Skip to main content
NCBI home page
Social Cognitive and Affective Neuroscience logoLink to Social Cognitive and Affective Neuroscience
. 2016 May 20;11(10):1627–1637. doi: 10.1093/scan/nsw074

Conditioned social dominance threat: observation of others’ social dominance biases threat learning

Jan Haaker 1,✉, Tanaz Molapour 1, Andreas Olsson 1
PMCID: PMC5040915  PMID: 27217107

Abstract

Social groups are organized along dominance hierarchies, which determine how we respond to threats posed by dominant and subordinate others. The persuasive impact of these dominance threats on mental and physical well-being has been well described but it is unknown how dominance rank of others bias our experience and learning in the first place. We introduce a model of conditioned social dominance threat in humans, where the presence of a dominant other is paired with an aversive event. Participants first learned about the dominance rank of others by observing their dyadic confrontations. During subsequent fear learning, the dominant and subordinate others were equally predictive of an aversive consequence (mild electric shock) to the participant. In three separate experiments, we show that participants’ eye-blink startle responses and amygdala reactivity adaptively tracked dominance of others during observation of confrontation. Importantly, during fear learning dominant vs subordinate others elicited stronger and more persistent learned threat responses as measured by physiological arousal and amygdala activity. Our results characterize the neural basis of learning through observing conflicts between others, and how this affects subsequent learning through direct, personal experiences.

Keywords: amygdala, mPFC, threat-relevance, fear extinction, learning bias, social conflict

Introduction

Social groups of humans and other animals are organized along dominance hierarchies regulating the access to valuable resources, such as food, mates, social support and determine who constitutes a physical threat to whom. Learning about others’ relative dominance, and to adaptively respond to their threat value, is therefore central to health and survival (Kaplan and Manuck, 1999; Sapolsky, 2004, 2005; Adler et al., 2008; Selten et al., 2013). Although modern humans are likely to live in relatively more egalitarian groups than our ancestors (Boehm, 2009), many social milieus—from the school yard to the office—still contain situations where the social hierarchy is experienced as a threat (Sapolsky, 2004; Due et al., 2005), which might play a role in the development of neuropsychiatric diseases (Selten and Cantor-Graae, 2005; Meyer-Lindenberg and Tost, 2012; Selten et al., 2013). Yet, the behavioral and neural mechanisms underlying the interaction between dominance and threat learning remain unknown. Here, we addressed this by investigating the impact of the relative social rank of others on aversive learning. We examined behavioral, psychophysiological and neural responses to conditioned social dominance threat by employing an established experimental model to learn about others’ dominance through observation of social conflicts (Jones et al., 2011) and extended this with personal aversive confrontations during aversive (Pavlovian) learning.

Whereas past research on dominance in humans has focused on static dominance cues, such as facial features and body postures to study responses to dominance in general (Perrett et al., 1998; Oosterhof and Todorov, 2008; Marsh et al., 2009), social dominance rank across species is often determined by the probability to win agonistic confrontation against in-group conspecifics (Rowell, 1974). In naturalistic settings, such confrontations are not only relevant to the individuals involved in the confrontation (e.g. the agonist and responder) but also to everyone in the group. Indeed, across species, learning about the value of stimuli and situations through the observation of interactions between conspecifics is common (Olsson and Phelps, 2007; van Schaik, 2010) and reduces the risk of injury for the observer. In non-human animals, observing dyadic confrontations between conspecifics elicits physiological and behavioral responses comparable to the direct experience of social defeat (Warren et al., 2013) and is sufficient to learn a dominance rank hierarchy (Grosenick et al., 2007). In humans, learning about dominance of others can be employed by observation of stereotypic ‘social conflicts’, consisting of an angry agonist and a fearful responder. In an earlier study (Jones et al., 2011), participants increased the attribution of dominance to the angry face when it was observed in a dyad confronting a fearful as compared with another angry face. Yet, it is unclear to what extent dominance rank in others (learned through observation) biases defensive responses and a direct aversive experience with the same individuals. Moreover, the neural circuits in humans mediating these responses have so far not been investigated.

Across species, the amygdala has a central role in the detection and learning about threats (LeDoux, 2000; Davis and Whalen, 2001), and these responses are modulated by its close interconnection with regions of the medial prefrontal cortex (mPFC, Etkin et al., 2011). Interestingly, these nodes of the neurobiological circuitry processing threat partly overlap with brain regions involved in the processing of salient social information (Oosterhof and Todorov, 2008; Adolphs, 2010; Todorov, 2012).

In non-human primates, lesions of the amygdala affect dominance-related behaviors (Bard and Mountcastle, 1948; Rosvold et al., 1954; Emery et al., 2001; Bauman et al., 2004, 2006; Machado and Bachevalier, 2006; Bliss-Moreau et al., 2013) and the interactive behavior in response to dominance signals in others (Emery et al., 2001; Amaral, 2003). Additionally, recent neurobiological models in rodents suggest that the interaction between amygdala and mPFC activity regulates adaptive responses toward dominant conspecifics (Wang et al., 2014). In humans, the formation of associations between specific others and their social behaviors engage the amygdala and the mPFC (Baron et al., 2011). In particular, the joint action of these regions might be involved in declarative learning and disambiguating social hierarchies (Zink et al., 2008; Kumaran et al., 2012) in concert with the dorsolateral PFC and inferior parietal regions, which are involved in the representation of ranks in general (Chiao, 2010).

In three experiments, we examined how the relative dominance of others learned through observations of their confrontations (Figure 1a) modified basic threat responses and new subsequent fear learning to the same individuals. To this aim, we used psychophysiological (Experiments 1 and 2) and neurobehavioral (Experiment 3) methods (see Supplementary Figure S1 for overview), focusing on responses in the Amygdala. Each experiment consisted of two consecutive stages: first, in the observation of confrontation stage participants learned about other individuals’ relative social dominance rank through observing their face-to-face confrontations. Second, in the direct confrontation stage, participants subsequently confronted and learned about the same individuals through personal aversive experiences (Pavlovian conditioning) to examine the impact of social rank knowledge on direct fear learning.

Fig. 1.

Fig. 1.

Open in a new tab

Observation of confrontations between others. (a) Face-to-face confrontation between others (b) enabled participants to discriminate between social dominance rank during interspersed test-trials. (c) The individual history of confrontations outcomes of each face (d) shapes hierarchical representation of social dominance rank evolving over time in Experiment 1.

