Null effect of anodal and cathodal transcranial direct current stimulation (tDCS) on own- and other-race face recognition

Abstract

Successful face recognition is important for social interactions and public security. Although some preliminary evidence suggests that anodal and cathodal transcranial direct current stimulation (tDCS) might modulate own- and other-race face identification, respectively, the findings are largely inconsistent. Hence, we examined the effect of both anodal and cathodal tDCS on the recognition of own- and other-race faces. Ninety participants first completed own- and other-race Cambridge Face Memory Test (CFMT) as baseline measurements. Next, they received either anodal tDCS, cathodal tDCS or sham stimulation and finally they completed alternative versions of the own- and other-race CFMT. No difference in performance, in terms of accuracy and reaction time, for own- and other-race face recognition between anodal tDCS, cathodal tDCS and sham stimulation was found. Our findings cast doubt upon the efficacy of tDCS to modulate performance in face identification tasks.

Full text

Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=psns20
Social Neuroscience
ISSN: (Print) (Online) Journal homepage: https://www.tandfonline.com/loi/psns20
Null effect of anodal and cathodal transcranial
direct current stimulation (tDCS) on own- and
other-race face recognition
Siew Kei Kho, David Keeble, Hoo Keat Wong & Alejandro J. Estudillo
To cite this article: Siew Kei Kho, David Keeble, Hoo Keat Wong & Alejandro J. Estudillo (16 Oct
2023): Null effect of anodal and cathodal transcranial direct current stimulation (tDCS) on own-
and other-race face recognition, Social Neuroscience, DOI: 10.1080/17470919.2023.2263924
To link to this article: https://doi.org/10.1080/17470919.2023.2263924
Published online: 16 Oct 2023.
Submit your article to this journal
Article views: 4
View related articles
View Crossmark data

RESEARCH PAPER
Null eect of anodal and cathodal transcranial direct current stimulation (tDCS)
on own- and other-race face recognition
Siew Kei Kho
a,b
, David Keeble
a
, Hoo Keat Wong
a
and Alejandro J. Estudillo
a
a
Department of Psychology, Bournemouth University, Poole, United Kingdom;
b
School of Psychology, University of Nottingham Malaysia,
Semenyih, Malaysia
ABSTRACT
Successful face recognition is important for social interactions and public security. Although some
preliminary evidence suggests that anodal and cathodal transcranial direct current stimulation
(tDCS) might modulate own- and other-race face identication, respectively, the ndings are
largely inconsistent. Hence, we examined the eect of both anodal and cathodal tDCS on the
recognition of own- and other-race faces. Ninety participants rst completed own- and other-race
Cambridge Face Memory Test (CFMT) as baseline measurements. Next, they received either anodal
tDCS, cathodal tDCS or sham stimulation and nally they completed alternative versions of the
own- and other-race CFMT. No dierence in performance, in terms of accuracy and reaction time,
for own- and other-race face recognition between anodal tDCS, cathodal tDCS and sham stimula-
tion was found. Our ndings cast doubt upon the ecacy of tDCS to modulate performance in face
identication tasks.
ARTICLE HISTORY
Received 8 March 2023
Revised 22 August 2023
Published online 19 October
2023
KEYWORDS
Transcranial electrical
stimulation; brain
stimulation; face perception;
face memory; tDCS
Introduction
Face recognition is important for many social interactions
that occur in our everyday life (Jack & Schyns, 2015).
Although face recognition is used extensively, research
has shown that we are not experts in recognizing unfa-
miliar faces (Bruce et al., 1999; Davis & Valentine, 2009;
Kemp et al., 1997; White et al., 2014; Young & Burton,
2018). For example, passport control ocers present
high error rates (14%) in face matching despite having
years of experience and having received specic training
in the task (White et al., 2014). Diculties in face identi-
cation are even more prominent with other-race faces
(Meissner & Brigham, 2001). The other-race eect (ORE)
in face recognition shows that humans tend to be better
at recognizing own-race faces compared to other-race
faces (Estudillo et al., 2020; Malpass & Kravitz, 1969;
Wong et al., 2021). The ORE has been found across dier-
ent tasks and countries, and even when the morphologi-
cal dierences across the faces are minor (McKone et al.,
2011), pointing to a very robust phenomenon.
Own and other race faces are recognized dierently
and potentially involve dierent neural mechanisms.
Prior research has reported greater activation to own-
race compared to other-race faces in dierent brain
areas such as the occipital face area, the fusiform gyrus,
the right inferior frontal gyrus and the right medial
frontal cortex (Feng et al., 2011; Golby et al., 2001; Kim
et al., 2006). Interestingly, although the activation in the
fusiform face area is initially stronger for own-race faces,
the activation for other-race faces increases over time,
eventually surpassing the response to own-race faces
(Natu et al., 2011). This suggests that own-race faces
are processed more automatically compared to other-
race faces. Furthermore, event-related potential (ERP)
research has generally found larger N170 amplitudes in
response to other-race compared to own-race faces
(Anzures & Mildort, 2021; Giménez-Fernández et al.,
2020; Yao & Zhao, 2019, but see Cassidy et al., 2014;
Senholzi & Ito, 2013; Wiese, 2013, for a reversed pattern).
This nding has been associated with a disruption of
congural face processing (Jacques & Rossion, 2010), as
it is comparable to the N170 face inversion eect, where
larger N170 amplitudes are observed for inverted faces
as opposed to upright faces (Eimer, 2000; Goaux et al.,
2003; Rossion et al., 1999). Other ERP components, such
as the P100 (Anzures & Mildort, 2021; Giménez-
Fernández et al., 2020) and P200 (Anzures & Mildort,
2021; Wiese, 2013), have also shown dierences
between own- and other-race faces (for a review, see
Serani & Pesciarelli, 2022).
Improvement of face recognition for own- and other-
race faces could be important for individuals with
CONTACT Siew Kei Kho kkho@bournemouth.ac.uk; Alejandro J. Estudillo aestudillo@bournemouth.ac.uk Department of Psychology,
Bournemouth University, Poole House Talbot Campus, Poole BH12, United Kingdom
SOCIAL NEUROSCIENCE
https://doi.org/10.1080/17470919.2023.2263924
© 2023 Informa UK Limited, trading as Taylor & Francis Group

