Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Jul 1.
Published in final edited form as: Clin Psychol Sci. 2015 Oct 30;4(4):675–682. doi: 10.1177/2167702615602672

Revisiting Depression Contagion as a Mediator of the Relation Between Depression and Rejection: A Speed-Dating Study

Madeline L Pe a, Ian H Gotlib b, Wim Van den Noortgate a, Peter Kuppens a
PMCID: PMC5006939  NIHMSID: NIHMS713648  PMID: 27595052

Abstract

Interpersonal theories of depression postulate that depressed individuals’ experience of social isolation is attributable, in part, to their tendency to behave in ways that elicit rejection from others. Depression contagion has been implicated as a factor that may account for the rejection of depressed individuals. The current study revisits this hypothesis using a controlled, but realistically motivated setting: speed-dating. Approximately two weeks before the speed-dating event, participants’ depression levels were assessed. During the event, participants had four-minute “dates” with opposite-sex partners. After each date, they responded to items measuring their affect and romantic attraction. At the end of the evening, participants indicated which partners they wanted to see again. Our results did not support depression contagion: after four minutes of interaction with partners with high levels of depressive symptoms, participants did not experience increased negative affect; instead, they experienced reduced positive affect, which led to the rejection of these partners.

Keywords: depression, interpersonal processes, rejection, depression contagion

INTRODUCTION

Depression is a debilitating mood disorder with enormous personal and societal costs. In fact, depression is the second leading cause of disability in the world (Ferrari et al., 2013). Given this alarming statistic, it is critical that we work to understand and prevent the development of depression.

One factor that has been implicated in the development of depression is social isolation (Cacioppo, Hughes, Waite, Hawkley, & Thisted, 2006). Interpersonal theories of depression posit that the experience of social isolation of depressed individuals is attributable, in part, to their tendency to behave in ways that elicit rejection from others (Coyne, 1976; Hames, Hagan, & Joiner, 2013). Indeed, several studies have supported this postulation: people tend to reject depressed individuals after interacting with them (depression-rejection relation; see reviews by Segrin & Dillard, 1992; Marcus & Nardone, 1992).

Coyne (1976) proposed that depression contagion may account for the rejection of depressed individuals by those with whom they interact; specifically, Coyne postulated that the rejection experienced by depressed individuals may be due to the negative affect they induce in others. Thus, not only do individuals with depressive symptoms experience the debilitating effects of their own depression, but they may also induce negative affect and other depressive symptoms in people with whom they interact. Indeed, a meta-analysis by Joiner and Katz (1999) provides support for the formulation that individuals report higher levels of negative affect and depressive symptoms after interacting with depressed persons. Moreover, theorists have posited that this induction of negative mood and depressive symptoms leads people to avoid or even ultimately reject depressed individuals (Hames et al., 2013).

The goal of the present study is to test whether depression contagion explains the rejection of depressed individuals by those with whom they interact. Specifically, we examined whether the negative affect induced in people through interactions with individuals who are experiencing depressive symptoms could explain the rejection of the depressed individuals. In conceptualizing depression contagion, we do not mean to imply that individuals “catch” the depressive symptoms of depressed partners. Instead, it is more likely that depressed partners behave in specific ways that induce negative affect in others (Hames et al., 2013). For example, depressed individuals tend to have poor social skills (Segrin, 2000); these behaviors may then act as conduits for the relation between depression and induced negative affect in, and rejection by, others. Although the behaviors by which depressed individuals may induce negative affect in others have not been elucidated, it is critical for the concept of depression contagion that people experience a change in negative affect after interacting with individuals with depressive symptoms. In other words, we posit that interacting with depressed individuals influences the negative affect experienced by others.