Materials and methods

Participants

In total, 84 participants were included and compensated for their participation in the study. In Experiment 1, 28 participants (17 females, 11 males) were included in the study, 23 participants in Experiment 2 (7 females, 16 males) and 23 (13 females, 10 males) in Experiment 3. All participants gave written informed consent and the local ethics committee in Stockholm approved all experiments.

Stimuli

Faces used in all experiments were depicted from the Karolinska directed emotional faces and used as a side-profile during the observation of confrontation stage or in a frontal view during aversive learning in the direct confrontation stage (Lundqvist et al., 1998) (items: M11, M31, M10). The neutral facial expression of all three faces have been rated comparable neutral in a previous study (hit-rate: 96.8–89.3%, Goeleven et al., 2008). The faces that were used as the Facedominant and Facesubordinate (M11 and M10) were counterbalanced between subjects and have been identified as equally angry (hit-rate for both: 100%) (Goeleven et al., 2008). To avoid differences in perception of upper-body strength [which has been shown to influence attribution of dominance (Sell et al., 2009)], all faces were mounted on the same torso.

Stimuli were presented using Presentation Software (NeuroBehavioral Systems, Albany, CA, USA). The unconditioned stimulus during the direct confrontation stage consisted of an electric stimulation (100 ms DC-pulse, STM200; Biopac Systems Inc, Santa Barbara, CA, USA) to the participants right wrist. Prior to the experiment, US intensity was individually adjusted to a level of maximum unpleasantness.

Experimental design

Observation of confrontation (social hierarchy learning)

Similar to a previous experiment (Jones et al., 2011), participants were presented with dyadic face-to-face confrontations (Figure 1a), presented for 3 s. The two faces were selected from a total of three facial stimuli, expressing anger or fear, depending on their position in the triadic dominance hierarchy. The highest ranked face (Facedominant) displayed an angry expression in each confrontation, whereas the Facesubordinate always displayed fear. Accordingly, the Faceintermediate always expressed fear when confronting the Facedominant and anger when confronting the Facesubordinate. To track the success of observational learning, participants were asked after each block (3 trials, all possible combinations of confrontations, 3 s durations) to select the face with the relatively higher dominance and rate their confidence in their selection on a scale from 1 (not sure) to 3 (very sure). Based on a previous study on learning of social hierarchies (Kumaran et al., 2012), we created an index of hierarchy knowledge by multiplying accuracy (0 = incorrect face/1 = correct face) with confidence-ratings (1–3).

Direct confrontation (fear conditioning)

Participants underwent a Pavlovian fear conditioning paradigm using the same faces as during the observation of confrontation stage, each presented for 6 s (intertrial interval 10–13 s). The Facedominant (now: CS+dominant) and the Facesubordinate (now: CS+subordinate) were presented (Experiment 2: 6 times; Experiment 3: 9 times) and followed by a mild electric shock as USA after 5.5 s (Experiment 2: 4 times; Experiment 3: 6 times). The Faceintermediate (now: CS−) served as a control stimulus and was never followed by US. The threat learning was followed by nine presentations of each CS without aversive stimulation (extinction phase) to study the persistence of learned fear once the Faces were ‘safe’. Additionally, in Experiment 3 return of fear responses were tested using a reinstatement procedure (3 CS presentation each after 3 unsignaled US presentations, for methodological details see Supplementary Methods and Haaker et al 2014).

In Experiment 2, all CSs were faces expressing anger, while in Experiment 3, neutral expression of all CSs were used.

Acquisition and data analysis

Subjective rating of fear, anger and trust

Before the observation of confrontation stage (baseline ratings), as well as before and after the aversive learning during the direct confrontation stage, participants provided explicit fear, anger and trust ratings on a screen showing a picture of each face and a visual analog scale [0 (none)–100 (maximum)]. Rating-values in each experiment were corrected for the baseline rating before the observation of confrontation to eliminate individual differences in the evaluation of facial cues (Oosterhof and Todorov, 2008). Two participants were excluded from the rating analysis in Experiment 3, due to technical difficulties with the response buttons within the functional magnetic resonance imaging (fMRI) environment.

Skin Conductance

Skin conductance was measured by a pair of Ag-AgCL electrodes attached to the distal phalanges of the index and middle finger of the left hand. The physiological signals were amplified and recorded using a Biopac 150 System (Biopac Systems Inc, Santa Barbara, CA, USA) and filtered between 0.05 and 5 Hz. Phasic skin conductance responses (SCRs) were scored as an increase in skin conductance within 0.5–4.5 s after stimulus onset and then logarithmized and range-corrected {ln[(SCR)/log(SCRmax)] + 1} as in previous studies (Haaker et al., 2013a,b).

Eye-blink startle

Startle reactions were measured by recording electromyographic (EMG) activity over the orbicularis oculi muscle beneath the left eye using miniature Ag/AgCl electrodes. The EMG signal was amplified and filtered through a BIOPAC MP-100 amplifier (BIOPAC Systems Inc, Goleta, CA, USA) and recorded with AcqKnowledge 4 software. Data were downsampled to 100 Hz and manually scored off-line from onset to peak within 20–120 ms after stimulus onset using a custom-made computer program, as described previously (Haaker et al., 2013b). Blink magnitudes were normalized using z-standardization and converted to T-scores to ensure that all participants contributed equally to the group mean. FPS data from a limited number of participants (4 participants) had insufficient data quality due to technical difficulties (as judged by two researches) and were thus excluded, leaving 24 participants (14 females, 10 males) in Experiment 1. Startle reactions were scored as missing if a blink occurred immediately before startle probe administration or due to obvious electrode artifacts.