developmental and neurological disorders that are asso-
ciated with face recognition decits such as prosopag-
nosia (Rossion, 2014), autism (Weigelt et al., 2012) and
schizophrenia (Marwick & Hall, 2008); but see (Bortolon
et al., 2015). Prosopagnosia, also known as face blind-
ness, is a visual impairment that aects face recognition
despite intact visual acuity and intelligence (Bate & Tree,
2017). Individuals with prosopagnosia may face dicul-
ties in recognizing unfamiliar faces (Duchaine et al.,
2006), familiar faces (Busigny & Rossion, 2010) and occa-
sionally their own face (Parketny et al., 2015). Failure in
recognizing familiar identities (e.g., family members and
friends) could contribute to negative consequences such
as feeling of embarrassment and guilt which may build
up anxiety, increase fear of social interaction and lower
levels of self-condence (Dalrymple et al., 2014; Yardley
et al., 2008). Own- and other-race face recognition
improvements could also be important in terms of pub-
lic security. In fact, errors in the identication of unfami-
liar faces in public security could lead to serious personal
and societal consequences such as wrongful conviction
of an innocent person, while the actual criminal remains
unrestrained. Given the catastrophic consequences of
inaccurate face recognition in terms of public security
and for individuals with developmental and neurological
disorders associated with face recognition decits, it is
important to develop eective ways of improving face
recognition skills.
One possible method of improving face recognition is
by using transcranial direct current stimulation (tDCS).
TDCS is a form of noninvasive brain stimulation techni-
que where a low-level intensity electrical current is deliv-
ered between two or more electrodes attached to the
scalp to modulate neuronal excitability (Reed & Cohen
Kadosh, 2018). TDCS brings the neurons closer to their
ring threshold without eliciting an action potential
(Bikson et al., 2004). During an action potential,
a change in voltage across the membrane is caused by
the ow of ions such as potassium, sodium and chloride
into and out of the neuron. TDCS modulates the resting
membrane potential of neurons by adjusting their state
to approach or move away from the threshold potential
(approximately −55 mV) necessary to generate an action
potential. In this manner, tDCS can enhance or diminish
neuronal excitability. TDCS produces opposing eects
on neuronal excitability depending on electrode polar-
ity. Anodal tDCS (a-tDCS) is thought to cause neuronal
depolarization which leads to an increase in neurons
ring rate and excitability, while cathodal tDCS (c-tDCS)
is thought to cause neuronal hyperpolarization which
leads to a decrease in neurons ring rate and excitability
(Nitsche & Paulus, 2000; Yamada & Sumiyoshi, 2021).
Although anodal stimulation often led to performance
enhancement, the eects from cathodal stimulation
were relatively inconsistent (Jacobson et al., 2012).
Improvement in own-race face processing has been
found following the application of a-tDCS to the occipital
area (Barbieri et al., 2016) and, more specically, to the
fusiform face area (Brunyé et al., 2017). For example,
participants who received online (i.e., stimulation applied
during task execution) 1.5 mA of a-tDCS to the right fusi-
form gyrus (i.e., PO10) during both study and test phase
showed improvement in face memory accuracy com-
pared to participants who received 0.5 mA of a-tDCS and
participants who received no stimulation (Brunyé et al.,
2017). Another study found that oine (i.e., stimulation
applied before task execution) 1.5 mA of a-tDCS to the
right occipital cortex (i.e., PO8) improved face perception
(measured by the Face Perception task) and face memory
(measured by the Cambridge Face Memory Test), while no
eect of online a-tDCS applied during both study and test
phase was found (Barbieri et al., 2016). This showed that
oine stimulation may work better compared to online
stimulation in terms of improving face processing.
However, the positive eects of a-tDCS on face identica-
tion are not always replicated (Willis et al., 2019).
In comparison to a-tDCS, research on the eects of
c-tDCS on face identication is scarce (Costantino et al.,
2017; Yang et al., 2014). One early study found that, com-
pared to sham stimulation, both anodal and cathodal o-
line 1.5 mA tDCS over occipito-temporal regions (i.e., left
occipito-temporal cortex: P7; right occipito-temporal cor-
tex: P8) reduced the N170 face-specic event-related
potential component (Yang et al., 2014). The ndings of
this study showed that the polarity of the current did not
alter the eect of the stimulation, suggesting that anodal
and cathodal tDCS elicit similar eects, at least in the face
domain. A more recent study found that oine 1.5 mA of
c-tDCS over the right occipital cortex (i.e., PO8) could
decrease recognition performance for other-race faces
(Costantino et al., 2017). Specically, this study tested
a group of non-Caucasian participants who lived in
a Caucasian-majority country and had extensive experience
with Caucasian faces. Interestingly, after c-tDCS, perfor-
mance to identify Caucasian faces decreased in the non-
Caucasian group, suggesting that c-tDCS elicited an ORE-
like behavior.
However, Costantino et al., (2017) study presents a few
important methodological drawbacks. In the study, stimu-
lation was a between-subjects variable where participants
either received sham stimulation or c-tDCS. However, while
2S. K. KHO ET AL.

the pre-stimulation assessment comprised a face percep-
tion test (i.e., Cambridge Face Perception Test, Duchaine
et al., 2007), the post-stimulation assessment comprised
a face memory test (i.e., the Cambridge Face Memory
Test, Duchaine & Nakayama, 2006). Interestingly, research
has shown that face perception (i.e., face identication
without memory component) and face memory (i.e., face
identication with memory component) are only moder-
ately correlated (Bate et al., 2019; Verhallen et al., 2017) and
dissociations between these two skills have been previously
reported (Barton, 2008; Behrmann et al., 2005; Dalrymple
et al., 2014; Estudillo & Bindemann, 2014; Weigelt et al.,
2014). Therefore, it is possible that there may be pre-
stimulation dierences in face memory performance
between the c-tDCS and the sham stimulation groups
which could potentially explain any post-stimulation dier-
ences. In addition, Costantino et al. (2017) only used c-tDCS,
so it is unknown whether anodal stimulation would pro-
duce similar eects in other-race faces.
Present study
The current study aims to further investigate the eect of
anodal and cathodal tDCS on the recognition of own- and
other-race faces. The stimulation will be applied in an o-
line manner since previous research using transcranial
electrical stimulation has shown that oine stimulation is
more eective compared to online stimulation in the work-
ing memory (Friehs & Frings, 2019) and face identication
domain (Barbieri et al., 2016; Estudillo et al., 2023), at least in
the neurotypical population (Hill et al., 2016). The ability to
recognize own- and other-race faces will be assessed
before the stimulation sessions to ensure that there are
no dierences in general face recognition ability between
the stimulation groups. To assess face recognition ability,
we employed the same face memory measure (the
Cambridge Face Memory Test) before and after stimula-
tion. Furthermore, both a-tDCS and c-tDCS will be included
to compare the eects of these two forms of stimulation.
As previous work examining the tDCS eects on face pro-
cessing showed inconsistent ndings (Barbieri et al., 2016;
Costantino et al., 2017; Willis et al., 2019; Yang et al., 2014),
we based our hypothesis on the neurophysiological
mechanism of tDCS (Nitsche & Paulus, 2000) where
a-tDCS should improve the recognition of own- and other-
race faces, while c-tDCS should impair the recognition of
own- and other-race faces.
Methods
The experiment was pre-registered via the Open Science
Framework (OSF) before data collection (https://osf.io/
6cf7w).
Participants
A power sensitivity analysis conducted using G*Power 3.1
(Faul et al., 2009) revealed that a 2 (CFMT type: own-race
vs. other-race) × 3 (simulation group: a-tDCS vs. c-tDCS vs.
sham) mixed ANOVA with 90 participants would be sensi-
tive to eects of η
2
= .015, f = .124 with 80% power (alpha
= 0.5). This means that the study would not be able to
reliably detect eects smaller than η
2
= .015. The eect
size reported in Costantino et al. (2017) was η
2
= .037, f
= .196, which exceeds the detectable eect size in the
current study.
Ninety Chinese Malaysian (67 females) were recruited.
Participants’ age ranged between 18 and 28 years (M =
21.11 years, SD = 1.97 years) and were students at the
University of Nottingham Malaysia. The participants
were assigned randomly to one of three stimulation
conditions: a-tDCS, c-tDCS or sham stimulation, with 30
participants in each condition. One-way ANOVA
revealed no age dierence between stimulation groups,
F(2, 87) = 1.9, p = .16. Prior to the experimental session,
all participants completed a screening form regarding
the inclusion and exclusion criteria concerning the appli-
cation of transcranial electrical stimulation and provided
informed consent. Participants were asked to sleep for
a minimum of 6 hours and refrain from consuming
alcohol (1 day before the experimental session) and
caeine (1 hour before the experimental session).
Participants were also asked to avoid using any hair
products (i.e., hair cream, hair gel) on the day of the
experimental session. Since hormone levels which uc-
tuate among females due to the menstrual cycle could
aect cortical excitability (Smith et al., 2002), female
participants were only recruited during the follicular
phase of the menstrual cycle, as in this phase the hor-
mone levels are most similar to males (for a similar pro-
cedure, see Barbieri et al., 2016).
A remuneration of RM20 was given for participation.
The study has been reviewed and approved by the
Science and Engineering Research Ethics Committee
(SEREC) at the University of Nottingham Malaysia
(approval code: KSK050320).
Cambridge face memory test (CFMT)
Two versions of the own-race CFMT (i.e., CFMT-Chinese,
McKone et al., 2017) and CFMT-Chinese Malaysian, Kho
et al. (2023) and two versions of the other-race CFMT
(i.e., CFMT-original, Duchaine & Nakayama, 2006, and
CFMT-Australian; McKone et al., 2011) were used in the
experiment. The CFMT-Chinese Malaysian was designed
to replicate the original CFMT using Chinese Malaysian
faces and has recently been validated (Kho et al., 2023).
SOCIAL NEUROSCIENCE 3

