Abstract

Serotonin is known to play an important role not only in regulating emotional behaviors, but also in the formation of social behavior traits. To determine the location and serotonin function of brain areas involved in social behavior traits, we tested serotonin transporter (SERT) binding and neural activity linked with the social behaviors of common marmosets with positron emission tomography using [11C]-3-amino-4-(2-dimetylaminomethyl-phenylsulfanyl)-benzonitrile and [18F]fluorodeoxyglucose, respectively. Factor analysis of behavioral measures during a direct encounter between unfamiliar adult males identified three classes of social behavioral traits: (1) aggressive, (2) anxious, and (3) unfriendly (opposite of friendly). Voxel-based analysis revealed a significant association between SERT binding with the social behavioral traits in the midline cortical subregions. Aggressive and friendly traits are localized to the posterior cingulate cortex, and the anxious trait is localized to the anterior cingulate cortex. In addition, neural activity and functional connectivity of the posterior and anterior cingulate cortices appear to be altered depending on the social situation. These results suggest that the midline cortical serotonergic system is crucial in social behavior traits and its subregions are functionally segregated in socio-emotional processing.

Introduction

A growing body of evidence highlights a role for serotonin in regulating emotional behaviors in humans and nonhuman primates. The role of serotonin has been demonstrated by studying genetic variations (Lesch et al. 1996; Hariri et al. 2002; Sen et al. 2004; Nomura et al. 2006), making neurochemical measurements (Coccaro 1992; Mehlman et al. 1994, 1995; Fairbanks et al. 2001), and manipulating serotonin (Cleare and Bond 1995; Moeller et al. 1996; Knutson et al. 1998; Fairbanks et al. 2001; Tse and Bond 2002; Cools et al. 2005; Anderson et al. 2007). Some studies have shown an inverse relationship between central serotonergic indices and anxiety temperament or impulsive-aggressive behavior (Coccaro 1992; Mehlman et al. 1994, 1995; Cleare and Bond 1995; Lesch et al. 1996; Fairbanks et al. 2001; Hariri et al. 2002; Sen et al. 2004; Nomura et al. 2006). Neuroimaging studies using specific positron emission tomography (PET) tracers have reported that the serotonergic neurotransmission in the thalamus and the prefrontal cortex is related to anxiety in healthy subjects (Tauscher et al. 2001; Takano et al. 2007). Although there is evidence that serotonin is involved in the expression of negative emotion and behaviors of social construction such as social affiliation, dominance, and trait openness (Mehlman et al. 1995; Knutson et al. 1998; Tse and Bond 2002; Kalbitzer et al. 2009), no direct correlation has been demonstrated between these behaviors and regional serotonin function. Several studies have focused on the amygdala–serotonin system in the anxious temperament (Oler et al. 2009) and the perception of others' emotional expression (Hariri et al. 2002; Cools et al. 2005; Anderson et al. 2007); however, it is unclear whether the amygdala is the only brain region in which the serotonergic system would affect or regulate socio-cognitive behaviors.

Recently, it has been suggested that social information processing is involved not only in subcortical limbic regions including amygdala, but also in the cerebral cortex. The midline cortical structures including the cingulate cortex and medial frontal cortex have been implicated as parts of a network for emotional processing and self-social cognition (Northoff and Bermpohl 2004; Amodio and Frith 2006; Buckner and Carroll 2007; Olsson and Ochsner 2008; Etkin et al. 2011). The anterior parts of these structures, for example, the anterior cingulate cortex (ACC), seems to be involved in negative aspects of emotion such as anxiety or pain (Amodio and Frith 2006; Etkin et al. 2011), whereas more posterior parts such as the posterior cingulate cortex (PCC) are likely to be involved in integrating self-referential stimuli in the context of one's own person (Northoff and Bermpohl 2004; Buckner and Carroll 2007). Although it is not well understood how the midline cortical areas process social information in association with serotonergic function, several lines of evidence suggest that the cortical serotonergic system affects social information processing. In rodents, serotonin was reported to regulate anxiety-like behaviors by acting not only in the deep brain (e.g., raphe nucleus) but also in the cortex (Gross et al. 2002; Weisstaub et al. 2006) and also to exert a powerful influence on the excitability of prefrontal neurons (Zhong and Yan 2011). In addition, in anxious (Lanzenberger et al. 2007) and autistic (Murphy et al. 2006; Nakamura et al. 2010) human subjects who suffered from impairments of social cognition, decreased binding for serotonin receptors or transporters has been shown in midline cortical regions such as the ACC or the PCC compared with normal controls.

In this study, we tested whether cortical serotonin transporter (SERT) binding and neural activity are associated with social behavior in adult male common marmosets, which are known as a primate with complex social behaviors. Marmosets have a high level of responsiveness to the behavior of others (Burkart, Hrdy et al. 2009; Pesendorfer et al. 2009) and spontaneous prosociality relying on cooperative breeding (Burkart et al. 2007, Burkart, Hrdy et al. 2009a), similar to humans. First, we evaluated the social behaviors of animals when they directly encountered unfamiliar conspecifics and identified three independent social behavior traits by factor analysis of behavioral measures. Then the regional serotonin function that is correlated with the behavioral traits was evaluated using PET with a specific probe for the SERT, [11C]-3-amino-4-(2-dimetylaminomethyl-phenylsulfanyl)-benzonitrile ([11C]DASB). Finally, after social intervention, we used PET with [18F]fluorodeoxyglucose ([18F]FDG) to obtain an index of neuronal activity (Reivich et al. 1979) in cortical regions in which SERT binding was correlated with behavioral traits.

Materials and Methods

Subjects

We used 12 male common marmosets (Callithrix jacchus) with age ranging from 2.5 to 5 years for behavioral assessment. The same subjects underwent PET experiments (see “Housing condition” in the Supplementary material and Table 3). The experimental protocol (Fig. 1A) was as follows: (1) The social behavior of 12 animals was observed during an encounter with other unfamiliar marmosets; (2) 8 of the 12 animals received PET scans for SERT binding under conscious resting conditions; and (3) 6 of the 8 animals were scanned for cerebral neural activity using PET imaging in three different social contexts: alone, unfamiliar, and familiar. Animals were maintained and handled in accordance with the recommendations of the United States National Institutes of Health, and all procedures were approved by the Animal Care and Use Committee of the Kobe Institute of Riken (MAH18-03-6).

Table 3

Peak voxels and number of voxels obtained from regression analysis of SERT binding with factor scores

 Direction of relationship Regions/hemisphere (x,y,z) Standard spacea, mm Voxelsb Maximum Zc 
F1: Aggressive Positive – – – – 
Negative Posterior cingulate area/R (0.2, −9.4, 5.6) 1490 5.16 
F2: Anxious Positive – – – – 
Negative Anterior cingulate area/L (−0.4, 5.2, 3.8) 371 4.74 
Lateral prefrontal area/R (9.0, 4.0, 1.0) 140 5.05 
F3: Unfriendly Positive Posterior cingulate area/L (−1.8, −6.6, 7.6) 610 5.24 
Negative – – – – 
 Direction of relationship Regions/hemisphere (x,y,z) Standard spacea, mm Voxelsb Maximum Zc 
F1: Aggressive Positive – – – – 
Negative Posterior cingulate area/R (0.2, −9.4, 5.6) 1490 5.16 
F2: Anxious Positive – – – – 
Negative Anterior cingulate area/L (−0.4, 5.2, 3.8) 371 4.74 
Lateral prefrontal area/R (9.0, 4.0, 1.0) 140 5.05 
F3: Unfriendly Positive Posterior cingulate area/L (−1.8, −6.6, 7.6) 610 5.24 
Negative – – – – 

aThe coordinates for the anterior commissure on the ac-pc line in the standard MRI of the common marmoset brain prepared in our laboratory.

b1 voxel = (0.2)3 mm; clusters with voxels >125 (1.0 mm³).

cVoxel level corrected to P < 0.05.

Figure 1.

Entire time course of experiment (A) and details of timeline of PET scans for [11C]DASB and [18F]FDG (B). While four subjects were omitted for [11C]DASB-PET and an additional two for [18F]FDG-PET, the same subjects were underwent all procedures. All the PET scans were performed under a conscious state. Dynamic PET data were obtained for 90 min immediately after the [11C]DASB injection to derive parametric images of binding potentials. Static PET data were obtained for 30 min as an index of local cerebral neural activity during a freely moving period of around 30 min after the [18F]FDG injection.

Figure 1.

Entire time course of experiment (A) and details of timeline of PET scans for [11C]DASB and [18F]FDG (B). While four subjects were omitted for [11C]DASB-PET and an additional two for [18F]FDG-PET, the same subjects were underwent all procedures. All the PET scans were performed under a conscious state. Dynamic PET data were obtained for 90 min immediately after the [11C]DASB injection to derive parametric images of binding potentials. Static PET data were obtained for 30 min as an index of local cerebral neural activity during a freely moving period of around 30 min after the [18F]FDG injection.

