Neuropsychologia 48 (2010) 441–447
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/neuropsychologia
Frontopolar activation during face-to-face conversation: An in situ study using
Masashi Suda, Yuichi Takei, Yoshiyuki Aoyama, Kosuke Narita, Toshimasa Sato,
Masato Fukuda∗, Masahiko Mikuni
Department of Psychiatry and Neuroscience, Gunma University Graduate School of Medicine, 3-39-22 Showa, Maebashi, Gunma 371-8511, Japan
a r t i c l e i n f o
Received 10 June 2009
Received in revised form 20 August 2009
Accepted 30 September 2009
Available online 9 October 2009
Brodmann area 10
a b s t r a c t
of brain function during social interactions such as face-to-face conversation compared with functional
for studying brain mechanisms underlying social interactions: subjects have to lie down on a bed in a
small gantry during examination. The purpose of this study was to validate the possible use of NIRS as
a functional brain imaging technique for studying social interactions in a natural setting; therefore, we
investigated frontal and temporal lobe activation during face-to-face conversation in healthy subjects in
the sitting position. The frontal and superior temporal regions were activated during face-to-face con-
versation, with higher activity in the speaking segments than in the mute segments during conversation
particularly in frontopolar NIRS channels. The magnitude of frontopolar activity negatively correlated
with the cooperativeness score of the subjects assessed using the temperament and character inventory.
These results demonstrated the successful monitoring of brain function during realistic social interac-
tions using NIRS and interindividual differences in frontopolar activity during conversation in relation to
the cooperativeness of an individual.
© 2009 Elsevier Ltd. All rights reserved.
The brain mechanisms underlying social interactions have
recently been one of the most enthusiastically discussed topics in
neuroscience, and the prefrontal cortex, superior temporal gyrus,
brain” (Frith, 2001, 2007). These studies have been made possi-
ble by the development of functional brain imaging techniques,
emission tomography (PET), and have focused on facial recogni-
adaptation (Adolphs, 2003). However, both fMRI and PET have
methodological constraints for studying brain mechanisms under-
lying social interactions: subjects have to lie down on a bed in a
small noisy gantry during examination. Owing to such method-
ological constraints, social interactions have been studied mainly
in an unusual and nonrealistic situation, such as using pictures or
computer graphics shown on a PC monitor as task stimuli. Singer
∗Corresponding author. Tel.: +81 27 220 8185; fax: +81 27 220 8187.
E-mail address: firstname.lastname@example.org (M. Fukuda).
et al. (2006) have taken up this issue and argued that truly interac-
tive mind paradigms should be employed in, for instance, the fMRI
setting. Such attempts have been made by Singer herself (Singer et
al., 2006), but also in studies that make use of paradigms base on
game theory (McCabe, Houser, Ryan, Smith, & Trouard, 2001). In
other approaches, the use of virtual characters has also been sug-
gested to be of particular interest and could be a useful tool for
further investigations (Sanchez-Vives & Slater, 2005; Schilbach et
al., 2006). Although these methods using virtual reality are useful
ing methodology that enables the monitoring of brain activation in
a more natural setting is expected to offer more informative data
regarding social interactions in a real-life situation.
Near-infrared spectroscopy (NIRS) is one of the more recently
available functional brain imaging techniques that employ near-
infrared light emitted and detected on the skull skin (Boas, Dale, &
2004; Sato et al., 2007). It allows the monitoring of cerebral
blood volume (CBV) changes in the neocortex, as an increase in
oxygenated hemoglobin concentration ([oxy-Hb]) and a decrease
in deoxygenated hemoglobin concentration ([deoxy-Hb]) using a
small apparatus. It enables brain activity measurement in a more
natural setting as compared with other functional brain imaging
0028-3932/$ – see front matter © 2009 Elsevier Ltd. All rights reserved.
M. Suda et al. / Neuropsychologia 48 (2010) 441–447
techniques: for example, subjects can undergo NIRS examination
in the sitting position, with their eyes open, and while speaking.
These characteristics of NIRS are considered to be particularly suit-
able for studies of social interactions. Thus far, NIRS has already
been successfully used to monitor brain function for delicate sub-
ment (Kameyama et al., 2006; Suto, Fukuda, Ito, Uehara, & Mikuni,
2004; Uehara et al., 2007).