The CFMT consists of three stages: learn (faces were
presented with same light and viewpoint condition),
novel (faces were presented with dierent light and
viewpoint condition) and novel-with-noise (faces were
presented with dierent light and viewpoint condition
with Gaussian noise applied).
A total of six target faces were employed for the task.
Only male identities were used in the task. This is
because recognition performance between males and
females is comparable for male faces, whereas females
typically exhibit an advantage when recognizing female
faces (Duchaine & Nakayama, 2006; Lewin & Herlitz,
2002). Participants were given three practice trials before
the actual trials to familiarize them with the procedure.
In the learning stage (18 trials), three viewing angles of
the target face were presented (frontal view, left 1/3
prole and right 1/3 prole) for 3 s each (Figure 1a).
The target face was then presented with two distractor
faces, and participants were required to select the target
face shown (Figure 1b). The distractor faces were
selected based on their similarity in appearance to the
target faces. The keys used to indicate the target face
were “1” if the target face was the left image, “2” if it was
the image in the middle and “3” if it was the right image
with no time limit. In the novel stage (30 trials, Figure 1c)
and novel-with-noise stage (24 trials, Figure 1d), partici-
pants were rst instructed to memorize the same six
target faces presented in the learning stage in frontal
view for 20 s. Next, participants were required to select
the target face presented with two distractor faces with
no time limit during the test phase for the novel and the
novel-with-noise stages.
Transcranial direct current stimulation (tDCS)
The stimulation was delivered using Starstim 8 (Starstim,
Neuroelectrics, Barcelona, Spain). The electrodes were
inserted into a neoprene cap in accordance with the
international 10–10 EEG system. For the cathodal condi-
tion, 1.5 mA was applied to PO8 (cathode) and FP1
(anode) by using a pair of surface sponge electrodes
(25 cm
2
) soaked in saline solution (0.9% NaCl).
Conversely, 1.5 mA was applied to PO8 (anode) and
FP1 (cathode) for anodal condition. The current was
ramped up and down for the rst and last 30 seconds
for anodal and cathodal stimulation. In the sham condi-
tion, the stimulation was only delivered for the rst and
last 30 seconds to evoke the sensation of stimulation,
without aecting neuronal excitability (Thair et al., 2017).
The parameters of the stimulations were in accordance
with the standard safety constraints (i.e., maximum total
injected current: 4 mA; maximum current for each elec-
trode: 2 mA). All stimulation conditions lasted for 20
minutes. Participants were monitored for any signs of
distress at all times for safety purposes.
Three study images in learning stage
presented in different views
Test trials in learning stage (faces are
presented with same light and viewpoint
condition as in study image)
Test trials in novel stage (faces are presented
with different lighting and cropping template)
Test trials in novel-with-noise stage (faces are
presented with different light and viewpoint condition
with Gaussian noise applied)
Figure 1. Sample CFMT stimuli (images were not used in the actual task).
4S. K. KHO ET AL.

Procedure
The CFMT was presented using PsychoPy (Peirce et al.,
2019). Own and other-race versions of the CFMT were
counterbalanced across participants. Participants rst com-
pleted one own- and one other-race CFMT as baseline
tasks. The baseline tasks were included to ensure that
there was no dierence in individual face recognition abil-
ity between the stimulation groups prior to the stimulation.
Upon completing the baseline tasks, participants pro-
ceeded with the stimulation session. At the beginning of
the stimulation session, a suitable neoprene cap size was
selected based on the participant’s head circumference
measurement. Next, the location of stimulation was
cleaned with alcohol solution using a cotton swab. The
sponge electrodes were then tted onto the neoprene
cap, and the electrical reference ear clip was xed onto
the participant’s ear lobe. The impedance level was
checked prior to the stimulation and monitored
throughout the stimulation session. Participants
received either sham stimulation, a-tDCS or c-tDCS for
20 minutes. A cartoon video was presented during the
stimulation session to reduce inter-participant variability
in visual sensation during the session (e.g., Renzi et al.,
2015, for a similar procedure).
After the stimulation, participants completed the
alternate versions of the own- and other-race CFMT. At
the end of the experiment, participants were asked to
complete a questionnaire related to tDCS sensations to
check if there was any dierence between the sensation
perceived from a-tDCS, c-tDCS and sham stimulation.
The experimental session lasted for approximately one
and a half hours for each session.
Results
All data were analyzed using JASP version 0.16.3 (JASP
Team, 2022).
Perceived sensation
Kruskal–Wallis test was conducted on the rating score (0 =
none, 1 = mild, 2 = moderate and 3 = strong) of perceived
sensation (itching, pain, burning, warmth/heat and fatigue/
decreased alertness) from the stimulation (a-tDCS vs.
c-tDCS vs. sham stimulation). A dierence in rating score
of itching between stimulation type was found (H(2) =
13.918, p < .001). Post-hoc Dunn test showed that itching
sensation for a-tDCS (M = 1.633, SD = .890) was higher than
c-tDCS (M = 1.233, SD = .774), p = .046. Itching sensation for
a-tDCS was also higher than sham stimulation (M = .833, SD
= .592), p < .001. The post-hoc also showed that the itching
sensation for c-tDCS was higher than sham stimulation, p
= .046. No dierence was found for rating score of pain (H
(2) = 1.233, p = .540), burning (H(2) = 1.851, p = .396),
warmth/heat (H(2) = 4.791, p = .091) and fatigue/decreased
alertness (H(2) = 1.002, p = .606) between stimulation type.
Kruskal–Wallis test also revealed no dierence between
stimulation type on the rating score for change in general
state after stimulation (0 = not at all, 1 = slightly, 2 = con-
siderably, 3 = much and 4 = very much), H(2) = .744, p
= .689. For additional remarks on the sensation of stimula-
tion (Table A1) and participants’ beliefs about whether they
had received real or sham stimulation (Table A2), refer to
Appendix A.
Baseline (pre-stimulation)
Accuracy is reported in proportion correct. A one-way
ANOVA revealed no accuracy dierence for own-race
recognition between stimulation groups, F(2, 87) = 1.532,
p = .222, η
2
= .034. Bayesian analysis indicated that the null
hypothesis (absence of accuracy dierences between sti-
mulation groups for own-race recognition) was 2.999 times
more likely than the presence of accuracy dierences
between stimulation groups for own-race recognition
(BF
01
= 2.999, anecdotal evidence for null hypothesis). In
terms of other-race recognition accuracy, a one-way
ANOVA revealed a signicant eect of stimulation group,
F(2, 87) = 3.417, p = .037, η
2
= .073. Bayesian analysis
revealed that the presence of accuracy dierences
between stimulation groups for other-race recognition
was 1.455 times more favored than the null hypothesis
(absence of accuracy dierences between stimulation
groups for other-race recognition) (BF
10
= 1.455, anecdotal
evidence for alternative hypothesis). However, Holm’s post
hoc test reveals no dierence between a-tDCS and c-tDCS
(p = .915, BF
01
= 3.794, moderate evidence for null hypoth-
esis), a-tDCS and sham (p = .069, BF
10
= 2.22, anecdotal
evidence for alternative hypothesis), c-tDCS and sham (p
= .069, BF
10
= 2.399, anecdotal evidence for alternative
hypothesis). Altogether, the results showed no dierence
in recognition ability for own- and other-race faces
between stimulation groups prior to receiving stimulation.
Post-stimulation performance
1
A mixed 2 (CFMT type: own-race vs. other-race) × 3 (simula-
tion group: a-tDCS vs. c-tDCS vs. sham) ANOVA was con-
ducted to examine if there was any dierence in accuracy
between stimulation groups (Figure 2a). Accuracy reported
1
Results of analysis by stage (2 (CFMT type: own-race vs. other-race) × 3 (CFMT stage: learn vs. novel vs. novel-with-noise) × 3 (simulation group: a-tDCS vs.
c-tDCS vs. sham) ANOVA) conducted on accuracy and reaction time are included in Appendix B.
SOCIAL NEUROSCIENCE 5

