Inhibition on irrelevant negative information alleviates the mediating role of psychological distress in the association between trait rumination and symptoms of depression and anxiety

Abstract

This study aims to examine whether psychological distress mediates the association between rumination and symptoms of depression-anxiety, and whether such a mediating role is moderated by the ability to inhibit irrelevant negative information (a moderated-mediation model). On-line questionnaires comprising the Ruminative Response Scale (RRS), Depression, Anxiety, and Stress Scale (DASS-21), and Negative Affective Priming (NAP) Task as a measure of inhibitory control (IC) on negative information were administered to 181 participants (M = 21.57 years old, 80.1% females). The results of the analyses showed (1) a significant negative association between psychological distress and the performance of inhibitory control on negative information, (2) a partial mediating role of psychological distress in the relationship between rumination and symptoms of depression and anxiety, and (3) that the mediating role was moderated by inhibitory control performance. The stronger the inhibitory control, the weaker a relationship between rumination and psychological distress, which is associated with the reduction in the mediating role of psychological distress on the symptoms of depression and anxiety. The implications of our findings will be discussed by considering the systemic dynamic Model for understanding depression and anxiety.

Keywords:

• rumination
• psychological distress
• depression
• anxiety
• inhibitory control
• moderated mediation

1. Introduction

Mood disorders such as anxiety and depression are major public health problems. According to the World Health Organization (WHO, 2017), anxiety affects more than 200 million individuals globally (3.6% of the population). Depression is the most common mood disorder, affecting 322 million people (4.4% of the world’s population), over half of whom are from Southeast Asia. Records from the Ministry of Health of the Republic of Indonesia’s Basic Health Research (Kementerian Kesehatan Republik Indonesia, 2018) indicate that among those aged 15 years and older, the incidence of mood disorders increased from 6% in 2013 to 9.8% in 2018. In the same year, 6.1% of the population had experienced depression.

Anxiety and depression disorders often occur together and are classified under a broader group known as internalizing disorders (Kalin, 2020). According to a global survey, 45.7% of people with major depressive disorder also had a lifetime history of one or more anxiety disorders (Kessler et al., 2015). Depressive disorders, which we refer to as depression, are characterized by persistent sadness, cognitive dysfunction, and somatic symptoms, whereas anxiety disorders involve excessive fear and worry. Depressive disorders are associated with high mortality, much of which is accounted for by suicide; however, it is not the only cause. Individuals frequently present with tearfulness, irritability, brooding, obsessive rumination, anxiety, phobias, excessive worry over physical health, and complaints of pain (e.g., headaches, joint, abdominal, or other pain). Anxiety disorders are often associated with muscle tension and vigilance in preparation for future danger and cautious or avoidant behaviors. However, depression and anxiety also share a number of symptoms such as irritability, difficulty concentrating, rumination, and sleep problems (American Psychiatric Association [APA], 2013). Thus it is important to investigate the mechanisms of both symptoms of depression and anxiety in parallel.

2. Rumination, psychological distress, and depression-anxiety

Depression and anxiety, known as “distress disorders,” are characterized by a broad range of negative affect and emotions (as a dimension of psychological distress) such as fear, sadness, anger, guilt, and disgust (Watson, 2005). On the other hand, perseverative cognitive processes, including worry and rumination, have also been identified as etiological factors that contribute to the maintenance of symptoms of depression and anxiety (Verkuil et al., 2010). According to previous findings from Brosschot et al. (2006), perseverative cognitions or chronic and repeated activations of thoughts of psychological stressors might play an important role in the escalation of negative affect and prolongation of psychological distress. Various additional empirical studies have corroborated Brosschot’s notion that chronic and repeated activation of negative thoughts, also known as rumination, can trigger symptoms of depression or anxiety (McLaughlin & Nolen-Hoeksema, 2011; Olatunji et al., 2013; Watkins, 2009).