It is important to note here that previous studies have failed to support the hypothesis that rejection of depressed individuals is due to induced negative affect (see Gurtman, 1986, for a review). For example, although Joiner, Alfano, and Metalsky (1992) found evidence for both rejection of depressed persons and the contagion effect in a naturalistic study of college roommates, they did not find that contagion mediated the relation between depression and rejection. The failure to find such a mediating effect, however, may be due to other factors, such as shared common history and familiarity (Joiner & Katz, 1999). That is, it is possible that, over time, people become accustomed or immune to the negative affect or the depressive symptoms of their depressed partners and, therefore, do not use these characteristics in making decisions about how much they like their partner (Joiner & Katz, 1999). In contrast, however, when making decisions about how much they like a stranger following a brief interaction, the lack of shared history with, or knowledge about, their interaction partner might lead people to rely more strongly on their affect as a basis for their decisions.

In this study we wanted to explore the possibility that, in addition to inducing negative affect (depression contagion), individuals with high levels of depressive symptoms may also induce reduced positive affect in others. Indeed, depression is characterized not only by increased negative affect, but also by reduced or blunted positive affect (anhedonia) (e.g., Clark & Watson, 1999). Paralleling the concept of depression contagion, we hypothesize that depressed individuals behave in ways that reduce positive affect in others, which then leads to rejection.

To examine whether increased negative affect and reduced positive affect would mediate the relation between depression and social rejection following interactions with partners who have high levels of depressive symptoms, we used a methodology that allowed us to observe short-term interactions among strangers in a controlled, but realistically motivated setting: speed-dating. In a speed-dating event, men and women are given an opportunity to talk to several potential partners in a series of four-minute “dates” before deciding whether or not they would like to pursue a possible relationship with a person with whom they interacted (Todd, Penke, Fasolo, & Lenton, 2007; Finkel, Eastwick, & Matthews, 2008). We asked all participants to complete a questionnaire assessing depressive symptoms two weeks before they attended the speed-dating event, and examined whether individuals’ depression scores were associated with their partners’ affect and romantic attraction to them following a relatively short dyadic interaction.

We predicted that after a brief interaction with partners with high levels of depressive symptoms, participants would experience more negative affect (depression contagion) and less positive affect, and would be less romantically attracted to these partners (depression-rejection relation). We also hypothesized that participants’ negative and positive affect would be related to their level of romantic attraction to their partners, such that lower negative affect and higher positive affect would be correlated with higher levels of romantic attraction. Finally, and critical to our investigation, we predicted that the depression-rejection relation would be accounted for (i.e., mediated) by the negative affect and positive affect participants experienced following the interaction with their partners.

METHOD

Participants and Procedure

We conducted six speed-dating sessions for 136 single heterosexual individuals who were recruited within and outside Leuven using posters, advertisements, Facebook and website posts, and mailing lists. Sessions were separated by age group: ages 21-26 (four sessions) and ages 27-32 (two sessions). Approximately two weeks before the event, participants signed the consent form and responded to online questionnaires that assessed their levels of depressive symptoms, among other constructs. When participants arrived at the event, they first responded to a baseline emotion questionnaire. Participants were given 4 minutes to interact with each of their 10 to 12 interaction-partners. After each “date,” participants immediately completed a two-minute interaction record that assessed their emotions and their evaluation of their current interaction-partner. At the end of the evening, they were given a sheet of paper with the photographs of their interaction-partners, and were asked to indicate by checking “yes” or “no” whether they wanted to meet their interaction-partner again. They were informed that, in the case of a match, the mutual contact details of the participants would be made available to them.

Materials

Severity of depressive symptoms

We measured depressive symptom severity using the Center for Epidemiologic Studies Depression Scale (CES-D; Radloff, 1977), in which participants rated how frequently they experienced a range of depressive symptoms (e.g., “I had crying spells”) over the past week on a four-point scale from 0 (rarely or none of the time) to 3 (most or all of the time). The sum of the CES-D items was used to index severity of depressive symptoms (Cronbach’s α=.86).

