Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Feb 1.
Published in final edited form as: J Trauma Stress. 2020 Aug 20;34(1):241–247. doi: 10.1002/jts.22577

Social Anhedonia is Associated with Low Social Network Diversity in Trauma-Exposed Adults

Elizabeth A Olson 1, Diego A Pizzagalli 1, Isabelle M Rosso 1
PMCID: PMC7903974  NIHMSID: NIHMS1623295  PMID: 32816343

Abstract

Social anhedonia has been proposed to contribute to social isolation in several psychiatric disorders, but has not been examined in relation to deficits in social connection that also characterize posttraumatic stress disorder (PTSD). A growing body of evidence emphasizes the health importance of structural features of social networks, including their size and complexity. The current study examined the relationship between social anhedonia and social network features, in a sample of trauma-exposed participants with and without PTSD, as well as healthy controls. Participants (N = 101: 37 healthy control, 23 trauma-exposed non-PTSD control, 41 lifetime PTSD) completed self-report measures of social anhedonia (Revised Social Anhedonia Scale) and structural social network features including social network size, diversity, and number of embedded networks (Social Network Index). Relative to healthy controls, PTSD participants had significantly lower social network size and number of embedded networks. In the combined trauma-exposed sample, greater social anhedonia was associated with lower social network diversity, r(62) = −.43, p < .001, an effect that remained statistically significant after controlling for PTSD and depression symptom severity. These results suggest that elevated social anhedonia in trauma-exposed participants may contribute to disruptions in social network structure consistent with social isolation.

Keywords: trauma, posttraumatic stress disorder, anhedonia, social networks

Background

Social anhedonia refers to a reduced ability to experience pleasure and reward from social interactions (Barkus & Badcock, 2019). While social anhedonia has been extensively studied in psychiatric disorders including schizophrenia and major depression (Chapman et al., 1976; Kwapil et al., 2008), a growing literature acknowledges the importance of social anhedonia in PTSD as a core component that may contribute to feelings of social detachment or estrangement (DSM-5 Criterion D6: Nawijn et al., 2015; Olson et al., 2018). However, social anhedonia has not been examined in relation to deficits in social connection that characterize PTSD. One approach to objectively measuring social connections is social network analysis (Bryant et al., 2017), which identifies the structural features of connections (networks) rather than focusing on the content or quality of social relationships (Hammer, 1981). These structural features assess social network size and complexity, including variables such as network density (number of network members who know other members), diversity (number of different social roles), and embeddedness (number of different high-contact social roles) (Cohen & Wills, 1985). In the general population, structural features of social networks are associated with poor mental health outcomes such as higher suicide risk (Handley et al., 2012; Sripada et al., 2015). These structural features also predict increased physical morbidity, including reduced immune response and heightened cardiovascular risk (Ford et al., 2006; Molesworth et al., 2015). Despite evidence that social anhedonia contributes to PTSD symptoms, and broader research examining social network features as predictors of emotional well-being, no study has examined social anhedonia in relation to social network structure in PTSD.

An emerging literature points to associations between PTSD diagnosis and alterations in structural features of social networks. Most prior studies of social connection in PTSD have focused on social support, which is an important function that social networks can provide. Kaniasty & Norris (2008) showed that greater social support predicts less symptom severity in the initial months after trauma exposure, and that PTSD symptoms lead to a progressive erosion of social support at later stages of illness. In a study that examined structural features of supportive relationships in a recently traumatized sample, Lee and Youm (2011) found that Korean refugees with a greater number of supportive connections had lower risk of later developing PTSD. In a study that separately examined structural social network features versus social support, PTSD diagnosis was more strongly associated with low social network diversity than with low perceived availability of social support (Platt et al., 2014). Finally, Bryant et al. (2017) demonstrated that people with PTSD were less likely to be named (“nominated”) by other people as a member of their social network. Altogether, this literature suggests that having an established PTSD diagnosis, or longstanding symptoms, is associated with a loss of social connections and deterioration of social network structure. Thus, identifying cognitive-affective processes that might underlie the association between PTSD diagnosis and altered social network features is an important goal. In healthy populations, social reward valuation has been proposed as a core process driving individual differences in social network size and complexity. Recent evidence suggests that affiliative processes influence individual differences in social network features (Bickart et al., 2012). Positive social stimuli (e.g., attractive faces) and positive outcomes of social interactions (e.g., approval, cooperation) are rewarding, and these social rewards may support social network size and complexity by increasing motivation for social interaction (Fareri & Delgado, 2014). A recent large online study of community adults demonstrated that social anhedonia was associated with social network features at the population level (Dodell-Feder et al., 2020). Together, these findings support examining the relationship between social anhedonia and social network features in posttraumatic stress samples.