Rumination is a common response to stressful situations as individuals attempt to cope with psychological distress (Nolen-Hoeksema, 1991). It is considered a maladaptive coping strategy characterized by passive and repetitive thinking about unpleasant memories and feelings, or perseverative cognitions (Nolen-Hoeksema et al., 2008). As a part of perseverative cognitionsrumination worsens psychological distress and is recognized as a transdiagnostic factor associated with both depression and anxiety (Watkins & Roberts, 2020).

Our study also focuses on psychological distress as a significant risk factor for the development of depression and anxiety, given its well-established connection to scientific literature. Psychological distress, as defined by the American Psychological Association (n.d.) refers to a range of painful mental and physical symptoms associated with normal mood fluctuations experienced by most individuals. It encompasses various negative emotional and mental experiences including symptoms related to anxiety and depression. While the occurrence of psychological distress is quite common among healthy, functioning people, a higher level of persistent psychological distress may be an indicator of mental disorders and may indicate prevalent mental disorders, such as depressive disorder and anxiety disorder (Cuijpers et al., 2009; Richter Levin & Xu, 2018).

Presently, most theorists posit that depression and anxiety arise from the systemic synchronization of various interoceptive, perceptual, cognitive, and motor components that mutually amplify each other. By adopting such a systemic approach, we can enhance our understanding of anxiety and depression as multi-component and dynamic processes that evolve continuously over time. These conditions are intrinsically interconnected due to their self-organizing and dynamic nature, exhibiting circular causality (Jeronimus, 2019). In this study, we adopted systemic dynamic model of depression and anxiety as a theoretical framework.

3. Systemic dynamic model

The systemic dynamic model is a valuable tool for understanding how various causal feedback mechanisms contribute to the development of systemic syndromes with various patient trajectories (Homer & Hirsch, 2006). This approach has provided support for understanding the causes of illnesses and associated trends as well as for the design of prevention, treatment, and policy interventions (Homer et al., 2016; Thompson et al., 2015). According to this model, depression (Jeronimus, 2019; Wittenborn et al., 2016) and anxiety (Jeronimus, 2019; Morris et al., 2010) can both be considered systemic syndromes.

Ruscio et al. (2015) proposed that rumination, psychological distress, and depression are closely interconnected. They suggested that rumination indirectly increases susceptibility to depression through its impact on psychological distress. Negative affect during rumination, dysfunctional behaviors (such as social disengagement, inactivity, and behavioral avoidance), and the persistence of memories of past stressors contribute to the link between rumination and psychological distress (Ruscio et al., 2015). The study by Lyubomirsky and Nolen-Hoeksema (1993) supports this notion, as dysphoric participants reported a reduced willingness to engage in enjoyable activities after engaging in rumination, despite anticipating pleasure from those activities. Hosseinichimeh et al. (2018) found that higher levels of rumination prolong the retention of stressful memories, leading to increased rumination. Consequently, psychological distress persists longer than necessary. Anxiety, based on research by Morris et al. (2010), can also be viewed as a systemic syndrome. In the current study, based on previous studies (Brosschot et al., 2006; Verkuil et al., 2010) and the systemic dynamic model (Ruscio et al., 2015) as our framework, we hypothesized that the relationship between rumination and symptoms of depression and anxiety is mediated by psychological distress.

Furthermore, Morris et al. (2010) proposed that rumination is a response to stress, which can progressively heighten psychological distress, increase the level of anxiety and disrupt cognitive functions, including inhibitory control. Previous studies have explored the interactions between rumination, psychological distress, depressive symptoms, anxiety symptoms, and the role of working memory. However, the impact of inhibition on negative information, which is linked to the working memory updating process (Joormann, 2008; Mayr & Keele, 2000), remains unclear within the framework of the systemic dynamic model of depression and anxiety. Consequently, there is still much to learn about the extent to which the inhibition on negative information influences the association strength between rumination, psychological distress, depressive symptoms, and anxiety symptoms.

