To reappraise or not to reappraise? Emotion regulation strategies moderate the association of loneliness during COVID-19 with depression and anxiety

ABSTRACT

Background

The COVID-19 pandemic resulted in heightened feelings of loneliness due to lockouts and social restrictions.

Objective

In the present study, we examined the association of loneliness during the pandemic with anxiety and depression, while exploring the moderating role of the tendency to use two emotion-regulation strategies (expressive suppression, cognitive reappraisal).

Design

We chose to examine these associations in a sample of older adults, because they faced higher risk for loneliness and health problems during the COVID-19 pandemic.

Methods

Specifically, 174 Israeli veterans and ex-prisoners of wars from the 1973 Yom Kippur war (mean age = 69) completed self-report scales tapping loneliness, depression, anxiety, and emotion regulation strategies at the beginning of the COVID-19 outbreak (April-May 2020).

Results

Findings revealed a stronger association between loneliness and depression among participants who had a greater tendency of using suppression. The tendency to use suppression did not significantly moderate the link between loneliness and anxiety. Additionally, a weaker association between loneliness and depression was found among participants who has a greater tendency of using reappraisal. However, these participants showed a stronger association between loneliness and anxiety.

Conclusions

These findings highlight the complexity of reappraisal and adds to the growing body of work on emotion regulation.

KEYWORDS:

• Loneliness
• depression
• anxiety
• COVID 19
• emotion regulation
• reappraisal

The intensity of stressors is often dependent on a person’s subjective appraisal and interpretation (Lazarus, 1966; Lazarus & Folkman, 1984). Nevertheless, there are a few characteristics of the stressor itself which may influence its perceived intensity. A stressor is considered more threatening and distressing when it has an impact on various aspects of one’s life, rather than just one; when it affects more important areas of life; when it is prolonged; and when it is perceived as uncontrollable (Snyder & Pulvers, 2015). The COVID-19 outburst, which was first declared as a public health emergency of international concern, and later progressed to be characterized as a pandemic, is an unprecedent intense stressor that affected individuals from all parts of the world (World Health Organization, 2020a, 2020b). It had an immense scope of negative influences across life domains – economic and financial extensive costs, vast impact on people’s social and family lives, alongside the huge threat to physical and mental health (Ahrendt et al., 2020; Aristovnik et al., 2020; Banerjee & Rai, 2020; Goodell, 2020; OECD, 2021a, 2021b). Additionally, the uncertainty and ferocity of the pandemic left people helpless and uncapable to control the unfolding events (Leone, 2021). Taking into account its ongoing consequences throughout many significant facets of life, and its perceived uncontrollable nature, COVID-19 can be considered as a severe stressor.

One of the most prominent changes caused by the COVID-19 pandemic is social distancing. Maintaining physical distance from others was amongst the first instructions published by the World Health Organization (Banerjee & Rai, 2020; World Health Organization, 2020c). Large international events as well as personal events and family gatherings got canceled or delayed, and even school and work attendance became virtual or called off altogether (Ahrendt et al., 2020; Aristovnik et al., 2020; Olympics.com, 2020; UNESCO, n.d.). In an attempt to contain the infection, almost all countries worldwide had imposed a local or national lockdown at least once since the pandemic’s outbreak, causing most of the population to stay at home, inevitably compelling people to social isolation (Our World in Data, n.d.). The majority of research thus far suggests that these unique circumstances led to an increase in feelings of loneliness (Dahlberg, 2021; Killgore et al., 2020; OECD, 2021b).