Affect

On each interaction record, using a scale from 1 (strongly disagree) to 5 (strongly agree), participants responded to a 10-item measure assessing their current affect (“After my last interaction, I now feel…). Composite scores of positive and negative affect were created by computing the mean of the positive emotion items (interested, happy, excited, relaxed, in love; between-subject reliability=.98) 1 and the negative emotion items (sad, irritated, bored, nervous, anxious; between-subject reliability=.94), respectively.

Romantic attraction

On each interaction record (adapted from Eastwick & Finkel, 2008), using a scale from 1 (strongly disagree) to 5 (strongly agree), participants responded to a 3-item measure of romantic attraction (“I felt attracted to my partner,” “I found my interaction-partner physically attractive,” “I would like to see my interaction-partner again”). A single romantic attraction score was obtained by computing the average rating across these three items (between-subject reliability=.97).

Actual choice

At the end of the evening, participants were given a sheet of paper with the photographs of their interaction-partners, and were asked to rate “yes” or “no” (coded as 1 and 0, respectively) to indicate whether they wanted to meet each interaction-partner again.

Statistical Models

To test our predictions, we utilized a cross-classified multilevel model; this model is recommended by Kenny and Kashy (2011) to analyze data with a social relations model design (SRM; round-robin). By using this model, we were able to take into account the non-independence of observations when analyzing our speed-dating data. That is, a) each observation is nested within both partner and participant; b) participants encounter different partners and vice versa, and c) there is a correlation between the two scores from members of the same dyad.

Of particular interest is the examination of individual differences in partners’ depressive symptoms as a predictor of romantic attraction. To conduct this analysis, we added partners’ depressive symptoms (participant-mean centered) as a fixed effects predictor of participant’s romantic attraction in the SRM model. We conducted similar SRM models to test other relations. For the contagion effect, we regressed participants’ negative affect on partners’ depressive symptoms. We conducted a similar model for positive affect. To examine whether participants’ affect was associated with romantic desire, we regressed romantic desire on participants’ negative (or positive) affect. Again, all predictors were participant-mean centered. All analyses were conducted using SAS PROC MIXED.

A similar set of analyses was also conducted with actual choice as a dependent variable. We used a logistic model however, because of the binary nature of the outcome variable. Analyses were conducted using SAS PROC GLIMMIX.

Within-Subject Mediation analysis

We used the multiple mediator model recommended by Mackinnon (2008) to examine whether the effect of partners’ depressive symptoms on participants’ romantic attraction was mediated by participants’ positive and negative affect. To estimate the relations at the within-subject level that is not confounded by between-subject differences, we first participant-mean centered all the predictors in the model. To take into account the cross-classified structure of the data, we estimated each relevant path for the mediation test within the cross-classified multilevel SRM model. Specifically, we first regressed participants’ romantic attraction on partners’ depressive symptoms. Then, we regressed participants’ romantic attraction on participants’ positive affect (b1) and negative affect (b2) while controlling for partners’ depression scores (c’). Next, we regressed positive affect on partners’ depression scores (a1). Finally, we regressed participants’ negative affect on partners’ depression scores (a2). Analyses were conducted using SAS PROC MIXED. Support for mediation was found when the indirect paths (a1b1 or a2b2) were significant (see also Mackinnon, Fairchild, & Fritz, 2008).

To test whether the indirect paths were significant, we estimated the 95% confidence intervals of the indirect effects (a1b1 or a2b2) using the distribution-of-the-product method (MacKinnon, Lockwood, Hoffmann, West, & Sheets, 2002). This method has been demonstrated to be more accurate in estimating confidence intervals for specific mediated effects than are other methods (Tofighi & MacKinnon, 2011). The indirect effect is statistically significant when the confidence interval does not include the value of zero. These tests were conducted using the PRODCLIN option from the R package RMediation (Tofighi & MacKinnon, 2011).

To examine whether the effect of partners’ depressive symptoms on participants’ actual choice at the end of the evening was mediated by positive or negative affect, we conducted a similar analysis to that described above, but with actual choice specified as a binary outcome variable. Analyses were conducted using SAS PROC GLIMMIX.

