areas and protein
expression in animals. For example, if we suspect that a particu-
lar gene is involved in fear extinction, we can only measure a
correlation between genotype and behaviour on a task in
humans. However, we can directly manipulate the expression
and presence of that gene in animals, and then assess how this
affects task performance, allowing stronger inference. Similarly,
if we suspect that particular neural circuitry is involved in a
behaviour (and, say, observe consistent neuroimaging patterns
across species), we can use techniques such as inactivation and
optogenetics to directly assess the effects of this circuitry on a
task in animals. Crucially, the equivalent human work can gen-
erally only indirectly assess neural circuitry involvement via the
blood-oxygen-level-dependent response in functional imaging,
except in the rare cases where patients have lesions or implanted
electrodes. Using common currency tasks in these experiments
therefore allows us to draw a more direct causal conclusion about
the mechanisms behind these symptoms, resulting in a better
understanding of the pathophysiology. Integration of causal clin-
ical, genetic and environmental information can ultimately build
a more complete picture of the underlying mechanistic changes,
and hence inform the most appropriate treatment strategies.
Furthermore, translational work in psychiatry is often stymied
by the fact that many of the features we study—anhedonia, wor-
ry, intrusive thoughts—do not lend themselves to easy translation
into animal work. Creating a veridical animal model of depres-
sion or psychosis is much more difficult than creating an animal
3Curr Behav Neurosci Rep (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

model of, for example, cancer or diabetes, resulting in poor con-
struct and face validity [20,30]. Moreover, psychiatric disorders
are highly heterogeneous and comorbid, with myriad possible
presentations and symptoms, and as such are unlikely to be cap-
tured by a single animal model [31]. However, common curren-
cy tasks offer a partial alternative: we can study the processes
that may underlie some of these complex symptoms, such as
reward learning [10,32], memory biases [33]orhabitformation
[34], instead, and use cross-species evidence to investigate our
hypotheses. This type of process-based, or individual symptom-
based approach, encompasses the search for ‘biomarkers’,
‘endophenotypes’[35]or‘research domains’[36], which form
key parts of the ‘experimental medicine’approach to drug dis-
covery [3].
Such process-based translational work might also allow
drug discovery to be focused on symptoms that are not com-
monly targeted by treatments, but which are nevertheless of
importance to patients. Many pharmacological agents target
the ‘primary’symptoms reported by patients—low mood in
depression, delusions and hallucinations in schizophrenia—
whilst other symptoms like concentration or memory receive
less attention. For example, in both depression and schizo-
phrenia, residual cognitive deficits are often still present after
successful treatment of the primary symptoms with pharma-
cological agents [37–41]. Indeed, prior to a push to develop
translationally valid tasks for cognition in schizophrenia, there
was previously no mechanism for the FDA in the USA to
approve a treatment for cognitive deficits in schizophrenia if
said treatment did not also treat psychosis [26,42,43]. The
development of further common currency tasks might enable
a similar focus on overlooked symptoms in mood and anxiety
disorders.
Finally, an indirect advantage of common currency tasks is
that the need to ensure accessibility across species can enforce
simplicity. As a result, these tasks tend to be more focused on a
single underlying construct, which may facilitate a more pre-
cise measurement of specific processes without confounds. For
example, a human neuropsychological task, the Wisconsin
Card Sorting Test, is designed to measure set-shifting, but also
implicitly requires the ability to perform successful visual
matching. In this task, participants must sort cards into piles
which share features such as shape, colour or number of items
on the card, by learning over time which of these features
should be matched for positive feedback. The target feature
may change throughout the task, and successful performance
following a change requires a ‘set-shift’: the participant must
shift their attention and choices to another feature of the cards.
Even once the target feature has been learnt, successful perfor-
mance on this task requires visual matching—a participant
must be able to identify the features present on the card in front
of them and match these features to the features present on the
four ‘piles’at the top of the screen [5]. Thus, failure on the task
could be due to impaired set-shifting (as is often inferred) or to
impaired matching to sample. However, one common curren-
cy equivalent of this task—the intra-extra dimensional set shift
task—was necessarily made simpler for use across species, but
as a result, it is a more precise measure of set-shifting ability,
which is less confounded by visual search ability as no visual
matching is required [5].
Three Examples of Common Currency Tasks
Affective Bias
Recent work from our group has focused on translating an
animal task that measures negative affective bias (a common
feature of mood and anxiety disorders [20,44]) into humans
[10,45]. This task, sometimes known as the ‘ambiguous-cue
interpretation task’, was originally reported in 2004 [8]: rodents
were trained to press a lever when they heard a tone that was
associated with a positive event and to avoid pressing the lever
when a tone was delivered that was associated with a negative
event (70 dB white noise). Their affective bias was measured
by how they subsequently responded to intermediate, non-
reinforced tones: pressing the lever to intermediate tones on a
lower proportion of trials indicated negative affective bias.