Loneliness has long been considered one of the most prominent issues of the twenty-first century. It is viewed as a public health problem, affecting a rising number of people, with a peak of more than 3 out of 5 Americans in 2019 (Cacioppo & Cacioppo, 2018; Daniel et al., 2018; Jeste et al., 2020; OECD Data, 2019; Renken, 2020). Loneliness is a subjective feeling resulting from perceived social isolation, though not necessarily coupled with objective social isolation, and defined as the “subjective distress resulting from a discrepancy between desired and perceived social relationships” (Jeste et al., 2020, p. 1). Loneliness and lack of social connection are substantial risk factors for mortality, comparable to well-established risk factors such as smoking, consuming alcohol, and even of greater risk than obesity and being physically inactive (e.g., Holt-Lunstad, 2017; Holt-Lunstad et al., 2010; Luo et al., 2012). Loneliness and social isolation also have negative implications for physical and mental health; they are linked to elevated blood pressure and cardiovascular diseases, weakened immune system, increased risk of Alzheimer’s Disease, cognitive decline, increased anxiety and depression, and decreased optimism and self-esteem (e.g., Hawkley & Cacioppo, 2010; Holt-Lunstad, 2017; Luo et al., 2012).

During the rise of the COVID-19 pandemic, social distancing instructions and repeated lockdowns were associated with an increase in loneliness (e.g., Killgore et al., 2020). In turn, loneliness during COVID-19 was found to be a dominant risk factor for depression and anxiety (Palgi et al., 2020). Around the world, many weren’t immune to the emotional impact of COVID-19 and the restrictions enforced due to it. As a result, feelings of loneliness and social isolation became prevalent and needed to be tackled and coped with (Dahlberg, 2021; Inkster, 2021; Killgore et al., 2020; OECD, 2021b).

The abrupt repercussions of COVID-19, like other severe stressors, demand people to try and regulate their rising negative emotions. Emotion regulation is defined as “attempts to influence which emotions one has, when one has them and how one experiences or expresses these emotions” (Gross, 2015, pp. 4–5). Humans often use different ways of emotion regulation to cope with negative emotions, such as fear, disgust, anger, and frustration (e.g., Gross, 1998a; Gross, 2002; Smith et al., 2021; Sutton et al., 2009; Tamir et al., 2008). Gross (1998b) presented a process model of emotion regulation, differentiating emotion regulation strategies by the timing of their effect on the emotion generative process. According to the model, some strategies are “antecedent-focused” in that they come earlier in the process, prior to the emotional response. Other strategies are “response-focused,” because they emerge after emotional arousal to modulate responses that are already in progress.

In the current study, we decided to focus on the tendency to use cognitive reappraisal and expressive suppression, because they are the most frequently studied and most commonly used strategies for regulating emotions (Gross, 2002; Gross & Ford, 2023). Cognitive reappraisal, a strategy which is considered antecedent-focused, involves cognitive changes in the way people interpret a stressful event that can reduce its potential emotional impact. Expressive suppression, on the other hand, is a response-focused strategy that arises later in the emotion generation process. It involves inhibition or suppression of overt expressions of emotions (e.g., physical gestures, facial expressions) in an attempt to modulate their intensity rather than preventing their occurrence (Gross, 2002).

Early studies suggested that cognitive reappraisal is a more effective emotion regulation strategy, whereas expressive suppression was considered less preferable and even maladaptive (e.g., Appleton et al., 2013; Gross, 1998a; Gross, 2002; Gross & John, 2003). In recent years, however, researchers claim that emotion regulation strategies are not necessarily good or bad per se, but the ability to flexibly alter between strategies according to situational demands and constraints is what makes emotion regulation adaptive (e.g., Bonanno & Burton, 2013; Levy-Gigi et al., 2020; Sheppes, 2020; Sheppes et al., 2011). Recent work also suggests that the characteristics of the situation are of great importance. For example, when a situation is very intense or hard to control, reappraisal can be a double-edged sword and yield opposite outcomes than intended, whether increasing negative emotion or resulting in maladaptive behavior (Ford & Troy, 2019; Levy-Gigi et al., 2016; Smith et al., 2021).

In the current paper, we investigate the links between perceived loneliness during the COVID-19 pandemic and reported intensity of anxiety and depression, while examining the potential moderating roles of the tendencies to use expressive suppression and cognitive reappraisal. Given that COVID-19 is a very intense and unique stressor, we are interested in understanding what will be the role that the tendency to suppress the expression of feelings or alternatively to reappraise them might play in moderating the association between loneliness and emotional problems during the pandemic and its associated social restrictions.