Encounter Trials

Using a group of 12 marmoset monkeys, we obtained a total of 60 encounter trials between two marmosets. These trials were in a round-robin format except for those between cage-mates and provided 120 data sets in which each animal faced 10 different unfamiliar animals. Unfamiliar animals were defined as those whom a subject of interest has never lived with in the same cage and visual access between them was restricted in the room in which they were housed. Tests were performed in the experimental room in the afternoon (13:00–15:00), with all animals undergoing one trial per day with a minimum intertrial interval of 2 days. For preparation before direct encounters, three pretest habituation phases were designed to avert excessive excitement of animals during the test trial. First, each animal, confined in a transfer box, was placed in a new cage with the same shape as its home cage for 5 min. Next, one animal was freed into the cage and another was boxed in for 5 min, and then handling of the animals was reversed for the next 5 min. Following the pretest habituation phases, the two animals had a direct encounter in the cage for 5 min as a test trial, which was videotaped via unattended recording. Of the 60 total trials, three were terminated prematurely because the animals fought one another continuously for more than 5 s. Within the remaining 57 trials, there were winners and losers in 29 trials determined by the winner's call of chatter or the loser's call of submission squeal (Epple 1968; Bezerra and Souto 2008). However, we did not find a winner or loser effect, in which animals that win or lose a competition win or lose subsequent ones over repeated encounters. There was no absolute winner or loser among these animals throughout the trials.

Identification of Traits of Social Behavior

We measured 17 rigorously defined items of behavior observed in a frame-by frame analysis of the test trial (Table 1), by reference to the previously reported ethograms of the common marmoset (Stevenson and Poole 1976; Harrison and Tardif 1989). We compiled the total time spent engaging in each behavior (Table 1). Durations of all behaviors were scored by a single rater via videotape analysis with the aid of behavioral observation software (Observer VT version 5.0; Noldus Information Technology, Wageningen, the Netherlands). Using Pearson's correlation coefficient, we estimated the intrarater reliability for 10 randomized trials, which were reevaluated by the same examiner after 1 month (r = 0.99 on average). For identifying the traits for each animal, we confirmed two types of test–retest reliability, that is, “variable-oriented consistency” as a correlation of between-individual variation across trials and “individual-oriented consistency” as a correlation of the individual profile of behavior across trials. They are estimated by Pearson's correlation coefficient using the odd–even split-half method, according to the basic concepts of identifying individual behavioral phenotypes (Uher 2011). Within all behavior items, we selected those with high scores of variable-oriented consistency (>0.576; P < 0.05) (Supplementary Table S1). Before entering factor analysis, the reliable behavioral measures of nine items were averaged for 10 trials individually, and the measure of sampling adequacy (MSA) using them revealed two items with very low MSA scores (<0.25), which were omitted for the following factor analysis (Supplementary Table S1). Ultimately, the factor analysis was conducted for the seven behavior items, that is, approach and contact, attack, chase, confrontation, locomotion, sniff perch & scent mark, and scent mark (Table 1). Using the selected seven items, we observed a high level of individual-oriented consistency (Pearson's correlation: r = 0.89 on average) (Supplementary Table S2). To identify the factors which influence these behaviors, maximum likelihood estimation analysis and varimax orthogonal rotation were applied, and then a score for each factor and for each animal was estimated by regression methods. Statistical analysis for behavioral measurements was performed using the statistical package R version 2.8.1 (http://www.r-project.org/). These factor scores for individuals were used as independent variables for statistical analysis of the [11C]DASB-PET images (Supplementary Table S3).

Table 1

Behaviors observed during encounter trials

Behaviorsa Description Time spent of expressionb 
Approach and contact Behavior of approach and subsequent soft body contact to another individual 54.78 ± 64.77 
Approach and sniff Behavior of approach and subsequent sniff to another individual 21.12 ± 25.82 
Attack Grabbing or attempting to grab, scratch, or bite another individual unilaterally 0.38 ± 1.55 
Backflip Rapid move of 360° rotations in the backward direction 1.89 ± 8.45 
Chase Locomotion* in which the animal follows another individual 6.32 ± 9.57 
Confrontation Stand facing another individual 3.17 ± 16.47 
Defecation Void solid or semisolid feces 0.32 ± 1.42 
Display Raising the tail to display the genitals 0.32 ± 2.10 
Drink Drink water from an automatic water dispenser 0.34 ± 1.80 
Escape Locomotion* in which the animal escapes from another individual 3.03 ± 9.23 
Fight Exchange of grabbing or attempting to grab, scratch, or bite another individual bilaterally 3.41 ± 17.02 
LocomotionTravelling from one place to another including walking, running, climbing, and jumping (without direction to another individual) 29.00 ± 22.44 
Sniff perch Sniff a perch 3.68 ± 11.62 
Sniff perch and scent mark Sniff a perch and subsequent rubbing the body on it 5.57 ± 17.23 
Scent mark Rubbing the body on a perch, a cage wall, or floor 3.60 ± 7.22 
Sit Sit stationary alone on a perch or a cage wall or floor 150.22 ± 72.19 
Sniff Sniff a cage wall or floor 12.85 ± 13.77 
Behaviorsa Description Time spent of expressionb 
Approach and contact Behavior of approach and subsequent soft body contact to another individual 54.78 ± 64.77 
Approach and sniff Behavior of approach and subsequent sniff to another individual 21.12 ± 25.82 
Attack Grabbing or attempting to grab, scratch, or bite another individual unilaterally 0.38 ± 1.55 
Backflip Rapid move of 360° rotations in the backward direction 1.89 ± 8.45 
Chase Locomotion* in which the animal follows another individual 6.32 ± 9.57 
Confrontation Stand facing another individual 3.17 ± 16.47 
Defecation Void solid or semisolid feces 0.32 ± 1.42 
Display Raising the tail to display the genitals 0.32 ± 2.10 
Drink Drink water from an automatic water dispenser 0.34 ± 1.80 
Escape Locomotion* in which the animal escapes from another individual 3.03 ± 9.23 
Fight Exchange of grabbing or attempting to grab, scratch, or bite another individual bilaterally 3.41 ± 17.02 
LocomotionTravelling from one place to another including walking, running, climbing, and jumping (without direction to another individual) 29.00 ± 22.44 
Sniff perch Sniff a perch 3.68 ± 11.62 
Sniff perch and scent mark Sniff a perch and subsequent rubbing the body on it 5.57 ± 17.23 
Scent mark Rubbing the body on a perch, a cage wall, or floor 3.60 ± 7.22 
Sit Sit stationary alone on a perch or a cage wall or floor 150.22 ± 72.19 
Sniff Sniff a cage wall or floor 12.85 ± 13.77 

aBehaviors used in factor analysis, according to the test–retest reliability and the MSA scores for them, are given in boldface (see the text and Supplementary Table S1 in detail).

bMean ± SD (s) of mean values for 12 animals. The asterisk has same meaning.

[11C]DASB-PET

Eight of the 12 animals assessed by the behavioral testing received [11C]DASB-PET scans. The [11C]DASB-PET scans were performed in awake animals (Fig. 1B), according to a previously published method (Yokoyama et al. 2009). Briefly, we used a PET scanner, the microPET Focus 220 (Siemens Medical Solutions, Inc., Knoxville, TN). After cannulation into a tail vein, animals were fixed via a head holder (see “Preparation for PET scans” in the Supplementary material) in a sitting position with the scanner tilted to 45°. [11C]DASB were prepared according to previously described methods (Wilson et al. 2002), and radiochemical purities were >95% with specific activities of 44.8 ± 13.9 (mean ± SD) GBq/μmol. After a transmission scan with a 68Ge–68Ga pin source for 30 min for attenuation correction, scans were performed for a period of 90 min after a bolus injection of 54.6 ± 4.9 MBq/body of [11C]DASB. From the PET scan dynamic data, parametric images of binding potentials (BPND) of [11C]DASB were calculated with the two-parameter linearized reference tissue model (MRTM2) (Ichise et al. 2003; Innis et al. 2007) using software for analyzing biomedical images, PMOD version 3.0 (PMOD Technologies, Zurich, Switzerland). In this model, the value of k2′ was estimated by parametric analysis using the cerebellum as a reference region, with the midbrain and thalamus as rich regions.