Rodents experiencing a stressor (‘unpredictable housing’)—
intended to create a state analogous to depression in
humans—displayed increased negative affective bias. The pos-
sibility of measuring negative affective bias in rodents was an
important advance, given that previous human work in this area
had no corresponding animal paradigms.
There have been a number of modifications to this task to
remove confounds and improve validity [9,20,45–47]. Our
direct human translation of this task used two differently sized
rewards (Fig. 2) and also analysed the data using the same com-
putational modelling approach adopted for the animal task [45].
This task can also be performed with visual rather than auditory
stimuli [48] in humans, which may be more ethologically rele-
vant even though it shows lower face validity [49].
There are a number of strengths of this common currency
task: firstly, the task can be performed by diverse species
including starlings [50], honeybees [51], drosophila [52]and
macaques [53], and secondly, there is an associated computa-
tional model that captures performance on this task [45]. The
computational model used is a version of a drift diffusion
model [54,55]. Using this model identifies (putatively mech-
anistic) latent parameters, rather than just summary statistics
such as ‘mean accuracy’, which can be compared between
species and across manipulations. However, whilst this task
seems to have good construct validity, it has not shown strong
predictive validity in acute-administration antidepressant stud-
ies in animals thus far, though this may relate to the delayed-
onset mechanism of action of most commonly prescribed an-
tidepressants [20].
4 Curr Behav Neurosci Rep (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Fig. 2 a–cThe designs of the three types of cognitive task mentioned: the
affective bias task, reversal learning task and two-step task. d–fExamples
of how data is typically collapsed and analysed for these tasks. g–i
Examples of the additional information that can be gained by taking a
computational approach. aIn the affective bias task, also known as the
‘ambiguous-cue interpretation task’, participants are first trained to press
either the left or right button in response to the extreme stimuli (large or
small circles in this example) which are 100% associated with either a £1
or £4 reward (associations counterbalanced across participants). In the
test phase, during different trials, participants are shown either one of
the original extreme stimuli or a novel, intermediate stimulus, to which
they must respond by pressing the button associated with the stimulus
they think it is closer to. On intermediate trials, there is a 50% chance of
receiving a £1 or £4 reward. dAffective bias is operationalised here as the
proportion of times participants press the button associated with the
higher reward stimulus on intermediate stimulus trials. gAn example of
the drift rate, which can be estimated using a drift diffusion model
(DDM), allowing us to account for participant accuracy and reaction
times. In our work using this task [10], we found that patients with
mood and anxiety disorders demonstrate a lower drift rate towards
classifying the mid-tone as high reward. bIn reversal learning tasks,
participants typically choose between two stimuli on screen by pressing
the corresponding button. One stimulus is associated with reward,
indicated by a smiley face, and the other with punishment, indicated by
a sad face. The contingencies are then reversed, so that the previously
rewarded stimulus is now punished and vice versa. eThe probability of
participants choosing a correct (rewarded) choice. hThe estimated
learning rate; the shallower learning curve and greater latency before
performance returns to high accuracy after a reversal is indicative of a
slower learning rate in patients here. cIn this example of a two-step task,
participants start in one state (shown here in grey), and choose between
two stimuli (star or hexagon), each of which result in a probabilistic
transition (here, high probabilities are represented with a thicker arrow,
and low probabilities—‘rare transitions’—with a thinner arrow) to a
second-level state (either pink or orange), at which point they can
choose between the two stimuli which are available to them in that
state. Here, imagine that a participant chooses the star, and
probabilistically moves to the orange state (on the right). They then
choose the circle, which results in a reward. To obtain this reward
again, the participant could perform in a ‘model-free’way, without
understanding the transitional structure of the stages, and simply choose
the star again. However, this ‘model-free’way of behaving is most likely
to take them to the pinkstate, rather than the orange one. A ‘model-based’
choice would entail choosing the hexagon in state one, which is more
likely to result in a transition to the desired orange state. When these
choices are repeated over many trials, logistic regression or
computational modelling can be used to demonstrate the extent to
which participants behave in a ‘model-based’way to seek out the best
second state, rather than simply repeating actions whichpreviously led to
reward. fThe probability of repeating the last trial, split by the outcome
and transition type of the previous trial. iA computational modelling
analysis of participant data (solid lines) can be used to estimate a
‘weight’for each participant, which represents the extent to which they
rely on model-based (dotted lines) and model-free (dashed lines)
strategies
5Curr Behav Neurosci Rep (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Reversal Learning
Another recent paper that used a common currency task to
collect complementary data across-species focused on the
CACNA1C gene (which encodes a subunit of a type of
voltage-gated calcium channel) and reversal learning [56].