is in proportion correct. Analysis revealed no main eect of
stimulation group on accuracy, F(2, 87) = 1.093, p = .34, η
p2
= .025. Bayesian analysis indicated that the null hypothesis
(absence of accuracy dierences between stimulation
groups) was 4.379 times more likely than the presence of
accuracy dierences between stimulation groups (BF
01
=
4.379, moderate evidence for null hypothesis). A main
eect of CFMT type was found, F(1, 87) = 160.809, p < .001,
η
p2
= .649, where own-race face recognition (M = .795, SD
= .132) had higher accuracy compared to other-race face
recognition (M = .660, SD = .121). Bayesian analysis
revealed that the presence of accuracy dierences in
CFMT type was 2.401e + 18 times more favored than the
null hypothesis (absence of accuracy dierences in CFMT
type) (BF
10
= 2.401e + 18, extreme evidence for alternative
hypothesis). No signicant interaction eect was found
between stimulation group and CFMT type on accuracy, F
(2, 87) = .861, p = .427, η
p2
= .019. In line with this, Bayesian
analysis showed that the null hypothesis (absence of inter-
action between stimulation group and CFMT type) was
4.238 times more favored than the interaction between
stimulation group and CFMT type (BF
01
= 4.238, moderate
evidence for null hypothesis).
A mixed 2 (CFMT type: own-race vs. other-race) × 3
(simulation group: a-tDCS vs. c-tDCS vs. sham) ANOVA
was also conducted on correct median reaction times
(Figure 2b). Analysis revealed no main eect of stimula-
tion group on reaction time, F(2, 87) = .892, p = .414, η
p2
= .02. Bayesian analysis indicated that the null hypoth-
esis (absence of reaction time dierences between sti-
mulation groups) was 2.532 times more likely than the
presence of reaction time dierences between stimula-
tion groups (BF
01
= 2.532, anecdotal evidence for null
hypothesis). A main eect of CFMT type was found, F(1,
87) = 32.247, p < .001, η
p2
= .27, where own-race face
recognition (M = 2.046 s, SD = .561 s) had shorter reac-
tion time compared to other-race face recognition (M =
2.269 s, SD = .630 s). Similarly, Bayesian analysis revealed
that the presence of reaction time dierences in CFMT
type was 78,786.398 times more favored than the null
hypothesis (absence of reaction time dierences in
CFMT type) (BF
10
= 78786.399, extreme evidence for
alternative hypothesis). No signicant interaction eect
was found between stimulation group and CFMT type
on reaction time, F(2, 87) = .265, p = .768, η
p2
= .006.
Bayesian analysis showed that the null hypothesis
(absence of interaction between stimulation group and
CFMT type) was 6.431 times more favored than the
interaction between stimulation group and CFMT type
(BF
01
= 6.431, moderate evidence for null hypothesis).
To explore the change in performance as
a consequence of stimulation type, we also calculated
the dierence in accuracy between post- and pre-
stimulation for each stimulation group and CFMT type
(ACCPost ACCPre). A higher value would indicate higher
improvement in accuracy after stimulation. We analyzed
these scores using a 2 (CFMT type: own-race vs. other-
race) × 3 (simulation group: a-tDCS vs. c-tDCS vs. sham)
ANOVA (Figure 3a). Analysis revealed no main eect of
stimulation group, F(2, 87) = .458, p = .634, η
p2
= .01, nor
a main eect of CFMT type, F(1, 87) = .063, p = .802, η
p2
= .0007, on accuracy improvement. In line with this,
Bayesian analysis showed that the null hypothesis was
8.696 times more favored than the presence of accuracy
improvement dierences between stimulation group
(BF
01
= 8.696, moderate evidence for null hypothesis)
Figure 2. Accuracy (proportion correct) and median reaction time for correct trials in the stimulation groups, separated by own- and
other-race CFMT versions. Post-stimulation measurements are represented by solid lines, while the pre-stimulation (baseline)
measurements are shown with dotted lines. Error bar represents 95% confidence interval.
6S. K. KHO ET AL.

and that the null hypothesis was 5.937 times more likely
that the presence of accuracy improvement dierences
between CFMT type (BF
01
= 5.937, moderate evidence
for null hypothesis). No signicant interaction eect was
found between stimulation group and CFMT type on
accuracy improvement, F(2, 87) = 2.688, p = .074, η
p2
= .058. Bayesian analysis revealed that the null hypoth-
esis (absence of interaction between stimulation group
and CFMT type) was 13.21 times more favored than the
interaction between stimulation group and CFMT type
(BF
01
= 13.21, strong evidence for null hypothesis).
We also calculated the dierence in correct median
reaction time between post- and pre-stimulation for
each stimulation group and CFMT type (RTPre RTPost).
A higher value would indicate higher improvement in
reaction times after stimulation. We analyzed these
scores using a 2 (CFMT type: own-race vs. other-race) ×
3 (simulation group: a-tDCS vs. c-tDCS vs. sham) ANOVA
(Figure 3b). Analysis revealed no main eect of stimula-
tion group, F(2, 87) = 1.782, p = .174, η
p2
= .039, nor
a main eect of CFMT type, F(1, 87) = .613, p = .436, η
p2
= .007, on reaction time improvement. Bayesian analysis
indicated that the null hypothesis was 3.409 times more
favored than the presence of reaction time improve-
ment dierences between stimulation group (BF
01
=
3.409, moderate evidence for null hypothesis) and that
the null hypothesis was 4.606 times more likely that the
presence of reaction time improvement dierences
between CFMT type (BF
01
= 4.606, moderate evidence
for null hypothesis). No signicant interaction eect was
found between stimulation group and CFMT type on
reaction time improvement, F(2, 87) = .114, p = .893,
η
p2
= .003. Similarly, Bayesian analysis revealed that the
null hypothesis (absence of interaction between
stimulation group and CFMT type) was 54.841 times
more favored than the interaction between stimulation
group and CFMT type (BF
01
= 54.841, very strong evi-
dence for null hypothesis).
Discussion
This study aimed to investigate the eect of anodal and
cathodal tDCS on the recognition of own- and other-race
faces. Based on the neurophysiological mechanism of
tDCS (Nitsche & Paulus, 2000), we expected to nd an
enhanced performance for own- and other-race face
recognition following a-tDCS and a reduced perfor-
mance for own- and other-race face recognition follow-
ing c-tDCS. Our ndings demonstrated that participants’
post-stimulation performance was similar across all sti-
mulation conditions (i.e., a-tDCS, c-tDCS and sham sti-
mulation). In addition, there were no dierences in the
performance change (calculated using baseline and
post-stimulation scores) between the dierent stimula-
tion conditions. Thus, overall, our results showed no
dierence in accuracy and reaction time for own- and
other-race face recognition after either a-tDCS, c-tDCS or
sham stimulation.
Contrary to our expectation, a-tDCS did not improve
own- or other-race face recognition. Our ndings are in
line with past work which have reported null eects of
a-tDCS on the occipital region involved in face proces-
sing (Willis et al., 2019). Interestingly, although the same
stimulation protocol (i.e., 20 min of oine 1.5 mA of
tDCS to the occipital region delivered using a 25 cm
2
sponge electrode) and face recognition measure (i.e.,
CFMT) were used in the current experiment and
Barbieri et al. (2016), we failed to replicate the face
Figure 3. Change in accuracy (post- minus pre-stimulation) and median reaction time for correct trials(pre- minus post-stimulation)
after stimulation, separated by stimulation group and own- and other-race CFMT versions. Error bar represents 95% confidence
interval.
SOCIAL NEUROSCIENCE 7

memory improvement eect found in their experiment.
We also found no impairment of own- or other-race face
recognition after c-tDCS. This is in line with previous
work suggesting that the eects from cathodal stimula-
tion are relatively inconsistent (Jacobson et al., 2012).
However, this contradicted ndings by Costantino et al.
(2017) where they suggested that c-tDCS impaired the
recognition of other-race faces. In this study, we used
the same stimulation protocol and face recognition
measure as Costantino et al. (2017). In addition, we also
used more comparable measures across baseline and
post-stimulation tasks (i.e., dierent versions of the
CFMT). However, we failed to replicate the impairment
of other-race recognition reported by Costantino et al.
(2017). Our ndings are in line with past studies showing
that cathodal stimulation does not always lead to
a decrease in neuronal excitability and performance
(Horvath et al., 2015; Wietho et al., 2014) and that its
eects on cognition can be inconsistent (Jacobson et al.,
2012).
Overall, our ndings support past research showing
that the eect of a-tDCS and c-tDCS may not always be
reliable (López-Alonso et al., 2014; Strube et al., 2015;
Wietho et al., 2014). For example, it has been reported
that more than half of the participants (55%) did not
show the expected excitatory eect on neuronal excit-
ability after a-tDCS, whereas the remaining 45% showed
the expected excitatory eect (López-Alonso et al.,
2014). In line with this, a dierent study reported that
50% of the participants showed little or no response to
tDCS, whereas the remaining participants responded
similarly to both c-tDCS and a-tDCS (Wietho et al.,
2014). Thus, it could be that the participants in our
experiment were less responsive to tDCS leading to the
null eects of both a-tDCS and c-tDCS on the face
recognition tasks.
In fact, the inter-individual dierences in the tDCS
eects are a known limitation of tDCS studies. The
lack of stimulation eect could be attributed to dif-
ferences in the biological substrate such as the pre-
existing neurotransmitter levels and dierences in
head size and scalp thickness (Krause & Cohen
Kadosh, 2014; Laakso et al., 2019). Therefore, some
participants might have received more or less stimu-
lation eect than others, leading to variability in the
eectiveness of tDCS. This issue, however, could not
be resolved by implementing a within-subjects
design as past work has also shown intra-individual
dierences in the eect of tDCS where the eect of
tDCS varies across dierent test sessions (Dyke et al.,
2016). Hence, intra- and inter-individual dierences
may have contributed to the inconsistent ndings
of tDCS studies.
In addition, the lack of stimulation eect observed
in our study may be due to the low focality of
stimulation to the target area, which is a common
limitation of tDCS studies that use a traditional two-
electrode montage. Research by Barbieri et al. (2016)
found that this type of stimulation did not produce
face-specic eects, as it improved both face and
object memory. In contrast, research targeting the
FFA with high-focality stimulation have shown selec-
tive enhancement of face memory (Brunyé et al.,
2017). The low focality stimulation used in our
study may have resulted in current spreading to non-
target regions, leading to noise in the data. To
address this limitation, future studies could use high
focality stimulation techniques such as high-
denition tDCS (Datta et al., 2009; Kuo et al., 2013;
Villamar et al., 2013) or multifocal tDCS (Fischer et al.,
2017), which rely on smaller electrodes to increase
focality and reduce current spread to non-target
regions.
To conclude, we found no eect of a-tDCS and
c-tDCS in the recognition of own- and other-race
faces. Our ndings showed that the eects of anodal
and cathodal tDCS may not always be reliable and
support the inconsistency of tDCS eects in face pro-
cessing (Willis et al., 2019). This is consistent with the
increasing number of studies that have failed to repli-
cate the positive eects of transcranial electrical stimu-
lation on mood and emotion (Koenigs et al., 2009),
working memory (Nilsson et al., 2015, 2017;
Westwood & Romani, 2018), verbal uency
(Vannorsdall et al., 2016; Westwood & Romani, 2018),
reading (Cummine et al., 2020), sustained attention
(Jacoby & Lavidor, 2018) and spatial attention
(Learmonth et al., 2017).
Disclosure statement
No potential conict of interest was reported by the author(s).
Funding
This work was supported by the Fundamental Research Grant
Scheme (FRGS) from the Ministry of Education (MOE) Malaysia
(Grant number: FRGS/1/2018/SS05/UNIM/02/4) and the Society
for Applied Research in Memory & Cognition (SARMAC)
Student Caucus Research Grant.
ORCID
Siew Kei Kho http://orcid.org/0000-0001-5675-9655
8S. K. KHO ET AL.

