Selective Medial Prefrontal Cortex Responses During ...

Brain Topogr DOI 10.1007/s10548-014-0414-2

ORIGINAL PAPER

Selective Medial Prefrontal Cortex Responses During Live Mutual Gaze Interactions in Human Infants: An fNIRS Study Susumu Urakawa • Kouichi Takamoto Akihiro Ishikawa • Taketoshi Ono • Hisao Nishijo

•

Received: 28 April 2014 / Accepted: 24 October 2014 Ó Springer Science+Business Media New York 2014

Abstract To investigate the role of the prefrontal cortex (PFC) in processing multimodal communicative ostensive signals in infants, we measured cerebral hemodynamic responses by using near-infrared spectroscopy (NIRS) during the social interactive play ‘‘peek-a-boo’’, in which both visual (direct gaze) and auditory (infant-directed speech) stimuli were presented. The infants (mean age, around 7 months) sat on their mother’s lap, equipped with an NIRS head cap, and looked at a partner’s face during ‘‘peek-a-boo’’. An eye-tracking system simultaneously monitored the infants’ visual fixation patterns. The results indicate that, when the partner presented a direct gaze, rather than an averted gaze, toward an infant during social play, the infant fixated on the partner’s eye region for a longer duration. Furthermore, hemodynamic activity increased more prominently dorsomedial prefrontal cortex (mPFC) in response to social play with a partner’s direct gaze compared to an averted gaze. In contrast, hemodynamic activity increased in the right dorsolateral prefrontal cortex (R-lPFC) regardless of a partner’s eye gaze direction. These results indicate that a partner’s direct gaze

S. Urakawa K. Takamoto T. Ono Department of Neurophysiotherapy, Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama, Sugitani 2630, Toyama 930-0194, Japan S. Urakawa H. Nishijo (&) Department of System Emotional Science, Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama, Sugitani 2630, Toyama 930-0194, Japan e-mail: [email protected] A. Ishikawa Research & Development Medical Systems Division, Shimadzu Corporation, Nishinokyoukuwabaracho 1, Nakagyouku, Kyoto, Japan

shifts an infant’s attention to the partner’s eyes for interactive communication, and specifically activates the mPFC. The differences in hemodynamic responses between the mPFC and R-lPFC suggest functional differentiation within the PFC, and a specific role of the mPFC in the perception of face-to-face communication, especially in mutual gaze, which is essential for social interaction. Keywords NIRS Prefrontal pole Dorsolateral prefrontal cortex Peek-a-boo Social communication Eye contact

Introduction For humans, the ability to perceive precise information from the face plays an important role in social communication. Face perception is essential, especially in infants, to develop the ability of social recognition. Within the first few days after birth, infants show a preference to the face or face-like patterns (Goren et al. 1975; Johnson et al. 1991; Valenza et al. 1996). Eyes and gaze direction are especially important for infants’ face preference (Farroni et al. 2002; Reid and Striano 2005; Senju and Csibra 2008). Mutual gaze (eye contact) provides the main mechanism of establishing a communicative context in social interaction (Hains and Muir 1996; Kleinke 1986). Therefore, it has been argued that early development of the perception of eye gaze serves as a major foundation for the later development of social skills (Baron-Cohen 1995; Csibra and Gergely 2006). Indeed, an impairment of face perception, and eye gaze detection in particular, might be one of the early signs of atypical social development, such as in autism (Phillips et al. 1992; Zwaigenbaum et al. 2005). Bryson et al. (2004) pointed out that one of the early signs

123

Brain Topogr

of autism is a lack of shared interest in social games such as ‘‘peek-a-boo’’. Noninvasive studies have revealed a complex neural system for face and eye gaze perception. In general, a dominance of the right temporal cortex, including the superior temporal sulcus (STS), for face perception has been supported by previous studies in adults by using functional magnetic resonance imaging (fMRI) (Kanwisher et al. 1997; Puce et al. 1996) and in infants by using nearinfrared spectroscopy (NIRS) (Nakato et al. 2009; Otsuka et al. 2007). Additionally in adults, the prefrontal cortex (PFC) has been implicated in cognitive facial processing, especially for eye gaze (Amodio and Frith 2006; Kuzmanovic et al. 2009; Wang et al. 2011). In adults, dorsomedial PFC (mPFC) has been found to be activated when gaze is direct, but not when it is averted (Kampe et al. 2003; Schilbach et al. 2006). Because the mPFC has been shown to play a distinct role in attributing the mental states to others (Amodio and Frith 2006; Gilbert et al. 2007), this brain region is considered a candidate core region for social cognition and social interaction (Amodio and Frith 2006). A recent study in adults showed that hearing a voice calling the subjects’ names also activated the mPFC (Kampe et al. 2003). These results suggest that ostensive communicative signals (i.e., direct gaze or calling someone’s name) with different modalities activate the same neural system, including the mPFC (Kampe et al. 2003). In adults, a psychological study showed that participants judged faces with an averted gaze as looking directly at them when they simultaneously heard their own name called (Stoyanova et al. 2010). In infants, vocal stimuli that accompany visual stimuli (i.e., faces) enhanced their responsiveness in a ‘‘peek-a-boo’’ game (Greenfield 1972). These results suggest that communicative ostensive signals of the auditory and visual modalities interact in the mPFC. Consistent with this idea, recent studies have shown that both visual (direct gaze) and/or auditory ostensive stimuli (calling) activated the infant PFC including the mPFC (Parise and Csibra 2013) and left lateral PFC (L-lPFC) (Grossmann et al. 2010, 2013; Grossmann and Johnson 2010). However, when these visual stimuli were presented together with ostensive auditory stimuli (infant-directed speech), there were no significant differences in EEG activity between direct and averted gazes (Parise and Csibra 2013). Since infants might be less sensitive to objects or human subjects in video compared with real things (Strouse and Troseth 2008), insensitivity to gaze direction in the composite condition (simultaneous presentation of the visual and auditory stimuli) might be ascribed to presentation by video. Thus, the role of the infant PFC in processing composite stimuli of simultaneous auditory and visual communicative ostensive signals has been unclear.