Functional magnetic resonance imaging

fMRI data were obtained with a 3 Tesla MR scanner (General Electrics 750) using an 8-channel head coil. Each functional image volume comprised 46 continuous axial slices (2.3 mm thick, no gap) that were acquired using a T2*-sensitive gradient echo-planar imaging sequence [repetition time: 3000 ms; echo time: 31 ms; flip angle: 85°; field of view: 96 × 96 mm, covering the temporal lobe up to the cingulate cortex, 3 × 3 mm in-plane resolution]. To account for T1 equilibrium effects, the first five volumes of each time series were discarded. High-resolution T1-weighted structural images (1 × 1 × 1 mm) were acquired after the experimental session. Preprocessing using Statistical parametric mapping [SPM8, (www.fil.ion.ucl.ac.uk/spm)] running on Matlab2013b (The MathWorks, Natick, MA)] involved realignment, unwarping coregistration and normalization to a sample-specific template, using DARTEL (Ashburner, 2007). Normalized data series were spatially smoothed with a 6 mm FWHM isotropic Gaussian kernel and manually inspected for excessive head movement. Further processing included temporal high-pass filtering (cut-off 128 s) and correction for temporal autocorrelations using first-order autoregressive modeling (see Supplementary Methods for details).

A general linear model was set up for statistical first-level (single-subject) analysis. For the observation of confrontation stage, we included two regressors for the confrontation of the faces (one for the 45° angle and one for the face-to-face confrontations) and two regressors for the test-trials (one during the view of the faces and one during the response phase). We added to each of these regressors two parametrical modulators (coding the posing rank of the viewed stimuli and the social discrimination index for the specific pairing of the faces).

For the direct confrontation stage, we modeled each presentation of the CSs as separate regressors for conditioning and extinction and reinstatement-test (Lonsdorf et al., 2014). To each of these regressors, we added a time modulation capturing responses changing quadric over time as in a previous study of conditioned fear (Buchel et al., 1998) and since the psychophysiological data in Experiment 2 indicated a stimulus*time interaction in a quadratic temporal fashion [F(1,1) = 6.35; P = 0.022]. In addition, two nuisance regressors were included to factor out experimental effects of no interest: one regressor modeled the whole duration (as a boxcar function) of each ITI (including the rest period after the reinstatement-USs) and another nuiscance regressor modeled the all USs (as a stick function).

All regressors were convolved with a canonical hemodynamic response function.

Random effect analysis on the group level was performed using SPM’s ‘full factorial’ model and focused on the parametric responses during the observation of confrontation stage as well as the comparisons between the CS+dominant and the CS+subordinate during the direct confrontation stage. Analyses of the parametric responses during the observation of confrontation stage were calculated using one sample t-tests. Separate analyses for fear conditioning, extinction and reinstatement-test included beta-estimates for each CS (1 factor, 3 levels), derived from individual single subjects general linear modeling.

The amygdala ROI was defined as a probabilistic anatomical mask (http://www.cma.mgh.harvard.edu; threshold 0.7, Desikan et al., 2006).

P-values inside the ROI were corrected for multiple testing (small volume correction) using family-wise error (FWE) correction. Additionally, hypothesis generating effects outside our ROIs with a FWE corrected P-value for all independent voxels in the brain were reported as well.

In all analysis, an alpha level of 0.05 was adopted but marginally significant results (P < 0.10) are also reported.

Results

Observing dyadic confrontation shapes hierarchical representation of dominance rank

In the first experiment (Experiment 1), we asked if the observation of confrontations between other individuals shaped the learning of a social dominance rank hierarchy.

As expected, the result of the logistic linear mixed model (see SI) revealed that the hierarchy knowledge index increased with increasing number of observed confrontations (P < 0.001, Figure 1d, Supplementary Table S1). This effect interacted with the social dominance rank of the observed face (P = 0.019), resulting in a lower discrimination between the faces with the lowest social rank (Facesubordinate vs Faceintermediate) as compared to the pairs of faces with higher social rank in the first half of the trials (P < 0.006).

After the observation of confrontation stage, we asked participants to estimate their own level of dominance relative to the faces. We found that the self-reported relative social status of each participant exerted a main effect on the discrimination of social dominance rank (P = 0.047), when included in the regression model. Participants with high, as compared with low, self-reported social rank displayed a lower hierarchy knowledge index, i.e. relied less on observational learning. Additional analysis including gender of the participant did not change these results and did not significantly contribute as a factor of the model (Supplementary Table S1).

We then examined how the learned social dominance rank influenced subjective responses (fear, anger and trust) toward the faces. We observed that higher social dominance rank of a face was positively related to subjectively rated fear (main effect of rank, P = 0.017), and marginally so to subjective anger (P = 0.064), but not trust (P = 0.129). Participants rated the most fear toward the Facedominant as compared with the two other faces (P < 0.048). The marginally significant effect for rated anger showed a concordant pattern: participants rated more anger toward the faces with the highest social dominance rank as compared with the Facesubordinate (P < 0.048, Supplementary Table S2). Again, entering gender as a covariate did not change these results (Supplementary Table S2).

Emergence of social dominance rank modulates defensive responses

In Experiment 1, we measured defensive responding by examining the potentiation of the eye-blink startle reflex, which has been shown to be sensitive to the presence of learned threats, as well as inherently aversive stimuli (Grillon and Davis, 1997). To test if the learning about social dominance rank modulates defensive responses, we conducted a logistic linear mixed model, testing if startle responses during the observation of confrontations predicted the social rank discrimination during test-trials (i.e. the hierarchy knowledge index).

We found that increasing hierarchy knowledge was associated with decreasing potentiation of startle responses (P = 0.017, Figure 2a, Supplementary Table S3, controlled for unspecific habituation). In other words, Participants exhibited lower defensive responding during confrontation of others if participants had acquired knowledge about the dominance structure on test-trials. In turn, incomplete knowledge about the dominance structure (i.e. suggesting the experience of an unstable social hierarchy) predicted higher defensive responding of the participants. Startle response were not modulated by the social dominance rank of the faces in the confrontation or the relative self-reported social dominance rank of the participants.

Fig. 2.

Fig. 2.

Open in a new tab

(a) Decreasing startle responses (Experiment 1, N = 22 in Category 0, N = 13 in Category 1, N = 26 in Category 2, N = 25 in Category 3) and (b) amygdala reactivity (Experiment 3, N = 19 in Category 0, N = 9 in Category 1, N = 20 in Category 2, N = 21 in Category 3) with increasing hierarchy knowledge (social discrimination score, 0 = wrong discrimination, 1–3 = discrimination with increasing confidence) during observation of confrontations in others [right amygdala: x,y,z (MNI): 16;−2;−18; t = 4.44; z = 3.66; left amygdala: x,y,z (MNI) −22;−4;−16; t = 3.38; z = 2.97]. Note the similarity in response pattern. Bars display the mean parameter estimate for category; error-bars indicate SEM. Activation superimposed on an average structural image. Threshold of P < 0.01 for illustrative purposes.