RESULTS

Participant Characteristics

Participants had a mean age of 25 years (SD = 3.24). There were approximately equal numbers of males (n=67) and females (n=69) in the study. The average level of severity of depressive symptoms on the CES-D was 11.17 (SD = 6.50), and 29 of the participants (15 females) scored at or above the clinical cutoff score of 16 proposed by Radloff (1977). There were no gender, t(134)=−.72, p=.47, d=−.12, or age differences, r = −.01, p=.93, in the severity of depressive symptoms.

Relations among Partners’ Depressive Symptoms, Partners’ Affect, and Participants’ Ratings of Partners’ Attractiveness

Table 1 presents the associations among partners’ depression levels, participants’ romantic attraction, actual choice, and levels of negative and positive affect (see slope estimates). We found partial support for the hypothesis that partner depressive symptoms would be directly associated with rejection. We found that the higher the depressive symptoms of the interaction partner, the less romantically attracted participants were to their partners following the interaction (see Table 1, Partners’ depressive symptoms → Participants’ romantic attraction); this was not the case, however, for actual choice at the end of the evening (see Table 1, Partners’ depressive symptoms → Participants’ actual choice).

Table 1.

Associations among Partners’ Depressive Symptoms, Participants’ Positive and Negative Affect, and Participants’ Romantic Attraction to the Interaction Partner.

Coef SE df t p 95% [CI]
Partners’ depressive symptoms → Participants’ romantic attraction
Intercept 2.88 .06 200 44.96 <.01 [2.75, 3.01]
Slope −.46 .18 122 −2.57 .01 [−.81, −.10]
Partners’ depressive symptoms → Participants’ actual choice
Intercept −.65 .15 144 −4.24 <.01 [−.95, −.34]
Slope −.60 .41 103.7 −1.48 .14 [−1.41, .20]
Partners’ depressive symptoms → Participants’ negative affect
Intercept 1.26 .02 146 60.68 <.01 [1.22, 1.30]
Slope .03 .02 124 1.43 .16 [−.01, .08]
Partners’ depressive symptoms → Participants’ positive affect
Intercept 2.65 .05 162 55.56 <.01 [2.55, 2.74]
Slope −.18 .06 118 −2.95 <.01 [−.29, −.06]
Participants’ NA → Participants’ romantic attraction
Intercept 2.87 .06 200 44.56 <.01 [2.74, 2.99]
Slope −.44 .08 1315 −5.72 <.01 [−.58, −.29]
Participants’ PA → Participants’ romantic attraction
Intercept 2.88 .05 216 55.20 <.01 [2.78, 2.98]
Slope 1.02 .04 1360 25.65 <.01 [.94, 1.10]
Participants’ NA → Participants’ actual choice
Intercept −.71 .15 145.4 −4.57 <.01 [−1.01, −.40]
Slope −1.40 .28 1313 −4.97 <.01 [−1.95, −.85]
Participants’ PA → Participants’ actual choice
Intercept −.75 .15 140 −5.05 <.01 [−1.04, −.45]
Slope 2.86 .20 1441 14.57 <.01 [2.48, 3.25]
Open in a new tab

We did not find support for the depression contagion hypothesis, in that the depressive symptoms of the interaction partner did not significantly predict participants’ negative affect (see Table 1, Partners’ depressive symptoms → Participants’ negative affect). We did find, however, that higher levels of depressive symptoms in the interaction partner were related to lower levels of positive affect in participants following the interaction (see Table 1, Partners’ depressive symptoms → Participants’ positive affect). Thus, although partners’ depressive symptoms did not increase participants’ negative affect, they appear to have dampened participants’ positive affect.