4. The relationship of rumination, inhibitory control, and psychological distress

Inhibitory control (IC) refers to the internal processes through which individuals inhibit the retrieval of irrelevant information while maintaining relevant information in their working memory (Greiff et al., 2015; Valle et al., 2019). Joormann and Gotlib (2010) suggested that poor IC on negative information may reduce the use of adaptive emotion regulation strategies such as reappraisal (i.e., suppressing the effect of task-irrelevant emotional information while reappraising a stressful situation; Cohen et al., 2014), resulting in maladaptive coping with negative emotions, such as rumination. Other studies have also shown that ruminators struggle to block pertinent information from working memory due to their rigid attentional bias towards negative information (Mayr & Keele, 2000). For example, DeJong et al. (2019) found that higher attentional switch costs were linked to more depressed brooding, suggesting that ruminators have a propensity for rigid biased processing of negative stimuli. Such rigidity might occur due to impairments in the attentional disengagement process (Grafton et al., 2016) among ruminators. Furthermore, impaired attentional disengagement might reduce the effectiveness of the inhibition process (Crawford et al., 2012), specifically in preventing irrelevant negative stimuli from entering working memory.

Based on the aforementioned studies, deficiencies in inhibitory control (IC) regarding negative information have been found to exacerbate attentional bias towards negative stimuli and foster persistent negative thoughts in working memory. Consequently, these factors contribute prolonging of psychological distress (Wittenborn et al., 2016). The amplification of negative bias can be attributed to the interplay between rumination tendencies in individuals with depression-anxiety and their poor performance in IC when confronted with negative information. In a similar vein, Hosseinichimeh et al. (2018) proposed that the impact of rumination on the escalation and perpetuation of psychological distress is closely linked to impaired inhibitory control. Therefore, to elucidate the role of inhibitory control within the systemic dynamic model of depression, our investigation sought to determine its significance as an underlying mechanism that moderates the strength of the association between rumination and psychological distress.

We used the Negative Affective Priming (NAP) Task (Joormann, 2004, 2006) to measure the performance of the inhibition on irrelevant negative stimuli. An NAP Task involves two Valence conditions, namely Positive and Negative. In this task, participants are shown positive and negative words simultaneously as distractors and targets, respectively. Participants are instructed to determine whether the target words are positive or negative while disregarding the distractor words displayed above them. What participants are not aware of is that each Valence condition includes both NAP and Control conditions. Within each NAP and Control condition, a pair of trials, known as prime and probe trials, are presented in sequence. The main distinction between the NAP and Control conditions lies in the relationship between distractors during prime trials and targets during probe trials. In the NAP conditions, the distractors during the prime trials and the targets during the probe trials share the same valence (either positive or negative words). However, under Control conditions, they have opposite valences.

To assess inhibitory control performance, the NAP Effect is calculated, which measures the difference in reaction time between probe trials in the NAP and Control Conditions. A positive NAP Effect indicates good performance, with larger values indicating better performance. The purpose of this calculation is to evaluate the ability to ignore distractor stimuli during the prime trial, which requires the inhibition of cognitive representations associated with the distractor and related words of similar valence in working memory. When the probe trial follows, participants in the NAP condition must respond to the target stimulus that shares the same valence as the previously inhibited distractor. This leads to a slower reaction time for the probe’s target due to the readjustment process required in the selection/inhibition mechanism of working memory. In contrast, the Control condition lacks this readjustment process because the valence of the prime’s distractor and the probe’s target is dissimilar. Consequently, a negative NAP Effect (shorter reaction times for probe trials in the NAP condition) or zero effect (no difference in reaction times between NAP and Control conditions) indicates poor inhibitory performance, indicating that inhibition has not effectively operated in the NAP conditions where a readjustment process would be unnecessary (Frings et al., 2015; Joormann, 2004).