Variants of this gene have been linked to risk for schizophre-
nia and bipolar disorder [57,58]. The authors demonstrated
that reversal learning (Fig. 2) was impaired in humans with
two different risk alleles in the CACNA1C gene, and also in
rats who were heterozygote knockouts for CACNA1C. They
subsequently demonstrated in data from post-mortem tissue
that the human risk alleles result in lower expression of BDNF
(brain-derived neurotrophic factor) in the prefrontal cortex.
Similarly, the rat heterozygote knockouts also displayed re-
duced prefrontal BDNF expression, according to results from
both in situ hybridization and qPCR. The authors subsequent-
ly recommend that reversal learning tasks should be used in
translational research targeting voltage-gated calcium
channels.
Using a common currency task in this study allowed the
authors to demonstrate convincingly in both humans and ro-
dents that variants in the CACNA1C gene result in impair-
ments in reversal learning, and they were able to use comple-
mentary methods in animals and humans to demonstrate that
this may be underpinned by reduced BDNF in the prefrontal
cortex. These converging sets of evidence are more convinc-
ing than data from either species would be alone. Furthermore,
if this study had just been performed in humans, it would not
be possible to directly manipulate the gene of interest, and
instead, the conclusions would have to rely on associations
between genotype and task performance.
However, the reversal-learning task used differed substan-
tially between animals and humans. In particular, in the hu-
man task, the stimuli were coloured squares, and in the animal
task, these were different shapes. In the human task, rewards
were ‘smiley’faces and 1p monetary gain, and punishments
‘frowny’faces and 1p monetary loss, which were passively
received. By contrast, in the animal task, the reward was 10%
sucrose solution which was actively obtained from the maga-
zine, and the punishment was a 10-s ‘time-out’periods. In the
human task, reversals occurred after 7–11 trials, whereas in
the animal task, there was only one reversal, which occurred
after 2 days of > 80% accuracy on the task. The level of train-
ing also differed—animals were trained on both the procedure
for reward collection (collecting sucrose from the magazine)
and the association of a nose-poke action with a reward,
whereas humans were not trained. The variable from the hu-
man task that was compared between genotype was accuracy
after the first reversal along with total earnings, whereas the
variable from the rodent task that was used was percentage of
animals of each genotype that completed each experimental
condition. Whilst these differences may be more related to
face validity than construct validity, future work may focus
on aligning these paradigms more closely.
Goal-Based Decision-Making
Another study that combines some of the strengths from the
first and second studies is a back-translation of the two-step
task (Fig. 2), commonly used to demonstrate disrupted goal-
based decision making in OCD [59,60], for use in rodents
[61]. Computational models akin to those used in humans
were fitted to rodent behaviour, and it was demonstrated that
muscimol inactivation of either the dorsal hippocampus or the
OFC caused a reduction in the use of model-based reinforce-
ment learning. This allowed the authors to conclude that these
brain regions causally contribute to this type of learning,
whereas MRI could only demonstrate an association. The
use of modelling in conjunction with a common currency task
also allowed the researchers to compare behaviour between
species in a more technical way: they stated in their discussion
that the lack of observed model-free planning inrats compared
with humans may be due to the increased training that rats
received on the task. It is possible that the number of hours
of training rodents receive could be adjusted until computa-
tional analyses, performed on choices generated by both
humans and rodents, show no difference in the extent to which
these species are using model-free learning strategies. This
may ensure more optimal construct validity.
Future Directions and Recommendations
The advent of novel technological solutions has brought new
options to the development of common currency tasks.
Technologies such as the use of a touch screen [49,62]allow
all species to have the same access to instructions and training,
whilst offering standardisation and higher throughput. Crucially,
both stimuli and responses can be in the same modality between
species, and touch-screens may ensure that all animals are using
the same strategy to complete the task [49]. The use of virtual
reality [7] allows for human participants to be placed in environ-
ments closer to those used in classical animal tasks—such as
mazes—without the corresponding needs for space and sophis-
ticated ethical controls. Virtual reality versions of the Morris
water maze have been used in both stress and schizophrenia
research in humans [63,64]. Recent work in humans has also
used virtual reality to create ethical, and precisely-controlled,
threatening and non-threatening contexts [65]. Furthermore, the
use of virtual reality has also been found to be beneficial in
animals, as well as humans, either to precisely control the envi-
ronment or to reduce animal motion when using techniques such
as two-photon imaging or fMRI [66–68].