Whereas the tendency to use suppression was found to be consistently related to increased depression and anxiety (see Aldao et al., 2010 for a meta-analytic review), findings are less conclusive regarding the effects of the tendency to use cognitive reappraisal. A large body of work throughout the years suggested that the tendency to use reappraisal is associated with decreased depression and anxiety (e.g., Gross, 1998a; Gross & John, 2003; Hofmann et al., 2009). In a meta-analytic review, however, reappraisal was found to have only a weak negative association with depression, having even a weaker negative association with anxiety (Aldao et al., 2010). Recent findings suggested that, at times, the tendency to use reappraisal can even increase distress (e.g., Ford & Troy, 2019; Levy-Gigi et al., 2016). These findings, taken together with COVID-19 being an intense and uncontrollable stressor, makes the mental-health effects of the tendency to use reappraisal in the current study ambiguous.

In the current study, we chose to focus on a sample of older combat men veterans and ex-POWs from the 1973 Yom Kippur War. Older adults, above the age of 60, were at higher risk for life-threatening effects of COVID-19 and tend to experience higher levels of social isolation and loneliness (e.g., Wethington & Pillemer, 2014). Therefore, this population was at risk for experiencing loneliness and distress due the COVID-19 outbreak and related social restrictions and may be susceptible to negative psychological outcomes.

We hypothesize that as perceived loneliness during the COVID-19 pandemic increases, so will the report of depression and anxiety. Considering the consistent findings regarding the effects of the tendency to use expressive suppression, we further hypothesize that this tendency will intensify the link between loneliness and the reported intensity of depression and anxiety. That is, loneliness will have a stronger association with depression and anxiety among people who tend to rely on more suppression for regulating negative emotions than among those who less rely on this emotion-regulation strategy. Given the diverse findings concerning reappraisal, we hypothesize that the tendency to use of cognitive reappraisal for regulating negative emotions will have a significant moderating effect, yet leave the direction of this effect open to exploration in the current study.

Method

Participants and procedure

For the present study, we used data from a longitudinal study focusing on the psychological implications of war among combat men veterans and ex-POWs from the 1973 Yom Kippur War (for full details, see Solomon et al., 2012). To locate the veterans, we used the Israel Defense Forces (IDF) records. Before filling out the questionnaires, participants signed an informed consent form.

Veterans’ data were collected at five times: 1991, 2003, 2008, 2015 and 2020 (after receiving approval from the research ethics committee of the university). The current study focused on COVID-19 implications, and therefore included only the data collected in 2020. Data collection was conducted between April 12th and May 20th, 2020. It started during the last week of the first lockdown in Israel and continued during the post-lockdown period when the government started to gradually lift restrictions. Social distancing restrictions as well as obligation to wear masks remained in effect throughout that period (COVID-19 pandemic in Israel, 2023). At that time, 185 veterans participated in the study and 174 participants who completed the scales we analyzed were included in the analyses. Participants received a link and filled the questionnaires online. All 174 participant were males, aged 65–90 (M = 69.00, SD = 3.86, Mdn = 68). Their averaged years of education was 14. 67 (SD = 3.72). Forty-three percent of participants (N = 75) worked at the time of the study, 92% (N = 160) were living with their spouse during the COVID-19 outbreak, and 71.8% (N = 125) described themselves as non-religious Jews. The observed power for detecting small-to-moderate effects (5% of explained variance; Faul et al., 2009) in the sample of the current study (N = 174) was 75%.

Measures

Loneliness during COVID-19. Participants completed five items particularly tailored to the COVID-19 experience, based on the UCLA loneliness scale (Russell et al., 1980). Participants were asked to rate on a 7-point scale, ranging from 1 (not at all) to 7 (very much), the extent to which they agreed with each of the items (e.g., “Ever since we were requested to stay at home, I feel lonelier than usual,” “It’s difficult not to meet close people due to the new restrictions,” “I find it hard to stay most of the time at home”). Loneliness during COVID-19 score was calculated by averaging the responses to all items, with higher scores indicating higher perceived loneliness during COVID-19. In the current sample, Cronbach α for the five items was .84, indicating high reliability.