Authors’ contributions
The authors conrm contribution to the paper as follows:
study conception and design: Siew Kei Kho and Alejandro
J. Estudillo; data collection: Siew Kei Kho; analysis and inter-
pretation of results: Siew Kei Kho and Alejandro J. Estudillo;
draft manuscript preparation: Siew Kei Kho, David
R. T. Keeble, Hoo Keat Wong and Alejandro J. Estudillo. All
authors reviewed the results and approved the nal version
of the manuscript.
Ethics approval
This study was performed in line with the principles of the
Declaration of Helsinki. Approval was granted by the Science
and Engineering Research Ethics Committee (SEREC) at the
University of Nottingham Malaysia (approval code: KSK050320).
Availability of data and materials
The data that support the ndings of this study are openly
available in Open Science Framework (OSF) at https://osf.io/
8t4zr/?view_only=654e85d7b3db43f38bf97385bea11a8e.
Open Practices Statements
The data for the experiment are available in Open Science
Framework (OSF) at https://osf.io/8t4zr/?view_only=
654e85d7b3db43f38bf97385bea11a8e and the experiment
was preregistered via the OSF before data collection (https://
osf.io/6cf7w).
References
Anzures, G., & Mildort, M. (2021). Do perceptual expertise and
implicit racial bias predict early face-sensitive ERP
responses? Brain & Cognition, 147(December 2020),
105671. https://doi.org/10.1016/j.bandc.2020.105671
Barbieri, M., Negrini, M., Nitsche, M. A., & Rivolta, D. (2016).
Anodal-tDCS over the human right occipital cortex
enhances the perception and memory of both faces and
objects. Neuropsychologia, 81, 238–244. https://doi.org/10.
1016/j.neuropsychologia.2015.12.030
Barton, J. J. S. (2008). Structure and function in acquired pro-
sopagnosia: Lessons from a series of 10 patients with brain
damage. Journal of Neuropsychology, 2(1), 197–225. https://
doi.org/10.1348/174866407X214172
Bate, S., Bennetts, R. J., Gregory, N., Tree, J. J., Murray, E.,
Adams, A., Bobak, A. K., Penton, T., Yang, T., & Banissy, M. J.
(2019). Objective patterns of face recognition decits in 165
adults with self-reported developmental prosopagnosia.
Brain Sciences, 9(6), 133. https://doi.org/10.3390/
brainsci9060133
Bate, S., & Tree, J. J. (2017). The denition and diagnosis of
developmental prosopagnosia. Quarterly Journal of
Experimental Psychology, 70(2), 193–200. https://doi.org/10.
1080/17470218.2016.1195414
Behrmann, M., Avidan, G., Marotta, J. J., & Kimchi, R. (2005).
Detailed exploration of face-related processing in
congenital prosopagnosia: 1. Behavioral ndings. Journal of
Cognitive Neuroscience, 17(7), 1130–1149. https://doi.org/10.
1162/0898929054475154
Bikson, M., Inoue, M., Akiyama, H., Deans, J. K., Fox, J. E.,
Miyakawa, H., & Jeerys, J. G. R. (2004). Eect of uniform
extracellular DC electric elds on excitability in rat hippo-
campal slices in vitro. Journal of Physiology, 557(1), 175–190.
https://doi.org/10.1113/jphysiol.2003.055772
Bortolon, C., Capdevielle, D., & Raard, S. (2015). Face recogni-
tion in schizophrenia disorder: A comprehensive review of
behavioral, neuroimaging and neurophysiological studies.
Neuroscience and Biobehavioral Reviews, 53, 79–107.
https://doi.org/10.1016/j.neubiorev.2015.03.006
Bruce, V., Henderson, Z., Greenwood, K., Hancock, P. J. B.,
Burton, A. M., & Miller, P. (1999). Verication of face identities
from images captured on video. Journal of Experimental
Psychology: Applied, 5(4), 339–360. https://doi.org/10.1037/
1076-898X.5.4.339
Brunyé, T. T., Moran, J. M., Holmes, A., Mahoney, C. R., &
Taylor, H. A. (2017). Non-invasive brain stimulation targeting
the right fusiform gyrus selectively increases working mem-
ory for faces. Brain and Cognition, 113, 32–39. https://doi.
org/10.1016/j.bandc.2017.01.006
Busigny, T., & Rossion, B. (2010). Acquired prosopagnosia
abolishes the face inversion eect. Cortex, 46(8), 965–981.
https://doi.org/10.1016/j.cortex.2009.07.004
Cassidy, K. D., Boutsen, L., Humphreys, G. W., & Quinn, K. A.
(2014). Ingroup categorization aects the structural encod-
ing of other-race faces: Evidence from the N170
event-related potential. Social Neuroscience, 9(3), 235–248.
https://doi.org/10.1080/17470919.2014.884981
Costantino, A. I., Titoni, M., Bossi, F., Premoli, I., Nitsche, M. A., &
Rivolta, D. (2017). Preliminary evidence of “other-race
eect”-like Behavior induced by cathodal-tDCS over the
right occipital cortex. The Absence of Overall Eects on
Face/Object Processing Frontiers in Neuroscience, 11, 661.
https://doi.org/10.3389/fnins.2017.00661
Cummine, J., Villarena, M., Onysyk, T., & Devlin, J. T. (2020).
A study of null eects for the use of transcranial direct
Current stimulation (tDCS) in adults with and without read-
ing impairment. Neurobiology of Language, 1(4), 434–451.
https://doi.org/10.1162/nol_a_00020
Dalrymple, K. A., Fletcher, K., Corrow, S., Barton, J. J. S.,
Yonas, A., & Duchaine, B. (2014). “A room full of strangers
every day”: The psychosocial impact of developmental pro-
sopagnosia on children and their families. Journal of
Psychosomatic Research, 77(2), 144–150. https://doi.org/10.
1016/j.jpsychores.2014.06.001
Dalrymple, K. A., Garrido, L., & Duchaine, B. (2014). Dissociation
between face perception and face memory in adults, but not
children, with developmental prosopagnosia.
Developmental Cognitive Neuroscience, 10, 10–20. https://
doi.org/10.1016/j.dcn.2014.07.003
Datta, A., Bansal, V., Diaz, J., Patel, J., Reato, D., & Bikson, M.
(2009). Gyri-precise head model of transcranial direct cur-
rent stimulation: Improved spatial focality using a ring elec-
trode versus conventional rectangular pad. Brain
Stimulation, 2(4), 201–207. https://doi.org/10.1016/j.brs.
2009.03.005
Davis, J. P., & Valentine, T. (2009). CCTV on trial: Matching video
images with the defendant in the dock. Applied Cognitive
Psychology: The Ocial Journal of the Society for Applied
SOCIAL NEUROSCIENCE 9