It has also been acknowledged that using computational
modelling could allow more of a mechanistic understanding
6 Curr Behav Neurosci Rep (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

of behaviour on tasks [69]. Typically, behavioural tasks are
analysed using summary statistics (Fig. 2d-f), which may cap-
ture differences between groups or relevant correlations, but
are atheoretical. Generative computational models contain not
only tractable summaries of data, but also contain within them
hypotheses about the (hidden) processes which led to the gen-
eration of the observed behaviour (such as, for example, learn-
ing rate, which cannot be directly observed—only inferred
from a set of responses collected over time). Methods for
adjudicating between different models enable researchers to
directly assess the evidence for different hypothesised data-
generating processes, and differences in computational param-
eters may reflect changes to the mechanisms which are
hypothesised to generate behaviour, such as learning rate, be-
havioural noise or prior beliefs about the world. Notably, two
of the papers discussed above in the example section used a
common currency task with an associated computational
model (Fig. 2g-i)[10,61]. The use of computational model-
ling allows hypotheses about the strategies used by different
species or participants to be quantitatively tested, and the fit of
different models (representing different strategies) compared.
It is yet to be seen whether using computational modelling
improves translational efficiency in psychiatric research, but
it may be a valuable avenue for exploration.
In Table 1, we therefore provide a summary of poten-
tial aspects that may influence validity that researchers
Table 1 Possible aspects of the tasks that are ‘common’in common currency tasks—for consideration when designing new tasks
Common aspects Type of validity Notes
Demographics
of sample
Face/construct Many animal studies only use males [71]—is this appropriate for the research question? Are
humans only recruited if they fit into certain demographics (age, medication)?
Developmental stage Face/construct Is the same developmental stage used in both human and animal research? e.g. adolescent vs
adult
Task difficulty Construct Does the task need to be simpler for animals to achieve the same level of performance? Do
overtrained animals perform better than humans?
Task duration Face/construct Do different species need different task durations? i.e. do humans get bored faster/produce
more varied behaviour so more data points are needed for accurate inference? Do animals
need many short sessions of a task, whereas humans can perform the task in one longer
session?
Motivation Construct Are animals water restricted to maximise their desire for (water) reward? Are humans
reimbursed more for good performance?
Training
Instructed Face/construct Are the instructions verbal?
Overtraining Face/construct Animals are frequently overtrained on tasks, whereas humans are usually not
Stimulus presentation
Modality Face/construct Visual, aural? Would construct validity be achieved better if stimuli are different between
species?
Actual stimuli Face/construct E.g. tones may be adjusted for different species’hearing ranges
Response Face/construct E.g. do animals and humans both press buttons, or do animals enter a nose-poke?
Feedback
Classification Face/construct Often, animals receive primary reinforcers such as sucrose or electrical shocks, and humans
receive points or money
Actual feedback Face/construct Even if both primary reinforcers, feedback may still differ: e.g. white noise in humans, and
electric shocks in animals
Strategy Construct Are animals and humans using the same ‘strategy’to complete the task? For example,
animals and humans may rely to different extents on spatial strategies in the Morris water
maze and the virtual-reality human equivalent [16]
Data preprocessing Construct Is data quality assessed in the same way between species? Are data cleaned in the same way?
Analysis Construct Are the primary outcome measures the same? Are they calculated in the same way?
Behavioural performance Construct Is behavioural performance (e.g. patterns of accuracy) similar between species?
Neural basis Construct Are homologous brain areas and circuits implicated in the performance of this task between
species?
Sensitivity to symptoms Face/construct/
predictive
Is behaviour on the task sensitive to psychiatric symptoms, e.g. do animal models of
anhedonia demonstrate a measurable change from healthy animals in the same way
anhedonic humans perform differently to healthy controls?
Effects of interventions Predictive Do pharmacological agents have the same effects on both human and animal
behaviour/neural activity in the task?
7Curr Behav Neurosci Rep (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

should consider when designing or translating tasks. We
recommend that future common currency tasks, rather
than trying to make tasks precisely identical in all the
ways shown, ascertain which elements of the task should
match in order to provide the highest likelihood that the
underlying cognitive strategies and neural mechanisms are
thesame[26,49]. Not only may species be using different
strategies to perform the same apparent behaviour [3,12,
70], animals may also be performing a different behaviour
altogether (resulting in poor construct validity, despite
accompanying good face validity).