Given the proposed role of social reward processing in influencing social network features via affiliative processes and evidence that social anhedonia relates to social network structure in the general population, we hypothesized that social anhedonia would be associated with altered social network features in the context of trauma-related psychopathology.

Method

Participants

Data from two studies of decision-making following trauma exposure were combined for the present analysis (total N = 101). The first study (N = 56) included healthy control (HC) participants (N = 15), trauma-exposed non-PTSD controls (TENC, N = 23, with no lifetime history of meeting full criteria for PTSD), and participants with current (N = 10, meeting full criteria for past month) or lifetime (N = 8, meeting full criteria for worst month but not past month, i.e., partially remitted) DSM-5 PTSD. Inclusion criteria were: age 20–50 years, no history of lifetime Axis I diagnosis for HC group, trauma exposure consistent with group status. Exclusion criteria were: history of neurological disorder; history of head trauma with loss of consciousness greater than 5 minutes; or history of psychotic disorder, bipolar disorder, eating disorder, intellectual disability, pervasive developmental disorder, or obsessive compulsive disorder. Psychotropic medications were exclusionary, aside from a stable (6-week) dose of antidepressant medications in trauma-exposed groups.

The second study (N = 45) included HC (N = 22) and symptomatic trauma exposed (STE) participants (21 meeting full criteria for current PTSD, 2 with current subthreshold symptoms). Inclusion criteria were: age 18–45 years; English as first language; and trauma exposure consistent with group status. Exclusion criteria were: history of neurological disorder; history of head trauma with loss of consciousness greater than 5 minutes; estimated full-scale intelligence quotient less than 70; history of attention deficit hyperactivity disorder; contraindications for MRI; left-handedness; and alcohol/substance use disorder in the past year. For HC participants, history of any DSM-5 disorder was exclusionary (aside from alcohol/substance use disorder, prior to the past 12 months); for STE participants, history of psychotic or bipolar disorder was exclusionary. For the STE group, a stable (6-week) dose of antidepressant medication was permitted; other past month psychotropic medication use was exclusionary.

The combined sample included 37 HC participants, 23 TENCs, and 41 participants with a history of current or lifetime DSM-5 PTSD. There were no group differences in gender distribution (26/37 participants in the HC group were female; 14/23 in the TENC group; 35/41 in the PTSD group, X2(2) = 5.11, p = .078, Cramer’s V = .225). Demographics are summarized in Table 1. Comorbid diagnoses and index traumatic events are presented in the Supplement.

Table 1.

Demographic and Clinical Characteristics by Group

HC (N = 37) TENC (N = 23) PTSD (N = 41) Statistical Test
Variable M SD M SD M SD
Age (years)   27.49 7.71   30.43 8.09   26.49 6.89 F (2, 98) = 2.09, p = .129, ηp2 = .041
WASI-II FSIQ (2 subtest)a 113.14 15.15 106.57 16.16 111.75 15.59 F (2, 97) = 1.33, p = .270, ηp2 = .027
Social anhedonia (RSAS)b  0.74 0.26  0.90 0.24  1.11 0.31 F (2, 98) = 17.61, p < .001c, ηp2 = .264
Social network size (SNI)b  1.28 0.29  1.18 0.22  1.09 0.28 F (2, 98) = 4.94, p = .009d, ηp2 = .092
Number of embedded networks (SNI)b  0.45 0.20  0.38 0.22  0.30 0.20 F (2, 98) = 5.18, p = .007d, ηp2 = .096
Social network diversity (SNI)b  0.69 0.14  0.68 0.15  0.62 0.13 F (2, 98) = 2.57, p = .081, ηp2 = .050
CAPS-5 Total Score  5.26 6.20   27.59 10.49 F(1,62) = 86.81, p < .001, ηp2 = .583
On antidepressant N = 0 N = 12
Race Asian (N = 16), Black / African American (N = 10), Native Hawaiian or Other Pacific Islander (N = 2), White (N = 53), Multiple races (N = 11), Other (N = 2), and Not Reported (N = 7)
Ethnicity 13 Hispanic ethnicity; 78 non-Hispanic; 10 participants did not respond.
Open in a new tab