Research in Memory & Cognition, 23(4), 482–505. https://doi.
org/10.1002/acp.1490
Duchaine, B., Germine, L., & Nakayama, K. (2007). Family resem-
blance: Ten family members with prosopagnosia and
within-class object agnosia. Cognitive Neuropsychology, 24
(4), 419–430. https://doi.org/10.1080/02643290701380491
Duchaine, B., & Nakayama, K. (2006). The Cambridge face
memory test: Results for neurologically intact individuals
and an investigation of its validity using inverted face stimuli
and prosopagnosic participants. Neuropsychologia, 44(4),
576–585. https://doi.org/10.1016/j.neuropsychologia.2005.
07.001
Duchaine, B., Yovel, G., Butterworth, E. J., & Nakayama, K.
(2006). Prosopagnosia as an impairment to face-specic
mechanisms: Elimination of the alternative hypotheses in
a developmental case. Cognitive Neuropsychology, 23(5),
714–747. https://doi.org/10.1080/02643290500441296
Dyke, K., Kim, S., Jackson, G. M., & Jackson, S. R. (2016). Intra-
subject consistency and reliability of response following 2
mA Transcranial direct Current stimulation. Brain
Stimulation, 9(6), 819–825. https://doi.org/10.1016/j.brs.
2016.06.052
Eimer, M. (2000). Eects of face inversion on the structural
encoding and recognition of faces. Cognitive Brain
Research, 10(1–2), 145–158. https://doi.org/10.1016/s0926-
6410(00)00038-0
Estudillo, A. J., & Bindemann, M. (2014). Generalization across
view in face memory and face matching. I-Perception, 5(7),
589–601. https://doi.org/10.1068/i0669
Estudillo, A. J., Lee, Y. J., Álvarez-Montesinos, J. A., &
García-Orza, J. (2023). High-frequency transcranial random
noise stimulation enhances unfamiliar face matching of high
resolution and pixelated faces. Brain and Cognition, 165,
105937. https://doi.org/10.1016/j.bandc.2022.105937
Estudillo, A. J., Lee, J. K. W., Mennie, N., & Burns, E. (2020). No
evidence of other‐race eect for Chinese faces in Malaysian
non‐Chinese population. Applied Cognitive Psychology, 34(1),
270–276. https://doi.org/10.1002/acp.3609
Faul, F., Erdfelder, E., Buchner, A., & Lang, A. G. (2009). Statistical
power analyses using G*Power 3.1: Tests for correlation and
regression analyses. Behavior Research Methods, 41(4),
1149–1160. https://doi.org/10.3758/BRM.41.4.1149
Feng, L., Liu, J., Wang, Z., Li, J., Li, L., Ge, L., Tian, J., & Lee, K.
(2011). The other face of the other-race eect: An fMRI
investigation of the other-race face categorization
advantage. Neuropsychologia, 49(13), 3739–3749. https://
doi.org/10.1016/j.neuropsychologia.2011.09.031
Fischer, D. B., Fried, P. J., Runi, G., Ripolles, O., Salvador, R.,
Banus, J., Ketchabaw, W. T., Santarnecchi, E., Pascual-Leone,
A., & Fox, M. D. (2017). Multifocal tDCS targeting the resting
state motor network increases cortical excitability beyond
traditional tDCS targeting unilateral motor cortex.
NeuroImage, 157, 34–44. https://doi.org/10.1016/j.neuro
image.2017.05.060
Friehs, M. A., & Frings, C. (2019). Oine beats online.
NeuroReport, 30(12), 795–799. https://doi.org/10.1097/WNR.
0000000000001272
Giménez-Fernández, T., Kessel, D., Fernández-Folgueiras, U.,
Fondevila, S., Méndez-Bértolo, C., Aceves, N.,
García-Rubio, M. J., & Carretié, L. (2020). Prejudice drives
exogenous attention to outgroups. Social Cognitive and
Aective Neuroscience, 15(6), 615–624. https://doi.org/10.
1093/scan/nsaa087
Goaux, V., Gauthier, I., & Rossion, B. (2003). Spatial scale con-
tribution to early visual dierences between face and object
processing. Cognitive Brain Research, 16(3), 416–424. https://
doi.org/10.1016/S0926-6410(03)00056-9
Golby, A. J., Gabrieli, J. D. E., Chiao, J. Y., & Eberhardt, J. L. (2001).
Dierential responses in the fusiform region to same-race
and other-race faces. Nature Neuroscience, 4(8), 845–850.
https://doi.org/10.1038/90565
Hill, A. T., Fitzgerald, P. B., & Hoy, K. E. (2016). Eects of anodal
transcranial direct Current stimulation on working memory:
A systematic review and meta-analysis of ndings from
healthy and neuropsychiatric populations. Brain
Stimulation, 9(2), 197–208. https://doi.org/10.1016/j.brs.
2015.10.006
Horvath, J. C., Vogrin, S. J., Carter, O., Cook, M. J., & Forte, J. D.
(2015). Eects of transcranial direct current stimulation on
motor evoked potential amplitude are neither reliable nor
signicant within individuals over 9 separate testing
sessions. Brain Stimulation, 8(2), 318–325. https://doi.org/
10.1016/j.brs.2015.01.033
Jack, R. E., & Schyns, P. G. (2015). The human face as a dynamic
tool for Social communication. Current Biology, 25(14), R621–
R634. https://doi.org/10.1016/j.cub.2015.05.052
Jacobson, L., Koslowsky, M., & Lavidor, M. (2012). TDCS polarity
eects in motor and cognitive domains: A meta-analytical
review. Experimental Brain Research, 216(1), 1–10. https://doi.
org/10.1007/s00221-011-2891-9
Jacoby, N., & Lavidor, M. (2018). Null tDCS Eects in a sustained
attention task: The modulating role of learning. Frontiers in
Psychology, 9, 9. https://doi.org/10.3389/fpsyg.2018.00476
Jacques, C., & Rossion, B. (2010). Misaligning face halves
increases and delays the N170 specically for upright faces:
Implications for the nature of early face representations.
Brain Research, 1318, 96–109. https://doi.org/10.1016/j.
brainres.2009.12.070
JASP Team. (2022). JASP (Version 0.16.3)[Computer Software].
https://jasp-stats.org/
Kemp, R., Towell, N., & Pike, G. (1997). When seeing should not
be believing: Photographs, Credit Cards and Fraud. Applied
Cognitive Psychology, 11(3), 211–222. https://doi.org/10.
1002/(SICI)1099-0720(199706)11:3<211:AID-ACP430>3.0.
CO;2-O
Kho, S. K., Leong, B. Q. Z., Keeble, D. R. T., Wong, H. K., &
Estudillo, A. J. (2023). A new Asian version of the CFMT: The
Cambridge face memory test – Chinese Malaysian. CFMT-MY).
Behavior Research Methods. https://doi.org/10.3758/
s13428-023-02085-6
Kim, J. S., Yoon, H. W., Kim, B. S., Jeun, S. S., Jung, S. L., &
Choe, B. Y. (2006). Racial distinction of the unknown facial
identity recognition mechanism by event-related fMRI.
Neuroscience Letters, 397(3), 279–284. https://doi.org/10.
1016/j.neulet.2005.12.061
Koenigs, M., Ukueberuwa, D., Campion, P., Grafman, J., &
Wassermann, E. (2009). Bilateral frontal transcranial direct
current stimulation: Failure to replicate classic ndings in
healthy subjects. Clinical Neurophysiology, 120(1), 80–84.
https://doi.org/10.1016/j.clinph.2008.10.010
Krause, B., & Cohen Kadosh, R. (2014). Not all brains are created
equal: The relevance of individual dierences in responsive-
ness to transcranial electrical stimulation. Frontiers in
10 S. K. KHO ET AL.