In addition to the task design, it is also important to
consider the experimental subjects. Most research in ani-
mals is performed using males [71] and is also often re-
stricted to specific inbred strains, which are not necessar-
ily representative of wild-type animals [72]. Limiting
translational research to highly standardised and
constrained populations is likely to reduce the
generalisability of findings to humans. For instance, in
human psychiatry research, there are significant gender
differences between the prevalence of different disorders,
with mood and anxiety disorders being more prevalent in
womenthaninmen[73]. Limiting animal work to male
subjects may drive differences between preclinical and
clinical findings and increase the chance of translational
failure. Similarly, many psychiatric disorders are thought
to be polygenic and have epigenetic influences, factors
which are hard to account for and study using a geneti-
cally close-to-identical sample [74].
We also recommend that evidence indicating the ex-
tent of construct validity is summarised in papers pre-
senting common currency tasks, whether this is behav-
ioural, neural or otherwise. Whilst face validity may be
easy to assess by comparing the methods used across
species, assessing construct validity is harder—and face
validity does not necessarily entail construct validity, as
described above.
A‘multifactorial’approach is also recommended, in-
cluding both behavioural and neural measures, as this
could increase confidence in translational results [3,11,
19]. In particular, as fMRI becomes a more common con-
comitant of human research, fMRI in animals shows in-
creasing promise as a directly translatable measure of the
neural effects of new pharmacological agents on common
currency tasks [19].
Finally, we recommend that future research and develop-
ment of common currency tasks should be bi-directional: ba-
sicresearchshouldbeusedtoinformclinicalpractice,and
clinical observations can inform basic research. Both transla-
tion and back-translation should be iterated over in order to
obtain tasks that are truly translationally valid. This work will
ensure that the promise of common currency tasks is truly
achievable.
Conclusions
In this review, we have discussed the definitions of com-
mon currency tasks and the aspects of tasks which may be
consistent across animals and humans. We have also
highlighted several benefits of common currency tasks:
the most important of which is that they may alleviate
the ‘bottleneck’in drug development work. Three recent
examples using common currency tasks are discussed in
detail, with their strengths and limitations. We conclude
by offering several recommendations for future work: in-
cluding focus on construct rather than face validity, use of
multifactorial methods and novel technological ap-
proaches, and the use of computational models. If prog-
ress in this field is sustained, common currency tasks may
offer a window of opportunity for significant advances in
translational work, hopefully heralding a new period of
psychiatric drug discovery.
Compliance with Ethical Standards
Conflict of Interest Alexandra C. Pike has served as a postdoctoral
research associate, funded by OJR’s fellowship with the Medical
Research Council, and has received both sponsored travel and non-finan-
cial support for industrial collaboration with Roche.
Human and Animal Rights All reported studies/experiments with hu-
man or animal subjects performed by the authors have been previously
published and complied with all applicable ethical standards (including
the Helsinki declaration and its amendments, institutional/national re-
search committee standards and international/national/institutional
guidelines).
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing, adap-
tation, distribution and reproduction in any medium or format, as long as
you give appropriate credit to the original author(s) and the source, pro-
vide a link to the Creative Commons licence, and indicate if changes were
made. The images or other third party material in this article are included
in the article's CreativeCommons licence, unless indicated otherwise in a
credit line to the material. If material is not included in the 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/.
References
1. Hyman SE. Revolution stalled. Sci Transl Med. 2012;4(155):
155cm11.
2. KesselheimAS, Hwang TJ, Franklin JM. Two decades of new drug
development for central nervous system disorders. Nat Rev Drug
Discov. 2015;14(12):815–6.
3. Markou A, Chiamulera C, Geyer MA, Tricklebank M, Steckler T.
Removing obstacles in neuroscience drug discovery: the future path
for animal models. Neuropsychopharmacology. 2009;34(1):74–89.
8 Curr Behav Neurosci Rep (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

4. Morris RG, Downes JJ, Sahakian BJ, Evenden JL, Heald A,
Robbins TW. Planning and spatial working memory in
Parkinson’s disease. J Neurol Neurosurg Psychiatry. 1988;51(6):
757–66.
5. Downes JJ, Roberts AC, Sahakian BJ, Evenden JL, Morris RG,
Robbins TW. Impaired extra-dimensional shift performance in
medicated and unmedicated Parkinson’s disease: evidence for a
specific attentional dysfunction. Neuropsychologia. 1989;27(11):
1329–43.
6. Dickson PE, Calton MA, Mittleman G. Performance of C57BL/6J
and DBA/2J mice on a touchscreen-based attentional set-shifting
task. Behav Brain Res. 2014;261:158–70.
7. Shipman SL, Astur RS. Factors affecting the hippocampal BOLD
response during spatial memory. Behav Brain Res. 2008;187(2):
433–41.