We also found significant associations between participants’ affect and their romantic attraction to and actual choices of partners (see Table 1, Participants’ negative (or positive) affect → Participants’ romantic attraction (or actual choice)). Participants who experienced lower negative and higher positive affect after interacting with their partner reported being more attracted to their interaction partner and were more likely to choose their partner at the end of the evening.2,3

Mediating Effect of Participants’ Affect on the Relation between Partners’ Depressive Symptoms on Participants’ Rating of Partners’ Romantic Attractiveness

Figure 1 displays the effect of partners’ depressive symptoms on participants’ rating of romantic attraction, and participants’ actual choice, mediated by participants’ levels of positive and negative affect after interacting with their partners. Positive affect was a significant mediator of the relation between depressive symptoms and level of romantic attraction (a1b1= −.18, SE=.06, 95% CI [−.30, −.06]); this was not the case with negative affect (a2b2= −.00, SE=.00, 95% CI [−.01, .00] (see Figure 1, top panel). A similar pattern of results was found for actual choice. Positive affect significantly mediated the relation between depressive symptoms and actual choice (a1b1= −.49, SE=.17, 95% CI [−.85, −.17]), whereas negative affect did not (a2b2=−.02, SE=.02, 95% CI [−.05, .01]) (see Figure 1, bottom panel). Participants reported lower levels of positive affect after they interacted with partners who reported higher levels of depressive symptoms. Further, participants’ experience of reduced positive affect was associated with less romantic attraction to, and higher likelihood of rejecting, interaction partners with higher levels of depressive symptoms.

Figure 1.

Figure 1

Open in a new tab

The top figure shows the relation between partner’s depressive symptoms and romantic attraction mediated by positive and negative affect. The bottom figure shows the relation between partners’ depressive symptoms and actual choice mediated by positive and negative affect.

DISCUSSION

Interpersonal theories of depression posit that individuals who are experiencing depressive symptoms induce negative affect in others (depression contagion), which leads them to be rejected (Coyne, 1976; Segrin & Dillard, 1992). In this study, we tested this hypothesis by using a speed-dating methodology, which allowed us to examine how individuals with varying levels of depressive symptoms interact with strangers in a controlled, but realistically motivated, setting (Todd, et al., 2007; Finkel et al., 2008).

First, with respect to the direct relation between depression and rejection, we found that participants were less romantically attracted to, but not necessarily more likely to reject, interaction partners with high levels of depressive symptoms. Second, with regard to the depression-contagion hypothesis, participants did not experience increased negative affect after interacting with partners with high levels of depressive symptoms; instead, they experienced reduced positive affect after interacting with partners with high levels of depressive symptoms. More important, however, we found a consistent mediating effect of reduced positive affect on the depression-rejection relation. Specifically, after brief four-minute interactions, individuals with high levels of depressive symptoms induced lower positive affect (but not higher negative affect) in others, which led them not only to be rated as less romantically attractive immediately following the interaction, but also to be rejected at the end of the evening.

These findings are not precisely what Coyne originally suggested; that is, we did not find support specifically for the depression contagion hypothesis. Participants’ negative affect did not increase after interacting with partners with high levels of depressive symptoms. Instead, participants’ positive affect declined after interacting with partners with high levels of depressive symptoms, and this decreased positive affect mediated the relation between depression and rejection. Although these findings do not support the depression contagion hypothesis in the strictest sense, they do demonstrate that participants tend to not enjoy their interactions with partners with high levels of depressive symptoms, increasing the likelihood that they will reject these partners at the end of the evening.

Given the short-term nature of the dyadic interactions in this study, these findings suggest that decreased positive affect (and not necessarily increased negative affect) is one of the first indicators of rejection. Indeed, this proposition is consistent with Baumeister and Leary’s (1995) theory concerning people’s motivation to form and maintain relationships. Their theory posits that the formation of social bonds is specific to the experience of positive, but not of negative, affect. Several studies provide additional support for this possibility: decreases in positive affect (but not in negative affect) have been found to be associated with less attraction and social activity (Watson, Clark, McIntyre, & Hamaker, 1992; Vittengl & Holt, 2000; Berrios, Totterdell, & Niven, 2015). This implies, therefore, that strangers who experienced reduced positive affect after an interaction would be less inclined to form a relationship with their interaction partner. Because individuals with depressive symptoms tend to induce lower levels of positive affect in strangers with whom they interact, the likelihood that they will be rejected is increased.