123

Recently, NIRS has been used in infant studies to reveal brain activity involved in the cognitive processing of faces (Nakato et al. 2009, 2011; Otsuka et al. 2007) and eye gaze direction (Grossmann et al. 2008, 2010; Lloyd-Fox et al. 2011). Furthermore, previous NIRS studies reported that activity in the temporal lobe was increased during presentation silent video of female actors, who moved eyes or performed a ‘‘peek-a-boo’’ game, in typically developing infants but not in infants at risk for autism (Lloyd-Fox et al. 2009, 2013). Lloyd-Fox et al. (2011) also reported that video clips of a female actor with movements of eyes, mouth and hand elicited specific patterns of hemodynamic responses in the frontal-temporal regions in 5-month-old infants. One of the great advantages of NIRS is that recording can be performed without any physical restraint, as opposed to the use of fMRI. Therefore, this method is useful for recording brain activity in awake infants, and to simultaneously measure various physiological parameters (Takeuchi et al. 2009). In the present study, to elucidate the role of the infant PFC in social interactive play that usually involves both visual and auditory ostensive signals, we analyzed hemodynamic responses in the infant PFC by using NIRS during the social interactive play ‘‘peek-a-boo’’ in which communicative ostensive signals in both auditory and visual modalities were presented to infants. Furthermore, in this study, a real human presenter, instead of video, played a ‘‘peek-a-boo’’ game, which has not been tested in the previous infant NIRS and EEG studies.

Materials and Methods Subjects Eleven healthy infants (7 boys and 4 girls, 211 ± 8.9 days old) and their mothers participated in the experiment. Seven additional infants were excluded because of insufficient performance in more than 6 trials because of crying, interrupting the measurements, or motion artifacts. The experiments were conducted according to the Declaration of Helsinki and United States Code of Federal Regulations: Protection of Human Participants, and approved by the ethical committee at the University of Toyama. The nature of the experimental procedures was explained, and written informed consent was obtained from all mothers. Behavioral Procedures A mother sat on a comfortable chair and held her infant on her lap (Fig. 1a). The experiment proceeded in 3 steps: eye-tracking calibration, NIRS calibration, and recording. During the recording, a young woman (partner) sat in front of the infant and performed a ‘‘peek-a-boo’’ experiment:

Brain Topogr

Fig. 1 Experimental set up for eye-tracking system and near-infrared spectroscopy (NIRS) recording. a A mother (m) held her infant (i) on her lap in front of a partner (p). The partner performed a ‘‘peek-aboo’’ play according to a recorded voice from speakers (s). Infant behavior was monitored by a camera (c1). Eye-tracking information from the Eye Tracker (e) and screen images of the partner from a monitor camera (c2) were integrated into movie images to indicate

eye fixation of the infant on the partner (see Fig. 3). b A head cap for infant NIRS recording and probe distribution. NIRS channels were located at mid points between the emitter (white circle) and detector (grey circle) probes. Black dashed lines indicate region of interests (ROIs) in the medial prefrontal cortex (mPFC; 12, 13 and 16 ch), and solid grey lines indicate ROIs in the right and left lateral prefrontal cortex (R-lPFC; 1, 4, 5 and 8 ch and L-lPFC; 3, 6, 7 and 10 ch)

Voice start Baseline -2

-1

0

Reactive time window 1

2

Inai, inai,

3

4

5

6

7

8

9

10 11

s

Baa, baa! Averted

Direct

Fig. 2 Time sequence of the ‘‘peek-a-boo’’ play. The partner performed ‘‘peek-a-boo’’ in Japanese, ‘‘Inai, inai, inai’’ for 3 s by hiding her face, and ‘‘Baa, baa!’’ for 5 s by exposing her face (facing period). Note that during the facing period, the partner displayed one of two gazes: an averted or direct gaze. The oxyhemoglobin (oxy-Hb)

and deoxyhemoglobin (deoxy-Hb) changes were analyzed during the reactive time window (the period from 3 to 10 s after the start of the task). Zero on the time scale indicates the start of the task (the start of play back of the recorded voice)

the presenter hid her face with both hands and called ‘‘Inai, inai, inai’’ (I am not here in Japanese) for 3 s, and then exposed her face with the call ‘‘Baa, baa!’’ (surprise onomatopoeia) for 5 s (facing period), and after that hid her face again (Fig. 2). The speakers behind the presenter produced recorded voices that would appear to have been

produced by the presenter (Fig. 1a). To eliminate the presenter’s emotional biases, the recorded vocal sound was presented from a speaker, and the presenter was well trained to perform a peek-a-boo play and move her lips in synchrony with the recorded voice. During the facing period, the presenter displayed an averted gaze (averted

123

Brain Topogr Fig. 3 Infants’ eye fixation on the face of the ‘‘peek-a-boo’’ partner. A Definition of 4 areas of interest (AOIs) for eye fixation: eyes, nose, mouth, and other (e.g., hand or forehead). B, C Differences in fixation time of the AOIs between the averted (B) and direct (C) gaze conditions. Representative examples of hotspot plotting (Ba averted condition; Ca direct condition) showed that infants fixated more on the partner’s eye region in the direct condition. The averaged ratio of fixation time of each AOI (Bb averted condition; Cb direct condition) indicated that infants significantly fixated on the partner’s eye region in the direct condition (*p \ 0.01)