[18F]FDG-PET

Six of the 8 animals that underwent the [11C]DASB-PET experiment received [18F]FDG-PET scans. We set three conditions that are distinct in terms of social situation of individual familiarity and conducted a total of 18 [18F]FDG-PET scans. We used another animal as a social stimulus in the experimental condition based on previous studies in marmosets (Cilia and Piper 1997; Kinnally et al. 2006), instead of the photographs of emotional faces widely used as stimuli in human functional magnetic resonance imaging (fMRI) studies (Hariri et al. 2002; Cools et al. 2005; Anderson et al. 2007; Gobbini and Haxby 2007). A timeline of the [18F]FDG-PET scan experiment is shown in Figure 1B. In the evening on the day preceding the [18F]FDG-PET scan, the test animal underwent a transmission scan for 30 min. The animal's position in the PET scanner was exactly duplicated for the subsequent [18F]FDG scan. Then the animal was placed alone overnight in a cage in another room different from the room it usually lived in, with a 12-h light/dark cycle. On the morning of the test day, animals were placed in the holder with a tail vein cannulated. Animals were placed in a test box immediately after a bolus injection of 75.6 ± 16.5 MBq/body of [18F]FDG. The test box was adjacent to an identical box containing an unfamiliar, familiar, or no animal in the experimental room. An unfamiliar animal was defined as an individual that had never been a cage mate (unfamiliar condition, UNF). A familiar animal was defined as one that the test animal had been housed with (familiar condition, FAM) and no animal in the adjacent cage was the alone condition (alone condition, ALN). The three PET scans were performed on different days with the order of conditions randomized across animals, and with a scan interval of 3 weeks. To identify the condition-dependent behavioral change during [18F]FDG uptake, we scored intrascan subject behavior such as self-playing, locomotion, and vocalizations which were overtly displayed by each animal for 30 min following the [18F]FDG injection (see “Behavior during [18F]FDG-PET” in the Supplementary material). After exposure to the experimental condition for 45 min, the test animal was transported and fixed in a sitting position via the head holder in the PET scanner. The PET scan was started 60 min after the [18F]FDG injection and lasted for 30 min. The static images of [18F]FDG accumulation were used for further analysis, as an index of local cerebral neural activity during a freely moving period of around 30 min after the [18F]FDG injection (Holschneider and Maarek 2004).

Statistical Analysis for PET Imaging

The PET images of both modalities ([11C]DASB BPND and [18F]FDG uptake) were deskulled manually, and coregistered between subjects in each modality by affine transformation with 12 degrees-of-freedom and a normalized correlation ratio cost function by means of FLIRT (FMRIB's Linear Image Registration Tool) of FSL (FMRIB's Software Library, http://www.fmrib.ox.ac.uk/fsl). The images were spatially smoothed using a 2.0 mm Gaussian kernel of full-width at half-maximum. The voxel-based statistics of PET images were performed using FEAT (FMRI Expert Analysis Tool, Version 5.98), part of FSL, based on general linear model using factor scores of behavioral traits as regressors for PET images of [11C]DASB BPND. For PET images of [18F]FDG uptake, the voxel-based statistics were performed using three conditions in a repeated-measures 1-factor 3-level analysis of variance (ANOVA) and the post hoc triplet paired t-test (see “Statistics” in the Supplementary material on design matrices set for PET images). All the results of voxel-based statistics were registered to the standard space of the template MRI image (see “MRI template” in the Supplementary material). The anatomical locations referred to a stereotaxic atlas (Stephan et al. 1980), and the subdivisions of the midline cortical structures were identified corresponding to recent histological reports (Palmer and Rosa 2006; Roberts et al. 2007; Burman and Rosa 2009).

In addition to voxel-based statistics, we applied regions of interest (ROIs) derived from the preceding [11C]DASB-behavior correlation analysis to the [11C]DASB-PET images and also to the coregistered [18F]FDG-PET images (see “Statistics” in the Supplementary material). The ROIs were the PCC including two adjoining areas, the ACC and the lateral prefrontal cortex (LPFC), coinciding with four clusters of voxels in which regional SERT binding significantly correlated with each of the social behavior traits (see Fig. 2). In the analysis using the coregistered [18F]FDG-PET images, two adjoining clusters of the PCC were combined into one ROI of the PCC, because they were nearly contiguous (see Fig. 3). Finally, we searched for functional connectivity of these three ROIs, the PCC, the ACC, and the LPFC. Using FDG-PET data, the activity of the seed ROI (R), the social situation (S), and also their interaction (R × S) were entered into the design matrix. Our interest was in the contrast of the interaction, which represents functional connectivity of the seed area, affected by social situations.

Figure 2.

Significant clusters in which SERT binding, shown in [11C]DASB BPND, was correlated with factor scores derived from behavioral analysis, are overlaid with the standard MRI brain images of marmosets (A). Clusters were detected in the PCC, the ACC, and the LPFC. These areas were designated as ROIs. Scatter graphs showed the correlation between residualized and standardized [11C]DASB BPND and one of the factor scores by the ROIs (B). The coordinates for the anterior commissure and Z-scores of all significant clusters are listed in Table 3.

Figure 2.

Significant clusters in which SERT binding, shown in [11C]DASB BPND, was correlated with factor scores derived from behavioral analysis, are overlaid with the standard MRI brain images of marmosets (A). Clusters were detected in the PCC, the ACC, and the LPFC. These areas were designated as ROIs. Scatter graphs showed the correlation between residualized and standardized [11C]DASB BPND and one of the factor scores by the ROIs (B). The coordinates for the anterior commissure and Z-scores of all significant clusters are listed in Table 3.

Figure 3.

Activated clusters for contrasts between different social situations in the [18F]FDG uptake are overlaid with the standard MRI brain images of marmosets (A). Two contrasts (UNF-ALN and FAM-ALN) and the overlapping parts of them are shown in the color legend, but no clusters for UNF-FAM were detected. The coordinates for the anterior commissure and Z-scores of all significant clusters are listed in Table 4. Functionally derived ROIs from the [11C]DASB–behavior correlation analysis, such as the posterior cingulate in which the two clusters were combined, the anterior cingulate, and the dorsolateral prefrontal areas are also presented in red on the standard brain, with symbols corresponding to those in Figure 2. Line graphs showed normalized [18F]FDG uptake values for each animal in the three conditions by the ROIs (B). Two-way ANOVA revealed a significant ROI by condition interaction, and repeated-measures ANOVA and post hoc Dunnett's test revealed a significant effect of condition between ALN and UNF in the posterior cingulate (indicated by asterisk). ALN, staying alone; FAM, staying with a familiar animal; UNF, staying with an unfamiliar animal.

Figure 3.

Activated clusters for contrasts between different social situations in the [18F]FDG uptake are overlaid with the standard MRI brain images of marmosets (A). Two contrasts (UNF-ALN and FAM-ALN) and the overlapping parts of them are shown in the color legend, but no clusters for UNF-FAM were detected. The coordinates for the anterior commissure and Z-scores of all significant clusters are listed in Table 4. Functionally derived ROIs from the [11C]DASB–behavior correlation analysis, such as the posterior cingulate in which the two clusters were combined, the anterior cingulate, and the dorsolateral prefrontal areas are also presented in red on the standard brain, with symbols corresponding to those in Figure 2. Line graphs showed normalized [18F]FDG uptake values for each animal in the three conditions by the ROIs (B). Two-way ANOVA revealed a significant ROI by condition interaction, and repeated-measures ANOVA and post hoc Dunnett's test revealed a significant effect of condition between ALN and UNF in the posterior cingulate (indicated by asterisk). ALN, staying alone; FAM, staying with a familiar animal; UNF, staying with an unfamiliar animal.

The statistical threshold was set at P < 0.05, a voxel level corrected for multiple comparisons for [11C]DASB-PET analysis of the across-subject correlations, or a cluster-level corrected for [18F]FDG-PET analysis of the social situation-dependent effects. For functional connectivity, the statistical threshold was set at P < 0.01 uncorrected (Z > 2.3). The latter two correction procedures for multiple testing permitting a relatively low statistical threshold were designed to explore the experimentally inferred area locations (Nichols and Poline 2009).

Results

Behavioral Characterization by Factor Analysis

We conducted factor analysis using the reliable behavioral measures of seven items of behavior responses during the direct encounter between unfamiliar adult male marmosets, which were averaged for 10 trials individually. The Kaiser–Meyer–Olkin measure was 0.59 and the Bartlett's test of sphericity was significant at P < 0.001 (χ2 = 69.11, df = 21), indicating the adequacy of the sample for factor analysis. Factor analysis identified three factors. The number of factors was determined depending on the decay pattern of the eigenvalues in the correlation matrix. The model showed goodness-of-fit from a test of the null hypothesis that three factors are sufficient (P = 0.389, χ2 = 3.0, df = 3). The percent variance explained by the first factor is 41.1%, 25.0% by the second, 13.4% by the third factor, and the total explained variance is 79.5% (Table 2). Factor 1 was referred to as the aggressive trait, because it was associated with chase, attack, and confrontation behaviors toward the other marmoset. Factor 2 was referred to as the anxious trait, because it was associated with scent-marking behaviors. Although wild common marmosets are known to exhibit scent-marking behavior as informative behavior rather than as dominance-related behavior (Lazaro-Perea et al. 1999; Heymann 2006), in a laboratory setting, common marmosets display scent-marking behavior frequently in the context of fearful and anxious situations. Scent-marking behavior is reportedly reduced by administration of anxiolytic agents (Cilia and Piper 1997; Barros et al. 2001). Factor 3 was referred to as an unfriendly trait, the opposite of friendly, because it was negatively associated with approach and subsequent soft body contact toward the other adult male marmoset with apparent social intimacy.