There are some features of this work that limit the conclusions we can draw about the mediating role of reduced positive affect in explaining the depression-rejection relation. First, this study is correlational; therefore, we cannot make causal claims about the relations among depressive symptoms, affect, and romantic attraction. Second, given that our findings are based on individuals’ self-reports about their levels of depressive symptomatology prior to participating in the speed-dating event, we cannot generalize our findings to clinically depressed individuals or to individuals experiencing other clinical disorders. Third, levels of depressive symptomatology were assessed two weeks before the speed-dating event, and may not reflect participants’ levels of depressive symptomatology during the day of the event itself. It is important to note, however, that the time interval between these two measurements could only add noise to the estimated relation. The fact that we still find evidence for the relation between individuals’ depressive symptoms (assessed two weeks before) and their partners’ positive affect (after a brief interaction) underscores the strength of this finding. Finally, individuals may have experienced reduced positive affect after interacting with depressed partners because their partners behaved in ways that were perceived to be unattractive. For example, researchers have shown that depressed people tend to have poor social skills (Segrin, 2000; Hames et al., 2013); it is possible, therefore, that the poor social skills of partners with depressive symptoms made them less attractive and caused individuals to experience reduced positive affect after interacting with them. Future research is needed not only to clarify this relation, but also to identify which behaviors of depressed individuals lead to reduced positive affect in others and increase their likelihood of being rejected.

One may also wonder whether the fact that participants had to make an explicit choice concerning whom they wanted to see again after a speed-dating interaction represents what occurs in real life. We contend that it does. In real life, we often make immediate and private judgments concerning whether or not we like someone (Willis & Todorov, 2006). We simulated these private judgments in the speed-dating procedure by telling participants that the choices they would make at the end of the evening would be kept confidential (and, therefore, private). Therefore, we do not think that participants’ responses during the speed-dating event would differ from the private judgments they would make in a real-life setting. This argument is speculative, however, and should be tested explicitly in future research.

To date, a significant amount of research has been conducted examining interpersonal processes in depression. To move the field forward, however, researchers should utilize situations that allow them to examine interactions of depressed individuals in a controlled, but ecologically valid, setting. For example, previous studies have used college roommates to examine depression contagion (Howes, Hokanson & Lowenstein, 1985; Joiner et al., 1992). The current study adds to this body of work by examining depressed individuals’ brief encounters in a speed-dating event. Brief interactions play a significant role in broadening one’s social support system. Individuals tend to gather information from such interactions, and this information is often used to decide whether to continue or stop a potential relationship (Miller & Todd, 1998). Undoubtedly, being rejected at such an early part of the process results in a lower probability of starting a new relationship. By gaining a better understanding of the behaviors of depressed individuals during brief encounters, we can provide them with tools to improve their interpersonal skills, which may then help them to increase their chances of broadening their social support system.

In conclusion, our findings demonstrate that the interpersonal problems of individuals with high levels of depressive symptoms extend beyond interactions with their significant others. Rejection by strangers likely contributes to their sense of social isolation by preventing them from forming new relationships and developing a broader social support system. The experience of rejection not only by their significant others (Gotlib & Lee, 1989), but also by strangers, in this case potential mates, may exacerbate individuals’ symptoms (including feelings of social isolation) and increase their risk of developing clinically significant depression.

Acknowledgments

The research leading to the results reported in this paper was supported in part by the Research Fund of KU Leuven (GOA/15/003; OT/11/031), by the Interuniversity Attraction Poles programme financed by the Belgian government (IAP/P7/06), by NIMH Grant (MH59259) to Ian H. Gotlib, and a research grant from the Fund for Scientific Research-Flanders to Peter Kuppens.