HC, Healthy Control; TENC, Trauma-Exposed Non-PTSD Control; PTSD, Posttraumatic Stress Disorder; CAPS, Clinician-Administered PTSD Scale; WASI-II, Wechsler Abbreviated Scale of Intelligence, Second Edition; FSIQ, Full-Scale IQ; RSAS, Revised Social Anhedonia Scale; SNI, Social Network Index.

a

WASI-II FSIQ is a two-subtest estimate (Vocabulary and Matrix Reasoning). One PTSD group participant is missing WASI data.

b

These are transformed scores: log(raw value + 1). The constant was added to all scores to avoid log0.

c

HC < TENC < PTSD

d

HC > PTSD

Procedure

Participants were recruited from advertisements in the Boston metropolitan area and provided written informed consent to protocols approved by the institutional review board of Partners Healthcare. Doctoral-level psychologists conducted interviews, including the Structured Clinical Interview for DSM-5 (SCID-5: First, Williams, Karg, & Spitzer, 2015) and, for trauma-exposed participants, the Clinician Administered PTSD Scale for DSM-5 (CAPS-5: Weathers et al., 2018).

Measures

The Revised Social Anhedonia Scale (RSAS; Eckblad et al., 1982) is a self-report questionnaire consisting of 40 true/false items (scored 0 or 1, 18 reverse-keyed), with higher scores reflecting greater social anhedonia (i.e., social withdrawal and/or apathy). Reliability was good within this data set, Cronbach’s alpha = .92.

The Social Network Index (SNI: Cohen, 1997) is a self-report questionnaire assessing the extent of participants’ social contact. Participants responded to 12 questions regarding social roles (family relationships, friendships, group/class membership, employment, neighbors, etc.). Derived scales were: 1) social network size as the number of people in the social network (total number of people that participants indicated having regular contact with, at least once every two weeks); 2) number of embedded networks (range: 0-8; number of different high-contact network domains); and 3) diversity of the social network (range: 0-12: number of social roles in which participants report contact at least every 2 weeks with at least one person. Given the structure of the questionnaire, Cronbach’s alphas were low (under .5): see Supplement for a discussion.

The Beck Depression Inventory, 2nd version (BDI-II: Beck et al., 1996) is a widely used self-report measure of depression. Total scores were used to assess overall depression severity.

Data Analysis

For t-tests, when Levene’s test for equality of variances was violated, adjusted degrees of freedom (equal variances not assumed) are reported. Missing data were handled as follows: on the SNI, missing data occurred on 3 of 101 cases. Two participants skipped an item pertaining to contact with in-laws, and a third participant indicated working but did not provide number of supervisees/work contacts. Because missing data were rare and there is no clear method for pro-rating, those missing data were scored ‘0’ (no contacts in that domain). For two TENC participants, a single CAPS-5 item was omitted; these items were scored 0. For the RSAS, 6 participants were missing one item out of 40. Total scores were prorated per participant as follows: prorated score = (40 * (raw total) / 39). Social network features and RSAS scores were not normally distributed and were therefore log-transformed. Because the raw values included 0, a constant (1) was added to every value prior to transformation to avoid log(0). The resulting transformed variables were normally distributed.

For between-group comparisons, ANOVAs were followed by least-significant difference post-hoc tests. We examined Pearson correlations between social network index measures, social anhedonia, and, for trauma-exposed participants, overall PTSD symptom severity. Significant associations between social network parameters and social anhedonia were then entered into a regression with PTSD symptom severity as an additional predictor, to assess whether those relationships remained statistically significant after accounting for overall symptom severity. The regression was repeated controlling for gender, age, and depression severity.