High social dominance rank biases responses to learned threat

In Experiment 2, we examined how the varying degree of social dominance rank of the face biased participants’ threat responses during direct confrontations. We employed a Pavlovian fear conditioning paradigm using the same faces as during the observation of confrontation stage. Participants learned to associate an aversive event (mild electric shock, US) with the Facedominant (now: CS+dominant) and the Facesubordinate (now: CS+subordinate). The Faceintermediate (now: CS−) served as a control stimulus and was never followed by US. The threat learning was followed by unreinforced presentations of each CS without aversive stimulation (extinction phase) to study the persistence of learned fear once the Face were ‘safe’. The analysis of the observation of confrontation stage conceptually replicated the results in Experiment 1 (Supplementary Tables S4 and S5). Whereas the main effect of rank on the subjective rating of fear was marginally significant (P = 0.06, no effect for subjective anger or trust, Supplementary Table S5), participants rated significantly more fear toward the Facedominant as compared with the Facesubordinate (paired comparison P = 0.033) confirming that participants effectively learned about the social dominance rank of others.

During the subsequent direct confrontation stage, we tested if the acquired knowledge of social dominance rank biased the observers’ conditioned fear learning. We found that responses were higher toward the CS+dominant as compared with the unpaired CS− (P = 0.010), and marginally higher toward the CS+subordinate vs the CS− (P = 0.063; Figure 3b, Supplementary Table S6). During the subsequent extinction, when the CS+s were not longer followed by US, we observed persistently higher responses to the CS+dominant as opposed to a decline (from Blocks 1 to 3) in responding to the CS+subordinate (stimulus by time interaction, P = 0.043; decrease in SCR from Block 1–3: CS+dominant P = 0.6; CS− P = 0.7; CS+subordinate P = 0.051; Figure 3b, Supplementary Table S6).

Fig. 3.

Fig. 3.

Open in a new tab

Experiment 2: (a) aversive learning (fear conditioning) through paring of the dominant and subordinate face with an aversive stimulus. (b) During extinction, SCRs were more sustained to the CS+dominant as compared with the CS+subordinate. The asterisks indicate a P-value of 0.043 for the interaction between stimulus-type and block.

This pattern indicated that conditioned responses to dominant as compared with subordinated faces were more persistent during extinction.

Neural corollaries of observational learning through others’ confrontations

In both Experiments 1 and 2, we showed that participants learned social dominance rank through the observation of confrontations, and that subsequent direct confrontations with highly ranked faces evoked sustained threat responses that resisted extinction (Experiment 2). Next, in Experiment 3 we used fMRI, to investigate the hemodynamic responses underlying the observational learning about social dominance rank, as well as subsequent threat responses in direct confrontations with highly ranked faces.

First, we again replicated the behavioral results of social rank discrimination during test-trials (as shown in Experiments 1 and 2), indicating the social dominance rank was learned through the observation of confrontation (Supplementary Tables S7 and S8). Similar to the analysis of the startle responses (Experiment 1), we included the individual time-course of social rank discrimination, as well as the dominance rank of the observed face as an additional (parametric) regressor in the fMRI-analysis. This was done by the inclusion of two parametrical modulators coding (i) the posing rank of the viewed faces and (ii) the individual social discrimination index for the faces during each confrontation.

The parametric regressor of the individual social discrimination index for the specific pairing during the confrontation of others revealed that social dominance rank discrimination was negatively associated with responses in the amygdala. Increasing social rank discrimination predicted decreasing responses in the right amygdala [P(FWE) = 0.014] and trend-wise in the left amygdala [P(FWE) = 0.069, Figure 2b and c].

The relationship between decreasing responses in the amygdala and increasing hierarchy knowledge might be more complex then a linear association, yet these results closely mimicked the decreasing startle potentiation as a function of increasing social rank discrimination as observed in Experiment 1 (Figure 2a).

Additionally, the parametric regressor coding the rank of the posing faces during the observation of confrontations revealed activity within the right dorsolateral PFC (outside our ROI but significant as corrected for independent voxels in the brain; P FWE, whole brain = 0.043, Supplementary Table S10)

Next, we analyzed the same parametric regressors during test-trials, and in support of previous research on transitive inference of social ranks (Kumaran et al., 2012), we showed that the increase in rank discrimination performance positively predicted amygdala responses [P(FWE) = 0.024, Supplementary Figure S4a]. Negative responses in the amygdala were found for test-trials with low discrimination performance and positive responses in test-trials with the highest discrimination.

In addition, the same parametric regressor coding the individual social rank discrimination during test-trials predicted responses in the mid-cingulate cortex (outside our ROI but significant as corrected for independent voxels in the brain; P FWE, whole brain = 0.023, Supplementary Table S10). Here, the pattern of activity was not following a linear increase of responses as a function of social rank discrimination (Supplementary Figure S4b). Instead, activity was higher during test-trials with successful discrimination with either no or high confidence. Lower activity was found during test-trials with wrong discrimination of social rank as well as successful discrimination with uncertain confidence.

Taken together, our results showed that learning about social dominance rank of others by observing their confrontation involves amygdala activity, responses in the dorsolateral PFC and mid-cingulate cortex of the observers.

The amygdala is responsive to dominant social threat

Next, we examined the hemodynamic responses underlying the effect of the social dominance rank on direct threat learning. Notably, we used neutral expression of all CSs to test that the effects obtained in Experiment 2 were independent of the facial expression of the CS. The psychophysiological results in Experiment 3 during threat acquisition closely mirrored our previous finding (Experiment 2) showing that SCRs were potentiated only toward the CS+dominant as compared with the CS−, (P = 0.042) (Supplementary Table S9 and Figure S1). Hemodynamic responses in the amygdala displayed a convergent pattern with the SCRs, i.e. higher responses to the CS+dominant as compared with the CS− in the right amygdala, whereas no activity in the amygdala was significantly greater to the CS+subordinate relative to the CS− [P(FWE) = 0.031, Supplementary Figure S4].