Conversation with other people is one of the most frequent
and nonverbal components. A number of functional brain imaging
studies related to conversation have focused on speech percep-
tion, a verbal component of conversation. Studies in which speech
stimuli were presented during passive listening tasks showed the
et al., 1994), and studies that used tasks similar to syllable discrim-
ination or identification showed prominent activations of the left
superior temporal regions (Demonet et al., 1992) and left inferior
frontal lobe (Zatorre, Evans, Meyer, & Gjedde, 1992). As a non-
verbal aspect, studies involving tasks such as facial recognition
or eye gaze perception showed that the activation of the medial
prefrontal cortex underlies both the perception of social commu-
nication indicated by facial expressions and the feeling of personal
involvement indicated by eye gaze (Schilbach et al., 2006), and that
the involvement of the medial prefrontal cortex associated with
outcome monitoring and mentalizing indicates higher-order social
cognitive processes related to the evaluation of the ongoing com-
municational input conveyed by direct gaze duration (Kuzmanovic
et al., 2009). In addition, in some recent studies, the integration of
information from face and voice has been investigated and found
to be associated with nonverbal components (Campanella & Belin,
2007; Kang et al., 2006). During realistic conversations, the non-
verbal components of communication are essential, and in some
cases, might be more important than the verbal components. How-
ever, to the best of our knowledge, there are only a few functional
that contains both verbal and nonverbal components was investi-
reciprocity. The main reason is the limitation in the experimental
setting of functional brain imaging, as described above.
In this study, by monitoring prefrontal and temporal lobe reac-
tivities using NIRS in healthy subjects in the sitting position during
their conversation opposite to another person, we investigated the
possible applicability of NIRS as a functional brain imaging tech-
nique for studying social interactions that include both verbal and
approach could help enhance our understanding of brain activity
during a social interaction in our everyday life and of the relation-
ship between verbal and nonverbal communication from a new
perspective. In addition, we investigated the relationship between
these brain activation and personality traits in healthy subjects.
We hypothesized that the face-to-face conversation is associated
with prefrontal and superior temporal cortex activation, the brain
areas implicated in social cognition, and a part of the personality
traits of healthy subjects reflects interindividual variations in brain
2. Materials and methods
Thirty healthy volunteers participated in this study (15 males and 15 females;
We had confirmed that they use the right hand when writing, drawing, using some
tools, and throwing, and had never changed their hand dominance. They had no
Fig. 1. Task procedures and picture of measurement settings. Two types of activa-
tional task, conversation condition and control condition, were employed in this
study. Both tasks consisted of three segments, pretask, task, and posttask. Under
the conversation condition, the subjects were required to talk with the interviewer
facing them during the task segment. The task period consisted of six cycles of such
a 30s talk, with the total conversation lasting up to 180s. Before starting and after
finishing the experiment, the subject and interviewer were separated by a parti-
tion so that they could not see each other. Under the control condition, the subjects
were instructed to repeat meaningless syllables during their turns to talk in the task
period. As shown in the picture, the subject wearing the near-infrared spectroscopy
probe on his forehead sat on the right side, whereas the interviewer sat on the left
them during the pretask and posttask segments.
history of any major psychiatric disorder, neurological disorder, substance abuse,
head injury, or major physical illness, and they were not taking any psychotropic
medications at the time of the study, although this had not been checked by brain
CT or MRI. This study was approved by the Institutional Review Board of the Gunma
University Graduate School of Medicine. Written informed consent was obtained
from all the subjects prior to the study.
2.2. Activation tasks (Fig. 1)
We employed two types of activation task, namely, conversation condition and
control condition. The order of the two tasks was counterbalanced among the sub-
2.2.1. Conversation condition
The conversation condition was designed to simulate usual conversations in
everyday life in an experimental setting. Each subject and an interviewer sat facing
each other one meter apart on comfortable chairs in a sunlit room with their eyes
open, and the probes of the NIRS machine were placed on the subject’s frontal and
interviewer were separated by a partition so that they could not see each other, and
the partition was removed during the experiment.
The experiment consisted of three periods: pretask, task, and posttask. During
the task period, the subjects were required to talk with the interviewer facing them.
To enable later data analyses, the subjects were instructed to carry out the conver-
sation during the task period following two criteria as follows. First, the time course
of the conversation was fixed a priori: the subject and interviewer were supposed
to talk in turn in this order every 15s, which was accomplished with cues pro-
vided by the experimenter’s voice every 5s. The task period consisted of six cycles
of 30s talk, with the entire conversation lasting up to 180s. Second, the topic of
the conversation was limited to anything related to food. The interviewers were
selected from three male psychiatrists who were not acquainted with the subjects.
During the pretask and posttask periods, the subjects were instructed to repeat the
syllables/a/,/i/,/u/,/e/, and/o/as the Japanese counterparts of A, B, and C in English
to exclude the effect of phonation and hence to stabilize the baseline conditions.
The images and voices of the subjects and interviewers were videotaped during the
experiment for further analyses.