Results

Group Differences

There was a significant group difference in CAPS-5 total scores, which were higher in the PTSD group than in the TENC group (Table 1). There also was a significant group difference in RSAS scores, F(2,98) = 17.61, p < .001, ηp2 = .264 (large effect size); posthoc tests showed that the PTSD group had significantly more social anhedonia than the TENC group, p = .004, and that the TENC group had significantly more social anhedonia than the HC group, p = .031.

Additionally, the groups differed significantly in some of the social network features assessed by the SNI (Table 1). First, there were group differences in social network size, F(2, 98) = 4.94, p = .009, ηp2 = .092 (a medium effect size), with significantly fewer people in their network for PTSD versus HC participants, p = .002. There were not significant differences in social network size between HC and TENC or between TENC and PTSD participants. Second, there were group differences in the number of embedded networks, F(2, 98) = 5.18, p = .007, ηp2 = .096 (medium effect size), with a significantly lower number of embedded networks in PTSD versus HC participants, p = .002, but no significant differences in number of embedded networks between HC and TENC participants or between TENC and PTSD participants. Finally, there was no significant difference in social network diversity between any of the groups.

Correlations Between Social Anhedonia and Social Network Features

As hypothesized, across the entire sample, RSAS social anhedonia score was associated with structural network features, including: smaller social network size, lower number of embedded networks, and lower social network diversity (all medium effect sizes; Table 2). When considering only the trauma-exposed participants (i.e., TENC and PTSD), the relationship between RSAS scores and lower social network size also was statistically significant, r(62) = −.25, p = .043, as was the relationship between RSAS scores and lower social network diversity, r(62) = −.43, p < .001. There also was a significant positive correlation between RSAS and CAPS-5 scores, r(62) = .44, p < .001.

Table 2.

Pearson Correlations Between Social Anhedonia and Social Network Features. Lower portion of table: in combined three-group sample (N = 101). Upper portion of table: in trauma-exposed sample, combined TENC and PTSD groups (N = 64)

Measure 1 2 3 4 CAPS-5
1. Social anhedonia (RSAS) -- −.25* −.24 −.43**  .44**
2. Social network size (SNI) −.32* -- .77** .61**   −.23
3. Number of embedded networks (SNI) −.35** .77** -- .40*   −.30*
4. Social network diversity (SNI) −.40** .61** .52** --   −.20
Open in a new tab

HC, Healthy Control; TENC, Trauma-Exposed Non-PTSD Control; PTSD, Posttraumatic Stress Disorder; CAPS, Clinician-Administered PTSD Scale; RSAS, Revised Social Anhedonia Scale; SNI, Social Network Index.

*

indicates p < .05

**

indicates p < .001

Regression Analysis

Because social anhedonia was related to both social network features (size, diversity) and overall PTSD symptom severity in the combined trauma-exposed sample, regressions were performed within this combined sample (N = 62) to identify separate effects of PTSD symptom severity (CAPS-5 total scores) and social anhedonia (RSAS score) on social network features (SNI scores). The model including both RSAS scores and CAPS-5 total scores as predictors of social network size was not significant, F(2,61) = 2.684, p = .076, R2 = .081, Cohen’s f2 = .09. However, the model including both RSAS scores and CAPS-5 total scores as predictors of social network diversity was significant, F(2,61) = 6.891, p = .002, R2 = .184, Cohen’s f2 = .23. Social anhedonia predicted social network diversity while accounting for PTSD symptoms; that is, the relationship between RSAS score and social network diversity score was statistically significant when controlling for total CAPS-5 score (Table 3). The model also was significant when additionally controlling for gender, age, and overall depression severity (BDI-II total scores), F(5,58) = 3.443, p = .009, R2 = .229, Cohen’s f2 = .30; and again, RSAS score was the only variable that contributed significantly to the prediction of social network diversity. Results were unchanged after omitting the 12 PTSD participants on antidepressants (Supplement).