We gratefully acknowledge Daisy Buttiens for helping us organize the speed-dating events, Jurgen Convents for helping us recruit participants, the volunteers for their assistance, and the participants for attending the events.

Footnotes

1

All between-subject reliability estimates were calculated using the recommendation of Shrout and Lane (2012).

2

We also conducted these analyses controlling for several possible confounding variables. First, we controlled for participants’ own depressive symptoms by including their depressive symptoms (CES-D scores were group-mean centered) as a fixed predictor in all the models; all results remained the same. Second, we controlled for possible carry-over effects of positive and negative affect from previous interactions by including positive (or negative) affect at time t-1 as a covariate (participant-mean centered) when predicting current positive (or negative affect) ; and all results remained the same. Third, we controlled for possible time effects when predicting positive or negative affect by including time in the analyses (centered at time point 7; total of 13 time points); all results remained the same. Finally, we controlled for initial levels of positive (or negative) affect when predicting current positive (or negative affect); again, all results remained the same.

3

We also examined whether gender moderated any of these associations by including gender and its interaction with the specific predictor of interest into the SRM model. Men and women were coded as 0 and 1 respectively. Only the relations between participants’ negative affect and participants’ romantic attraction (β=−.55, SE=.15, t=−3.76, p<.01, and positive affect and actual choice (β=1.05, SE=.40, t=2.63, p<.01) were significantly moderated by gender. Disentangling these significant interaction effects, we found a significant negative relation between negative affect and romantic attraction in females but not in males; the relation between participants’ positive affect and actual choice was more strongly positively associated in females than in males.