Task performances of the conversation were evaluated in two ways. First, the
amount of talk by the subjects was evaluated quantitatively as talk time, which cor-
Second, the content of the talk was evaluated qualitatively in receiving and sending
aspects. The receiving aspect indicates the appropriateness of the talk in the con-
text of the conversation: the subjects’ replies to the preceding interviewer’s talk
were scored as (1) inappropriate, (2) partially inappropriate, (3) partially appro-
priate, and (4) appropriate. The sending aspect indicates the productivity of new
topics: the subjects’ questions to the interviewer were scored as (1) no new topic,
(2) nearly the same topic, (3) partially new topic, and (4) completely new topic. The
the contents individually at first, and then cross-checked the results.
M. Suda et al. / Neuropsychologia 48 (2010) 441–447
Fig. 2. The locations of NIRS channels were probabilistically estimated and anatomically labeled in the standard brain space in accordance with Tsuzuki et al. (2007). Shaded
gray channels indicated the channels that were excluded because of clear motion artifacts.
2.2.2. Control condition
To examine brain activation and artifact contamination induced only by phona-
tion, the control condition was used in addition to the experimental condition. The
subjects were instructed to repeat meaningless syllables such as “a”, “ka”, “sa”, “ta”,
and “na” during their turns to talk in the task period. All the subjects were able to
repeat such syllables without interruption during their turns.
2.3. Assessments of personality of subjects
The personality of the subjects was assessed using the Japanese version of the
temperament and character inventory (TCI: Cloninger, Svrakic, & Przybeck, 1993;
Kijima, Tanaka, Suzuki, Higuchi, & Kitamura, 2000). The TCI is a self-questionnaire
consisting of 240 items developed by Cloninger et al. The TCI assumes that per-
sonality consists of four biological dimensions (“temperament”) and three social
dimensions (“character”). The four dimensions of temperament are novelty seek-
ing, harm avoidance, reward dependence, and persistence. The three dimensions of
character are self-directedness, cooperativeness, and self-transcendence.
2.4. NIRS measurement
In this study, changes in [oxy-Hb] and [deoxy-Hb] were measured using a
52-channel NIRS machine (Hitachi ETG-4000) at two wavelength of near-infrared
light (i.e., 780 and 830nm) whose absorption was measured, and [oxy-Hb] and
changes were evaluated. The distance between the pair of emission and detector
probes was 3.0cm, and it was considered that the machine measured points at a
depth of 2–3cm from the scalp, that is, the surface of the cerebral cortex (Hock et
al., 1997; Toronov et al., 2001).
The probes of the NIRS machine were placed on the subject’s frontal region. The
frontal probes measured [Hb] changes at 52 measurement points in a 6cm×30cm
area, with the lowest probes positioned along the Fp1–Fp2 line in accordance with
the international 10/20 system used in electroencephalography. The measurement
points were labeled Ch 1 to Ch 52 from top to bottom. The correspondence between
by a multisubject study of anatomical craniocerebral correlation (Okamoto et al.,
(Fig. 2: Tsuzuki et al., 2007).
The absorption of near-infrared light was measured with a time resolution of
0.1s. The obtained data were analyzed using the “integral mode”. The pretask base-
line was determined as the mean across the last 10s of the 30-s pretask period, the
posttask baseline was determined as the mean across the last 10s of the 30-s post-
task period, and linear fitting was applied to the data between these two baselines.
The moving average method was used to exclude short-term motion artifacts in the
analyzed data (moving average window: 5s).
2.5. Data analyses
The waveforms of [Hb] changes in all 52 channels under the two conditions of
tasks were calculated for all the subjects. The NIRS data of channels that clearly con-
tained motion artifacts, as determined by close observation of the subjects, namely,
Chs 1–21, were excluded from further analyses. The most common cause of motion
artifact is hair. The probe placed on an area with a lot of hair cannot be sufficiently
fastened onto the head, and can be displaced easily.