Table 3.

Linear Regression Model of the Effects of Social Anhedonia and CAPS-5 Scores on Social Network Diversity in Trauma-Exposed Sample (Combined TENC & PTSD, N = 64)

Variable β B 95% CI B   t p
Constant .849 [.733, .966] 14.54 < .001
Social Anhedonia −.421 −.197 [−.317, −.076] −3.26    .002
CAPS-5 Total −.018 .000 [−.003, .002] −0.114    .892
Open in a new tab

TENC, Trauma-Exposed Non-PTSD Control; PTSD, Posttraumatic Stress Disorder; CAPS, Clinician-Administered PTSD Scale

To examine whether the relationship between social anhedonia and social network diversity varied between groups, we used a general linear model with social network diversity as the dependent variable and group (3 categories), RSAS score, and group by RSAS interaction as predictors. There was a significant effect of RSAS score, F(1, 95) = 14.006, p < .001, ηp2 = .128. The effect of group was not significant, F(1, 95) = 1.041, p = .357, ηp2 = .021, nor was the group by RSAS interaction, F(1, 95) = 0.875, p = .420, ηp2 = .018. Thus, the strength of the relationship between social anhedonia and social network diversity did not differ between groups.

Because there is no existing literature on the relationship between social anhedonia and social network features, it was not possible to estimate the expected effect size or to perform an a priori power analysis. Observed (post hoc) power for the correlation between RSAS score and social network diversity (r = −.43, N = 64) was 0.95 (two-tailed; calculated in G*Power 3.1).

Conclusions

In this study, individuals with PTSD endorsed significantly more social anhedonia than trauma-exposed non-PTSD controls, who endorsed significantly more social anhedonia than healthy controls. Compared to healthy controls, PTSD patients had altered structural features of their social networks, including smaller social network size and lower number of embedded networks. Across trauma-exposed participants, social anhedonia was associated with less social network diversity, an effect that was statistically significant even after accounting for overall PTSD and depression symptom severity. In the context of longitudinal studies demonstrating that PTSD symptoms predicts deterioration of social relationships over time (Kaniasty & Norris, 2008; King et al., 2006), our results suggest that social anhedonia may explain certain aspects of social isolation in posttraumatic stress samples. Specifically, social anhedonia following trauma exposure may contribute to individual differences in the diversity of one’s social roles. This finding may be particularly important given evidence that similar social network metrics are associated with adverse health outcomes including suicidality (Handley et al., 2012).

Social network diversity may be more closely related to social anhedonia as a dimensional process that is independent of psychiatric diagnosis, than to the clinical diagnosis of PTSD. Indeed, we did not find a significant group difference in social network diversity between PTSD, TENC, and HC participants. However, lower social network diversity was significantly correlated with higher social anhedonia across the whole sample and within the trauma-exposed participants. Our results are consistent with findings in healthy adults indicating that affiliative processes contribute to individual differences in social networks (Bickart et al., 2012; Fareri & Delgado, 2014). Our results also extend these findings to a trauma-exposed sample. To our knowledge there are no prior reports of a relationship between social anhedonia and structural features of social networks in PTSD. This motivates further inquiry into whether aberrant social reward processing leads to social isolation across a range of trauma exposure and psychopathology.

Trauma-exposed controls were more socially anhedonic than healthy controls, a finding that may be related to multiple factors. Unlike the HC group, some members of the TENC group had other psychiatric disorders that can involve anhedonia such as major depressive disorder and alcohol/substance use disorders. Moreover, trauma exposure itself is associated with alterations in mood and social factors (Barzilay et al., 2019). While these questions were beyond the scope and statistical power of this study, they motivate further inquiry into these possibilities.

The present work has several limitations. First, assessment of social anhedonia was limited to self-report. It will be important to extend the present findings to behavioral measures of social anhedonia. Second, we did not collect information on income or socioeconomic status, and possible effects of these additional demographic/socioeconomic variables should be examined in future studies. Finally, most participants (52 out of 62 trauma-exposed with time to trauma data available) were assessed 2 or more years post-exposure, so we were not able to examine how relationships between social anhedonia and social network structure may evolve over the early posttraumatic period. Future longitudinal studies examining how social anhedonia, social network disruption, and PTSD symptomatology evolve are warranted.