The NAP task has been widely used, particularly in the study of psychopathology such as depression and schizophrenia (Frings et al., 2015; Joormann et al., 2007), as well as grief (Delespaux & Zech, 2015). As discussed above, inhibition deficits make it harder for depressed ruminators to prevent irrelevant negative information from entering the working memory. Therefore, the strength of the association between rumination and psychological distress may be reduced by a higher ability to control attentional bias to negative information. This aligns with the mechanism of inhibitory control measured using the NAP task.

5. Aims of the study

Examining current components and novel interactions between existing constructs may help us to better understand the processes that underpin depression and anxiety, as there is still some confusion regarding the variables that may explain the dynamics of depression and anxiety. We proposed an explanation for the connection between rumination, psychological distress, inhibitory control, and depression/anxiety using a moderated mediation model (Figure 1). According to Jeronimus (2019), the systemic dynamic model and evidence of connections between the aforementioned variables may provide new explanations for the frequent co-occurrence of anxiety and depression. Thus, we assume no differences in the pathways between depression and anxiety. Later in the discussion depression and anxiety were examined separately.

Figure 1. Moderated mediation model of depression-anxiety.

This figure illustrates the moderated mediation model tested in our study. The links between the independent variable (Rumination), the mediator (Inhibitory Control on Negative Stimuli), and the dependent variable (Depression-Anxiety) are represented by solid arrows. Additionally, there is another arrow (A2) representing the interaction effect between the moderator (Inhibitory Control on Negative Stimuli) and the independent variable (Rumination) on the mediator (Psychological Distress). This model hypothesized an indirect link between rumination, psychological distress, and depression-anxiety would be conditional to the level of Inhibitory Control on Negative Stimuli.

Two research questions were addressed in this study: (a) whether rumination and depression or anxiety are related in a way that is indirectly mediated by psychological distress and (b) whether inhibitory control over irrelevant negative stimuli can act as a moderator in the association between rumination and psychological distress.

6. Methods

Approval was granted by the Ethics Committee of the Faculty of Psychology at Universitas Indonesia (093/FPsi.Komite Etik/PDP.04.00/2022). The study was correlational, comprising measures of rumination (independent variable), psychological distress (mediator), inhibitory control (moderator), and depression-anxiety (dependent variable).

6.1. Participants

The data sample was obtained from the general public with a minimum age of 18 years (N = 237). Data collection was carried out using an online survey that had been provided, in which there was an explanation of the research and informed consent that was agreed upon at the beginning. Participants were informed that their participation was confidential and that they were free to withdraw at any time. The average data collection time for each participant was around 25–30 minutes. Incentives were offered to encourage participation in the form of an e-wallet of IDR 25,000.00 (Estimated 1.7 USD).

6.2. Measures

6.2.1. Sociodemographic details

The first section of the questionnaire asked for sociodemographic details such as age, education level, gender, occupation, psychological evaluation history, and domicile. In this research, we use age as a covariate as it is associated with rumination and depression (Tong et al., 2021). Studies have shown that rumination tendencies and the prevalence of depression and anxiety can vary across age groups (Nolen-Hoeksema & Watkins, 2011). Research has consistently found gender differences in the prevalence, presentation, and symptomatology of psychological distress, depression, and anxiety disorders (Altemus et al., 2014; Kuehner, 2017). Women tend to experience higher rates of depression and anxiety compared to men. Meanwhile, education level is associated with cognitive abilities and intellectual functioning (Mirowsky & Ross, 2017). Higher levels of education are generally linked to improved problem-solving skills, critical thinking abilities, and better cognitive resources. These cognitive factors can influence the way individuals process and cope with rumination, depression, or anxiety.

6.2.2. Rumination

Rumination was measured using the Indonesian version of the Ruminative Response Scale (RRS) (Rahman, 2020) which was originally developed by Treynor et al. (2003). Participants were instructed to rate each item on a scale ranging from 1 (almost never) to 4 (almost always). Higher scores indicate greater rumination. Previous research showed that the Cronbach’s Alpha of the items was α = .87 (Rahman, 2020). In this study, the reliability of the items was α = .90.