[Hb] changes in the remaining 31 channels were analyzed in two steps. In the
first step, the channels with significantly greater [Hb] changes under the conver-
sation condition were identified. The individually averaged [Hb] data were divided
into the following three time segments for each condition: the ‘pretask’ segment
for 10s before the task period, the ‘task segment’ for 180s during the task period,
and the ‘posttask’ segment for 30s after the task period. The averages of [Hb] data
of ‘task segment’ of each condition were calculated and analyzed using two-way
repeated-measure analysis of variance (ANOVA) with “task” (conversation and con-
trol condition) and “channel” (31 channels) as independent variables, followed by a
In the second step, the relationships of [Hb] changes during conversation with
the personality of the subjects and task performances of the conversation were
investigated. For the channels with significantly greater [Hb] increases under the
conversation condition than under the control condition, as determined in the first
step, multiple regression analysis (stepwise procedure) was conducted with Bon-
ferroni correction: the averaged [Hb] changes in the task period as a dependent
variable, and gender and the seven scores of items in the TCI, namely, novelty
seeking, harm avoidance, reward dependence, persistence, self-directedness, coop-
control condition, as determined in the first step, multiple regression analysis was
conducted with Bonferroni correction: the averaged [Hb] changes in the task period
as a dependent variable, and three parameters for the conversation, namely, talk
time, receiving score, and sending score for the content of the talk as independent
3.1. Characteristics of participants and behavioral data (Table 1)
The TCI scores and task performances of the participants are
shown in Table 1. The scores of five out of the seven items in the
TCI, except harm avoidance and self-transcendence, were within
the standard values in the TCI (Cloninger et al., 1993). Among the
conversation, namely, talk time, receiving score, and sending score
for the content of the talk, significant correlations were obtained
only between the TCI scores and the three parameters for the con-
Characteristics of participants.
TCI Task performances
NSHA RDP SDC ST Time (s)RSSS
NS, novelty seeking; HA, harm avoidance; RD, reward dependence; P, persistence; SD, self-directedness; C, cooperativeness; ST, self-transcendence; time, speaking time; RS,
scoring of qualitative evaluation of receiving aspects; SS, scoring of qualitative evaluation of sending aspects; SV mean scores of TCI of standard values.
M. Suda et al. / Neuropsychologia 48 (2010) 441–447
Fig. 3. Grand averaged waveforms of oxygenated hemoglobin concentrations ([oxy-Hb]) under conversation condition (red line) and under control condition (blue line) in
31 channels over frontal and temporal regions measured by NIRS. The channels framed in bold black indicate those channels that showed significantly larger activations
under the conversation condition than under the control condition. Gray regions in a scale-up channel indicate the interval during which the subject kept silent under both
conditions. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
3.2. [Hb] changes during conversation and control tasks (Fig. 3,
and see supplementary figure online)
As for [oxy-Hb] changes, the two-way repeated-measure
ANOVA showed significant main effects of task (F=5.5, P=0.019)
and channel (F=2.1, P<0.001), and significant two-way inter-
actions for task by channel (F=25.6, P<0.001). The post hoc
paired t-test of grand averaged [oxy-Hb] changes during 180-s
task segments between the conversation and control condi-
tions demonstrated significantly larger [oxy-Hb] increases under
the conversation condition in 27 channels, namely, Chs 23–30,
32, 34–40, and 42–52 (t=3.6–8.7, P<0.0016), with Bonferroni
correction of multichannel comparisons (P<0.0016). The robust
activation was demonstrated over the frontal and superior tem-
poral regions under the conversation condition, but not under the
ferences between the conversation and control conditions (Chs 22,
31, 33, and 41) were located mainly around Broca’s area.
As for [deoxy-Hb] changes, the two-way repeated-measures
ANOVA showed the significant main effects of task (F=50.4,
P<0.001) and channel (F=1.5, P=0.037), and the significant two-
way interactions for task by channel (F=2.1, P<0.001). The post
hoc paired t-test of grand averaged [oxy-Hb] changes during
180-s task segments between the conversation and control con-
ditions demonstrated significantly larger [deoxy-Hb] decreases
under the conversation condition in 5 channels, namely, Ch 24
(t=4.0, P<0.001), Ch 25 (t=3.5, P=0.001), Ch 27 (t=3.7, P<0.001),
Ch 28 (t=4.1, P<0.001), and Ch 38 (t=3.6, P<0.001), with Bonfer-
roni correction of multichannel comparisons (P<0.0016).
3.3. Relationship between personality or task performance and
brain activation (Figs. 4 and 5)
and personality of the subjects were investigated. For the channels
with significantly greater [oxy-Hb] changes under the conversa-
tion condition than under the control condition determined in the
Bonferroni correction of multichannel comparisons (P<0.0016).
[oxy-Hb] increases in Ch 37 during the task period significantly
negatively correlated with the cooperativeness score in the TCI
The scatter graph of averaged [oxy-Hb] changes during the task
period in Ch 37 in relation to the cooperativeness score is shown
Fig. 4. Significant correlation between personality assessed by TCI and brain acti-
vation measured by NIRS: cooperativeness scores of TCI vs. oxygenated hemoglobin
concentration ([oxy-Hb]) changes in Ch 37 over medial prefrontal cortex.