Table 2

Factor loadings of behavior items during the encounter trials used in factor analysis

Behavior Factor 1: Aggressive trait Factor 2: Anxious trait Factor 3: Unfriendly trait 
Chase 0.918 0.382 0.086 
Attack 0.892 0.162 0.371 
Confront 0.848 0.326 0.378 
Scent mark 0.093 0.885 −0.162 
Sniff perch and scent mark 0.505 0.829 0.230 
Approach and contact −0.299 0.071 0.716 
Locomotion 0.408 0.035 0.237 
% variance explained 41.1 25.0 13.4 
% total variance 79.5   
Behavior Factor 1: Aggressive trait Factor 2: Anxious trait Factor 3: Unfriendly trait 
Chase 0.918 0.382 0.086 
Attack 0.892 0.162 0.371 
Confront 0.848 0.326 0.378 
Scent mark 0.093 0.885 −0.162 
Sniff perch and scent mark 0.505 0.829 0.230 
Approach and contact −0.299 0.071 0.716 
Locomotion 0.408 0.035 0.237 
% variance explained 41.1 25.0 13.4 
% total variance 79.5   

Note: Boldface indicates absolute value of factor loading >0.70.

Regional Serotonergic Transporter Binding Associated with Traits of Social Behavior

For PET images of [11C]DASB BPND, voxel-based regression analysis revealed regional SERT binding correlated with each social behavior trait at P < 0.05 with the voxel level corrected (Table 3 and Fig. 2A). SERT binding was negatively correlated with aggressive behaviors in the PCC (area 23, Z = 5.16; maximum at x = 0.2, y = −9.4, z = 5.6), with anxious behavior in the ACC (area 32, Z = 4.74; maximum at x = −0.4, y = 5.2, z = 3.8) and the LPFC (area 9, Z = 5.05; maximum at x = 9.0, y = 4.0, z = 1.0). On the other hand, SERT binding was positively correlated with unfriendly behaviors in the PCC (area 23, Z = 5.24; maximum at x = −1.8, y = −6.6, z = 7.6). Scatter graphs for the correlation between residualized and standardized [11C]DASB BPND and one-factor scores are shown in Figure 2B for all the significant clusters.

Regional Neural Activity in Different Social Contexts

In PET images of [18F]FDG, voxel-based repeated-measures ANOVA revealed a significant effect of social situation in the bilateral PCC (Z = 3.87; maximum at x = −1.1, y = −10.8, z = 7.4, Z = 3.66; maximum at x = 0.3, y = −6.4, z = 7.2) and in the occipital cortex (Z = 3.51; maximum at x = 0.4, y = −19.8, z = −1.5, Z = 3.16; maximum at x = 4.7, y = −16.7, z = − 1.5) at P < 0.05 with cluster level corrected (see Supplementary Table S4). The results from the post hoc triplet-paired t-statistics for three conditions are shown in Table 4 and in Figure 3A. In the UNF versus ALN contrast, higher activity was shown in the bilateral PCC (Z = 4.32; maximum at x = −1.1, y = −10.8, z = 7.4, Z = 4.00; maximum at x = 0.5, y = −6.4, z = 7.2) and in the occipital cortex (Z = 3.36; maximum at x = −1.6, y = −19.2, z = −1.1, Z = 3.39; maximum at x = 4.7, y = −16.9, z = −1.5). In the FAM versus ALN contrast, however, a significant effect was found in the parietal (Z = 3.48; maximum at x = −4.1, y = −7.3, z = 6.4, Z = 3.49; maximum at x = 2.8, y = − 10.7, z = 5.8) and occipital cortices (Z = 3.42; maximum at x = −2.2, y = −21.3, z = 0.9, Z = 4.01; maximum at x = 0.4, y = −19.8, z = − 1.5). No significant activity differences were found in any area in the UNF versus FAM contrast.

Table 4

Local maxima within a cluster showing a significant contrast in metabolic activity between different social situations

Contrasta Regions/hemisphere (x,y,z) Standard spaceb, mm Maximum Zc 
UNF-ALN Posterior cingulate area/L (−1.1, −10.8, 7.4) 4.32 
Posterior cingulate area/R (0.5, −6.4, 7.2) 4.00 
Occipital cortex/R (4.7, −16.9, −1.5) 3.39 
Occipital cortex/L (−1.6, −19.2, −1.1) 3.36 
UNF-FAM —   
FAM-ALN Occipital cortex/R (0.4, −19.8, −1.5) 4.01 
Parietal cortex, posterior/R (2.8, −10.7, 5.8) 3.49 
Parietal cortex, posterior/L (−4.1, −7.3, 6.4) 3.48 
Occipital cortex/L (−2.2, −21.3, 0.9) 3.42 
Contrasta Regions/hemisphere (x,y,z) Standard spaceb, mm Maximum Zc 
UNF-ALN Posterior cingulate area/L (−1.1, −10.8, 7.4) 4.32 
Posterior cingulate area/R (0.5, −6.4, 7.2) 4.00 
Occipital cortex/R (4.7, −16.9, −1.5) 3.39 
Occipital cortex/L (−1.6, −19.2, −1.1) 3.36 
UNF-FAM —   
FAM-ALN Occipital cortex/R (0.4, −19.8, −1.5) 4.01 
Parietal cortex, posterior/R (2.8, −10.7, 5.8) 3.49 
Parietal cortex, posterior/L (−4.1, −7.3, 6.4) 3.48 
Occipital cortex/L (−2.2, −21.3, 0.9) 3.42 

aALN, staying alone; FAM, staying with a familiar animal; UNF, staying with an unfamiliar animal.

aSee footnote of Table 3.

cCluster level corrected to P < 0.05.

Using normalized [18F]FDG uptake values in ROIs that derived from [11C]DASB–behavior correlation analysis, two-way ANOVA revealed a significant ROI by social situation interaction (F(4,15) = 3.97, P = 0.01). Next, repeated-measures ANOVA on each ROI revealed a significant effect of social situation in neural activity of the PCC (F(2,10) = 9.28, P = 0.01) but not of the ACC (F(2,10) = 2.25, P = 0.16) or the LPFC (F(2,10) = 0.08, P = 0.92) (Fig. 3B). In the PCC, a post hoc Dunnett's test for multiple comparisons revealed a significant difference in the contrast of UNF versus ALN (P < 0.01), but not FAM versus ALN or UNF versus FAM.

Functional Connectivity in Different Social Contexts

When the seed ROI was placed in the PCC, voxel-based ANOVA revealed a significant social situation-dependent functional connectivity with the right insular cortex (Z = 2.75; maximum at x = 7.8, y = 0.0, z = 1.4) and with the left hippocampus (Z = 3.18; maximum at x = −6.2, y = −6.6, z = −3.8) at P < 0.01, uncorrected (Table 5 and Fig. 4A). Interregional correlation (estimates) of the PCC with these significant clusters revealed that neural activity of the PCC was tightly coupled with that of the right insular cortex in UNF and with the left hippocampus in ALN or FAM (Fig. 4A). We also found several areas showing a significant social situation-dependent change in functional connectivity with the ACC, including the bilateral orbitofrontal cortex (Z = 3.18; maximum at x = −4.2, y = 3.6, z = 1.0, Z = 3.17; maximum at x = 3.6, y = 4.8, z = 1.8) and the amygdala (Z = 2.66; maximum at x = −3.0, y = −0.8, z = −6.6, Z = 2.67; maximum at x = 4.0, y = −2.4, z = −5.6) (Table 5 and Fig. 4B). Interregional correlations of the ACC with these clusters revealed that neural activity of the ACC was tightly coupled with that of the orbitofrontal cortex in UNF and with that of the amygdala in FAM and ALN (Fig. 4B). There were no areas showing significant social situation-dependent changes in functional connectivity of the LPFC.