A

Averted

Ba

Direct

Ca

Other Eye

Other

27.0%

36.3%

37.0%

Mouth 2.7%

* Eye 55.8%

Nose Mouth Nose 9.7%

Bb condition) or a direct gaze (direct condition) to the infant (Fig. 2). A ‘‘peek-a-boo’’ trial in the averted or direct condition was repeated several times in random order, and the start of each sequential recorded voice (total 8 s) was decided upon depending on the infant’s state judged by live video monitoring (Fig. 1a-c1, Evrio GZ-MG275, Victor, Kanagawa, Japan) in the next room. The interval between each trial was more than 20 s. The presenter put her face on a stand in order to maintain a fixed distance from the infant (60–70 cm) and eye-tracking system. Eye-Tracking Measurement A Tobii X120 Eye Tracker (Fig. 1a-e) and TobiiStudio software (Tobii Technology, Danderyd, Sweden) were used to measure and analyze the infant’s fixation. The eye-

123

14.6%

17.0%

Cb tracking system consisted of an infrared light source in order to capture the eye position (Tobii X120 Eye Tracker) and a monitoring camera (Fig. 1a-c2, DCR-HC37E, Sony, Tokyo, Japan) to obtain screen images of what the infant saw. These were integrated into one video-image to indicate the fixation points the infant looked at by using TobiiStudio (Fig. 3Ba, Ca). The Eye Tracker recorded the infant’s gaze position on a screen image as X–Y coordinates at 60 Hz. Calibration procedures were run using the TobiiStudio program for infant calibration on a 17-inch monitor. If the calibration succeeded, the fixation stand for the presenter was set in the same position as the monitor, and the distance between the presenter and the infant was set at 60–70 cm. Finally, NIRS calibration and recording were performed.

Brain Topogr

NIRS Recording The NIRS instrument (OMM 3000, Shimadzu Corporation, Kyoto, Japan) was used to measure hemodynamic changes in oxyhemoglobin (oxy-Hb), deoxyhemoglobin (deoxyHb), and total hemoglobin (total-Hb) with 0.1 s time resolution (Takamoto et al. 2013; Takeuchi et al. 2009). Seventeen channels were assigned by 6 emitters and 6 detectors (Fig. 1b), which were positioned across from each other at 2 cm intervals, because Taga et al. (2007) reported that a 2 cm emitter-detector distance is best for the analysis of infant hemodynamic responses. The midpoints between emitters and detectors were called ‘‘NIRS channels’’. Three different wavelengths (780, 805, and 830 nm) with a pulse of 5 ms were used to detect hemodynamic responses. The mean total irradiation power was less than 1 mW. We used a newly developed head cap to fix the optical fiber probe of emitters and detectors for infant subjects (Shimadzu, Fig. 1b) since it is light and provides softer contact with the skin and scalp. The head cap consisted of receptacles for the optical fiber probes and a black-colored plastic sheet that covered the PFC. It is noted that a mechanical spring in each probe gently pushed each optical fiber onto the scalp. The head cap with this mechanical spring system allowed simultaneous recording of NIRS and eye fixation by Eyetracker because it prevented extraneous noise of infrared light from Eyetracker. In each infant, the most ventral probes (emitters 5, 6 and detectors 5, 6) were placed on the Fp1 and Fp2 level according to the International 10–20 system (Jasper 1958) so that the NIRS channels were placed on the PFC. The NIRS instrument automatically checked ambient light or whether the contact was adequate to measure. The channels were eliminated from the analysis if adequate contact to infant scalp could not be achieved because of ambient light or interference by hair. Data Analysis Throughout the experiment, each infant’s behavior was monitored in the next room and video-recorded. The data were discarded when the infant did not look at the partner or interrupted the measure during ‘‘peek-a-boo’’. Furthermore, we removed data in trials that included movement artifacts offline, in which sharp changes in NIRS signals were found corresponding to sudden movements of infants in the synchronized video in Tobii Eyetracker system. To evaluate the infant’s fixation on the partner’s face during ‘‘peek-a-boo’’, the integrated video-images (fixation on the screen images) during the facing period were analyzed by TobiiStudio software. We defined 4 areas of interest (AOIs): eyes, nose, mouth, and other area (e.g.,

hand or forehead) on each screen image (Fig. 3A). Because infant fixation was not always detected during the facing period (i.e., pupil corneal reflection techniques depend on the size and contrast of each infant’s eye), the ratio of fixation time of each AOI was estimated using the following formula: fixation time of each AOI/total fixation time during the facing period. Hotspot plottings (Fig. 3Ba, Ca) were also created using TobiiStudio. Ratios of fixation time of each AOI were compared by paired t tests with Bonferroni correction between the averted and direct gaze conditions. To analyze the NIRS data, we focused on changes in oxy-Hb, which have been reported to be sensitive to neurohemodynamic relationships in NIRS studies (Hoshi et al. 2001; Strangman et al. 2002; Yamamoto and Kato 2002). Furthermore, we also analyzed changes in deoxy-Hb, which have been frequently evaluated in infant NIRS studies (see a review by Cristia et al. 2013). In each direct and averted condition, these NIRS data were summed and averaged in reference to the zero-onset (start of ‘‘peek-aboo’’ with play back of the voice, Fig. 2) of each trial. The averaged NIRS data from -2 to 2 s were analyzed as the baseline activity. Previous studies have showed that hemodynamic responses typically lag a few seconds after stimulus onset in infants (Csibra et al. 2004; Meek et al. 1998). Therefore, we defined the period from 3 to 10 s after the start of ‘‘peek-a-boo’’ as the oxy-Hb and deoxy-Hb reactive time window (Fig. 2). We defined the regions of interest (ROIs) in the PFC as follows: ch 1, 4, 5, and 8 as the right lateral PFC (R-lPFC); ch 3, 6, 7, and 10 as the left lateral PFC (L-lPFC); and ch 12, 13, and 16 as the mPFC (Fig. 1b). To analyze the NIRS data using a within-subject design, statistical analyses were carried out by two-way repeated-measures ANOVA with factors of gaze direction (averted condition vs. directed condition) and period (baseline vs. reactive-time-window). The mean values of the NIRS data (oxy-Hb, deoxy-Hb) during the reactive time window from 3 to 10 s after the start of ‘‘peek-a-boo’’ were estimated in each channel in each condition in each subject, and then the mean NIRS data in the channels within the same ROIs were averaged. Finally, grand mean of these NIRS data in each ROI across the subjects was computed. All values are expressed as mean values ± standard error of the mean (SEM). All statistical analyses were performed using the software package SPSS (v19, IBM, Somer, NY, USA). Differences were considered statistically significant at p B 0.05. Furthermore, we analyzed the data based on effect sizes (Schroeter et al. 2003), which could adjust different path length factors across subjects and ROIs. The effect size of hemodynamic responses during the reactive time window was defined by the following formula; [(mean oxy-Hb level during the reactive time window) - (mean oxy-Hb level

