1. The Journal of Neuroscience, March 20, 2013 • 33(12):5081–5082 • 5081
Journal Club
Editor’s Note: These short, critical reviews of recent papers in the Journal, written exclusively by graduate students or postdoctoral
fellows, are intended to summarize the important findings of the paper and provide additional insight and commentary. For more
information on the format and purpose of the Journal Club, please see http://www.jneurosci.org/misc/ifa_features.shtml.
On the Same Wavelength: Face-to-Face Communication
Increases Interpersonal Neural Synchronization
Kyongsik Yun
Computation and Neural Systems, Division of Biology, California Institute of Technology, Pasadena, California 91125
Review of Jiang et al.
Understanding neural mechanisms of social by quantifying the changes in oxyhemoglo- munication; and (4) analysis of spatiotem-
interaction is important for understanding bin concentration. Given that the left hemi- poral functional connectivity.
human social nature and for developing sphere is dominant for language function Validation of synchronization was strict
treatments for social deficits related to and that the left inferior frontal cortex and and appropriate in this study. Jiang et al.
disorders such as autism. However, conven- inferior parietal cortex have been known to (2012) verified that the increase in neural
tional cognitive and behavioral neurosci- be closely related to action understanding synchronization was specific for pairs of
ence has concentrated on developing novel and imitation, i.e., mirror neuron system, participants using a cross-validation entail-
experimental paradigms and investigating fNIRS was placed only on the left hemi- ing shuffling between participants and cal-
human– computer interactions, rather than sphere. The inferior frontal cortex has been culating interbrain correlations. They found
studying interpersonal interaction per se. particularly associated with empathy and no significant increase in synchronization
To fully understand neural mechanisms between randomly paired participants in
social cognition (Farrow et al., 2001). Jiang
of human interpersonal interaction, we any of the communication conditions. The
and colleagues found a significant increase
will likely have to investigate human be- results indicate that the interbrain corre-
in cross-correlation between changes in ac-
havior and neural processes in face-to-face lated activity is pair-specific and the in-
social interaction rather than human– com- tivity of two participants’ left inferior frontal
cortices only during a dialog in which sub- creased correlation was not by chance or
puter interaction. Recently, simultaneous artifacts. Moreover, the time-series of the
EEG or functional near-infrared spectros- jects faced each other (face-to-face), but
coherence values were randomly separated
copy (fNIRS) has been used to record brain none during other communication condi-
into two parts and the averaged coherence
activity of two participants in a face-to-face tions, including dialog in which subjects
values were compared between two parts.
setting (i.e., hyperscanning) to investigate faced away from each other (back-to-back),
No significant difference was found, indi-
human social interaction in a more natural- monologue with subjects facing each other,
cating that the coherence values are consis-
istic context (Jiang et al., 2012; Yun et al., and monologue with subjects facing away
tent across time. Last, Jiang et al. (2012) used
2012). from each other. Moreover, quality of com- other mirror neuron regions, including the
In a recent article published in The Jour- munication in each condition was assessed premotor area and inferior parietal cortices,
nal of Neuroscience, Jiang et al. (2012) inves- by self-report, and face-to-face dialog repre- to test whether the neural synchronization is
tigated interbrain neural synchronization sented a higher quality of communication region-specific. They found no significant
during face-to-face communication using than back-to-back dialog. These results difference in any of the conditions. The re-
fNIRS hyperscanning. fNIRS measures suggest that face-to-face conversation has sults suggest that neural synchronization is
changes in the regional cerebral blood flow special features that other types of com- specific to the inferior frontal cortex and
munication lack and that interbrain corre- that this synchrony was primarily contrib-
lated activity may be an underlying neural uted by face-to-face social interaction rather
Received Jan. 7, 2013; revised Jan. 31, 2013; accepted Feb. 4, 2013. process of successful face-to-face communi-
Many thanks to Shinsuke Shimojo for helpful comments on this
than mere mirroring the action of speaking.
manuscript. cation. In this Journal Club article, I con- The pair and region specificity of the results
Correspondence should be addressed to Kyongsik Yun, Computa- sider four important points regarding the and the temporal consistency of the coher-
tion and Neural Systems, Division of Biology, California Institute of study: (1) validity of the interbrain neural ence values successfully validate the robust-
Technology 139-74, 1200 E. California Blvd., Pasadena, CA 91125.
synchrony; (2) limitations that arise from ness of the data.
E-mail: yunks@caltech.edu.
DOI:10.1523/JNEUROSCI.0063-13.2013 using verbal communication; (3) aptness of It is important to note that face-to-face
Copyright©2013theauthors 0270-6474/13/335081-02$15.00/0 fNIRS methodology to face-to-face com- dialog represented a higher quality of

2. 5082 • J. Neurosci., March 20, 2013 • 33(12):5081–5082 Yun • Journal Club
communication than back-to-back dialog dialog. Moreover, fMRI gradient noise of- veloping treatment of social deficits asso-
(Arnal et al., 2009). Therefore, the stron- ten exceeds 100 dB and thus interferes ciated with autism.