Despite these limitations, the finding that social anhedonia is associated with lower social network diversity has important theoretical and possible clinical implications. It may be important to consider the contribution of social anhedonia to individual differences in social network features. Social anhedonia and low social network diversity may reciprocally drive each other over time, progressively exacerbating social isolation. From a clinical perspective, our results identify social anhedonia as a potential target in addressing social dysfunction in individuals exposed to trauma. Forms of therapy designed to increase social enjoyment (e.g., behavioral activation) may help prevent deterioration of social networks after trauma exposure.

Supplementary Material

Supplement

Acknowledgments

This research was supported by an Eleanor and Miles Shore 50th Anniversary Fellowship, Harvard Medical School (EAO), by a Brain and Behavior Research Foundation NARSAD Young Investigator Award (EAO), and by NIMH K23 MH112873 (EAO). DAP was partially supported by NIMH R37 MH068376 and R01 MH095809. IMR was partially supported by NIMH R01MH096987 and a Brain and Behavior Research Foundation NARSAD Independent Investigator Award. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Disclosures: Over the past 3 years, Dr. Pizzagalli has received consulting fees from Akili Interactive Labs, BlackThorn Therapeutics, Boehringer Ingelheim, Compass Pathway, Otsuka Pharmaceuticals, and Takeda Pharmaceuticals; one honorarium from Alkermes, and research funding from NIMH, Dana Foundation, Brain and Behavior Research Foundation, Millennium Pharmaceuticals. In addition, he has received stock options from BlackThorn Therapeutics. No funding from these entities was used to support the current work, and all other authors declare no competing interests. All views expressed are solely those of the authors.