123

Brain Topogr

during baseline)]/(standard deviation of oxy-Hb level during the baseline). The data of effect sizes were statistically compared by Wilcoxon signed-rank tests.

Results Eye Tracking We obtained eye-tracking and neural data from 11 infants who looked at a presenter during an experimental period of more than 6 trials. The mean numbers of trials were 6.91 ± 0.48 for the direct condition, and 7.09 ± 0.53 for the averted condition. There were no significant differences in the number of trials between conditions (paired t test, p = 0.68). Figure 3Ba and Ca show hotspot plotting (ratio of fixation time) during the facing period in typical trials of the averted and direct conditions. In these trials, the infant fixated on facial regions other than the eye region in the averted condition, while the infant fixated on the eye region of the partner in the direct condition. Figure 3Bb and Cb show more clearly the differences between the two conditions. The mean ratios of fixation time for the eye region

2 -2

-2

0

2

4

6

8

10 s

Deoxy-Hb

-4 -6 -8

-10

6

Oxy-Hb

4 2 -2

-2

-4 -6 -8

-10

0

2

4

6

8

Deoxy-Hb

10 s

Relative change (µM cm)


E

Averted

8

0

L-lPFC

C

Direct

8 6

Oxy-Hb

4 2 0 -2

-2

0

2

4

6

8

10s

-4 -6

Deoxy-Hb

-8

10-3 10

Oxy-Hb

6 4 2 -2

-2

0

2

-4 -6 -8 -10

Fig. 4 Mean oxyhemoglobin (oxy-Hb) and deoxyhemoglobin (deoxy-Hb) changes in the R-lPFC (a, d), mPFC (b, e), and L-lPFC (c, f) in the direct (a–c) and averted (d–f) conditions of the ‘‘peek-aboo’’ play. Data represent relative mean oxy-Hb (black solid lines) and deoxy-Hb (gray solid lines) changes in reference to the zero-onset

123

F

Averted

8

0

10-3 10

Direct

8 6

Oxy-Hb

4 2 0 -2

-2

0

2

4

6 8 Deoxy-Hb

10 s

-4 -6 -8 -10

-10

10-3 10

D

10-3 10

4

6

8

Deoxy-Hb

10 s


0

Figure 4 shows mean relative change of oxy-Hb and deoxy-Hb throughout a trial period, during which time 3–10 enclosed by dashed line indicates the reactive time window. The oxy-Hb changes in mPFC were greatly increased when the presenter’s gaze was directed towards the infant (B), but less in the averted condition (E). Figure 5 shows the average hemodynamic responses during the time window 3-10 s in the R-lPFC, mPFC, and L-lPFC. Statistical analyses revealed that there was a significant main effect of period (baseline vs. reactive-time-window) (F1,10 = 8.77, p = 0.014) with no significant main effect


Oxy-Hb

4



B

Direct

8 6

NIRS Recording

mPFC

R-lPFC 10-3 10

A

were significantly larger in the direct condition than in the averted condition (paired t test with Bonferroni correction, p = 0.008). On the other hand, there were no differences in the ratios of fixation time between the two conditions in the other 3 areas (paired t test with Bonferroni correction, p = 0.96 for nose, 0.11 for mouth and 0.10 for other stimuli). Furthermore, the averaged total time in the eye region was significantly larger in the direct condition than in the averted condition (2.38 ± 0.35 s for direct, 1.54 ± 0.35 s for averted, p = 0.002).

10-3 10

Averted

8 6

Oxy-Hb

4 2 0 -2 -4

-2

0

2

4

6

8

10 s

Deoxy-Hb

-6 -8 -10

(time 0). Note that oxy-Hb changes in the mPFC gradually increased more prominently in the direct (b) than averted (e) gaze conditions, as well as compared to the direct gaze condition in the other areas (a, c). Error bars indicate SEM. Arrows represent the reactive time window

Brain Topogr

B

6 4 2 0 -2

Baseline

Face

Mean oxy-Hb change (µM cm)

8

C

mPFC 8 6 4 2 0 -2

Baseline

Fig. 5 Distinct hemodynamic responses during oxyhemoglobin (oxyHb) reactive time window (3–10 s) in the R-lPFC (a), mPFC (b), and L-lPFC (c). a In the R-lPFC, exposure of the partner’s face significantly increased the mean oxy-Hb changes regardless of the partner’s gaze direction (*, p \ 0.05). b In the mPFC, exposure of the partner’s face significantly increased the mean oxy-Hb changes