In addition, outside our ROIs, we found patterns of activity similar to the one observed in the amygdala and the midbrain (resembling the substantia nigra/ventral tegmental area), displaying greater responses to the CS+dominant as compared with the CS− [P(FWE, whole brain) = 0.002, Supplementary Table S11]. Importantly, we found a difference in the temporal dynamics of the responses between the CS+dominant and the CS+subordinate. Comparing the quadratic change of responses over time, we observed that the CS+dominant showed higher exponentially increasing responses over trials as compared with the CS+subordinate in the right amygdala [P(FWE) = 0.010 and P(FWE) = 0.039, Figure 4a]. A comparable pattern emerged in the left amygdala during extinction, consisting of higher exponentially increasing responses over time toward the CS+dominant as compared with the CS+subordinate [P(FWE) = 0.034, Figure 4b]. This pattern mirrored the psychophysiological findings in Experiment 2, as well as the higher SCRs toward the CS+dominant as compared with the CS+subordinate (P = 0.020) in the fMRI Experiment (see earlier and Supplementary Table S9).

Fig. 4.

Fig. 4.

Open in a new tab

Experiment 3: (a) amygdala activity was more sustained during acquisition of threat toward the CS+dominant as compared with the CS+subordinate [right amygdala: x,y,z (MNI) 28;0;−14; t = 3.75; z = 3.56; right amygdala: P(FWE) = 0.010; x,y,z (MNI) 18;−6;−20; t = 3.23; z = 3.10] and (b) more sustained to the CS+dominant as compared with the CS+subordinate during extinction of conditioned threat [left amygdala: x,y,z (MNI) −30,−6,−20; t = 3.19; z = 3.06]. The asterisks indicate significant differences within the amygdala ROI (small volume corrected, P = 0.01 for acquisition and P = 0.034 for extinction). Activation superimposed on an average structural image. Threshold of P < 0.01 for illustrative purposes. Bars display the mean parameter estimate for each CS; error-bars indicate SEM.

In Experiment 3, we additionally administered three unexpected shocks (reinstatement) after extinction to induce the return of threat responses. We observed a CS+ specific increase (for both, CS+dominant and the CS+subordinate) of SCRs from extinction to the reinstatement-test (P = 0.006), displaying a reinstated pattern of conditioned fear with higher responses to the CS+dominant (P = 0.001), but not the CS+subordinate (P = 0.124, Supplementary Table S9 and Figure S3), when compared with the CS−. During reinstatement-test, we did not observe any differences between the CS+s in the fMRI analysis inside the amygdala ROI. However, outside the ROI, responses in the anterior cingulate cortex were marginally larger to the CS+dominant as compared with the CS+subordinate [P(FWE, whole brain) = 0.074, Supplementary Table S11].

Discussion

We show in three independent experiments how the observed relative dominance level of others is threat relevant. Consistent with previous research, observing confrontations between others is sufficient to learn a linear hierarchy of social dominance rank (Jones et al., 2011). Accordingly, the observation of confrontations modulated basic defensive responses (eye-blink startle) and amygdala activity in the observer, and biased subsequent threat learning through direct, aversive experiences.

In addition to being influenced by the social rank of the confronting others’, observational learning was influenced by the self-reported (relative) social rank of the observer. Indeed, participants were better in learning the social rank of individuals above themselves in social rank, consistent with the fact that humans (and other primates) pay more attention to conspecifics with higher rank (Shepherd et al., 2006; Klein et al., 2009; Jones et al., 2010). Moreover, participants that placed themselves in a high position relative to the confronting others’ showed a weaker learning of the social hierarchy, supporting previous research showing that aggressive dominant humans rely relatively less on social learning experiences (Cook et al., 2014). Moreover, our results showed that observational learning about outcome of social conflicts between others (i.e. inferring the social rank of others) influenced the defensive responding (i.e. potentiated eye-blink startle reflex) and amygdala activity during the observation of confrontations. In two independent samples, we showed that defensive responses (Experiment 1) and amygdala reactivity (Experiment 3) decreased with increasing knowledge about the social rank. These results dovetail with the observation that unstable hierarchies are perceived as stressful in primates, even in animals that hold a high social rank position (Sapolsky, 2005). In humans, competition with others within an unstable hierarchy engages more amygdala activity as compared with a stable hierarchy (Zink et al., 2008). Our findings suggest that the amygdala serves to integrate information about others’ relative dominance to resolve ambiguous social information (Davis and Whalen, 2001).

Consistent with this interpretation, uncertainty about the social status of the faces during confrontations might have contributed to the observed responses in the amygdala. Following this, responses in this region could reflect an additional recruitment of attention to resolve the uncertainty of the social situation (Todorov, 2012).

During test-trials, when participants succeeded to discriminate between levels of dominance, the amygdala activity increased as a function of accumulated knowledge about the hierarchy. This finding mirrors the effect described by Kumaran et al. (2012), where the effect of increasing amygdala activity as function of hierarchy knowledge was compared with a non-social control. Additionally, we found that the mid cingulum coded hierarchy knowledge in concert with the amygdala, yet with a different pattern that might be related to uncertainty.

To sum up, we show that when participants had less knowledge about the social hierarchy, their amygdala was engaged during the observation of confrontations. As the knowledge about the social hierarchy developed, the amygdala was no longer involved during the observation of confrontation but during the discriminations between the social ranks of others. In our study, the attributions of social rank to others were acquired through observing the outcome history of dyadic confrontations, offering an associative learning account for the establishment of social dominance. According to this account, amygdala activity during observing confrontations in others should decrease, while its activity to the associatively learned attribution of social rank should increase. Indeed, this is what we found. A corresponding pattern has been observed in associative learning paradigms (e.g. fear conditioning), were expectancy and anticipation is reflected by activity of the amygdala (McNally et al., 2011).

Although the results in our study are framed within the context of dominance learning, we cannot disentangle categorization of dominance from processes elicited by emotional expression and/or saliency of the facial stimuli during the confrontations (Oosterhof and Todorov, 2008). As such, psychophysiological responses and activity in the amygdala might be biased by the emotional expressions of the faces. Accordingly, the relatively greater saliency of the angry faces might have triggered more attention [along with amygdala activation (Davis and Whalen, 2001; Todorov, 2012)], and thus enhanced later learning. Nevertheless, these explanations are in line with our conclusion that the observation of social conflicts in others biases direct experiences with the observed conspecifics. Here, we frame the results in the context of dominance categorization.