8. Harding EJ, Paul ES, Mendl M. Cognitive bias and affective state.
Nature. 2004;427(6972):312.
9. Enkel T, Gholizadeh D, von Bohlen und Halbach O, Sanchis-
Segura C, Hurlemann R, Spanagel R, et al. Ambiguous-cue inter-
pretation is biased under stress- and depression-like states in rats.
Neuropsychopharmacology. 2010;35(4):1008–15.
10. Aylward J, Hales C, Robinson E, Robinson OJ. Translatinga rodent
measure of negative bias into humans: the impact of induced anx-
iety and unmedicated mood and anxiety disorders. Psychol Med.
2020;50(2):237–46.
11. Der-Avakian A, Pizzagalli DA. Translational assessments of re-
ward and anhedonia: a tribute to Athina Markou. Biol Psychiatry.
2018;83(11):932–9.
12. Young JW, Jentsch JD, Bussey TJ, Wallace TL, Hutcheson DM.
Consideration of species differences in developing novel molecules
as cognition enhancers. Neurosci Biobehav Rev. 2013;37(9):2181–
93.
13. Belzung C, Lemoine M. Criteria of validity for animal models of
psychiatric disorders: focus on anxiety disorders and depression.
Biol Mood Anxiety Disord. 2011;1(1):9.
14. Willner P. The validity of animal models of depression.
Psychopharmacology. 1984;83(1):1–16.
15. Cronbach LJ, Meehl PE. Construct validity in psychological tests.
Psychol Bull. 1955;52(4):281–302.
16. Schoenfeld R, Schiffelholz T, Beyer C, Leplow B, Foreman N.
Variants of the Morris water maze task to comparatively assess
human and rodent place navigation. Neurobiol Learn Mem.
2017;139:117–27.
17. Delgado MR, Jou RL, Phelps EA. Neural Systems Underlying
Aversive Conditioning in Humans with Primary and Secondary
Reinforcers. Front Neurosci [Internet]. 2011;5:71. Available from.
https://doi.org/10.3389/fnins.2011.00071/full.
18. Sescousse G, Caldú X, Segura B, Dreher J-C. Processing of primary
and secondary rewards: a quantitative meta-analysis and review of
human functional neuroimaging studies. Neurosci Biobehav Rev.
2013 May 1;37(4):681–96.
19. Silverman JL, Nithianantharajah J, Der-Avakian A, Young JW,
Sukoff Rizzo SJ. Lost in translation: at the crossroads of face va-
lidity and translational utility of behavioral assays in animal models
for the development of therapeutics. Neurosci Biobehav Rev.
2020;116:452–3.
20. Robinson ESJ, Roiser JP. Affective Biases in Humans and
Animals. Curr Top Behav Neurosci. 2016;28:263–86.
21. Hendrie C, Pickles A. The failure of the antidepressant drug dis-
covery process is systemic. J Psychopharmacol (Oxf). 2013;27(5):
407–16.
22. Kirlic N, Young J, Aupperle RL. Animal to human translational
paradigms relevant for approach avoidance conflict decision mak-
ing. Behav Res Ther. 2017;96:14–29.
23. Aupperle RL, Sullivan S, Melrose AJ, Paulus MP, Stein MB. A
reverse translational approach to quantify approach-avoidance con-
flict in humans. Behav Brain Res. 2011;225(2):455–63.
24. Collins AGE, Frank MJ. How much of reinforcement learning is
working memory, not reinforcement learning? A behavioral, com-
putational, and neurogenetic analysis: working memory in rein-
forcement learning. Eur J Neurosci. 2012;35(7):1024–35.
25. Geyer MA. Developing translational animal models for symptoms
of schizophrenia or bipolar mania. Neurotox Res. 2008;14(1):71–8.
26. Young J, Geyer M. Developing treatments for cognitive deficits in
schizophrenia: the challenge of translation. J Psychopharmacol
(Oxf). 2015;29(2):178–96.
27. Hafizi S, Chandra P, Cowen PJ. Neurokinin-1 receptor antagonists
as novel antidepressants: trials and tribulations. Br J Psychiatry.
2007;191(4):282–4.
28. Ebner K, Singewald N. The role of substance P in stress and anxiety
responses. Amino Acids. 2006;31(3):251–72.
29. Ironside M, Amemori K, McGrath CL, Pedersen ML, Kang MS,
Amemori S, et al. Approach-avoidance conflict inmajor depressive
disorder: congruent neural findings in humans and nonhuman pri-
mates. Biol Psychiatry. 2020;87(5):399–408.
30. Nestler EJ, Hyman SE. Animal models of neuropsychiatric disor-
ders. Nat Neurosci. 2010;13(10):1161–9.