Table 5

Regions showing social situation-dependent functional coupling with the midline cortical structures

Seed regionsa Functional coupling regions/Hemisphere (x,y,z) Standard spaceb, mm Voxelsc Maximum Zd 
Posterior cingulate area (a + d) Hippocampus/L (−6.2, −6.6, −3.8) 390 3.18 
Insular cortex/R (7.8, 0.0, 1.4) 164 2.75 
Anterior cingulate area (b) Orbitofrontal cortex/L (−4.2, 3.6, 1.0) 930 3.18 
Orbitofrontal cortex/R (3.6, 4.8, 1.8) 649 3.17 
Auditory cortex/L (−9.8, −0.4, 0.4) 294 2.49 
Superior temporal cortex/L (−11.6, −7.6, −1.6) 1152 2.70 
Superior temporal cortex/R (10.8, −2.8, 1.4) 753 2.59 
Visual area 3/L (−9, −11.4, −4.0) 226 2.67 
Visual area 3/R (9.4, −10.0, −4.2) 449 2.54 
Amygdala/L (−3.0, −0.8, −6.6) 189 2.66 
Amygdala/R (4.0, −2.4, −5.6) 179 2.67 
PCC/L (−1.6, −6.4, 6.2) 346 3.04 
Seed regionsa Functional coupling regions/Hemisphere (x,y,z) Standard spaceb, mm Voxelsc Maximum Zd 
Posterior cingulate area (a + d) Hippocampus/L (−6.2, −6.6, −3.8) 390 3.18 
Insular cortex/R (7.8, 0.0, 1.4) 164 2.75 
Anterior cingulate area (b) Orbitofrontal cortex/L (−4.2, 3.6, 1.0) 930 3.18 
Orbitofrontal cortex/R (3.6, 4.8, 1.8) 649 3.17 
Auditory cortex/L (−9.8, −0.4, 0.4) 294 2.49 
Superior temporal cortex/L (−11.6, −7.6, −1.6) 1152 2.70 
Superior temporal cortex/R (10.8, −2.8, 1.4) 753 2.59 
Visual area 3/L (−9, −11.4, −4.0) 226 2.67 
Visual area 3/R (9.4, −10.0, −4.2) 449 2.54 
Amygdala/L (−3.0, −0.8, −6.6) 189 2.66 
Amygdala/R (4.0, −2.4, −5.6) 179 2.67 
PCC/L (−1.6, −6.4, 6.2) 346 3.04 

aRegions that derived from [11C]DASB–behavior correlation analysis. Symbols in parentheses correspond to those in Figures 2 and 3.

b,cSee footnote of Table 3.

dUncorrected to P < 0.01.

Figure 4.

Z-static maps from the F-test using the [18F]FDG-PET images that showed social context-dependent functional coupling with the seed regions of the PCC (A) and the ACC (B), respectively. Bar graphs showed estimated coefficients and the standard errors in the three conditions for each region. Clusters showing altered functional coupling with the PCC were detected in the right insular cortex and the left hippocampus (A). Clusters showing altered functional coupling with the ACC were detected in the right orbitofrontal cortex and the left amygdala (B). The coordinates for the anterior commissure and Z-scores of all significant clusters are listed in Table 5. ALN, staying alone; FAM, staying with a familiar animal; UNF, staying with an unfamiliar animal.

Figure 4.

Z-static maps from the F-test using the [18F]FDG-PET images that showed social context-dependent functional coupling with the seed regions of the PCC (A) and the ACC (B), respectively. Bar graphs showed estimated coefficients and the standard errors in the three conditions for each region. Clusters showing altered functional coupling with the PCC were detected in the right insular cortex and the left hippocampus (A). Clusters showing altered functional coupling with the ACC were detected in the right orbitofrontal cortex and the left amygdala (B). The coordinates for the anterior commissure and Z-scores of all significant clusters are listed in Table 5. ALN, staying alone; FAM, staying with a familiar animal; UNF, staying with an unfamiliar animal.

Discussion

Similar to humans, common marmosets are known as a primate species with complex social behaviors such as a high level of social learning (Burkart, Strasser et al. 2009; Pesendorfer et al. 2009) and unsolicited prosociality relying on cooperative breeding (Burkart et al. 2007, Burkart, Hrdy et al. 2009). In the present experiment, we performed the quantification of social behavior traits of adult male common marmosets while they directly encountered unfamiliar conspecific males and found individual variation in the patterns of their social behaviors composed of three independent dimensions of aggressive, anxious, and friendly traits. The particular combination of social response patterns of an individual could be regarded as “personality” in humans and other animals (Uher 2011), though social behavior traits derived from the data of marmosets in the laboratory environment would not be completely analogous to human personality by self-report questionnaires (Tauscher et al. 2001; Takano et al. 2007; Kalbitzer et al. 2009), which is influenced by linguistic, conceptual, and cultural factors (McCrae et al. 2010). The present results may be associated with fundamental aspects of social behavior in human and nonhuman primates. Neurobiological research on nonhuman primates is important to extract basic and valuable information for understanding human social cognition (Fitch et al. 2010).

The present results highlight the midline cortical SERT binding that is associated with individual-specific social behavior traits. Our voxel-based statistical analysis uncovered distinct midline cortical subregions associated with different social behavior traits. The aggressive and friendly traits were negatively correlated with SERT binding in the PCC and the anxious trait was negatively associated with SERT binding in the ACC. Our results specifically indicate that the higher anxiety trait was associated with lower cortical SERT binding activity in the ACC. In marmosets, administration of selective inhibitors of SERT has been reported to reduce behavioral inhibition during a confrontation with a conspecific stranger (Kinnally et al. 2006), which implies that serotonin neurotransmission in the ACC may play a role in behavioral inhibition of anxiety-related behavior. In rodent studies, cortical serotonin receptors have been shown to play a role in control of anxiety-related behaviors (Gross et al. 2002; Weisstaub et al. 2006), though the location of action has not been specified. Recently, human neuroimaging studies reported that serotonin 1A receptor binding in the ACC is inversely correlated with anxiety in healthy volunteers and is significantly reduced in social anxiety disorder (Tauscher et al. 2001; Lanzenberger et al. 2007). These results from human subjects indicate that low functional serotonin neurotransmission in the ACC is associated with anxiety-related behavior.

In contrast, we found SERT binding to be negatively associated with aggressive and friendly traits in the PCC, which are also part of the midline cortical structures. While serotonin has been reported to be associated with social affiliation or social dominance (Mehlman et al. 1995; Knutson et al. 1998; Tse and Bond 2002; Kalbitzer et al. 2009), and the midline cortical structures have been implicated in self-social cognition in humans (Northoff and Bermpohl 2004; Buckner and Carroll 2007), we have not known how the midline cortical serotonergic system is involved in self-social cognition relevant to aggressive and friendly traits. It has been reported that SERT binding activity correlates positively with the neuroticism dimension and negatively with openness of personality inventory measures, which is localized to subcortical structures, such as the thalamus, the striatum, and the midbrain (Takano et al. 2007; Kalbitzer et al. 2009). In this study, we did not detect a significant correlation of SERT binding activity in these subcortical regions with the friendly trait. However, serotonin in the neural network including these regions which have anatomical connections with the PCC could play a role in regulating affiliative behavior. On the other hand, abnormally higher neural responsivity in the PCC to serotonergic stimulation and threat stimuli has been reported in human subjects with impulsive aggression and violent offenders (New et al. 2002; Lee et al. 2009). These results from human studies suggest that serotonergic neurotransmission in the PCC could be involved in aggressive behavior.

It has been noted that abnormality in the various midline regions is associated with major depressive disorder (Northoff et al. 2011), in which SERT is one of the major therapeutic targets. Autistic subjects showed reduced SERT and serotonin 2A receptor binding in midline cortical areas, which correlated with impairments of social cognition (Murphy et al. 2006; Nakamura et al. 2010). Although the present results indicate the normal range of variation in social behavior traits for marmosets, the data may actually be relevant to our understanding of human pathological conditions such as depression and autism.

Without referring to serotonin function, the midline cortical structures could be functionally segregated in terms of the contexts of social cognition in humans, that is, the anterior regions including the ACC have been thought to be involved in emotional processing and mentalizing (Amodio and Frith 2006; Etkin et al. 2011), while the posterior region, the PCC, is thought to have a more outward-directed social focus (Northoff and Bermpohl 2004; Buckner and Carroll 2007). The results of our [18F]FDG-PET experiment support a distinct role for the subregions in the midline cortical structures for socio-emotional processing. The neural activity in the PCC but not in the ACC changed depending on the social situation, showing a significant contrast of UNF versus ALN. Thus, the PCC could be involved in neural processing related to social saliency in staying with an unfamiliar conspecific. Data from human subjects and the present results are suggestive of a dominant role of the PCC as a nexus of self-social and cognitive processes in primates.

Other than the midline cortical structures, we found that SERT binding in the LPFC was negatively associated with anxiety trait. The LPFC is known to be critical for inhibitory control in attentional selection in marmosets (Dias et al. 1997). Marmosets with selective serotonin depletion of the prefrontal cortex, including the LPFC, displayed perseverative responding (cognitive inflexibility) (Walker et al. 2006; Clarke et al. 2007). Recently, excitotoxic lesion of the LPFC has reported to heighten behavioral and autonomic fear response, and anxious behavior in marmosets (Agustín-Pavón et al. 2012). Therefore, it is reasonable to consider that the LPFC played a role in inhibitory control of fear and anxiety, being regulated by serotonergic activity.