Apart from the described activity in the amygdala, we found that during the discrimination of social rank, activity within the dorsolateral PFC was modulated by the social ranks of the confronting individuals. Studies in rodents have suggested that the PFC is involved in the representation of a dominance rank and is an important regulatory structure to gate adaptive behavior toward dominant and subordinate conspecifics (Wang et al., 2014). Additionally, we found responses in the mid-cingulate cortex that coded hierarchy knowledge during the discrimination (test-trials) of the dominance status of others. The pattern of the mid-cingulate was different as compared with the activity in the amygdala and it is tempting to speculate that the combination of cortico-limbic interaction between these regions mediate the learning and the representation of a social dominance hierarchy in humans. We can, however, not conclude that the aforementioned results are exclusive for hierarchical learning of social stimuli, since no non-social control of hierarchical learning was employed in this study.

Following the establishment of a dominance hierarchy, our behavioral, psychophysiological and hemodynamic measures consistently showed that high social dominance rank in others serves as a potent social threat, especially when paired with a direct aversive experience (here a shock). Previous studies show that conditioned fear to threat-relevant social stimuli, such as angry and fearful faces (Orr and Lanzetta, 1980; Dimberg and Öhman, 1996), and faces belonging to social out-group members (Olsson et al., 2005), results in strong learned threat responses that are resistant to change. In these studies, the social stimuli themselves contained threat-relevant visual cues, such as negative emotional expressions or out-group features. Importantly, and in contrast to these studies, in the current experiment, both angry (Exp. 2) and neutral (Exp. 3) looking faces were presented during the direct confrontation stage, which implies that the enhanced threat responses to high dominance faces must have been acquired through the preceding observation of confrontations. The stronger threat learning toward the dominant, as compared with subordinate, face was reflected by potentiated SCR responses and sustained activity in the amygdala during acquisition of conditioned threat a pattern that was resistant to the extinction of conditioned fear. Whereas a similar resistance to extinction has been previously found to male faces from social out-groups (Olsson et al., 2005), our experiments are the first to show that relative dominance rank bias threat learning toward social stimuli.

Our study used only male faces as stimuli but the sample consisted of both genders. Fear conditioning has shown to be affected by gender (Milad et al., 2009; Merz et al., 2010, 2011; Lonsdorf et al., 2015) but we did not observe any effects of participant gender on conditioned dominance threats in our results. The absence of this effect is consistent with previous studies suggesting that fear responses toward male fear-relevant dominance cues (angry faces) are enhanced in both, males and females, whereas female angry faces only showed enhanced responding in female participants (Öhman and Dimberg, 1978; Dimberg and Öhman, 1996). To our knowledge, our findings are the first to show that watching social confrontations not only transmits information about relative dominance of the antagonists to bystanders but also biases future threat learning about these individuals as a function of their relative dominance.

Taken together, the results obtained in three separate experiments indicate that social knowledge about others biases both our direct experience and our subsequent learning about these individuals. Our results might be explained within a framework of prepared learning and belongingness between dominance attributes and threat learning (Seligman, 1971; Öhman and Dimberg, 1978; Öhman and Mineka, 2001) but future studies are needed to investigate the underlying mechanisms in more detail. This pattern of biased responding might contribute to our understanding the emergence and maintenance of the organizational structure of social groups. A biased responding to real or imagined dominant others might not always be adaptive, and has been suggest as a hallmark of neuropsychiatric disorders (Selten and Cantor-Graae, 2005; Meyer-Lindenberg and Tost, 2012; Selten et al., 2013). Moreover, the biased persistence of social dominant threat could mark one step within a cascade of stress responses to inequalities of social rank that are thought to result in maladaptive cardiovascular and pro-inflammatory responses and changes in neurogenesis (Kaplan and Manuck, 1999; Sapolsky, 2004, 2005; McEwen, 2012). Such a learning bias might have been adaptive in our ancestral environment and might still be so in certain contexts. Yet, this response pattern might also represent what go awry in dysfunctional social avoidance, and experiences of fear and defeat that are commonly seen in clinical social anxiety and depression. Our results can also be linked to recent findings on the relevance of the experience of being low in social status for the stress responses as implicated in neuropsychiatric illness (Akdeniz et al., 2014).

Our findings describe basic biological processes underlying the learning and responding to social dominance, which is central to the individual’s health and survival, as well as to the functional organization of our society at large. We hope that our novel experimental model of the formation of a social dominance hierarchy can be used in future research on the processes underlying this universal social phenomenon in humans, and possibly also in other primates.

Supplementary Material

Supplementary Data
supp_11_10_1627__index.html (910B, html)

Acknowledgements

The authors thank Emily Holmes, Erno Hermans, Lars Hall and Mats Lekander for comments on an earlier version of this article.

Funding

This research was supported by an Independent Starting Grant (284366; Emotional Learning in Social Interaction) from the European Research Council to A.O.

Supplementary data

Supplementary data are available at SCAN online.

Conflict of interest. None declared.