M. Suda et al. / Neuropsychologia 48 (2010) 441–447
Fig. 5. Contrast of the differences in cooperativeness score. The subjects were divided into high- and low-cooperativeness-score subgroups using 30 as the cut-off score, the
mean of the cooperativeness score of TCI: high-score group (more than 30 points, 14 subjects, mean 33.5, SD 3.1) and low-score group (less than 30 points, 16 subjects, mean
26.6, SD 2.2). Pink lines indicate the grand averaged waveforms of [oxy-Hb] during conversation of the high-score group, and red lines indicate those of the low-score group.
(For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)
in Fig. 4, and the averaged [oxy-Hb] waveforms are shown in Fig. 5
dividing all the subjects into high- and low-cooperativeness-score
subgroups employing a cut-off score of 30, the mean of coopera-
There were no significant correlations between [oxy-Hb]
changes during conversation and the three scores of task per-
formances. [deoxy-Hb] changes during conversation did not
significantly correlate with the seven scores of items in the TCI and
the three scores of task performances.
4.1. Summary of obtained results
In this study, in which brain activation during conversation
was examined, the obtained results demonstrated that (1) brain
activation can be monitored time-dependently during a real face-
to-face conversation in a natural setting using a multichannel NIRS
machine; (2) the face-to-face conversation was accompanied by a
robust activation, [oxy-Hb] increases, over the frontal and supe-
rior temporal regions compared with the control condition; and
(3) such [oxy-Hb] increases negatively correlated with the cooper-
ativeness score in the TCI in the frontopolar channel.
4.2. In situ monitoring of brain function during conversation
In this study, and to the best of our knowledge for the first time,
ting. Such in situ functional brain imaging during conversation was
possible owing to the features of NIRS in which the subjects are
not obliged in this context to lie motionless in a gantry as opposed
to those of functional MRI, PET, and SPECT. Although NIRS has a
limitation, that is, it can be used to investigate only cerebral cortex
activation, it has been shown to be advantageous for elucidating
brain function during conversation, and hence, during more com-
plicated social interactions in situ.
In addition, NIRS is also useful for monitoring activation in the
frontopolar region. The monitoring of the frontopolar region by
fMRI and PET is considered difficult because of artifacts due to the
skull bone and frontal sinus. NIRS is assumed to be less affected by
such bone and frontal sinus artifacts and is suitable for the mea-
surement of the frontopolar region.
4.3. Brain activation during conversation
medial prefrontal and superior temporal cortices, which are the
brain areas implicated in social cognition. These findings are con-
sistent with previous findings. The medial prefrontal cortex plays
a unique role in social interaction (Amodio & Frith, 2006). It is sug-
gested that the anterior region of the rostral medial prefrontal lobe
has access to information about both actions and outcomes. The
characterization of this area comprises roughly three different cat-
egories: self-knowledge, person knowledge, and mentalizing. At
the same time, the anterior temporal region is also considered to
be related to social cognition such as mentalizing (Kampe, Frith,
Dolan, & Frith, 2001) and social concepts (Zahn et al., 2007). A face-
to-face conversation must require such social cognitive function,
because it is impossible to act and adapt to circumstances with-
out mentalizing, i.e., the ability to think about others’ thoughts and
mental states to predict their intentions and actions.
There was one functional brain imaging study in which the neu-
used functional MRI to compare cortical activation while normal
subjects casually talked to an actor with that while describing an
action performed in video clips. Significantly larger activations in
the medial prefrontal cortex and bilateral anterior superior tempo-
ral gyri were observed during the former communication task than
regions are important in the behavioral aspects of communica-
tive speech production. These results are in good agreement with
our results, that is, activation of the frontal and temporal regions
during the conversation. In addition, the channels over the dorso-
lateral prefrontal regions were also significantly activated under
the conversation condition in this study. These additional activa-
tions shown in our NIRS study might be associated with some sort
of nonverbal component of conversation, because of the difference
in the experimental setting (facing a gantry vs. face-to-face).
Differences between [oxy-Hb] changes under the conversation
and control conditions showed no statistically significant differ-
ences in the two channels located around Broca’s area, as shown in
Fig. 3. Because both conditions were similar in their composition of
alternating repetition of 15s talk and 15s mute phases, these brain
areas were considered to reflect Broca’s area function activated by
M. Suda et al. / Neuropsychologia 48 (2010) 441–447
4.4. Frontopolar activation along the time course of speaking and
In addition to the blockwise activations of the frontal and tem-
poral lobes during the 180-s conversation, rhythmic activations
superimposed them: accumulating activation during the talk sub-
segment and its resolution during the mute subsegment in the
conversation task, as shown in Fig. 3. Such 30-s-cycled rhythmic
frontal activation across the talk and mute subsegments was suc-
cessfully clarified owing to the high time resolution of NIRS. These
additional rhythmic activations were more prominent over the
frontopolar region than over other brain regions such as supe-
rior temporal regions. Such frontopolar dominant activation may
reflect a realistic nature of face-to-face conversation that requires
multitasking and ambiguous and uncertain judgments all the time,
because the frontal pole is assumed to be involved in high-level
judgment and intention decision (Koechlin & Hyafil, 2007), mul-
titasking (Burgess, 2000), working with ambiguity or uncertainty
(Yoshida & Ishii, 2006), and hence, possibly in conversation.