The present study revealed that functional connectivities with the PCC and with the ACC changed depending on the social situation. The functional connectivity analysis confirms previous human and nonhuman primate studies for both anatomical and functional connections of the PCC with the insular cortex (Augustine 1996; Cauda et al. 2011; Deen et al. 2011) and the hippocampus (Kobayashi and Amaral 2003, 2007; Andrews-Hanna et al. 2010), the ACC and the orbitofrontal cortex (Carmichael and Price 1996; Roberts et al. 2007; Beckmann et al. 2009), and the amygdala (Stefanacci and Amaral 2000, 2002; Roberts et al. 2007; Beckmann et al. 2009). Recent evidence from network analysis brings out the joint activation of the PCC and the hippocampus in the resting state or “default mode” in humans (Buckner et al. 2008), which may be comparable to the present results in staying alone or staying with a familiar conspecific. On the other hand, functional coupling of the PCC with the insula in the UNF condition implies that the passive exposition to an unfamiliar conspecific may be relevant to self-referential social salience. Recent accumulated evidence from human studies brings out the essential role of the insula in self-awareness of feelings (Craig 2009) or switching between externally oriented attention and internally oriented cognition (Menon and Uddin 2010). Functional coupling of the ACC with the orbitofrontal cortex in the UNF condition implies the interconnection of cognitive and emotional processing for socially salient stimuli (Rolls and Grabenhorst 2008, Rushworth et al. 2008). On the other hand, the amygdala is well known for its role in integrating and coordinating emotional expressions (LeDoux 2000; Phelps and LeDoux 2005). The altered connectivity of the ACC with the amygdala may be related to differential processing of emotional information among social situations.

Several limitations may be suggested in our results. First, we tested just one particular condition, the encounter between unfamiliar animals, among many social situations, though laboratory conditions of encountering others have been used as a fundamental empirical approach for studying social responses in marmosets (Cilia and Piper 1997, Kinnally et al. 2006). Second, our small sample size presents possible issues of sampling bias and low statistical power (Nichols and Poline 2009), though our interpretation rested on the statistical results that showed significance at a threshold corrected for multiple comparisons in the condition analysis. The result of the correlation of anxiety with SERT binding might have also occurred due to the small sample size, particularly, driven by one animal with a very high anxiety score. Further confirmatory study is required by increasing the number of test animals to analyze the SERT binding–anxiety correlation. Finally, we revealed the association between cortical SERT binding and social behavior traits, rather than the direct behavioral effects of local serotonergic manipulation. Specific serotonergic depletions (Walker et al. 2006; Clarke et al. 2007) and iontophoresis of serotonergic receptor ligands (Williams et al. 2002) within the prefrontal cortex subregions such as the dorsolateral prefrontal and/or orbital areas have been reported to have distinct effects on cognitive behavior in nonhuman primates. Further studies using such experimental interventions are needed to provide additional evidence of the serotonergic regulation on social information processing in the midline cortical structures.

In conclusion, we have reported that the social behavioral traits were associated with SERT binding in the midline cortical subregions of marmoset monkeys. The midline cortical subregions showing distinctive serotonin-behavior correlations were functionally segregated in terms of their role in socio-emotional processing. These findings stress the importance of serotonergic regulation in midline cortical subregions for the process of interindividual interaction as well as sociality-based emotional responses.

Supplementary Material

Supplementary material can be found at: http://www.cercor.oxfordjournals.org/.

Funding

This work was supported by the Molecular Imaging Research Program from the Ministry of Education, Culture, Sports, Science and Technology of the Japanese Government, and a grant to C.Y. from KAKENHI (Grant-in-Aid for Scientific Research (C) No. 19591388).

Notes

We thank Mr Y Wada and Mss E Hayashinaka, K Onoe, and C Takeda for technical help with PET imaging and data analysis; and Mss. H Nagata and T Mori for providing the PET tracer (from RIKEN). We gratefully acknowledge Dr J Uher from Freie Universität Berlin for initial advice with factor analysis for behavioral measures. Conflict of Interest: None declared.

References

Agustín-Pavón
C
Braesicke
K
Shiba
Y
Santangelo
AM
Mikheenko
Y
Cockroft
G
Asma
F
Clarke
H
Man
MS
Roberts
AC
Lesions of ventrolateral prefrontal or anterior orbitofrontal cortex in primates heighten negative emotion
Biol Psychiatry
 , 
2012
 