Depression and anxiety. Participants’ depressive and anxiety symptoms were measured using the Brief Symptom Inventory (BSI; Derogatis & Melisaratos, 1983). Participants were asked to rate the extent to which they had suffered from each symptom during the previous month on a 5-point scale ranging from 0 (not at all) to 4 (very much). The questionnaire includes 53 items, of which six tap depression (e.g., “feelings of worthlessness”) and six tap anxiety (e.g., “feeling so restless you could not sit still”). Depression and anxiety scores were calculated by averaging the responses to all items in each subscale, with higher scores indicating more severe depressive and anxiety symptoms. In the current sample, Cronbach αs were .90 and .93 for depressive symptoms and anxiety symptoms, respectively.

Cognitive reappraisal and expressive suppression. Participants’ tendency to use of cognitive reappraisal and expressive suppression was measured using the Emotion Regulation Questionnaire (ERQ; Gross & John, 2003). The questionnaire includes 10 items, of which six tap the tendency to rely on cognitive reappraisal (e.g., “I control my emotions by changing the way I think about the situation I’m in”), and four tap the tendency to rely on expressive suppression (e.g., “I control my emotions by not expressing them”) for regulating negative emotions. Participants were asked to rate on a 7-point scale, ranging from 1 (not at all) to 7 (very much), the extent to which they agreed with each item. Cognitive reappraisal and expressive suppression scores were calculated by averaging the responses to all items in each subscale, with higher scores indicating higher tendency to use the strategy. In the current sample, Cronbach αs were .92 and .83 for cognitive reappraisal and expressive suppression, respectively.

Data analysis

First, we conducted Pearson correlations examining bivariate associations between the study measures – loneliness during COVID-19, depression, anxiety, reappraisal and expressive suppression (see Table 1). Next, to examine the effects of loneliness during COVID-19 on depression and anxiety and whether the tendency to use an emotion regulation strategy (reappraisal, expressive suppression) moderated these effects, we conducted a series of hierarchical regression analyses. To check whether the model meets the assumptions of linear regression, we conducted the Shapiro–Wilk test of normality on the deviance residuals in the model. The test results indicated that the residuals were not normally distributed. Therefore, to address non-normal distribution of residuals, we examined the effects of loneliness during COVID-19 on depression and anxiety and whether emotion regulation strategies (reappraisal and expressive suppression) moderated these effects with robust hierarchical regression with MM-estimator (Chen, 2002), using the software RStudio, version 2023.06.0 + 421. This technique is known for its robustness against outliers and heavy-tailed distributions. In the first step of the analyses, we controlled for anxiety while predicting depression, and for depression while predicting anxiety. In the second step of the analyses, we added loneliness during COVID-19 and reappraisal and expressive-suppression scores as predictors. Lastly, in the third step of the analyses, we added the interaction between reappraisal and loneliness during COVID-19 as well as the interaction between expressive suppression and loneliness during COVID-19. To avoid multicollinearity and to facilitate interpretation of results, loneliness, reappraisal and expressive-suppression scores were mean centered prior to calculating the interaction terms. Given that all interactions involve two continuous variables, the Johnson-Neyman technique was employed for the interpretation of the interactions in the study. Regression coefficients are presented in Table 2.

Table 1. Means, SDs, and Pearson correlations between study’s measures.

		1	2	3	4	5
1	Loneliness during COVID19
2	Depression	.69**
3	Anxiety	.75**	.85**
4	Reappraisal	.21**	.15*	.17*
5	Expressive Suppression	.28**	.39**	.30**	.54**
	Mean	3.42	0.88	1.05	3.88	3.70
	SD	1.60	0.91	1.01	1.71	1.60

Note: * p < .05, ** p < .01.

Table 2. Regression analyses for anxiety and depression as a function of loneliness and emotion regulation strategies.