of gaze condition (F1,10 = 0.003, p = 0.96) nor significant interaction between gaze condition and period (F1,10 = 0.146, p = 0.71) in the R-lPFC (Fig. 5a). These findings indicate that hemodynamic responses increased in the oxy-Hb reactive time window regardless of gaze direction. In the mPFC, there were significant main effects of gaze condition (F1,10 = 11.4, p = 0.007) and period (F1,10 = 5.65, p = 0.039), and a significant interaction between gaze condition and period (F1,10 = 17.6, p = 0.002). Post-hoc tests indicated that the mean changes in oxy-Hb were significantly larger during a direct gaze than with an averted gaze (Bonferroni test, p = 0.007). These findings indicate that hemodynamic responses increased in response to the partner’s direct gaze in the mPFC (Fig. 5b). Furthermore, we analyzed effects of subject age (days) on these hemodynamic responses in the mPFC. Statistical analyses by Pearson’s correlation coefficients indicated that there was no significant correlation between the hemodynamic responses in the mPFC and subject ages (direct condition, r = -0.26, p = 0.44; averted condition, r = -0.41, p = 0.21). These results indicated that the hemodynamic responses to the direct gaze in the mPFC were not dependent on subject ages. In contrast, there were no significant differences in hemodynamic responses in the L-lPFC (Fig. 5c); no significant main effects of period (F1,10 = 1.58, p = 0.24) and gaze condition (F1,10 = 0.760, p = 0.40), nor significant interaction (F1,10 = 0.339, p = 0.57). Furthermore, the same data in the oxy-Hb reactive time window were analyzed based on effect sizes (Fig. 6). The mean effect sizes of hemodynamic responses were significantly larger in the direct gaze condition than the averted condition in the mPFC (p = 0.006). In addition, the mean effect sizes of the hemodynamic responses in the direct gaze condition were significantly larger in the mPFC than the other regions [R-lPFC (p = 0.05) and L-lPFC

L-lPFC 8 Direct 6

Averted

4 2 0 -2

Baseline

Face

Face

(*p \ 0.05). Furthermore, the partner’s direct gaze, rather than averted gaze, significantly increased the mean oxy-Hb changes ( p \ 0.01). c In the L-lPFC, no significant differences in oxy-Hb changes were observed. R-lPFC right dorsolateral prefrontal cortex, mPFC anterior dorsomedial prefrontal cortex, L-lPFC left dorsolateral prefrontal cortex

a

9 8

Averted Direct

7

Effect sizes

R-lPFC



A

b

6

c

5 4 3 2 1 0 -1

R-lPFC

mPFC

L-lPFC

Fig. 6 Effect sizes of hemodynamic responses (oxy-Hb changes) during the reactive time window. Note that effect size in the direct gaze condition in the mPFC is significantly greater than that in the averted gaze condition in mPFC, and also greater than those in the direct gaze condition in the other areas (R-lPFC and L-lPFC). a p \ 0.01, bp = 0.05, cp \ 0.05

(p = 0.033)]. These results indicated that activity in the mPFC was enhanced by direct gaze in the ‘‘peek-a-boo’’ play. We also analyzed the changes in deoxy-Hb. Statistical analyses revealed that there was a significant main effect of period (F1,10 = 6.18, p = 0.032) with no significant main effect of gaze condition (F1,10 = 1.73, p = 0.218) nor significant interaction between gaze condition and period (F1,10 = 1.07, p = 0.326) in the mPFC. These findings indicate that the deoxy-Hb changes decreased in the reactive time window regardless of gaze direction. In the other PFC areas, there was no significant difference (data not shown).

Discussion The aim of this study was to analyze hemodynamic responses in the infant PFC to composite (visual and

123

Brain Topogr

auditory) ostensive stimuli directed to infants during the natural social interactive play. It is emphasized that (i) we applied a real human partner instead of a video to the social interactive play since infants were reported to be more sensitive to a real human model than a video (see Introduction), and (ii) eye tracking behaviors of the infants were simultaneously recorded along with measurement of hemodynamic responses during the social interactive play. The results indicated that the infants spent more time looking to the eye region of the partner’s face with direct gaze, and that hemodynamic activity in the mPFC was more prominently increased in the direct gaze condition under the ostensive auditory stimuli while hemodynamic activity of the R-lPFC was increased during the social interactive play regardless of gaze direction.

difference in age of the infants; the infants were older in the present (7 months old) than in the previous (5 months old) study. Since PFC responses to social stimuli in 5-month-old infants were different from those of adults (Grossmann et al. 2010), the PFC in 7-month-old infants might be more similar to the adult PFC. Another difference is that the stimuli were more naturalistic in the present study; instead of a video, a real female experimenter was used as the partner. This suggests that the visual stimulus might be more ostensive in the present study. Further studies would be required to investigate these possibilities. The present results at least indicate that the mPFC could process composite auditory and visual ostensive signals.

Hemodynamic Activity in the mPFC

Although the mPFC has been shown to play a fundamental role in social cognitive abilities in adults (Amodio and Frith 2006), very little information is available on social cognition during early infant development (Grossmann 2013a). In the adult, the anterior part of the mPFC including has reciprocal connections with brain regions that are involved in emotional processing, memory formation, and higher cognitive and sensory regions. Because of such broad anatomical connections, the mPFC has been shown to play a wide range of roles in social cognitive functions such as interactive communication, personal perception, and theory of mind (Amodio and Frith 2006). Previous neurophysiological studies have demonstrated that not only visual stimuli directed to infants (direct gaze) but also infant-directed auditory ostensive stimuli activated the PFC, including the mPFC (Grossmann et al. 2010; Parise and Csibra 2013). The present study indicated that composite (visual and auditory) stimuli, both of which were directed to infants, activated the mPFC. Furthermore, the averted gaze with the ostensive auditory stimulus also induced weak activation in the mPFC. It is noted that an averted gaze without ostensive auditory stimulus did not elicit responses in the infant PFC (Grossmann et al. 2007). Psychological studies have reported that auditory ostensive stimuli enhanced ostensiveness of visual stimuli (Greenfield 1972; Stoyanova et al. 2010). These present and previous results suggest that the infant mPFC also plays an important role in interpreting social signals directed towards infants to initiate social communication. It has been reported that infants were more likely to follow the adult’s gaze when such an act was preceded by eye contact (adult’s direct gaze) compared with averted gaze (Senju and Csibra 2008). The present results along with these previous findings suggest that the mPFC might constitute an important part of neural circuits for gaze following, and direct gaze during face-to-face communication might induce gaze following by activating the mPFC.