31. Berton O, Hahn C-G, Thase ME. Are we getting closer to valid
translational models for major depression? Science.
2012;338(6103):75–9.
32. Der-Avakian A, D’Souza MS, Pizzagalli DA, Markou A.
Assessment of reward responsiveness in the response bias proba-
bilistic reward task in rats: implications for cross-species transla-
tional research. Transl Psychiatry. 2013;3(8):e297.
33. Stuart SA, Butler P, Munafò MR, Nutt DJ, Robinson ES. A trans-
lational rodent assay of affective biases in depression and antide-
pressant therapy. Neuropsychopharmacology. 2013;38(9):1625–
35.
34. McKim TH, Shnitko TA, Robinson DL, Boettiger CA.
Translational research on habit and alcohol. Curr Addict Rep.
2016;3(1):37–49.
35. Gould TD, Gottesman II. Psychiatric endophenotypes and the de-
velopment of valid animal models. Genes Brain Behav. 2006;5(2):
113–9.
36. Insel T, Cuthbert B, Garvey M, Heinssen R, Pine DS, Quinn K,
et al. Research domain criteria (RDoC): toward a new classification
framework for research on mental disorders. Am Psychiatric Assoc.
2010;167:748–51.
37. Keefe RSE. Treating cognitive impairment in depression: an unmet
need. Lancet Psychiatry. 2016;3(5):392–3.
38. Millan MJ, Agid Y, Brüne M, Bullmore ET, Carter CS, Clayton
NS, et al. Cognitive dysfunction in psychiatric disorders: character-
istics, causes and the quest for improved therapy. Nat Rev Drug
Discov. 2012;11(2):141–68.
39. Miyamoto S, Miyake N, Jarskog LF, Fleischhacker WW,
Lieberman JA. Pharmacological treatment of schizophrenia: a crit-
ical review of the pharmacology and clinical effects of current and
future therapeutic agents. Mol Psychiatry. 2012;17(12):1206–27.
40. Shilyansky C, Williams LM, Gyurak A, Harris A, Usherwood T,
Etkin A. Effect of antidepressant treatment on cognitive impair-
ments associated with depression: a randomised longitudinal study.
Lancet Psychiatry. 2016;3(5):425–35.
41. Trivedi MH, Greer TL. Cognitive dysfunction in unipolar depres-
sion: implications for treatment. J Affect Disord. 2014;152–154:
19–27.
42. Marder SR. The NIMH-MATRICS project for developing
cognition-enhancing agents for schizophrenia. Dialogues Clin
Neurosci. 2006;8(1):109–13.
43. Marder SR, Fenton W. Measurement and treatment research to
improve cognition in schizophrenia: NIMH MATRICS initiative
9Curr Behav Neurosci Rep (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

to support the development of agents for improving cognition in
schizophrenia. Schizophr Res. 2004;72(1):5–9.
44. Mathews A, MacLeod C. Cognitive vulnerability to emotional dis-
orders. Annu Rev Clin Psychol. 2005;1:167–95.
45. Hales CA, Robinson ESJ, Houghton CJ. Diffusion modelling re-
veals the decision making processes underlying negative judgement
bias in rats. PLoS One. 2016;11(3):e0152592.
46. Anderson MH, Hardcastle C, Munafò MR, Robinson ESJ.
Evaluation of a novel translational task for assessing emotional
biases in different species. Cogn Affect Behav Neurosci.
2012;12(2):373–81.
47. Hales CA, Stuart SA, Anderson MH, Robinson ESJ. Modelling
cognitive affective biases in major depressive disorder using ro-
dents. Br J Pharmacol. 2014;171(20):4524–38.
48. Daniel-Watanabe L, McLaughlin M, Gormley S, Robinson OJ.
Association between a directly translated cognitive measure of neg-
ative bias and self-reported psychiatric symptoms. Biol Psychiat
Cogn Neurosci Neuroimag. 2020;4:S2451902220300537.
49. Talpos J, Steckler T. Touching on translation. Cell Tissue Res.
2013;354(1):297–308.
50. Bateson M, Matheson S. Performance on a categorisation task sug-
gests that removal of environmental enrichment induces ‘pessi-
mism’in captive European starlings (Sturnus vulgaris). Anim
Welf. 2007;16:33–6.
51. Bateson M, Desire S, Gartside SE, Wright GA. Agitated honeybees
exhibit pessimistic cognitive biases. Curr Biol. 2011;21(12):1070–
3.
52. Deakin A, Mendl M, Browne WJ, Paul ES, Hodge JJL. State-
dependent judgement bias in Drosophila: evidence for evolution-
arily primitive affective processes. Biol Lett. 2018;14(2):20170779.