		Depression			Anxiety
		Β	β	Tolerance	Β	β	Tolerance
Step 1	Anxiety/Depression	.77**	.86**	1	1.01**	.9**	1
		[R^2 = .79**]			[R^2 = .79**]
Step 2	Reappraisal	−.08*	−.09*	.69	.05	.05	.68
	Expressive Suppression	.13**	.14**	.64	−.04	−.04	.61
	Loneliness during COVID19	.08	.08	.44	.25**	.25**	.49
	Anxiety/Depression	.69**	.77**	.43	.81**	.73**	.46
		[ΔR^2 = .02**]			[ΔR^2 = .02**]
Step 3	Reappraisal	−.11**	−.12**	.67	.08*	.08*	.65
	Expressive Suppression	.13**	.15**	.63	−.06	−.06	.59
	Loneliness during COVID19	.06	.07	.43	.23**	.15**	.48
	Anxiety/Depression	.75**	.78**	.42	.88**	.79**	.43
	Loneliness X Reappraisal	−.16**	−.18**	.64	.15**	.15**	.59
	Loneliness X Expressive Suppression	.11*	.12**	.64	−.08	−.08	.61
		[ΔR^2 = .03**]			[ΔR^2 = .02*]

Note: * p < .05, ** p < .01. In the model predicting depression, anxiety was entered as covariate. In the model predicting anxiety, depression was entered as a covariate.

Results

Bivariate associations between study variables

As can be seen in Table 1, Pearson correlations revealed that loneliness during COVID-19 was significantly associated with both depression and anxiety. As expected, the lonelier participants felt during COVID-19, the more depressed and anxious they were. Additionally, fitting past findings (e.g., Hooley et al., 2017), a significant and strong association was found between depression and anxiety. Loneliness was also significantly associated with reappraisal and expressive-suppression scores; such that the lonelier participants felt during COVID-19, the more they tended to rely on both reappraisal and expressive suppression for regulating negative emotions. Finally, Pearson correlations revealed that the more intense the tendency to use either reappraisal or expressive suppression, the higher the anxiety and depression.

The moderating effects of emotion regulation strategies

As predicted, the robust hierarchical regression conducted on ratings of depression revealed that both reappraisal and expressive-suppression scores moderated the association between loneliness during COVID-19 and depression (see Table 2). First, we examined the source of the significant interaction for loneliness and reappraisal with the Johnson-Neyman technique. The results indicated that the association of loneliness and depression became significant when the tendency to use reappraisal was below a z-score of −0.02 or above a z-score of 1.48. For lower reappraisal scores (below z = −0.02), a positive association was found between loneliness and depression: The higher the loneliness score, the higher the depression score. For higher reappraisal scores (above z = 1.48), however, the association was negative: The higher the loneliness score, the lower the depression score. This suggests that the observed positive association between loneliness and depression (see Table 2, Step 2) was significant only when the tendency to use reappraisal recedes from a z-score of −0.02.

Regarding the interaction for loneliness and suppression, the results indicated that the association of loneliness and depression was significant when the tendency to use suppression surpassed a z-score of 0.02. Above this suppression score, there was a significant positive association: The higher the loneliness score, the higher the depression score. The observed positive association between loneliness and depression was not significant among participants who did not tend to rely on suppression (z-scores lower than 0.02).

As can be seen in Figures 1 and 2, loneliness had a stronger association with depression among participants who had a stronger tendency of using expressive suppression and among those who had a weaker tendency of using reappraisal. That is, whereas the tendency to use expressive suppression seemed to intensify the link between loneliness and depression, the tendency to use reappraisal seemed to buffer it.

Figure 1. Johnson-Neyman plot for the effect of loneliness during COVID19 on depression as moderated by reappraisal.

Figure 2. Johnson-Neyman plot for the effect of loneliness during COVID19 on depression as moderated by expressive suppression.

Regarding anxiety, the multiple regression analyses revealed that the tendency to use reappraisal, but not the tendency to use suppression, significantly moderated the association between loneliness during COVID-19 and anxiety (see Table 2). This moderation effect, however, was opposite to the effect observed when predicting depression (see Figure 3). The results of the Johnson-Neyman analysis indicated that the association of loneliness on anxiety became significant when the tendency to use reappraisal surpassed a z-score of −1.09. Above this reappraisal score, a significant positive association was found between loneliness and anxiety: The higher the loneliness score, the higher the anxiety score. As can be seen in Figure 3, loneliness had a stronger association with anxiety among participants who had a stronger tendency of using on reappraisal. That is, the tendency to use reappraisal seemed to intensify the link between loneliness and anxiety.