In the present study, a direct gaze, rather than an averted gaze, significantly increased hemodynamic responses in the mPFC with an auditory ostensive stimulus. Consistent with the NIRS results, a direct gaze, rather than an averted gaze, significantly increased fixation time on the eye region of the partner. These results are consistent with those of previous studies in which infants responded differently to changes in facial features during ‘‘peek-a-boo’’ (Elsabbagh et al. 2014; Montague and Walker-Andrews 2001), and confirm that direct gaze powerfully modulates infants’ attentional focus and even cues it to the eye region (see Introduction). This increase in fixation time on the direct gaze may also explain increased activation of the mPFC region in the direct gaze condition compared to the averted gaze condition, since the mPFC is reported to be sensitive to direct gaze (see Introduction). Furthermore, a video of a mother smiling at infants (around 12 month old) induced greater activity in the mPFC compared to a video of a stranger smiling at infants (Minagawa-Kawai et al. 2009), suggesting that the communicative signals are also enhanced by familiarity of the presenter. Taken together, mPFC activity reflects ostensive communicative signals, which might be further modulated by previous experience of infants. In contrast with the present study, a previous study showed that there were no significant differences in gamma oscillation nor event related potentials between direct and averted gazes, when these visual stimuli were presented together with ostensive auditory stimuli (infant-directed speech) (Parise and Csibra 2013). However, the same study showed that there were significant differences between direct and averted gazes when these visual stimuli were presented without auditory stimuli (Parise and Csibra 2013). The differences between the previous (Parise and Csibra 2013) and present study might be ascribed to the

123

Role of the mPFC

Brain Topogr

Hemodynamic Activity in the R-lPFC Activity in the R-lPFC significantly increased during the reactive time window regardless of the partner’s gaze direction, suggesting that the R-lPFC might respond to the facial stimuli regardless of gaze direction. However, it is also possible that the R-lPFC responded to the ostensive auditory stimuli (‘‘baa, baa’’) presented during this period, but not to the facial stimuli. However, it is unlikely since mean peak oxy-Hb values during the reactive time window (time 3 to 10 s) with the visual (facial) and auditory stimuli were significantly greater than those in the preceding period from 0 to 3 s after the start of the voice with the auditory stimuli only (‘‘inai inai’’) (paired t test; p = 0.003 in the direct condition; p = 0.019 in the averted condition). These results suggest that the facial stimuli increased hemodynamic activity in the R-lPFC at least under the ostensive auditory stimuli. The present results, in which the R-lPFC and mPFC were activated in the peek-a-boo game, were not consistent with the previous studies in which the L-lPFC was activated (Grossmann et al. 2010, 2013; Grossmann and Johnson 2010). Previous studies in adults also reported inconsistent results regarding laterality of the activated areas in the PFC; one study reported L-lPFC activation (Druzgal and D’Esposito 2001) while other studies reported R-lPFC activation (Bhatt et al. 2009; Opitz et al. 2000) depending on the tasks imposed in the studies although face images were similarly used in these studies. The differences between the present and previous studies might be ascribed to the difference in the task designs and characteristics of the stimuli. The present study presented the naturalistic stimuli (i.e., real human partner), while the previous studies presented a video (Grossmann et al. 2010, 2013; Grossmann and Johnson 2010). Furthermore, the present study analyzed the PFC activity in the peek-a-boo game while the previous studies analyzed PFC activity during more complex behaviors such as joint attention or gaze following in which the partner moved eyes and turned the head (Grossmann et al. 2010, 2013). These differences in tasks and stimuli may underlie these differences in activation patterns between the present and previous studies. Possible Role of the R-lPFC The above results are consistent with those of previous studies. Neuroimaging studies suggest that the right hemisphere might be more involved in face perception not only in adults (Tong et al. 2000), but also in infants (Grossmann et al. 2008; Nakato et al. 2009; Otsuka et al. 2007). Recent studies have emphasized the role of the right temporal and PFC in processing dynamic facial changes such as eye gaze, facial expression, and facial orientation (Allison et al. 2000; Grossmann et al. 2008). The present

results also showed that hemodynamic responses in the R-lPFC increased following the dynamic change of facial stimuli (covering the face by hand and re-exposing the face accompanied with the voice). The present data are consistent with this idea that the R-lPFC processes dynamic facial changes regardless of gaze direction. Further studies are required to prove or disprove this idea.

Conclusion In the present study, we performed the interactive play ‘‘peek-a-boo’’ with as much similarity to a real life situation as possible. In the R-lPFC, hemodynamic responses significantly increased during the partner’s face exposure regardless of their gaze direction under the ostensive auditory stimuli. In contrast, the partner’s direct gaze, rather than an averted gaze, significantly enhanced hemodynamic responses during face exposure in the mPFC. These differences in response characteristics between the mPFC and R-lPFC might be associated with the functional differentiation between the mPFC and lPFC. The mPFC is generally associated with social and affective (hot) processes while the lPFC is associated with emotionally neutral and cognitive (cold) processes (Zelazo and Mu¨ller 2002; Grossmann 2013b). The eye-tracking data are consistent with this idea in that the eye region with direct gaze attracted infant’s attention, which is an important process to initiate social communication while hemodynamic activity in the mPFC was increased. Thus, the present results suggest that, 7-month-old infants, the mPFC plays an important role in communicative social interaction by processing ostensive visual stimuli (direct gaze with eye contact) under ostensive auditory stimuli. These results provide the first evidence showing a distinct role of the infant mPFC in mutual gaze perception under ostensive auditory stimuli, while the R-lPFC plays a role in face perception. However, the number of the samples was relatively small (n = 11), and further studies are required to confirm the results. Acknowledgments We thank Ms. Natsuko Sakai for special assistance in the ‘‘peek-a-boo’’ task. This research was supported in part by the JSPS Asian Core Program, the Ministry of Education, Science, Sports and Culture, Grants-in-Aid for Scientific Research (B) (25290005), and the MEXT, Japan. Mr. Akihiro Ishikawa is an employee of the company, which made the NIRS apparatus used in the present study. The other authors of this article have no conflicts of interest to declare.