53. Bethell EJ,Holmes A, MacLarnon A, Semple S. Cognitive bias in a
non-human primate: husbandry procedures influence cognitive in-
dicators of psychological well-being in captive rhesus macaques.
Anim Welf. 2012;21(2):185–95.
54. Ratcliff R, Rouder JN. Modeling response times for two-choice
decisions. Psychol Sci. 1998;9(5):347–56.
55. Ratcliff R. A theory of memory retrieval. Psychol Rev. 1978;85(2):
59–108.
56. Sykes L, Haddon J, Lancaster TM, Sykes A, Azzouni K, Ihssen N,
et al. Genetic variation in the psychiatric risk gene CACNA1C
modulates reversal learning across species. Schizophr Bull.
2019;45(5):1024–32.
57. Ripke S, Neale BM, Corvin A, Walters JTR, Farh K-H, Holmans
PA, et al. Biological insights from 108 schizophrenia-associated
genetic loci. Nature. 2014;511(7510):421–7.
58. Consortium C-DG of the PG. Identification of risk loci with shared
effects on five major psychiatricdisorders:a genome-wide analysis.
Lancet. 2013;381(9875):1371.
59. Daw ND, Gershman SJ, Seymour B, Dayan P, Dolan RJ. Model-
based influences on humans’choices and striatal prediction errors.
Neuron. 2011;69(6):1204–15.
60. Gillan CM, Kosinski M, Whelan R, Phelps EA, Daw ND.
Characterizing a psychiatric symptom dimension related to deficits
in goal-directed control. eLife. 2016;5:e11305.
61. Miller KJ, Botvinick MM, Brody CD. Dorsal hippocampus con-
tributes to model-based planning. Nat Neurosci. 2017;20(9):1269–
76.
62. Bussey TJ, Holmes A, Lyon L, Mar AC, McAllister KAL,
Nithianantharajah J, et al. New translational assays for preclinical
modelling of cognition in schizophrenia: the touchscreen testing
method for mice and rats. Neuropharmacology. 2012;62(3):1191–
203.
63. van Gerven DJH, Ferguson T, Skelton RW. Acute stress switches
spatial navigation strategy from egocentric to allocentric in a virtual
Morris water maze. Neurobiol Learn Mem. 2016;132:29–39.
64. Hanlon FM, Weisend MP, Hamilton DA, Jones AP, Thoma RJ,
Huang M, et al. Impairment on the hippocampal-dependent virtual
Morris water task in schizophrenia. Schizophr Res. 2006;87(1):67–
80.
65. Alvarez RP, Chen G, Bodurka J, Kaplan R, Grillon C. Phasic and
sustained fear in humans elicits distinct patterns of brain activity.
Neuroimage. 2011;55(1):389–400.
66. Thurley K, Ayaz A. Virtual reality systems for rodents. Curr Zool.
2017;63(1):109–19.
67. Lopatina OL, Morgun AV, Gorina YV, Salmin VV, Salmina AB.
Current approaches to modeling the virtual reality in rodents for the
assessment of brain plasticity and behavior. J Neurosci Methods.
2020;335:108616.
68. Dombeck DA, Harvey CD, Tian L, Looger LL, Tank DW.
Functional imaging of hippocampal place cells at cellular resolu-
tion during virtual navigation. Nat Neurosci. 2010;13(11):1433–
40.
69. Ferrante M, Redish AD, Oquendo MA, Averbeck BB, Kinnane
ME, Gordon JA. Computational psychiatry: a report from the
2017 NIMH workshop on opportunities and challenges. Mol
Psychiatry. 2019;24(4):479–83.
70. Badre D, Frank MJ, Moore CI. Interactionist neuroscience. Neuron.
2015;88(5):855–60.
71. Zucker I, Beery AK. Males still dominate animal studies. Nature.
2010;465(7299):690.
72. Würbel H. Behaviour and the standardization fallacy. Nat Genet.
2000;26(3):263.
73. Grant BF, Weissman MM. Gender and the prevalence of psychiat-
ric disorders. In: Age and gender considerations in psychiatric di-
agnosis: a research agenda for DSM-V. Arlington: American
Psychiatric Publishing, Inc.; 2007. p. 31–45.
74. Moore SJ, Murphy GG, Cazares VA. Turning strains into strengths
for understanding psychiatric disorders. Mol Psychiatry. 2020;13:
1–14.
Publisher’sNoteSpringer Nature remains neutral with regard to jurisdic-
tional claims in published maps and institutional affiliations.
10 Curr Behav Neurosci Rep (2021) 8:1–10
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com