Figure 3. Johnson-Neyman plot for the effect of loneliness during COVID19 on anxiety as moderated by reappraisal.

Discussion

In the current study, we hypothesized that perceived loneliness during the COVID-19 pandemic will be associated with heightened anxiety and depression. Additionally, based on previous findings, we sought to explore the role of emotion regulation strategies of suppression and reappraisal. Whereas the tendency to use suppression was hypothesized to intensify the link between loneliness and anxiety and depression, we expected the tendency to use reappraisal to buffer this link.

The findings partially supported the study’s predictions. First, as expected, perceived loneliness during COVID-19 was associated with higher level of anxiety and depression. Second, unexpectedly, loneliness was also associated with higher tendencies to use reappraisal and expressive suppression. These associations highlight the possibility that lonelier participants were more distressed and made more efforts to manage their negative emotions during COVID-19 using various strategies. Third, tendencies to use expressive suppression and reappraisal were both positively associated with depression and anxiety. Whereas associations for the tendency to use expressive suppression were consistent with past findings, the positive associations between the tendency to use reappraisal and emotional problems were at odds with these findings (e.g., Dryman & Heimberg, 2018; Schäfer et al., 2017). One possible explanation can be related to the fact that the COVID-19 outbreak was a highly uncontrollable stressor that involved a lot of uncertainty and hence harder to successfully reappraise. As a result, trying to positively reappraise COVID-19 might have been a so frustrating task that leaves the distress unmitigated, thereby exacerbating rather than buffering depression and anxiety. Fourth, tendencies to use suppression and reappraisal were found to moderate the affective manifestations of loneliness. However, the specific moderating effects of the emotion-regulation scores were more complex than expected.

As predicted, a stronger association between feelings of loneliness and depression was found among participants who had a greater tendency of using expressive suppression. This finding aligns with previous findings suggesting that expressive suppression tends to be a maladaptive emotion regulation strategy (e.g., Appleton et al., 2013; Gross, 1998a; Gross, 2002). However, the tendency to use expressive suppression didn’t significantly moderate the relationship between loneliness and anxiety. This finding implies that the tendency to use suppression might have different consequences for the subjective experience of depression and anxiety among lonelier people, with a more detrimental impact on depressive symptoms.

The tendency to use Reappraisal is mostly considered an adaptive and favorable emotion regulation strategy (e.g., Gross, 1998a; Gross, 2002; Gross & John, 2003), yet the current study yielded mixed results regarding its role in moderating the relationship of loneliness with depression and anxiety. Fitting past findings on the distress-buffering role of cognitive appraisal, the association between loneliness and depression was weaker among participants who had a greater tendency of using this emotion-regulation strategy. However, regarding anxiety, for those who reported using more reappraisal, the association between loneliness and anxiety was stronger. This finding implies that the tendency to use reappraisal might have some maladaptive consequences for the experience of anxiety during COVID-19.

The current findings suggest that although depression and anxiety are known to be highly comorbid and correlated (e.g., Hooley et al., 2017), the tendency to rely on reappraisal for regulating negative emotions might lead to different results in anxiety and depression. This suggestion fits previous findings, which highlighted variations on the effects of emotion regulation strategies on different emotional problems (Aldao et al., 2010). This distinction is also consistent with the growing body of work claiming that there are no “good” or “bad” emotion regulation strategies, and the ability to be flexible and switch between strategies, according to situational and personal resources and limitations, is what makes a strategy adaptive.

Another possible explanation for the results can lay in the unique situation discussed in the current paper. The COVID-19 pandemic was an uncontrollable stressor and the ability to alter the situation and the loneliness it imposed was mostly very limited. Previous studies have shown that reappraisal is less effective when the distress-eliciting situation is more intense and harder to control (Ford & Troy, 2019; Levy-Gigi et al., 2016; Smith et al., 2021). This highlights the importance of considering not only the appropriate strategy per emotional problem, but also the appropriate strategy in the context of a specific situation.