Systems Neuroscience, 8, 25. https://doi.org/10.3389/fnsys.
2014.00025
Kuo, H.-I., Bikson, M., Datta, A., Minhas, P., Paulus, W., Kuo, M., &
Nitsche, M. A. (2013). Comparing cortical plasticity induced
by conventional and high-denition 4 × 1 ring tDCS:
A neurophysiological study. Brain Stimulation, 6(4),
644–648. https://doi.org/10.1016/j.brs.2012.09.010
Laakso, I., Mikkonen, M., Koyama, S., Hirata, A., & Tanaka, S.
(2019). Can electric elds explain inter-individual variability
in transcranial direct current stimulation of the motor
cortex? Scientic Reports, 9(1), 626. https://doi.org/10.1038/
s41598-018-37226-x
Learmonth, G., Felisatti, F., Siriwardena, N., Checketts, M.,
Benwell, C. S. Y., Märker, G., Thut, G., & Harvey, M. (2017).
No interaction between tDCS Current strength and baseline
performance: A conceptual replication. Frontiers in
Neuroscience, 11, 11. https://doi.org/10.3389/fnins.2017.
00664
Lewin, C., & Herlitz, A. (2002). Sex dierences in face recogni-
tion - Women’s faces make the dierence. Brain and
Cognition, 50(1), 121–128. https://doi.org/10.1016/S0278-
2626(02)00016-7
López-Alonso, V., Cheeran, B., Río-Rodríguez, D., & Fernández-
Del-Olmo, M. (2014). Inter-individual variability in response
to non-invasive brain stimulation paradigms. Brain
Stimulation, 7(3), 372–380. https://doi.org/10.1016/j.brs.
2014.02.004
Malpass, R. S., & Kravitz, J. (1969). Recognition for faces of own
and other race. Journal of Personality and Social Psychology,
13(4), 330–334. https://doi.org/10.1037/h0028434
Marwick, K., & Hall, J. (2008). Social cognition in schizophrenia:
A review of face processing. British Medical Bulletin, 88(1),
43–58. https://doi.org/10.1093/bmb/ldn035
McKone, E., Hall, A., Pidcock, M., Palermo, R., Wilkinson, R. B.,
Rivolta, D., Yovel, G., Davis, J. M., & O’Connor, K. B. (2011).
Face ethnicity and measurement reliability aect face recog-
nition performance in developmental prosopagnosia:
Evidence from the Cambridge face memory test–
Australian. Cognitive Neuropsychology, 28(2), 109–146.
https://doi.org/10.1080/02643294.2011.616880
McKone, E., Wan, L., Robbins, R., Crookes, K., & Liu, J. (2017).
Diagnosing prosopagnosia in East Asian individuals: Norms
for the Cambridge face memory test–Chinese. Cognitive
Neuropsychology, 34(5), 253–268. https://doi.org/10.1080/
02643294.2017.1371682
Meissner, C. A., & Brigham, J. C. (2001). Thirty years of investi-
gating the own-race Bias in memory for faces: A
meta-Analytic review. Psychology, Public Policy, & Law, 7(1),
3–35. https://doi.org/10.1037/1076-8971.7.1.3
Natu, V., Raboy, D., & O’Toole, A. J. (2011). Neural correlates of
own- and other-race face perception: Spatial and temporal
response dierences. NeuroImage, 54(3), 2547–2555. https://
doi.org/10.1016/j.neuroimage.2010.10.006
Nilsson, J., Lebedev, A. V., & Lövdén, M. (2015). No signicant
eect of prefrontal tDCS on working memory performance
in older adults. Frontiers in Aging Neuroscience, 7, 7. https://
doi.org/10.3389/fnagi.2015.00230
Nilsson, J., Lebedev, A. V., Rydström, A., & Lövdén, M. (2017).
Direct-current stimulation does little to improve the out-
come of working memory training in older adults.
Psychological Science, 28(7), 907–920. https://doi.org/10.
1177/0956797617698139
Nitsche, M. A., & Paulus, W. (2000). Excitability changes induced
in the human motor cortex by weak transcranial direct
current stimulation. Journal of Physiology, 527(3), 633–639.
https://doi.org/10.1111/j.1469-7793.2000.t01-1-00633.x
Parketny, J., Towler, J., & Eimer, M. (2015). The activation of
visual face memory and explicit face recognition are delayed
in developmental prosopagnosia. Neuropsychologia, 75,
538–547. https://doi.org/10.1016/j.neuropsychologia.2015.
07.009
Peirce, J., Gray, J. R., Simpson, S., MacAskill, M.,
Höchenberger, R., Sogo, H., Kastman, E., & Lindeløv, J. K.
(2019). PsychoPy2: Experiments in behavior made easy.
Behavior Research Methods, 51(1), 195–203. https://doi.org/
10.3758/s13428-018-01193-y
Reed, T., & Cohen Kadosh, R. (2018). Transcranial electrical
stimulation (tES) mechanisms and its eects on cortical
excitability and connectivity. Journal of Inherited Metabolic
Disease, 41(6), 1123–1130. https://doi.org/10.1007/s10545-
018-0181-4
Renzi, C., Ferrari, C., Schiavi, S., Pisoni, A., Papagno, C., Vecchi, T.,
Antal, A., & Cattaneo, Z. (2015). The role of the occipital face
area in holistic processing involved in face detection and
discrimination: A tDCS study. Neuropsychology, 29(3),
409–416. https://doi.org/10.1037/neu0000127
Rossion, B. (2014). Understanding face perception by means of
prosopagnosia and neuroimaging. Frontiers in Bioscience, 6
(2), 706. https://doi.org/10.2741/e706
Rossion, B., Gauthier, I., Tarr, M. J., Pierenne, D., Debatisse, D., &
Despland, P. A. (1999). The N170 occipito-temporal compo-
nent is delayed to inverted faces but not to inverted objects:
Electrophysiological evidence of face-specic processes in
the human brain. NeuroImage, 9(6 PART II), 69–74. https://
doi.org/10.1097/00001756-200001170-00014
Senholzi, K. B., & Ito, T. A. (2013). Structural face encoding: How
task aects the N170’s sensitivity to race. Social Cognitive
and Aective Neuroscience, 8(8), 937–942. https://doi.org/10.
1093/scan/nss091
Serani, L., & Pesciarelli, F. (2022, October). Neural timing of the
other-race eect across the lifespan: A review.
Psychophysiology, 2022(4), 1–46. https://doi.org/10.1111/
psyp.14203
Smith, M. J., Adams, L. F., Schmidt, P. J., Rubinow, D. R., &
Wassermann, E. M. (2002). Eects of ovarian hormones on
human cortical excitability. Annals of Neurology, 51(5),
599–603. https://doi.org/10.1002/ana.10180
Strube, W., Bunse, T., Malchow, B., & Hasan, A. (2015). Ecacy
and interindividual variability in motor-cortex plasticity fol-
lowing anodal tDCS and paired-associative stimulation.
Neural Plasticity, 2015, 530423. https://doi.org/10.1155/
2015/530423
Thair, H., Holloway, A. L., Newport, R., & Smith, A. D. (2017).
Transcranial direct Current stimulation (tDCS): A Beginner’s
Guide for design and implementation. Frontiers in
Neuroscience, 11, 641. https://doi.org/10.3389/fnins.2017.00641
Vannorsdall, T. D., Van Steenburgh, J. J., Schretlen, D. J.,
Jayatillake, R., Skolasky, R. L., & Gordon, B. (2016).
Reproducibility of tDCS results in a randomized trial:
Failure to replicate ndings of tDCS-induced enhancement
of verbal uency. Cognitive and Behavioral Neurology, 29(1),
11–17. https://doi.org/10.1097/WNN.0000000000000086
Verhallen, R. J., Bosten, J. M., Goodbourn, P. T., Lawrance-
Owen, A. J., Bargary, G., & Mollon, J. D. (2017). General and
SOCIAL NEUROSCIENCE 11