High-cooperativeness-scores in the TCI are considered to
reflect social acceptance, empathy, helpfulness, compassion, and
pure-heartedness (Cloninger et al., 1993). In this study, the coop-
erativeness scores significantly negatively correlated with the
anterior medial prefrontal activation during conversation. There is
one structural MRI study (Yamasue et al., 2008) and one functional
MRI study (Decety, Jackson, Sommerville, Chaminade, & Meltzoff,
2004) that found a brain correlation of cooperativeness in the TCI.
In the structural MRI study, the cooperativeness score positively
correlated with the anterior medial prefrontal cortex volume. In
the functional MRI study in which the effects of cooperation and
competition with another player in a computer game were exam-
ined, the anterior medial cortex activation was larger under the
results are apparently inconsistent with those of the present study
small brain activation. This discrepancy may be explained by the
feature of the task design in this study: the subjects were forced to
talk in turn in accordance with the experimenter’s cue, instead of a
spontaneous natural conversation. Such an unusual setting of the
conversation may result in subjects with a low-cooperativeness-
score requiring larger brain activation because they had difficulty
in talking with others cooperatively.
Moreover, the “default system” of the brain, which has been
described as a set of regions that are ‘activated’ during rest and
‘deactivated’ during cognitively demanding tasks (Raichle, 2001),
and its putative role in social cognition related to the medial pre-
frontal lobe (Schilbach, Eickhoff, Rotarska-Jagiela, Fink, & Vogeley,
2008) might be useful to explain these findings. The findings on
the relative small activations over the medial prefrontal lobe of
using this “default system”, that is, the physiological ‘baseline’ of
the brain might not be equal for minds at rest, but might be linked
with a psychological ‘baseline’ related to social cognition such as
self-consciousness and other-person perception. A highly coopera-
at rest; therefore, his/her brain activation during social cognitive
tasks might be less affected by such factors as the “ceiling effect”.
This study has some methodological limitations. First, the task
design was still unnatural and did not sufficiently reproduce an
ordinary social communication situation. We imposed unnatural
regulations such as a conversation cycle of 15s and the topic of
conversation being limited to food. Such regulations were required
to control the task conditions to analyze and examine the method-
ological adequacy. The average time of the conversation task was
minus 2.1s every 15-s turn. Such reduction in the speaking time
might also be caused by unnatural regulations such as a conversa-
tion cycle of 15s. It might be attributable to the subject’s difficulty
to speak for 15s in every 15-s speaking segment during the inter-
active conversation, because some subjects might think of the next
topic before opening their mouth and others might come up with a
topic before the end of the 15-s speaking segment. In addition, the
evaluation of the contents of the conversation was a bit subjective,
we will develop a more sophisticated task design. Second, the con-
trol condition that we employed was only performing phonation.
sation and control conditions as brain activation for conversation.
However, such a task design cannot exclude the in-depth effects
other than the conversation such as nonverbal communication. In
future studies, other additional conditions, such as a conversation
via telephone or a face-to-face situation without speaking, will be
helpful in examining the effects of nonverbal communication in
detail. Third, NIRS has some methodological limitations, such as
a low spatial resolution of approximately 3cm, i.e., nearly equal
to one gyrus of the brain, and ineffectiveness to assess deep brain
structures. When placing a NIRS probe on a head in accordance
with 10–20 standard positions, the cerebro-cranial correlation is
considered to vary within the range of 1cm; therefore, the corre-
spondence at the level of the gyrus is not affected by it (Okamoto
et al., 2004). NIRS methodology is still a developing technology;
tion. One of the major impediments is the difficulty in determining
the exact path length for each subject. In activational studies, it
is important to define a relationship between NIRS data and fMRI
ence frame. In that case, there is a major concern about differences
in analyzing brain activation, in addition to the correspondence of
anatomical references. fMRI is based on BOLD fMRI, a method of
observing which areas of the brain are active at any given time.