6.2.3. Depression, anxiety, and psychological distress

The third part of the questionnaire measured depression, anxiety, and distress as assessed by the DASS-21 (Depression, Anxiety, Stress Scale-21) translated by Kinanthi et al. (2020), first developed by Lovibond and Lovibond (1995). This 21-item scale is applicable in both clinical and non-clinical settings and has been used to measure the negative emotions of individuals in the most recent week. The DASS is a state measure, not a trait measure (Lovibond & Lovibond, 1995). Each subscale contains seven items. Participants were asked to respond on how relatable the item was to them in the past week. Using the Likert four-level scoring system, with 0 to 3 points representing non-conformity (0) to very consistent (3). The higher the score, the higher the level of negative emotion. A Cronbach’s alpha of α = .90 was found in the current study for the overall DASS-21 scale. The Cronbach’s alpha for each subscale was also reported: depression (α = .88), anxiety (α = .75), and stress (α = .76). Psychological distress was measured using the Stress subscale. In addition, the results of testing the validity of the structure through factor analysis using CFA produced a model that is in accordance with Lovibond’s theoretical model and is supported by items representing each DASS-21 dimension, especially in Indonesian population (Arjanto, 2022; Nada et al., 2022)

6.2.4. Inhibitory control

In this study, the Negative Affective Priming (NAP) Task was employed to measure inhibitory control (IC) on irrelevant positive and negative information. The task was programmed using OpenSesame 3.3.12 software (Mathôt et al., 2011) and uploaded to JATOS (a website for online experiments) created by Lange et al. (2015).

Participants were shown two words simultaneously on the upper and lower sides of thescreen. Both words’ positions were in the center of the screen with a distance of 1 cm between the upper-side words and lower-side words. One word had a black background and the other had a white background. We used a Sans-type font with a size of 40px. The participants’ task was to read the two words shown together, followed by judging the valence (positive or negative) of the word with a white background (target) and ignoring the word with a black background (distractor). Participants judged the targets of the displayed target word by pressing the “K” button for negative words and “L” button for positive words. The participants were also asked to respond as quickly and accurately as possible.

For the stimuli of the NAP task, we selected 65 words, each with positive (Wise, Happy) or negative (Envy, Cruel) valence from an Indonesian word norming study (Sianipar et al., 2016) and a previous study by Putri et al. (2023). The words were selected based on two requirements: First, they must not differ in terms of length, number of syllables and word familiarity. In this study, the words’ length (t (128) = .88, p > .10), number of syllables (t (128) = 1.24, p > .10), and word familiarity (t (128) = 1.8, p > .05) also did not differ significantly. Second, positive and negative words differ in their valence levels. In this study, the selected words’ valence (Mean positive words = 7.03, mean negative words = 2.96, t (128) = 41.21, p < .001) differed significantly, with positive words showing higher levels of valence in comparison to negative words. The complete list of stimuli can be seen through this link: https://doi.org/10.6084/m9.figshare.21947405

The total number of trials conducted to measure the NAP task was 168, consisting of 8 practice trials (4 prime and 4 probe trials) and 160 main trials (80 prime and 80 probe trials). As with other studies using an NAP task, we created two conditions, namely NAP and Control conditions by pairing prime-probe trials based on the valence level of the distractor in the prime trials and the valence level of the target in the probe trials. Each of the NAP and Control conditions was divided into two groups based on the valence of the target words in the probe trials (positive and negative trials). See Table 1 for task design.