[Epub ahead of print]
Amodio
DM
Frith
CD
Meeting of minds: the medial frontal cortex and social cognition
Nat Rev Neurosci
 , 
2006
, vol. 
7
 (pg. 
268
-
277
)
Anderson
IM
Del-Ben
CM
McKie
S
Richardson
P
Williams
SR
Elliott
R
Deakin
JF
Citalopram modulation of neuronal responses to aversive face emotions: a functional MRI study
Neuroreport
 , 
2007
, vol. 
18
 (pg. 
1351
-
1355
)
Andrews-Hanna
JR
Reidler
JS
Sepulcre
J
Poulin
R
Buckner
RL
Functional-anatomic fractionation of the brain's default network
Neuron
 , 
2010
, vol. 
65
 (pg. 
550
-
562
)
Augustine
JR
Circuitry and functional aspects of the insular lobe in primates including humans
Brain Res Rev
 , 
1996
, vol. 
22
 (pg. 
229
-
244
)
Barros
M
Mello
EL
Huston
JP
Tomaz
C
Behavioral effects of buspirone in the marmoset employing a predator confrontation test of fear and anxiety
Pharmacol Biochem Behav
 , 
2001
, vol. 
68
 (pg. 
255
-
262
)
Beckmann
MH
Johansen-Berg
H
Rushworth
MFS
Connectivity-based parcellation of human cingulate cortex and its relation to functional specialization
J Neurosci
 , 
2009
, vol. 
29
 (pg. 
1175
-
1190
)
Bezerra
BM
Souto
A
Structure and usage of the vocal repertoire of Callithrix jacchus
Int J Primatol
 , 
2008
, vol. 
29
 (pg. 
671
-
701
)
Buckner
RL
Andrews-Hanna
JR
Schacter
DL
The brain's default network: anatomy, function, and relevance to disease
Ann NY Acad Sci
 , 
2008
, vol. 
1124
 (pg. 
1
-
38
)
Buckner
RL
Carroll
DC
Self-projection and the brain
Trends Cogn Sci
 , 
2007
, vol. 
11
 (pg. 
49
-
57
)
Burkart
JM
Fehr
E
Efferson
C
van Schaik
CP
Other-regarding preferences in a non-human primate: common marmosets provision food altruistically
Proc Natl Acad Sci USA
 , 
2007
, vol. 
104
 (pg. 
19762
-
19766
)
Burkart
JM
Hrdy
SB
Van Schaik
CP
Cooperative breeding and human cognitive evolution
Evol Anthropol
 , 
2009
, vol. 
18
 (pg. 
175
-
186
)
Burkart
JM
Strasser
A
Foglia
M
Trade-offs between social learning and individual innovativeness in common marmosets, Callithrix jacchus
Anim Behav
 , 
2009
, vol. 
77
 (pg. 
1291
-
1301
)
Burman
KJ
Rosa
MG
Architectural subdivisions of medial and orbital frontal cortices in the marmoset monkey (Callithrix jacchus)
J Comp Neurol
 , 
2009
, vol. 
514
 (pg. 
11
-
29
)
Carmichael
ST
Price
JL
Connectional networks within the orbital and medial prefrontal cortex of macaque monkeys
J Comp Neurol
 , 
1996
, vol. 
371
 (pg. 
179
-
207
)
Cauda
F
D'Agata
F
Sacco
K
Duca
S
Geminiani
G
Vercelli
A
Functional connectivity of the insula in the resting brain
NeuroImage
 , 
2011
, vol. 
55
 (pg. 
8
-
23
)
Cilia
J
Piper
DC
Marmoset conspecific confrontation: an ethologically based model of anxiety
Pharmacol Biochem Behav
 , 
1997
, vol. 
58
 (pg. 
85
-
91
)
Clarke
HF
Walker
SC
Dalley
JW
Robbins
TW
Roberts
AC
Cognitive inflexibility after prefrontal serotonin depletion is behaviorally and neurochemically specific
Cereb Cortex
 , 
2007
, vol. 
17
 (pg. 
18
-
27
)
Cleare
AJ
Bond
AJ
The effect of tryptophan depletion and enhancement on subjective and behavioural aggression in normal male subjects
Psychopharmacology
 , 
1995
, vol. 
118
 (pg. 
72
-
81
)
Coccaro
EF
Impulsive aggression and central serotonergic system function in humans: an example of a dimensional brain–behavior relationship
Int Clin Psychopharmacol
 , 
1992
, vol. 
7
 (pg. 
3
-
12
)
Cools
R
Calder
AJ
Lawrence
AD
Clark
L
Bullmore
E
Robbins
TW
Individual differences in threat sensitivity predict serotonergic modulation of amygdala response to fearful faces
Psychopharmacology
 , 
2005
, vol. 
180
 (pg. 
670
-
679
)
Craig
AD
How do you feel now? The anterior insula and human awareness
Nature Rev Neurosci
 , 
2009
, vol. 
10
 (pg. 
59
-
70
)
Dias
R
Robbins
TW
Roberts
AC
Dissociable forms of inhibitory control within prefrontal cortex with an analog of the Wisconsin Card Sort Test: restriction to novel situations and independence from “on-line” processing
J Neurosci
 , 
1997
, vol. 
17
 (pg. 
9285
-
9297
)
Deen
B
Pitskel
NB
Pelphrey
KA
Three systems of insular functional connectivity identified with cluster analysis
Cereb Cortex
 , 
2011
, vol. 
21
 (pg. 
1498
-
1506
)
Epple
G
Comparative studies on vocalization in marmoset monkeys (Hapalidae)
Folia Primatologia
 , 
1968
, vol. 
8
 (pg. 
1
-
40
)
Etkin
A
Egner
T
Kalisch
R
Emotional processing in anterior cingulate and medial prefrontal cortex
Trends Cogn Sci
 , 
2011
, vol. 
15
 (pg. 
85
-
93
)
Fairbanks
LA
Melega
WP
Jorgensen
MJ
Kaplan
JR
McGuire
MT
Social impulsivity inversely associated with CSF 5-HIAA and fluoxetine exposure in vervet monkeys
Neuropsychopharmacology
 , 
2001
, vol. 
24
 (pg. 
370
-
378
)
Fitch
WT
Huber
L
Bugnya
T
Social cognition and the evolution of language: constructing cognitive phylogenies
Neuron
 , 
2010
, vol. 
65
 (pg. 
795
-
814
)
Gobbini
MI
Haxby
JV
Neural systems for recognition of familiar faces
Neuropsychologia
 , 
2007
, vol. 
45
 (pg. 
32
-
41
)
Gross
C
Zhuang
X
Stark
K
Ramboz
S
Oosting
R
Kirby
L
Santarell
IL
Beck
S
Hen
R
Serotonin1A receptor acts during development to establish normal anxiety-like behaviour in the adult
Nature
 , 
2002
, vol. 
416
 (pg. 
396
-
400
)
Hariri
AR
Mattay
VS
Tessitore
A
Kolachana
B
Fera
F
Goldman
D
Egan
MF
Weinberger
DR
Serotonin transporter genetic variation and the response of the human amygdala
Science
 , 
2002
, vol. 
297
 (pg. 
400
-
403
)
Harrison
ML
Tardif
SD
Species differences in response to conspecific intruders in Callithrix jacchus and Saguinus oedipus
Int J Primatol
 , 
1989
, vol. 
10
 (pg. 
343
-
362
)
Heymann
EW
Scent marking strategies of new world primates
Am J Primatol
 , 
2006
, vol. 
68
 (pg. 
650
-
661
)
Holschneider
DP
Maarek
JM
Mapping brain function in freely moving subjects
Neurosci Biobehav Rev
 , 
2004
, vol. 
28
 (pg. 
449
-
461
)
Ichise
M
Liow
JS
Lu
JQ
Takano
A
Model
K
Toyama
H
Suhara
T
Suzuki
K
Innis
RB
Carson
RE
Linearized reference tissue parametric imaging methods: application to [11C]DASB positron emission tomography studies of the serotonin transporter in human brain
J Cereb Blood Flow Metab
 , 
2003
, vol. 
23
 (pg. 
1096
-
1112
)
Innis
RB
Cunningham
VJ
Delforge
J
Fujita
M
Gjedde
A
Gunn
RN
Holden
J
Houle
S
Huang
S-C
Ichise
M
, et al.  . 
Consensus nomenclature for in vivo imaging of reversibly binding radioligands
J Cereb Blood Flow Metab
 , 
2007
, vol. 
27
 (pg. 
1533
-
1539
)
Kalbitzer
J
Frokjaer
VG
Erritzoe
D
Svarer
C
Cumming
P
Nielsen
FA
Hashemi
SH
Baare
WF
Madsen
J
Hasselbalch
SG
, et al.  . 
The personality trait openness is related to cerebral 5-HTT levels
Neuroimage
 , 
2009
, vol. 
45
 (pg. 
280
-
285
)
Kinnally
EL
Jensen
HA
Ewing
JH
French
JA
Serotonin function is associated with behavioral response to a novel conspecific in marmosets
Am J Primatol
 , 
2006
, vol. 
68
 (pg. 
812
-
824
)
Knutson
B
Wolkowitz
OM
Cole
SW
Chan
T
Moore
EA
Johnson
RC
Terpstra
J
Turner
RA
Reus
VI
Selective alteration of personality and social behavior by serotonergic intervention
Am J Psychiatry
 , 
1998
, vol. 
155
 (pg. 
373
-
379
)
Kobayashi
Y
Amaral
DG
Macaque monkey retrosplenial cortex. II. Cortical afferents
J Comp Neurol
 , 
2003
, vol. 
466
 (pg. 
48
-
79
)
Kobayashi
Y
Amaral
DG
Macaque monkey retrosplenial cortex. III. Cortical efferents
J Comp Neurol
 , 
2007
, vol. 
502
 (pg. 
810
-
833
)
Lanzenberger
RR
Mitterhauser
M
Spindelegger
C
Wadsak
W
Klein
N
Mien
L-K
Holik
A
Attarbaschi
T
Mossaheb
N
Sacher
J
, et al.  . 
Reduced serotonin-1A receptor binding in social anxiety disorder
Biol Psychiatry
 , 
2007
, vol. 
61
 (pg. 
1081
-
1089
)
Lazaro-Perea
C
Snowdon
CT
de Fatima Arruda
M
Scent-marking behavior in wild groups of common marmosets (Callithrix jacchus)
Behav Ecol Sociobiol
 , 
1999
, vol. 
46
 (pg. 
313
-
324
)
LeDoux
JE
Emotion circuits in the brain
Annu Rev Neurosci
 , 
2000
, vol. 
23
 (pg. 
155
-
184
)
Lee
TM
Chan
SC
Raine
A
Hyperresponsivity to threat stimuli in domestic violence offenders: a functional magnetic resonance imaging study
J Clin Psychiatry
 , 
2009
, vol. 
70
 (pg. 
36
-
45
)
Lesch
KP
Bengel
D
Heils
A
Sabol
SZ
Greenberg
BD
Petri
S
Benjamin
J
Muller
CR
Hamer
DH
Murphy
DL
Association of anxiety-related traits with a polymorphism in the serotonin transporter gene regulatory region
Science
 , 
1996
, vol. 
274
 (pg. 
1527
-
1531
)
McCrae
RR
Terracciano
A
De Fruyt
F
De Bolle
M
Gelfand
MJ
Costa
PT
Jr
Project CotAPPoC
The validity and structure of culture-level personality scores: data from ratings of young adolescents
J Pers
 , 
2010
, vol. 
78
 (pg. 
815
-
838
)
Mehlman
PT
Higley
JD
Faucher
I
Lilly
AA
Taub
DM
Vickers
J
Suomi
SJ
Linnoila
M
Correlation of CSF 5-HIAA concentrations with sociality and the timing of emigration in free-ranging primates
Am J Psychiatry
 , 
1995
, vol. 
152
 (pg. 
907
-
913
)
Mehlman
PT
Higley
JD
Faucher
I
Lilly
AA
Taub
DM
Vickers
J
Suomi
SJ
Linnoila
M
Low CSF 5-HIAA concentrations and severe aggression and impaired impulse control in nonhuman primates
Am J Psychiatry
 , 
1994
, vol. 
151
 (pg. 
1485
-
1491
)
Menon
V
Uddin
LQ
Saliency, switching, attention and control: a network model of insula function
Brain Struct Funct
 , 
2010
, vol. 
214
 (pg. 
655
-
667
)
Moeller
FG
Dougherty
DM
Swann
AC
Collins
D
Davis
CM
Cherek
DR
Tryptophan depletion and aggressive responding in healthy males
Psychopharmacology
 , 
1996
, vol. 
126
 (pg. 
97
-
103
)
Murphy
DG
Daly
E
Schmitz
N
Toal
F
Murphy
K
Curran
S
Erlandsson
K
Eersels
J
Kerwin
R
Ell
P
, et al.  . 
Cortical serotonin 5-HT2A receptor binding and social communication in adults with Asperger's syndrome: an in vivo SPECT study
Am J Psychiatry
 , 
2006
, vol. 
163
 (pg. 
934
-
936
)
Nakamura
K
Sekine
Y
Ouchi
Y
Tsuji
M
Yoshikawa
E
Futatsubashi
M
Tsuchiya
KJ
Sugihara
G
Iwata
Y
Suzuki
K
, et al.  . 
Brain serotonin and dopamine transporter bindings in adults with high-functioning autism
Arch Gen Psychiatry
 , 
2010
, vol. 
67
 (pg. 
59
-
68
)
New
AS
Hazlett
EA
Buchsbaum
MS
Goodman
M
Reynolds
D
Mitropoulou
V
Sprung
L
Shaw
RBJ
Koenigsberg
H
Platholi
J
, et al.  . 
Blunted prefrontal cortical 18fluorodeoxyglucose positron emission tomography response to meta-chlorophenylpiperazine in impulsive aggression
Arch Gen Psychiatry
 , 
2002
, vol. 
59
 (pg. 
621
-
629
)
Nichols
TE
Poline
J-B
Commentary on Vul et al.'s (2009). “Puzzlingly High Correlations in fMRI Studies of Emotion, Personality, and Social Cognition”
Perspect Psychol Sci
 , 
2009
, vol. 
4
 (pg. 
291
-
293
)
Nomura
M
Kusumi
I
Kaneko
M
Masui
T
Daiguji
M
Ueno
T
Koyama
T
Nomura
Y
Involvement of a polymorphism in the 5-HT2A receptor gene in impulsive behavior
Psychopharmacology
 , 
2006
, vol. 
187
 (pg. 
30
-
35
)
Northoff
G
Bermpohl
F
Cortical midline structures and the self
Trends Cog Sci
 , 
2004
, vol. 
8
 (pg. 
102
-
107
)
Northoff
G
Wiebking
C
Feinberg
T
Panksepp
J
The ‘resting-state hypothesis’ of major depressive disorder: a translational subcortical–cortical framework for a system disorder
Neurosci Biobehav Rev
 , 
2011
, vol. 
35
 (pg. 
1929
-
1945
)
Oler
JA
Fox
AS
Shelton
SE
Christian
BT
Murali
D
Oakes
TR
Davidson
RJ
Kalin
NH
Serotonin transporter availability in the amygdala and bed nucleus of the stria terminalis predicts anxious temperament and brain glucose metabolic activity
J Neurosci
 , 
2009
, vol. 
29
 (pg. 
9961
-
9966
)
Olsson
A
Ochsner
KN
The role of social cognition in emotion
Trends Cogn Sci
 , 
2008
, vol. 
12
 (pg. 
65
-
71
)
Palmer
SM
Rosa
MG
Quantitative analysis of the corticocortical projections to the middle temporal area in the marmoset monkey: evolutionary and functional implications
Cereb Cortex
 , 
2006
, vol. 
16
 (pg. 
1361
-
1375
)
Pesendorfer
MB
Gunhold
T
Schiel
N
Souto
A
Huber
L
Range
F
The maintenance of traditions in marmosets: individual habit, not social conformity? A field experiment
PLoS ONE
 , 
2009
, vol. 
4
 pg. 
e4472
 