ger interbrain correlation in face-to-face with auditory stimulation, which is crucial
dialog could result from either a higher for communication experiments (Plichta et References
quality of communication or from the al., 2011). Some argue that fNIRS may not Ahissar E, Nagarajan S, Ahissar M, Protopapas A,
face-to-face setting, in which various non- Mahncke H, Merzenich MM (2001) Speech
be appropriate for studies of the adult
comprehension is correlated with temporal re-
verbal cues were present. To deal with this human brain, because the light path is sponse patterns recorded from auditory cortex.
issue, Jiang et al. (2012) performed an ad- grossly affected by the CSF; understanding Proc Natl Acad Sci U S A 98:13367–13372.
ditional analysis wherein they separated the optical properties of each layer of the CrossRef Medline
time points showing the nonverbal com- head, however, has allowed accurate model- Arnal LH, Morillon B, Kell CA, Giraud AL (2009)
munication between participants, such as ing of fNIRS characteristics in the adult hu- Dual neural routing of visual facilitation in
turn-taking behavior and body language. speech processing. J Neurosci 29:13445–13453.
man brain (Hoshi, 2003). CrossRef Medline
Time points in which facial expression In future studies, it may be worthwhile Bunce SC, Izzetoglu M, Izzetoglu K, Onaral B, Pour-
and gestures occurred showed significant to compute neural synchronization be- rezaei K (2006) Functional near-infrared spec-
neural synchronization compared with tween different regions of paired partici- troscopy. IEEE Eng Med Biol Mag 25:54 – 62.
other time points in the face-to-face dia- pants. Jiang et al. (2012) only compared CrossRef Medline
log condition only. The results suggest synchrony between the same regions of
Chandrasekaran C, Trubanova A, Stillittano S,
that the increased interbrain correlation Caplier A, Ghazanfar AA (2009) The natural
each paired participant. Critically, how- statistics of audiovisual speech. PLoS Comput
mainly reflected nonverbal interaction. ever, the neural synchronization does not Biol 5:e1000436. CrossRef Medline
However, it is still possible that the quality have to occur either at the same region or Farrow TFD, Zheng Y, Wilkinson ID, Spence SA,
of communication influences interbrain Deakin JFW, Tarrier N, Griffiths PD, Wood-
at the same time across the paired partic-
neural synchronization. A previous study ruff PWR (2001) Investigating the func-
ipants. Rather two brains may form a
suggests that changes in speech amplitude tional anatomy of empathy and forgiveness.
more complex dynamic system; for exam- Neuroreport 12:2433–2438. CrossRef
can be synchronized with the brain activ-
ple, when two people interact, activity in Medline
ity of a listener (Chandrasekaran et al.,
the perceptual system of one brain is likely Hasson U, Ghazanfar AA, Galantucci B, Garrod S,
2009). Successful temporal synchrony be- Keysers C (2012) Brain-to-brain coupling: a
to be correlated, with some time delay,
tween two participants’ dialog and brain mechanism for creating and sharing a social
with activity in the motor system of the
activity increased the signal-to-noise ratio world. Trends Cogn Sci 16:114 –121. CrossRef
of neural signals and thus helped to im- other brain (Hasson et al., 2012). Medline
prove quality of communication (Ahissar Face-to-face communication offers a Hoshi Y (2003) Functional near-infrared optical
et al., 2001; Luo and Poeppel, 2007). superior form of communication in the imaging: utility and limitations in human brain
A major advantage of fNIRS is that the context of this study. We communicate mapping. Psychophysiology 40:511–520.
over the phone and by e-mail, but mes- CrossRef Medline
instrument is more portable and inex- Jiang J, Dai B, Peng D, Zhu C, Liu L, Lu C (2012)
pensive for functional neuroimaging sages tend to get misinterpreted and a so-
Neural synchronization during face-to-face
than functional magnetic resonance im- cial connection can hardly be established communication. J Neurosci 32:16064 –16069.
aging (fMRI). In addition, fNIRS is robust or maintained. Jiang et al. (2012) found CrossRef Medline
to movement artifacts compared with that face-to-face interaction increased the Luo H, Poeppel D (2007) Phase patterns of neuro-
quality of communication as well as inter- nal responses reliably discriminate speech in hu-
EEG and fMRI, allowing investigation of
brain correlated activity, suggesting there man auditory cortex. Neuron 54:1001–1010.
language processes, infants, and various CrossRef Medline
neuropsychiatric patients who cannot re- is some literal truth to the expression “we Plichta MM, Gerdes AB, Alpers GW, Harnisch W,
main sufficiently still for fMRI (Bunce et are on the same wavelength”. The study Brill S, Wieser MJ, Fallgatter AJ (2011) Au-
al., 2006). fNIRS is especially appropriate suggests that face-to-face communication ditory cortex activation is modulated by emo-
for a face-to-face dialog experimental set- has important neural and behavioral fea- tion: a functional near-infrared spectroscopy
ting in that fMRI does not allow us the tures that other types of communication (fNIRS) study. Neuroimage 55:1200 –1207.
CrossRef Medline
study of a face-to-face condition and EEG cannot rival, and the interbrain correla-
Yun K, Watanabe K, Shimojo S (2012) Interper-
is vulnerable to movement artifacts, in- tion results may have implications for un- sonal body and neural synchronization as a
cluding those produced by vocalization, derstanding the neural mechanisms of marker of implicit social interaction. Sci Rep
facial expression, and gestures involved in social interaction and diagnosing and de- 2:959. Medline