References Allison T, Puce A, McCarthy G (2000) Social perception from visual cues: role of the STS region. Trends Cogn Sci 4:267–278

123

Brain Topogr Amodio DM, Frith CD (2006) Meeting of minds: the medial frontal cortex and social cognition. Nat Rev Neurosci 7:268–277 Baron-Cohen S (1995) Mind blindness: an essay on autism and theory of mind. The MIT Press, London Bhatt S, Mbwana J, Adeyemo A, Sawyer A, Hailu A, Vanmeter J (2009) Lying about facial recognition: an fMRI study. Brain Cogn 69:382–390 Bryson SE, Zwaigenbaum L, Roberts W (2004) The early detection of autism in clinical practice. Paediatr Child Health 9:219–221 Cristia A, Dupoux E, Hakuno Y, Lloyd-Fox S, Schuetze M, Kivits J, Bergvelt T, van Gelder M, Filippin L, Charron S, MinagawaKawai Y (2013) An online database of infant functional near infrared spectroscopy studies: a community-augmented systematic review. PLoS One 83:e58906 Csibra G, Gergely G (2006) Social learning and social cognition: the case for pedagogy. In: Munakata Y, Johnson MH (eds) Processes of change in brain and cognitive development. Attention and performance XXI. Oxford University Press, New York, pp 249–274 Csibra G, Henty J, Volein A, Elwell C, Tucker L, Meek J, Johnson MH (2004) Near infrared spectroscopy reveals neural activation during face perception in infants and adults. J Pediatr Neurol 2:85–89 Druzgal TJ, D’Esposito M (2001) A neural network reflecting decisions about human faces. Neuron 32:947–955 Elsabbagh M, Bedford R, Senju A, Charman T, Pickles A, Johnson MH; The BASIS Team (2014) What you see is what you get: contextual modulation of face scanning in typical and atypical development. Soc Cogn Affect Neurosci 9:538–543 Farroni T, Csibra G, Simion F, Johnson MH (2002) Eye contact detection in humans from birth. Proc Natl Acad Sci USA 99:9602–9605 Gilbert SJ, Williamson ID, Dumontheil I, Simons JS, Frith CD, Burgess PW (2007) Distinct regions of medial rostral prefrontal cortex supporting social and nonsocial functions. Soc Cog Affect Neurosci 2:217–226 Goren CC, Sarty M, Wu PYK (1975) Visual following and pattern discrimination of face-like stimuli by newborn infants. Pediatrics 56:544–549 Greenfield PM (1972) Playing peekaboo with a four-month-old: a study of the role of speech and nonspeech sounds in the formation of a visual schema. J Psychol 82:287–298 Grossmann T (2013a) The role of medial prefrontal cortex in early social cognition. Front Hum Neurosci 7:340 Grossmann T (2013b) Mapping prefrontal cortex functions in human infancy. Infancy 18:303–324 Grossmann T, Johnson MH (2010) Selective prefrontal cortex responses to joint attention in early infancy. Biol Lett 6:540–543 Grossmann T, Johnson MH, Farroni T, Csibra G (2007) Social perception in the infant brain: gamma oscillatory activity in response to eye gaze. Soc Cogn Affect Neurosci 2:284–291 Grossmann T, Johnson MH, Lloyd-Fox S, Blasi A, Deligianni F, Elwell C, Csibra G (2008) Early cortical specialization for faceto-face communication in human infants. Proc Biol Sci 275:2803–2811 Grossmann T, Parise E, Friederici AD (2010) The detection of communicative signals directed at the self in infant prefrontal cortex. Front Hum Neurosci 4:1–5 Grossmann T, Lloyd-Fox S, Johnson MH (2013) Brain responses reveal young infants’ sensitivity to when a social partner follows their gaze. Dev Cogn Neurosci 6:155–161 Hains SM, Muir DW (1996) Infant sensitivity to adult eye direction. Child Dev 67:1940–1951 Hoshi Y, Kobayashi N, Tamura M (2001) Interpretation of nearinfrared spectroscopy signals: a study with a newly developed perfused rat brain model. J Appl Physiol 90:1657–1662