References

  1. Adler N., Singh-Manoux A., Schwartz J., Stewart J., Matthews K., Marmot M.G. (2008). Social status and health: a comparison of British civil servants in Whitehall-II with European-and African-Americans in CARDIA. Social Science & Medicine, 66, 1034–45. [DOI] [PubMed] [Google Scholar]
  2. Adolphs R. (2010). What does the amygdala contribute to social cognition? Annals of the New York Academy of Sciences, 1191, 42–61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Akdeniz C., Tost H., Streit F., et al. (2014). Neuroimaging evidence for a role of neural social stress processing in ethnic minority-associated environmental risk. Journal of the American Medical Association Psychiatry, 71, 672–80. [DOI] [PubMed] [Google Scholar]
  4. Amaral D.G. (2003). The amygdala, social behavior, and danger detection. Annals of the New York Academy of Sciences, 1000, 337–47. [DOI] [PubMed] [Google Scholar]
  5. Ashburner J. (2007). A fast diffeomorphic image registration algorithm. NeuroImage, 38, 95–113. [DOI] [PubMed] [Google Scholar]
  6. Bard P., Mountcastle V.B. (1948). Suppression and facilitation of angry behavior by certain forebrain mechanisms. Science, 107, 457.. [PubMed] [Google Scholar]
  7. Baron S.G., Gobbini M.I., Engell A.D., Todorov A. (2011). Amygdala and dorsomedial prefrontal cortex responses to appearance-based and behavior-based person impressions. Social Cognitive and Affective Neuroscience, 6, 572–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bauman M.D., Lavenex P., Mason W.A., Capitanio J.P., Amaral D.G. (2004). The development of social behavior following neonatal amygdala lesions in rhesus monkeys. Journal of Cognitive Neuroscience, 16, 1388–411. [DOI] [PubMed] [Google Scholar]
  9. Bauman M.D., Toscano J.E., Mason W.A., Lavenex P., Amaral D.G. (2006). The expression of social dominance following neonatal lesions of the amygdala or hippocampus in rhesus monkeys (Macaca mulatta). Behavioral Neuroscience, 120, 749–60. [DOI] [PubMed] [Google Scholar]
  10. Bliss-Moreau E., Moadab G., Bauman M.D., Amaral D.G. (2013). The impact of early amygdala damage on juvenile rhesus macaque social behavior. Journal of Cognitive Neuroscience, 25, 2124–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Boehm C. (2009). Hierarchy in the Forest: The Evolution of Egalitarian Behavior. Harvard: Harvard University Press. [Google Scholar]
  12. Buchel C., Morris J., Dolan R.J., Friston K.J. (1998). Brain systems mediating aversive conditioning: an event-related fMRI study. Neuron, 20, 947–57. [DOI] [PubMed] [Google Scholar]
  13. Chiao J.Y. (2010). Neural basis of social status hierarchy across species. Current Opinion in Neurobiology, 20, 803–9. [DOI] [PubMed] [Google Scholar]
  14. Cook J.L., den Ouden H.E., Heyes C.M., Cools R. (2014). The social dominance paradox. Current Biology, 24, 2812–6. [DOI] [PubMed] [Google Scholar]
  15. Davis M., Whalen P.J. (2001). The amygdala: vigilance and emotion. Molecular Psychiatry, 6, 13–34. [DOI] [PubMed] [Google Scholar]
  16. Desikan R.S., Ségonne F., Fischl B., et al. (2006). An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. NeuroImage, 31, 968–80. [DOI] [PubMed] [Google Scholar]
  17. Dimberg U., Öhman A. (1996). Behold the wrath: psychophysiological responses to facial stimuli. Motivation and Emotion, 20, 149–82. [Google Scholar]
  18. Due P., Holstein B.E., Lynch J., et al. (2005). Bullying and symptoms among school-aged children: international comparative cross sectional study in 28 countries. European Journal of Public Health, 15, 128–32. [DOI] [PubMed] [Google Scholar]
  19. Emery N.J., Capitanio J.P., Mason W.A., Machado C.J., Mendoza S.P., Amaral D.G. (2001). The effects of bilateral lesions of the amygdala on dyadic social interactions in rhesus monkeys (Macaca mulatta). Behavioral Neuroscience, 115, 515.. [PubMed] [Google Scholar]
  20. Etkin A., Egner T., Kalisch R. (2011). Emotional processing in anterior cingulate and medial prefrontal cortex. Trends in Cognitive Sciences, 15, 85–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Goeleven E., De Raedt R., Leyman L., Verschuere B. (2008). The Karolinska directed emotional faces: a validation study. Cognition and Emotion, 22, 1094–118. [Google Scholar]
  22. Grillon C., Davis M. (1997). Fear-potentiated startle conditioning in humans: explicit and contextual cue conditioning following paired versus unpaired training. Psychophysiology, 34, 451–8. [DOI] [PubMed] [Google Scholar]
  23. Grosenick L., Clement T.S., Fernald R.D. (2007). Fish can infer social rank by observation alone. Nature, 445, 429–32. [DOI] [PubMed] [Google Scholar]
  24. Haaker J., Gaburro S., Sah A., et al. (2013a). Single dose of L-dopa makes extinction memories context-independent and prevents the return of fear. Proceedings of the National Academy of Sciences of the United States of America, 110, E2428–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Haaker J., Golkar A., Hermans D., Lonsdorf T.B. (2014). A review on human reinstatement studies: an overview and methodological challenges. Learning & Memory, 21, 424–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Haaker J., Lonsdorf T.B., Thanellou A., Kalisch R. (2013b). Multimodal assessment of long-term memory recall and reinstatement in a combined cue and context fear conditioning and extinction paradigm in humans. PLoS One, 8, e76179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Jones B.C., DeBruine L.M., Little A.C., Watkins C.D., Feinberg D.R. (2011). “Eavesdropping” and perceived male dominance rank in humans. Animal Behaviour, 81, 1203–8. [Google Scholar]
  28. Jones B.C., DeBruine L.M., Main J.C., et al. (2010). Facial cues of dominance modulate the short-term gaze-cuing effect in human observers. Proceedings of the Royal Society of London. Series B, Biological sciences, 277, 617–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kaplan J.R., Manuck S.B. (1999). Status, stress, and atherosclerosis: the role of environment and individual behavior. Annals of the New York Academy of Sciences, 896, 145–61. [DOI] [PubMed] [Google Scholar]
  30. Klein J.T., Shepherd S.V., Platt M.L. (2009). Social attention and the brain. Current Biology, 19, R958–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Kumaran D., Melo H.L., Duzel E. (2012). The emergence and representation of knowledge about social and nonsocial hierarchies. Neuron, 76, 653–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. LeDoux J.E. (2000). Emotion circuits in the brain. Annual Review of Neuroscience, 23, 155–84. [DOI] [PubMed] [Google Scholar]