Due to the cross-sectional nature of the current study, one cannot dismiss another alternative explanation. It is possible that participants who experienced greater loneliness and anxiety felt a higher demand to engage in reappraisal to regulate their emotions, and therefore reported a greater tendency of using it more often. Further prospective, longitudinal studies following trajectories of loneliness, cognitive reappraisal, and anxiety over time would be able to test the validity of this interpretation.

We can also understand the current findings as a reflection of the role that feelings of helplessness play in depression and anxiety. Helplessness (i.e., the subjective experience of lack of control over the occurrence and outcomes of external events) is closely associated with rising levels of depression (e.g., Mikulincer, 1994). Furthermore, helplessness – specifically in the context of COVID-19 – was found to have a curvilinear relationship with anxiety: People reporting high or low levels of helplessness in attempting to avoid COVID-19 infection reported lower levels of anxiety than those who reported having some moderate level of control over the infection risk (Lifshin et al., 2020). Thus, it is possible that when people felt somewhat potent in dealing with the COVID-19 threat and therefore less depressed, they might experience heightened anxiety. In fact, Hayes and Hubley (2017) found that experimental inductions aimed at lowering depression and feelings of helplessness resulted in heightened anxiety. This might help in explaining the current results: Regulating distress through cognitive reappraisal might weaken the experience of helplessness and lead to lower levels of depression despite experiencing heightened loneliness during the COVID-19 outbreak. However, at the same time, the tendency to use reappraisal might heighten anxiety due, perhaps, to the objective uncontrollable nature of COVID-19 outbreak and related social restrictions.

Along with its contributions, the current study also has several limitations to take into consideration. First, all the study variables were assessed simultaneously in a cross-sectional design with self-report measures. Therefore, we cannot deal with problems related to shared-method variance and cannot make any conclusion about the direction of causality of the observed associations (e.g., whether loneliness leads to heightened depression and anxiety or whether depression and anxiety result in more loneliness). In addition, due to the cross-sectional design of the current study, we cannot dismiss the possibility that previous traumas and emotional problems influenced participants’ choice of emotion regulation strategies and loneliness levels during the COVID-19 pandemic.

Additionally, the sample consisted of Israeli ex-POWs and combat veterans, all older males, preventing the generalizability of the findings to other populations differing in gender, age, nationality, and life experiences from the current sample. The fact that all participants were old men should be taken into consideration. Older adults tend to feel lonelier and more socially isolated (e.g., Berg-Weger & Morley, 2020) and to have more difficulties in using cognitive reappraisal (Opitz et al., 2012). Moreover, there is some evidence that men are less likely to report using emotion-regulation strategies than women (Nolen-Hoeksema & Aldao, 2011). The fact that all participants in the current study were exposed to past traumas (e.g., war, captivity) should be also taken into consideration, because previous studies have revealed modifications in neural processes associated with the use of emotion regulation strategies, such as reappraisal, among traumatized patients (Butler et al., 2019). Future research should attempt to replicate the current findings across diverse age groups and non-traumatized people to enhance the generalizability of the findings.

Furthermore, the present study focused on the tendency to use two extensively researched emotion regulation strategies – expressive suppression and reappraisal. To gain a more comprehensive understanding of the relationship between loneliness and emotional problems, future research should explore the contribution of alternative emotion regulation strategies, such as seeking social support, that might be highly relevant for regulating emotions in response to loneliness. The current study also focused on a specific type of loneliness – perceived loneliness during the COVID-19 pandemic, further limiting our ability to generalize the results. An additional limitation concerns the relatively low mean levels of depression and anxiety. The current study was conducted in a non-clinical sample, and then future studies should attempt to replicate its findings in clinical samples.

Albeit these limitations, the current study contributes to the large body of research on emotional regulation and mental health. The current findings are particularly interesting due to the unique time the study was conducted (during the first months of the COVID-19 outbreak), and they help us shed light on the relationship between emotion regulation, loneliness, depression, and anxiety when facing a severe, uncontrollable stressor.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data presented in the paper is available upon request from the authors.