123

Jasper HH (1958) The ten-twenty electrode system of the International Federation. Electroencephalogr Clin Neurophysiol 10:371–375 Johnson MH, Dziurawiec S, Ellis H, Morton J (1991) Newborns’ preferential tracking of face-like stimuli and its subsequent decline. Cognition 40:1–19 Kampe KK, Frith CD, Frith U (2003) ‘‘Hey John’’: signals conveying communicative intention toward the self activate brain regions associated with ‘‘mentalizing’’, regardless of modality. J Neurosci 23:5258–5263 Kanwisher N, McDermott J, Chun MM (1997) The fusiform face area: a module in human extrastriate cortex specialized for face perception. J Neurosci 17:4302–4311 Kleinke CL (1986) Gaze and eye contact: a research review. Psychol Bull 100:78–100 Kuzmanovic B, Georgescu AL, Eickhoff SB, Shah NJ, Bente G, Fink GR, Vogeley K (2009) Duration matters: dissociating neural correlates of detection and evaluation of social gaze. Neuroimage 46:1154–1163 Lloyd-Fox S, Blasi A, Volein A, Everdell N, Elwell CE, Johnson MH (2009) Social perception in infancy: a near infrared spectroscopy study. Child Dev 80:986–999 Lloyd-Fox S, Blasi A, Everdell N, Elwell CE, Johnson MH (2011) Selective cortical mapping of biological motion processing in young infants. J Cogn Neurosci 23:2521–2532 Lloyd-Fox S, Blasi A, Elwell CE, Charman T, Murphy D, Johnson MH (2013) Reduced neural sensitivity to social stimuli in infants at risk for autism. Proc Biol Sci 280:20123026 Meek JH, Firbank M, Elwell CE, Atkinson J, Braddick O, Wyatt JS (1998) Regional hemodynamic responses to visual stimulation in awake infants. Pediatr Res 43:840–843 Minagawa-Kawai Y, Matsuoka S, Dan I, Naoi N, Nakamura K, Kojima S (2009) Prefrontal activation associated with social attachment: facial-emotion recognition in mothers and infants. Cereb Cortex 19:284–292 Montague DP, Walker-Andrews AS (2001) Peekaboo: a new look at infants’ perception of emotion expressions. Dev Psychol 37:826–828 Nakato E, Otsuka Y, Kanazawa S, Yamaguchi MK, Watanabe S, Kakigi R (2009) When do infants differentiate profile face from frontal face? A near-infrared spectroscopic study. Hum Brain Mapp 30:462–472 Nakato E, Otsuka Y, Kanazawa S, Yamaguchi MK, Honda Y, Kakigi R (2011) I know this face: neural activity during mother’s face perception in 7- to 8-month-old infants as investigated by nearinfrared spectroscopy. Early Hum Dev 87:1–7 Opitz B, Mecklinger A, Friederici AD (2000) Functional asymmetry of human prefrontal cortex: encoding and retrieval of verbally and nonverbally coded information. Learn Mem 7(2):85–96 Otsuka Y, Nakato E, Kanazawa S, Yamaguchi MK, Watanabe S, Kakigi R (2007) Neural activation to upright and inverted faces in infants measured by near infrared spectroscopy. Neuroimage 34:399–406 Parise E, Csibra G (2013) Neural responses to multimodal ostensive signals in 5-month-old infants. PLoS One 8:72360 Phillips W, Baron-Cohen S, Rutter M (1992) The role of eye-contact in the detection of goals: evidence from normal toddlers, and children with autism or mental handicap. Dev Psychopathol 4:375–383 Puce A, Allison T, Asgari M, Gore JC, McCarthy G (1996) Differential sensitivity of human visual cortex to faces, letterstrings, and textures: a functional magnetic resonance imaging study. J Neurosci 16:5205–5215 Reid VM, Striano T (2005) Adult gaze influences infant attention and object processing: implications for cognitive neuroscience. Eur J Neurosci 21:1763–1766

Brain Topogr Schilbach L, Wohlschlaeger AM, Kraemer NC, Newen A, Shah NJ, Fink GR, Vogeley K (2006) Being with virtual others: neural correlates of social interaction. Neuropsychologia 44:718–730 Schroeter ML, Zysset S, Kruggel F, von Cramon DY (2003) Age dependency of the hemodynamic response as measured by functional near-infrared spectroscopy. Neuroimage 19:555–564 Senju A, Csibra G (2008) Gaze following in human infants depends on communicative signals. Curr Biol 18:668–671 Stoyanova RS, Ewbank MP, Calder AJ (2010) ‘‘You talkin’ to me?’’ Self-relevant auditory signals influence perception of gaze direction. Psychol Sci 21:1765–1769 Strangman G, Culver JP, Thompson JH, Boas DA (2002) A quantitative comparison of simultaneous BOLD fMRI and NIRS recordings during functional brain activation. NeuroImage 17:719–731 Strouse GA, Troseth GL (2008) ‘‘Don’t try this at home’’: toddlers’ imitation of new skills from people on video. J Exp Child Psychol 101:262–280 Taga G, Homae F, Watanabe H (2007) Effects of source-detector distance of near infrared spectroscopy on the measurement of the cortical hemodynamic response in infants. NeuroImage 38:452–460 Takamoto K, Hori E, Urakawa S, Katayama M, Nagashima Y, Yada Y, Ono T, Nishijo H (2013) Thermotherapy to the facial region in and around the eyelids altered prefrontal hemodynamic responses and autonomic nervous activity during mental arithmetic. Psychophysiology 50:35–47

Takeuchi M, Hori E, Takamoto K, Tran AH, Satoru K, Ishikawa A, Ono T, Endo S, Nishijo H (2009) Brain cortical mapping by simultaneous recording of functional near infrared spectroscopy and electroencephalograms from the whole brain during right median nerve stimulation. Brain Topogr 22:197–214 Tong F, Nakayama K, Moscovitch M, Weinrib O, Kanwisher N (2000) Response properties of the human fusiform face area. Cogn Neuropsychol 17:257–280 Valenza E, Simion F, Macchi Cassia V, Umilta` C (1996) Face preference at birth. J Exp Psychol Hum Percept Perform 22:892–903 Wang Y, Ramsey R, de C Hamilton AF (2011) The control of mimicry by eye contact is mediated by medial prefrontal cortex. J Neurosci 31:12001–12010 Yamamoto T, Kato T (2002) Paradoxical correlation between signal in functional magnetic resonance imaging and deoxygenated hemoglobin content in capillaries: a new theoretical explanation. Phys Med Biol 47:1121–1141 Zelazo PD, Mu¨ller U (2002) Executive function in typical and atypical development. In: Goswami U (ed) Handbook of childhood cognitive development. Blackwell, Oxford, pp 445–469 Zwaigenbaum L, Bryson S, Rogers T, Roberts W, Brian J, Szatmari P (2005) Behavioral manifestation of autism in the first year of life. Int J Dev Neurosci 23:143–152

123