Phelps
EA
LeDoux
JE
Contributions of the amygdala to review emotion processing: from animal models to human behavior
Neuron
 , 
2005
, vol. 
48
 (pg. 
175
-
187
)
Reivich
M
Kuhl
D
Wolf
A
Greenberg
J
Phelps
M
Ido
T
Casella
V
Fowler
J
Hoffman
E
Alavi
A
, et al.  . 
The [18F]fluorodeoxyglucose method for the measurement of local cerebral glucose utilization in man
Circ Res
 , 
1979
, vol. 
44
 (pg. 
127
-
137
)
Roberts
AC
Tomic
DL
Parkinson
CH
Roeling
TA
Cutter
DJ
Robbins
TW
Everitt
BJ
Forebrain connectivity of the prefrontal cortex in the marmoset monkey (Callithrix jacchus): an anterograde and retrograde tract-tracing study
J Comp Neurol
 , 
2007
, vol. 
502
 (pg. 
86
-
112
)
Rolls
ET
Grabenhorst
F
The orbitofrontal cortex and beyond: from affect to decision-making
Prog Neurobiol
 , 
2008
, vol. 
86
 (pg. 
216
-
244
)
Rushworth
MFS
Behrens
EJ
Rudebeck
PH
Walton
ME
Contrasting roles for cingulate and orbitofrontal cortex in decisions and social behaviour
Trends Cogn Sci
 , 
2008
, vol. 
11
 (pg. 
168
-
176
)
Sen
S
Burmeister
M
Ghosh
D
Meta-analysis of the association between a serotonin transporter promoter polymorphism (5-HTTLPR) and anxiety-related personality traits
Neuropsychiat Genet
 , 
2004
, vol. 
127B
 (pg. 
85
-
89
)
Stefanacci
L
Amaral
DG
Some observations on cortical inputs to the macaque monkey amygdala: an anterograde tracing study
J Comp Neurol
 , 
2002
, vol. 
451
 (pg. 
301
-
323
)
Stefanacci
L
Amaral
DG
Topographic organization of cortical inputs to the lateral nucleus of the macaque monkey amygdala: a retrograde tracing study
J Comp Neurol
 , 
2000
, vol. 
421
 (pg. 
52
-
79
)
Stephan
H
Baron
G
Schwerdtfeger
WK
The brain of the common marmoset (Callithrix jacchus): a stereotaxic atlas
 , 
1980
Berlin
Springer
Stevenson
MF
Poole
TE
An ethogram of the common marmosets (Calithrix jacchus jacchus): general behavioral repertoire
Anim Behav
 , 
1976
, vol. 
24
 (pg. 
428
-
451
)
Takano
A
Arakawa
R
Hayashi
M
Takahashi
H
Ito
H
Suhara
T
Relationship between neuroticism personality trait and serotonin transporter binding
Biol Psychiatry
 , 
2007
, vol. 
62
 (pg. 
588
-
592
)
Tauscher
J
Bagby
RM
Javanmard
M
Christensen
BK
Kasper
S
Kapur
S
Inverse relationship between serotonin 5-HT(1A) receptor binding and anxiety: a [(11)C]WAY-100635 PET investigation in healthy volunteers
Am J Psychiatry
 , 
2001
, vol. 
158
 (pg. 
1326
-
1328
)
Tse
WS
Bond
AJ
Serotonergic intervention affects both social dominance and affiliative behaviour
Psychopharmacology
 , 
2002
, vol. 
161
 (pg. 
324
-
330
)
Uher
J
Individual behavioral phenotypes: an integrative meta-theoretical framework. Why “behavioral syndromes” are not analogs of “personality”
Dev Psychobiol
 , 
2011
, vol. 
53
 (pg. 
521
-
548
)
Walker
SC
Mikheenko
YP
Argyle
LD
Robbins
TW
Roberts
AC
Selective prefrontal serotonin depletion impairs acquisition of a detour-reaching task
European J Neurosci
 , 
2006
, vol. 
23
 (pg. 
3119
-
3123
)
Weisstaub
NV
Zhou
M
Lira
A
Lambe
E
Gonzalez-Maeso
J
Hornung
JP
Sibille
E
Underwood
M
Itohara
S
Dauer
WT
, et al.  . 
Cortical 5-HT2A receptor signaling modulates anxiety-like behaviors in mice
Science
 , 
2006
, vol. 
313
 (pg. 
536
-
540
)
Williams
GV
Rao
SG
Goldman-Rakic
PS
The physiological role of 5-HT2A receptors in working memory
J Neurosci
 , 
2002
, vol. 
22
 (pg. 
2843
-
2854
)
Wilson
AA
Ginovart
N
Hussey
D
Meyer
J
Houle
S
In vitro and in vivo characterisation of [11C]-DASB: a probe for in vivo measurements of the serotonin transporter by positron emission tomography
Nucl Med Biol
 , 
2002
, vol. 
29
 (pg. 
509
-
515
)
Yokoyama
C
Yamanaka
H
Onoe
K
Kawasaki
A
Nagata
H
Shirakami
K
Doi
H
Onoe
H
Mapping of serotonin transporters by positron emission tomography with [11C]-DASB in conscious common marmosets: comparison with rhesus monkeys
Synapse
 , 
2009
, vol. 
64
 (pg. 
594
-
601
)
Zhong
P
Yan
Z
Differential regulation of the excitability of prefrontal cortical fast-spiking interneurons and pyramidal neurons by serotonin and fluoxetine
PLoS ONE
 , 
2011
, vol. 
6
 pg. 
e16970