REFERENCES

  1. Baumeister RF, Leary MR. The need to belong: desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin. 1995;117:497–529. [PubMed] [Google Scholar]
  2. Berrios R, Totterdell P, Niven K. Why do you make us feel good? Correlates and interpersonal consequences of affective presence in speed-dating. European Journal of Personality. 2015;29:72–82. doi: 10.1002/per.1944. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Cacioppo JT, Hughes ME, Waite LJ, Hawkley LC, Thisted RA. Loneliness as a specific risk factor for depressive symptoms: cross-sectional and longitudinal analyses. Psychology and aging. 2006;21:140–151. doi: 10.1037/0882-7974.21.1.140. [DOI] [PubMed] [Google Scholar]
  4. Clark LA, Watson D. Tripartite model of anxiety and depression: Psychometric evidence and taxonomic implications. Journal of Abnormal Psychology. 1991;100:316–336. doi: 10.1037//0021-843x.100.3.316. [DOI] [PubMed] [Google Scholar]
  5. Coyne JC. Depression and the response of others. Journal of Abnormal Psychology. 1976;85:186–193. doi: 10.1037//0021-843x.85.2.186. [DOI] [PubMed] [Google Scholar]
  6. Eastwick PW, Finkel EJ. Sex differences in mate preferences revisited: do people know what they initially desire in a romantic partner? Journal of Personality and Social Psychology. 2008;94:245–264. doi: 10.1037/0022-3514.94.2.245. [DOI] [PubMed] [Google Scholar]
  7. Ferrari AJ, Charlson FJ, Norman RE, Patten SB, Freedman G, Murray CJL, Whiteford HA. Burden of depressive disorders by country, sex, age, and year: Findings from the global burden of disease study 2010. PLoS Medicine. 2013;10:e1001547. doi: 10.1371/journal.pmed.1001547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Finkel EJ, Eastwick PW, Matthews J. Speed-dating as an invaluable tool for studying romantic attraction: A methodological primer. Personal Relationships. 2007;14:149–166. [Google Scholar]
  9. Gotlib IH, Lee CM. The social functioning of depressed patients: A longitudinal assessment. Journal of Social and Clinical Psychology. 1989;8:223–237. [Google Scholar]
  10. Gurtman MB. Depression and the response of others: reevaluating the reevaluation. Journal of Abnormal Psychology. 1986;95:99–101. doi: 10.1037//0021-843x.95.1.99. [DOI] [PubMed] [Google Scholar]
  11. Hames JL, Hagan CR, Joiner TE. Interpersonal processes in depression. Annual Review of Clinical Psychology. 2013;9:355–377. doi: 10.1146/annurev-clinpsy-050212-185553. [DOI] [PubMed] [Google Scholar]
  12. Howes MJ, Hokanson JE, Loewenstein DA. Induction of depressive affect after prolonged exposure to a mildly depressed individual. Journal of Personality and Social Psychology. 1985;49:1110–1113. doi: 10.1037//0022-3514.49.4.1110. [DOI] [PubMed] [Google Scholar]
  13. Joiner TE, Alfano MS, Metalsky GI. When depression breeds contempt: reassurance seeking, self-esteem, and rejection of depressed college students by their roommates. Journal of Abnormal Psychology. 1992;101:165–173. doi: 10.1037//0021-843x.101.1.165. [DOI] [PubMed] [Google Scholar]
  14. Joiner TE, Katz J. Contagion of Depressive Symptoms and Mood: Meta-analytic Review and Explanations From Cognitive, Behavioral, and Interpersonal Viewpoints. Clinical Psychology: Science and Practice. 1999;6:149–164. [Google Scholar]
  15. Kenny DA, Kashy DA. Dyadic data analysis using multilevel modeling. In: Hox JJ, Roberts JK, editors. The Handbook of Advanced Multilevel Analysis. Taylor & Francis Group; New York, NY: 2010. pp. 355–371. [Google Scholar]
  16. MacKinnon DP. Introduction to statistical mediation analysis. Taylor and Francis Group; New York, NY: 2008. [Google Scholar]
  17. MacKinnon DP, Lockwood CM, Hoffman JM, West SG, Sheets V. A comparison of methods to test mediation and other intervening variable effects. Psychological Methods. 2002;7:83–104. doi: 10.1037/1082-989x.7.1.83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Marcus DK, Nardone ME. Depression and interpersonal rejection. Clinical Psychology Review. 1992;12:433–449. [Google Scholar]
  19. Radloff LS. The CES-D scale a self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401. [Google Scholar]
  20. Segrin C. Social skills deficits associated with depression. Clinical Psychology Review. 2000;20:379–403. doi: 10.1016/s0272-7358(98)00104-4. [DOI] [PubMed] [Google Scholar]
  21. Segrin C, Dillard JP. The interactional theory of depression: A meta-analysis of the research literature. Journal of Social and Clinical Psychology. 1992;11:43–70. [Google Scholar]
  22. Shrout PE, Lane SP. Psychometrics. In: Mehl MR, Conner TS, editors. Handbook of Research Methods for Studying Daily Life. Guilford; New York: 2012. pp. 302–320. [Google Scholar]
  23. Todd PM, Penke L, Fasolo B, Lenton AP. Different cognitive processes underlie human mate choices and mate preferences. Proceedings of the National Academy of Sciences. 2007;104:15011–15016. doi: 10.1073/pnas.0705290104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Tofighi D, Mackinnon DP. RMediation: An R package for mediation analysis confidence intervals. Behavior Research Methods. 2011;43:692–700. doi: 10.3758/s13428-011-0076-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Vittengl JR, Holt CS. Getting acquainted: The relationship of self-disclosure and social attraction to positive affect. Journal of Social and Personal Relationships. 2000;17:53–66. [Google Scholar]
  26. Watson D, Clark LA, McIntyre CW, Hamaker S. Affect, personality, and social activity. Journal of Personality and Social Psychology. 1992;63:1011–1025. doi: 10.1037//0022-3514.63.6.1011. [DOI] [PubMed] [Google Scholar]
  27. Willis J, Todorov A. First impressions making up your mind after a 100-ms exposure to a face. Psychological Science. 2006;17:592–598. doi: 10.1111/j.1467-9280.2006.01750.x. [DOI] [PubMed] [Google Scholar]

RESOURCES