  33. Lundqvist D., Flykt A., Öhman A. (1998). The Karolinska Directed Emotional Faces - KDEF, CD ROM from Department of Clinical Neuroscience, Psychology section, Karolinska Institutet, ISBN 91-630-7164-9. [Google Scholar]
  34. Lonsdorf T.B., Haaker J., Kalisch R. (2014). Long-term expression of human contextual fear and extinction memories involves amygdala, hippocampus and ventromedial prefrontal cortex: a reinstatement study in two independent samples. Social Cognitive and Affective Neuroscience, 9(12), 1973–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Lonsdorf T.B., Haaker J., Schümann D., et al. (2015). Sex differences in conditioned stimulus discrimination during context-dependent fear learning and its retrieval in humans: the role of biological sex, contraceptives and menstrual cycle phases. Journal of Psychiatry & Neuroscience, 40, 140336.. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Machado C.J., Bachevalier J. (2006). The impact of selective amygdala, orbital frontal cortex, or hippocampal formation lesions on established social relationships in rhesus monkeys (Macaca mulatta). Behavioral Neuroscience, 120, 761.. [DOI] [PubMed] [Google Scholar]
  37. Marsh A.A., Blair K.S., Jones M.M., Soliman N., Blair R.J.R. (2009). Dominance and submission: the ventrolateral prefrontal cortex and responses to status cues. Journal of Cognitive Neuroscience, 21, 713–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. McEwen B.S. (2012). Brain on stress: how the social environment gets under the skin. Proceedings of the National Academy of Sciences of the United States of America, 109(Suppl 2), 17180–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. McNally G.P., Johansen J.P., Blair H.T. (2011). Placing prediction into the fear circuit. Trends in Neurosciences, 34, 283–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Merz C.J., Tabbert K., Schweckendiek J., et al. (2010). Investigating the impact of sex and cortisol on implicit fear conditioning with fMRI. Psychoneuroendocrinology, 35, 33–46. [DOI] [PubMed] [Google Scholar]
  41. Merz C.J., Tabbert K., Schweckendiek J., et al. (2011). Neuronal correlates of extinction learning are modulated by sex hormones. Social Cognitive and Affective Neuroscience, 7, 819–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Meyer-Lindenberg A., Tost H. (2012). Neural mechanisms of social risk for psychiatric disorders. Nature Neuroscience, 15, 663–8. [DOI] [PubMed] [Google Scholar]
  43. Milad M.R., Igoe S.A., Lebron-Milad K., Novales J.E. (2009). Estrous cycle phase and gonadal hormones influence conditioned fear extinction. Neuroscience, 164, 887–95. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Öhman A., Dimberg U. (1978). Facial expressions as conditioned stimuli for electrodermal responses: a case of “preparedness”? Journal of Personality and Social Psychology, 36, 1251.. [DOI] [PubMed] [Google Scholar]
  45. Öhman A., Mineka S. (2001). Fears, phobias, and preparedness: toward an evolved module of fear and fear learning. Psychological Review, 108, 483–522. [DOI] [PubMed] [Google Scholar]
  46. Olsson A., Ebert J.P., Banaji M.R., Phelps E.A. (2005). The role of social groups in the persistence of learned fear. Science, 309, 785–7. [DOI] [PubMed] [Google Scholar]
  47. Olsson A., Phelps E.A. (2007). Social learning of fear. Nature Neuroscience, 10, 1095–102. [DOI] [PubMed] [Google Scholar]
  48. Oosterhof N.N., Todorov A. (2008). The functional basis of face evaluation. Proceedings of the National Academy of Sciences of the United States of America, 105, 11087–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Orr S.P., Lanzetta J.T. (1980). Facial expressions of emotion as conditioned stimuli for human autonomic responses. Journal of Personality and Social Psychology, 38, 278–82. [DOI] [PubMed] [Google Scholar]
  50. Perrett D.I., Lee K.J., Penton-Voak I., et al. (1998). Effects of sexual dimorphism on facial attractiveness. Nature, 394, 884–7. [DOI] [PubMed] [Google Scholar]
  51. Rosvold H.E., Mirsky A.F., Pribram K.H. (1954). Influence of amygdalectomy on social behavior in monkeys. Journal of Comparative and Physiological Psychology, 47, 173.. [DOI] [PubMed] [Google Scholar]
  52. Rowell T.E. (1974). The concept of social dominance. Behavioral Biology, 11, 131–54. [DOI] [PubMed] [Google Scholar]
  53. Sapolsky R.M. (2004). Social status and health in humans and other animals. Annual Review of Anthropology, 33, 393–418. [Google Scholar]
  54. Sapolsky R.M. (2005). The influence of social hierarchy on primate health. Science, 308, 648–52. [DOI] [PubMed] [Google Scholar]
  55. van Schaik C.P. (2010). Social learning and culture in animals In: Kappeler P., editor. Animal Behaviour: Evolution and Mechanisms, Berlin: Springer, pp. 623–53. [Google Scholar]
  56. Seligman M.E.P. (1971). Phobias and preparedness. Behavior Therapy, 2, 307–20. [DOI] [PubMed] [Google Scholar]
  57. Sell A., Cosmides L., Tooby J., Sznycer D., von Rueden C., Gurven M. (2009). Human adaptations for the visual assessment of strength and fighting ability from the body and face. Proceedings of the Royal Society of London. Series B, Biological Sciences, 276, 575–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Selten J.P., Cantor-Graae E. (2005). Social defeat: risk factor for schizophrenia? The British Journal of Psychiatry: The Journal of Mental Science, 187, 101–2. [DOI] [PubMed] [Google Scholar]
  59. Selten J.P., van der Ven E., Rutten B.P.F., Cantor-Graae E. (2013). The social defeat hypothesis of schizophrenia: an update. Schizophrenia Bulletin, 39, 1180–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Shepherd S.V., Deaner R.O., Platt M.L. (2006). Social status gates social attention in monkeys. Current Biology, 16, R119–20. [DOI] [PubMed] [Google Scholar]
  61. Todorov A. (2012). The role of the amygdala in face perception and evaluation. Motivation and Emotion, 36, 16–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Wang F., Kessels H.W., Hu H. (2014). The mouse that roared: neural mechanisms of social hierarchy. Trends in Neuroscience, 37, 674–82. [DOI] [PubMed] [Google Scholar]
  63. Warren B.L., Vialou V.F., Iñiguez S.D., et al. (2013). Neurobiological sequelae of witnessing stressful events in adult mice. Biological Psychiatry, 73, 7–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Zink C.F., Tong Y., Chen Q., Bassett D.S., Stein J.L., Meyer-Lindenberg A. (2008). Know your place: neural processing of social hierarchy in humans. Neuron, 58, 273–83. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data
supp_11_10_1627__index.html (910B, html)
supp_nsw074_scan-16-031-File006.docx (1.9MB, docx)

Articles from Social Cognitive and Affective Neuroscience are provided here courtesy of Oxford University Press

RESOURCES