In contrast, most NIRS studies have not employed a homodynamic
tion of [Hb] changes. In addition, in this study, NIRS probes covered
and deep brain structures that might be particularly important for
of the NIRS methodology to assess deep brain structures. Fourth,
assessments of the behavioral data of the tasks and personality
traits of the participants were inadequate. The behavioral aspects
of the conversation were assessed only in terms of its verbal com-
ponent. Other aspects such as eye movement or facial expression
should be assessed in future studies. In addition, the personality
of the participants was assessed using only the TCI. Some reports
demonstrated racial differences in the standard scores of the TCI;
Japanese subjects scored lower in cooperativeness than American
subjects (Kijima et al., 2000). Because this problem arises from the
self-administered nature of the TCI, an objective evaluation of the
personality of the subjects should be carried out in future studies.
In summary, we monitored the frontal lobe activation during
face-to-face conversation in situ by NIRS, and demonstrated the
relationship between brain activation and the cooperativeness of
the subjects. Because other functional brain imaging modalities
such as fMRI, PET, and SPECT suffer from methodological limita-
M. Suda et al. / Neuropsychologia 48 (2010) 441–447 Download full-text
tions in experimental settings, studies in which brain function in
situ under natural conditions is examined by NIRS may deepen our
understanding of social and interpersonal cognition and behavior.
Gunma University (Drs. Fukuda and Mikuni) and Hitachi Group
(Advanced Research Laboratory, Hitachi Ltd. and the Research and
Developmental Center, Hitachi Medical Corporation) have had an
official contract for a collaborative study on the clinical applica-
tion of near-infrared spectroscopy in psychiatric disorders since
2002. The Hitachi Group has provided a project grant and material
pital and students of Gunma University Faculty of Medicine for
participating in this study.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
Neuroscience, 4, 165–178.
Amodio, D. M., & Frith, C. D. (2006). Meeting of minds: The medial frontal cortex and
social cognition. Nature Reviews Neuroscience, 7, 268–277.
Binder, J. R., Rao, S. M., Hammeke, T. A., Yetkin, F. Z., Jesmanowicz, A., Bandettini, P.
A., Wong, E. C., Estkowski, L. D., Goldstein, M. D., Haughton, V. M., et al. (1994).
Functional magnetic resonance imaging of human auditory cortex. Annals of
Neurology, 35, 662–672.
Boas, D. A., Dale, A. M., & Franceschini, M. A. (2004). Diffuse optical imaging of
brain activation: Approaches to optimizing image sensitivity, resolution, and
accuracy. Neuroimage, 23(Suppl. 1), S275–S288.
Burgess, P. W. (2000). Strategy application disorder: The role of the frontal lobes in
human multitasking. Psychological Research, 63, 279–288.
Campanella, S., & Belin, P. (2007). Integrating face and voice in person perception.
Trends in Cognitive Sciences, 11, 535–543.
Cloninger, C. R., Svrakic, D. M., & Przybeck, T. R. (1993). A psychobiological model of
temperament and character. Archives of General Psychiatry, 50, 975–990.
Decety, J., Jackson, P. L., Sommerville, J. A., Chaminade, T., & Meltzoff, A. N. (2004).
The neural bases of cooperation and competition: An fMRI investigation. Neu-
roimage, 23, 744–751.
Demonet, J. F., Chollet, F., Ramsay, S., Cardebat, D., Nespoulous, J. L., Wise, R., Ras-
col, A., & Frackowiak, R. (1992). The anatomy of phonological and semantic
processing in normal subjects. Brain, 115, 1753–1768.
Fallgatter, A. J., Ehlis, A. C., Wagener, A., Michel, T., & Herrmann, M. J. (2004). Near-
infrared spectroscopy in psychiatry. Nervenarzt, 75, 911–916.
Frith, C. D. (2007). The social brain? Philosophical Transactions of the Royal Society of
London. Series B Biological Sciences, 362, 671–678.
Frith, U. (2001). Mind blindness and the brain in autism. Neuron, 32, 969–979.
Hock, C., Villringer, K., Muller Spahn, F., Wenzel, R., Heekeren, H., Schuh Hofer, S.,
Hofmann, M., Minoshima, S., Schwaiger, M., Dirnagl, U., & Villringer, A. (1997).
Decrease in parietal cerebral hemoglobin oxygenation during performance of
a verbal fluency task in patients with Alzheimer’s disease monitored by means
of near-infrared spectroscopy (NIRS)—correlation with simultaneous rCBF-PET
measurements. Brain Research, 755, 293–303.
Kameyama, M., Fukuda, M., Yamagishi, Y., Sato, T., Uehara, T., Ito, M., Suto, T., &
Mikuni, M. (2006). Frontal lobe function in bipolar disorder: A multichannel
near-infrared spectroscopy study. Neuroimage, 29, 172–184.
and gaze. Nature, 413, 589.