References

  1. Barkus E, & Badcock JC (2019). A transdiagnostic perspective on social anhedonia. Frontiers in Psychiatry, 10:216 http://doi.org/ggdmj9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Beck AT, Steer RA, & Brown GK (1996). Manual for the Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation. [Google Scholar]
  3. Bickart KC, Hollenbeck MC, Barrett LF, & Dickerson BC (2012). Intrinsic amygdala-cortical functional connectivity predicts social network size in humans. The Journal of Neuroscience, 32(42), 14729–14741. http://doi.org/dnkt [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brewin CR, Andrews B, & Valentine JD (2000). Meta-analysis of risk factors for posttraumatic stress disorder in trauma-exposed adults. Journal of Consulting and Clinical Psychology, 68(5), 748–766. [DOI] [PubMed] [Google Scholar]
  5. Bryant RA, Gallagher HC, Gibbs L, Pattison P, MacDougall C, Harms L, … Lusher D (2017). Mental health and social networks after disaster. American Journal of Psychiatry, 174(3), 277–285. http://doi.org/f9vhx5 [DOI] [PubMed] [Google Scholar]
  6. Chapman LJ, Chapman JP, & Raulin ML (1976). Scales for physical and social anhedonia. Journal of Abnormal Psychology, 85(4), 374–382. http://doi.org/bnmvjt [DOI] [PubMed] [Google Scholar]
  7. Cohen S (1997). Social Ties and Susceptibility to the Common Cold. JAMA: The Journal of the American Medical Association, 277(24), 1940 http://doi.org/cwwh43 [PubMed] [Google Scholar]
  8. Cohen S, & Wills TA (1985). Stress, social support, and the buffering hypothesis. Psychological Bulletin, 98(2), 310–357. http://doi.org/b4k [PubMed] [Google Scholar]
  9. Dodell-Feder D, Shovestul B, Woodyatt J, Popov V, & Germine L (2020). Social anhedonia, social networks, and psychotic-like experiences: A test of social deafferentation. Psychiatry Research, 284 http://doi.org/dv3b [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Fareri DS, & Delgado MR (2014). Social rewards and social networks in the human brain. The Neuroscientist : A Review Journal Bringing Neurobiology, Neurology and Psychiatry, 20(4), 387–402. http://doi.org/f6cnff [DOI] [PubMed] [Google Scholar]
  11. First MB, Williams JBW, Karg RS, & Spitzer RL (2015). Structured clinical interview for DSM-5. American Psychiatric Association, Washington D.C. [Google Scholar]
  12. Ford ES, Loucks EB, & Berkman LF (2006). Social integration and concentrations of C-reactive protein among US adults. Annals of Epidemiology, 16(2), 78–84. http://doi.org/csnf4g [DOI] [PubMed] [Google Scholar]
  13. Handley TE, Inder KJ, Kelly BJ, Attia JR, Lewin TJ, Fitzgerald MN, & Kay-Lambkin FJ (2012). You’ve got to have friends: The predictive value of social integration and support in suicidal ideation among rural communities. Social Psychiatry and Psychiatric Epidemiology, 47(8), 1281–1290. http://doi.org/d7n8vf [DOI] [PubMed] [Google Scholar]
  14. Kaniasty K, & Norris FH (2008). Longitudinal linkages between perceived social support and posttraumatic stress symptoms: sequential roles of social causation and social selection. Journal of Traumatic Stress, 21(3), 274–281. http://doi.org/dwcq48 [DOI] [PubMed] [Google Scholar]
  15. King DW, Taft C, King LA, Hammond C, & Stone ER (2006). Directionality of the association between social support and posttraumatic stress disorder: A longitudinal investigation. Journal of Applied Social Psychology, 36(12), 2980–2992. http://doi.org/dqzq58 [Google Scholar]
  16. Kwapil TR, Barrantes-Vidal N, & Silvia PJ (2008). The dimensional structure of the Wisconsin Schizotypy Scales: factor identification and construct validity. Schizophrenia Bulletin, 34(3), 444–457. http://doi.org/fjg9d5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Lee B, & Youm Y (2011). Social network effects on post-traumatic stress disorder (PTSD) in female North Korean immigrants. Journal of Preventive Medicine and Public Health = Yebang Uihakhoe Chi, 44(5), 191–200. http://doi.org/fstj8p [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Molesworth T, Sheu LK, Cohen S, Gianaros PJ, & Verstynen TD (2015). Social network diversity and white matter microstructural integrity in humans. Social Cognitive and Affective Neuroscience, 10(9), 1169–1176. http://doi.org/f72j7z [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Nawijn L, van Zuiden M, Frijling JL, Koch SBJ, Veltman DJ, & Olff M (2015). Reward functioning in PTSD: a systematic review exploring the mechanisms underlying anhedonia. Neuroscience and Biobehavioral Reviews, 51, 189–204. http://doi.org/ggdmj3 [DOI] [PubMed] [Google Scholar]
  20. Olson EA, Kaiser RH, Pizzagalli DA, Rauch SL, & Rosso IM (2018). Anhedonia in trauma-exposed individuals: functional connectivity and decision-making correlates. Biological Psychiatry: Cognitive Neuroscience and Neuroimaging, 3(11), 959–967. http://doi.org/ggmcz4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Platt J, Keyes KM, & Koenen KC (2014). Size of the social network versus quality of social support: which is more protective against PTSD? Social Psychiatry and Psychiatric Epidemiology, 49(8), 1279–1286. http://doi.org/f6b86z [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Sripada RK, Bohnert ASB, Teo AR, Levine DS, Pfeiffer PN, Bowersox NW, … Valenstein M (2015). Social networks, mental health problems, and mental health service utilization in OEF/OIF National Guard veterans. Social Psychiatry and Psychiatric Epidemiology, 50(9), 1367–1378. http://doi.org/f7r6x3 [DOI] [PubMed] [Google Scholar]
  23. Weathers FW, Bovin MJ, Lee DJ, Sloan DM, Schnurr PP, Kaloupek DG, … Marx BP (2018). The clinician-administered ptsd scale for DSM-5 (CAPS-5): Development and initial psychometric evaluation in military veterans. Psychological Assessment. http://doi.org/gdfc5x [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

Supplement
NIHMS1623295-supplement-Supplement.docx (41.3KB, docx)

RESOURCES