specic factors in the processing of faces. Vision Research,
141, 217–227. https://doi.org/10.1016/j.visres.2016.12.014
Villamar, M. F., Volz, M. S., Bikson, M., Datta, A., DaSilva, A. F., &
Fregni, F. (2013). Technique and considerations in the use of
4x1 ring high-denition transcranial direct Current stimula-
tion (HD-tDCS). Journal of Visualized Experiments, 77(77),
e50309. https://doi.org/10.3791/50309
Weigelt, S., Koldewyn, K., Dilks, D. D., Balas, B., Mckone, E., &
Kanwisher, N. (2014). Domain-specic development of face
memory but not face perception. Developmental Science, 17
(1), 47–58. https://doi.org/10.1111/desc.12089
Weigelt, S., Koldewyn, K., & Kanwisher, N. (2012). Face identity
recognition in autism spectrum disorders: A review of beha-
vioral studies. Neuroscience and Biobehavioral Reviews, 36(3),
1060–1084. https://doi.org/10.1016/j.neubiorev.2011.12.008
Westwood, S. J., & Romani, C. (2018). Null eects on working
memory and verbal uency tasks when applying anodal
tDCS to the inferior frontal gyrus of healthy participants.
Frontiers in Neuroscience, 12, 12. https://doi.org/10.3389/
fnins.2018.00166
White, D., Kemp, R. I., Jenkins, R., Matheson, M., Burton, A. M., &
Guo, K. (2014). Passport ocers’ errors in face matching. PLoS
ONE, 9(8), e103510. https://doi.org/10.1371/journal.pone.
0103510
Wiese, H. (2013). Do neural correlates of face expertise vary with
task demands? event-related potential correlates of own- and
other-race face inversion. Frontiers in Human Neuroscience, 7
(DEC). https://doi.org/10.3389/fnhum.2013.00898
Wietho, S., Hamada, M., & Rothwell, J. C. (2014). Variability in
response to transcranial direct current stimulation of the
motor cortex. Brain Stimulation, 7(3), 468–475. https://doi.
org/10.1016/j.brs.2014.02.003
Willis, M. L., Costantino, A. I., Nitsche, M. A., Palermo, R., &
Rivolta, D. (2019). Anodal tDCS and high-frequency tRNS
Targeting the occipitotemporal cortex Do not always
enhance face perception. Frontiers in Neuroscience, 13, 78.
https://doi.org/10.3389/fnins.2019.00078
Wong, H. K., Estudillo, A. J., Stephen, I. D., & Keeble, D. R. T.
(2021). The other-race eect and holistic processing across
racial groups. Scientic Reports, 11(1), 8507. https://doi.org/
10.1038/s41598-021-87933-1
Yamada, Y., & Sumiyoshi, T. (2021). Neurobiological mechanisms
of transcranial direct Current stimulation for psychiatric dis-
orders; neurophysiological, Chemical, and anatomical
considerations. Frontiers in Human Neuroscience, 15, 631838.
https://doi.org/10.3389/fnhum.2021.631838
Yang, L. Z., Zhang, W., Shi, B., Yang, Z., Wei, Z., Gu, F., Zhang, J.,
Cui, G., Liu, Y., Zhou, Y., Zhang, X., Rao, H., & Hsiao, J. (2014).
Electrical stimulation over bilateral occipito-temporal
regions reduces N170 in the right hemisphere and the com-
posite face eect. PLoS ONE, 9(12), e115772. https://doi.org/
10.1371/journal.pone.0115772
Yao, Q., & Zhao, L. (2019). Using spatial frequency scales for
processing own-race and other-race faces: An ERP analysis.
Neuroscience Letters, 705(January), 167–171. https://doi.org/
10.1016/j.neulet.2019.04.059
Yardley, L., McDermott, L., Pisarski, S., Duchaine, B., &
Nakayama, K. (2008). Psychosocial consequences of devel-
opmental prosopagnosia: A problem of recognition. Journal
of Psychosomatic Research, 65(5), 445–451. https://doi.org/
10.1016/j.jpsychores.2008.03.013
Young, A. W., & Burton, A. M. (2018). Are we face experts?
Trends in Cognitive Sciences, 22(2), 100–110. https://doi.org/
10.1016/j.tics.2017.11.007
12 S. K. KHO ET AL.

Appendix A
Appendix B
Mixed ANOVA was conducted to examine the dierence in accuracy and median reaction time for correct trials for own- and other-
race CFMT task among the stimulation groups (a-tDCS vs. c-tDCS vs. sham stimulation). When Mauchly’s test indicated that the
assumption of sphericity had been violated, the degrees of freedom were corrected using Greenhouse–Geisser estimates of
sphericity.
Accuracy
A mixed 2 (CFMT type: own-race vs. other-race) × 3 (CFMT stage: learn vs. novel vs. novel-with-noise) × 3 (simulation group:
a-tDCS vs. c-tDCS vs. sham) ANOVA was conducted to examine if there was any dierence in accuracy between stimulation
groups. Accuracy reported is in proportion correct. Analysis revealed no main eect of stimulation group on accuracy,
F(2, 87) = 1.152, p = .321, η
p2
= .026. A main eect of CFMT type was found, F(1, 87) = 171.982, p < .001, η
p2
= .664, where
own-race face recognition (M = .795, SD = .132) had higher accuracy compared to other-race face recognition (M = .660, SD
= .121). No signicant interaction eect was found between stimulation group and CFMT type on accuracy, F(2, 87) = .737, p
= .481,
η
p2
= .017.
A main eect of CFMT stage was found, F(1.562, 135.921) = 389.430, p < .001, η
p2
= .817. Post-hoc Holm – Bonferroni
test revealed that the learning stage (M = .958, SD = .068) had higher accuracy compared to the novel stage (M = .685, SD
= .178), p < .001, d = 1.894. The learning stage also had higher accuracy compared to the novel-with-noise stage (M = .608,
SD = .210), p < .001, d = 2.430. Accuracy for novel stage was higher compared to novel-with-noise stage, p < .001, d = .536.
No signicant interaction eect was found between stimulation group and CFMT stage on accuracy, F(4, 174) = .846, p
= .498, η
p2
= .019.
A signicant interaction eect was found between CFMT type and CFMT stage on accuracy, F(2, 174) = 94.308, p < .001,
η
p2
= .520. Simple main eect analysis revealed that scores of the own-race CFMT was higher than other-race CFMT in the
learning stage (own-race: M = .967, SD = .063, other-race: M = .949, SD = .071), F(1, 89) = 4.961, p = .028, η
2
= .053, novel
stage (own-race: M = .743, SD = .181, other-race: M = .627, SD = .156), F(1, 89) = 51.356, p < .001, η
2
= .366 and novel-with-
noise stage (own-race: M = .731, SD = .174, other-race: M = .486, SD = .167), F(1, 89) = 262.331, p < .001, η
2
= .747. No
Table A1. Additional remarks on the sensation of stimulation.
Stimulation Additional remarks
a-tDCS Random electrical pinch on wrist
Very slight itch experienced
Fatigue (3 participants)
I felt a little sleepy half way of watching the video
Slightly more alert
I had a slight decrease in terms of awareness. Mild dizziness starting only toward the end of the video.
c-tDCS Less focus and face recognition skills reduced
Fatigue
Feeling a little bit sleepy, but otherwise no difference to usual tiredness before bed time
Felt fatigue in the middle of the experiment
Pin prickling sensation
Sham stimulation Dizzy
Feeling a bit tired and loss of focus after the stimulation
Not much sensation from the device but feeling a bit dizzy at the initial moment
I felt the itching at the beginning and the end, not in the middle
Note. Remarks provided by 17 participants.
Table A2. Participants’ beliefs about whether they had received real or placebo
stimulation.
Number of participants
Stimulation Real Placebo Not sure
a-tDCS 21 3 6
c-tDCS 17 1 12
Sham stimulation 15 4 11
Note. Each stimulation group had 30 participants.
SOCIAL NEUROSCIENCE 13

signicant interaction eect was found between stimulation group, CFMT stage and CFMT type on accuracy, F(4, 174) =
1.431, p = .226, η
p2
= .032.
Reaction time
A second mixed ANOVA was conducted to examine if there were any dierence in median reaction time for correct trials
between stimulation group. Analysis revealed no main eect of stimulation group on reaction time, F(2, 87) = .953, p = .390,
η
p2
= .021. A main eect of CFMT type was found, F(1, 87) = 47.650, p < .001, η
p2
= .354, where own-race face recognition (M
= 2.046 s, SD = .561 s) had shorter reaction time compared to other-race face recognition (M = 2.269 s, SD = .630 s). No
signicant interaction eect was found between stimulation group and CFMT type on reaction time, F(2, 87) = .036, p = .964,
η
p2
= .001.
A main eect of CFMT stage was found, F(1.659, 144.348) = 98.913, p < .001, η
p2
= .532. Post-hoc test revealed that the
learning stage (M = 1.639 s, SD = .488 s) had shorter reaction time compared to the novel stage (M = 2.534 s, SD = .849 s),
p < .001, d = 1.275. The learning stage also had shorter reaction time compared to the novel-with-noise stage (M = 2.547 s,
SD = 1.120 s), p < .001, d = 1.293. No dierence was found in reaction time for novel and novel-with-noise stage, p = .867,
d = .018. No signicant interaction eect was found between stimulation group and CFMT stage on reaction time, F(4,
174) = 1.415, p = .231, η
p2
= .032.
A signicant interaction eect was found between CFMT type and CFMT stage on accuracy, F(1.678, 145.977) = 4.823, p = .014,
η
p2
= .053. Simple main eect analysis revealed that reaction time of the own-race CFMT was shorter than other-race CFMT in the
learning stage (own-race: M = 1.545 s, SD = .439 s, other-race: M = 1.732 s, SD = .517 s), F(1, 89) = 28.026, p < .001, η
2
= .239, novel
stage (own-race: M = 2.396 s, SD = .861 s, other-race: M = 2.672 s, SD = .819 s), F(1, 89) = 19.618, p < .001, η
2
= .181 and novel-with-
noise stage (own-race: M = 2.307 s, SD = .803 s, other-race: M = 2.786 s, SD = 1.328 s), F(1, 89) = 22.830, p < .001, η
2
= .204. No
signicant interaction eect was found between stimulation group, CFMT stage and CFMT type on reaction time, F(4, 174) = .384, p
= .820, η
p2
= .009.
14 S. K. KHO ET AL.