Kang, E., Lee, D. S., Kang, H., Hwang, C. H., Oh, S. H., Kim, C. S., Chung, J. K., & Lee,
M. C. (2006). The neural correlates of cross-modal interaction in speech percep-
tion during a semantic decision task on sentences: A PET study. Neuroimage, 32,
Kijima, N., Tanaka, E., Suzuki, N., Higuchi, H., & Kitamura, T. (2000). Reliability and
validity of the Japanese version of the temperament and character inventory.
Psychological Reports, 86, 1050–1058.
decision-making. Science, 318, 594–598.
Kuzmanovic, B., Georgescu, A. L., Eickhoff, S. B., Shah, N. J., Bente, G., Fink, G. R., &
and evaluation of social gaze. Neuroimage, 46, 1154–1163.
McCabe, K., Houser, D., Ryan, L., Smith, V., & Trouard, T. (2001). A functional imag-
ing study of cooperation in two-person reciprocal exchange. Proceedings of the
National Academy of Sciences of the United States of America, 98, 11832–11835.
Okamoto, M., Dan, H., Sakamoto, K., Takeo, K., Shimizu, K., Kohno, S., Oda, I., Isobe,
S., Suzuki, T., Kohyama, K., & Dan, I. (2004). Three-dimensional probabilistic
anatomical cranio-cerebral correlation via the international 10–20 system ori-
ented for transcranial functional brain mapping. Neuroimage, 21, 99–111.
Raichle, M. E. (2001). Cognitive neuroscience: Bold insights. Nature, 412, 128–130.
Sanchez-Vives, M. V., & Slater, M. (2005). From presence to consciousness through
virtual reality. Nature Review Neuroscience, 6, 332–339.
Sassa, Y., Sugiura, M., Jeong, H., Horie, K., Sato, S., & Kawashima, R. (2007). Cortical
mechanism of communicative speech production. Neuroimage, 37, 985–992.
Sato, T., Ito, M., Suto, T., Kameyama, M., Suda, M., Yamagishi, Y., Ohshima, A., Uehara,
T., Fukuda, M., & Mikuni, M. (2007). Time courses of brain activation and their
ing finger tapping. Neuroscience Research, 58, 297–304.
Schilbach, L., Wohlschlaeger, A. M., Kraemer, N. C., Newen, A., Shah, N. J., Fink, G.
R., & Vogeley, K. (2006). Being with virtual others: Neural correlates of social
interaction. Neuropsychologia, 44, 718–730.
Schilbach, L., Eickhoff, S. B., Rotarska-Jagiela, A., Fink, G. R., & Vogeley, K. (2008).
Minds at rest? Social cognition as the default mode of cognizing and its putative
relationship to the “default system” of the brain. Consciousness and Cognition,
Empathic neural responses are modulated by the perceived fairness of others.
Nature, 439, 466–469.
Subjective feeling of psychological fatigue is related to decreased reactivity in
ventrolateral prefrontal cortex. Brain Research, 1252, 152–160.
& Mikuni, M. (2008). Decreased cortical reactivity underlies subjective daytime
light sleepiness in healthy subjects: A multichannel near-infrared spectroscopy
study. Neuroscience Research, 60, 319–326.
Suto, T., Fukuda, M., Ito, M., Uehara, T., & Mikuni, M. (2004). Multichannel
near-infrared spectroscopy in depression and schizophrenia: Cognitive brain
activation study. Biological Psychiatry, 55, 501–511.
Toronov, V., Webb, A., Choi, J. H., Wolf, M., Michalos, A., Gratton, E., & Hueber, D.
(2001). Investigation of human brain hemodynamics by simultaneous near-
infrared spectroscopy and functional magnetic resonance imaging. Medical
Physics, 28, 521–527.
Tsuzuki, D., Jurcak, V., Singh, A. K., Okamoto, M., Watanabe, E., & Dan, I. (2007).
Virtual spatial registration of stand-alone fNIRS data to MNI space. Neuroimage,
Uehara, T., Fukuda, M., Suda, M., Ito, M., Suto, T., Kameyama, M., Yamagishi, Y., &
spectroscopy. Eat Weight Disorder, 12, 183–190.
Yamasue, H., Abe, O., Suga, M., Yamada, H., Rogers, M. A., Aoki, S., Kato, N., & Kasai, K.
(2008). Sex-linked neuroanatomical basis of human altruistic cooperativeness.
Cerebral Cortex, 18, 2331–2340.
Zahn, R., Moll, J., Krueger, F., Huey, E. D., Garrido, G., & Grafman, J. (2007). Social
concepts are represented in the superior anterior temporal cortex. Proceed-
ings of the National Academy of Sciences of the United States of America, 104,
Zatorre, R. J., Evans, A. C., Meyer, E., & Gjedde, A. (1992). Lateralization of phonetic
and pitch discrimination in speech processing. Science, 256, 846–849.