Table 1. Control and NAP conditions in negative affective priming task

	Prime Trials		Probe Trials
Control Conditions	Distractor	Target	Distractor	Target
Positive Trials	-	-	-	+
Negative Trials	+	+	+	-
NAP Conditions
Positive Trials	+	-	-	+
Negative Trials	-	+	+	-

+ = positive words

- = negative words

In this study, we created a stimulus list consisting of 5 blocks. Each block consisted of 32 trials (16 prime trials and 16 probe trials), comprising 8 prime-probe pairs for the NAP condition and 8 prime-probe pairs for the control condition. Each condition consisted of the same number of positive and negative trials. In each block, the order of the positive and negative trials in the NAP and Control conditions was randomized. The distractor and target were repeated after a minimum of 8 trials, with a maximum of 3 repetitions of the target words in probe trials across the blocks.

We collected the reaction times (RTs in milliseconds) of correct answers for probe trials in the NAP Task. Trials with RTs faster than 250 ms and slower than 3000 ms (Schrobsdorff et al., 2012) were removed. There were 39 participants whose data were not used because 30% (maximum 48 trials removed per participant) of their reaction time data were discarded (McLennan et al., 2019). There were 17 participants who did not follow the instructions; therefore, so they were not included in the analysis, leading to 181 participants being included in the analysis.

The performance of IC on negative information for each participant was measured using the score of the NAP effect by subtracting the average RTs for Control conditions with RTs for NAP conditions (Joormann, 2004) for negative targets. A positive NAP effect would indicate good IC performance, whereby the inhibition of a negative distractor (in prime trials) slows down negative target processing (in probe trials) in NAP conditions, resulting in slower RTs for NAP conditions than those for Control conditions. IC performance was used in the moderated-mediation analysis to test its moderating effect on the relationship between rumination and psychological distress.

6.3. Procedure

Participants were asked to agree and fill out an informed consent to participate in the study, those who were willing to be sent a link to access the first part, which contained the DASS-21 & RRS measuring instrument. A link was distributed to the participants shortly after they completed the RRS and the DASS-21. Participants were instructed to perform the NAP task only on a desktop computer or laptop, and not on a cell phone or tablet. On average, the participants took 20–25 minutes to complete the task. The progress of each participant was monitored intensively to ensure that no participant was working on the first and second parts with a long pause.

6.4. Statistical analysis

This study used SPSS v.26 and PROCESS Hayes v.4 software (Hayes, 2022). First, a descriptive analysis (mean and SD) and correlation of each variable (i.e. rumination, depression, anxiety, psychological distress, and inhibitory control to negative information) were conducted. Then, we performed mediation-moderation analyses (PROCESS Model 7) for depression and anxiety as the dependent variables, separately. The direct effect of rumination on depression and anxiety (path c), and the indirect effect of psychological distress (path a and b) will be tested. Sex, age, and education were included as covariates. A bootstrapping technique was used with a value of 5000.

The significance of Model 7 was determined by an index of moderated-mediation, which shows that the confidence interval did not include 0 and the p-value <0.05. In this study, a significant model suggests that the relationship between rumination and depression-anxiety can be mediated by psychological distress, and the mediation effect is also influenced by the individual’s inhibitory control ability.

7. Result

7.1. Sample description

Data were collected to get an overview of the participants in this study. Table 2 summarizes the characteristics of the participants.

Table 2. Sample description (age, gender, and education)

Number of Participants	181
Age
Mean	21.57
Standard Deviation	4.29
Education
Junior High School	1	0.6%
High School	111	61.3%
Vocational Study	3	1.7%
Undergraduate	64	35.4%
Graduate	2	1%
Gender
Female	145	80.1%
Male	36	19.9%

7.2. Bivariate associations

The bivariate correlational analysis showed that rumination was significantly correlated with psychological distress (r = .654, p < .01), anxiety (r = .579, p < .01), and depression (r = .543, p < .01). A significant correlation was also found between psychological distress and anxiety (r = .739, p<.01), depression (r = .617, p < .01), and inhibitory control (r = −.149, p < .05). Anxiety also correlated with depression (r = .598, p < .01). There was no significant relationship between inhibitory control and rumination, anxiety, or depression (see. Table 3).

Table 3. Descriptive data and bivariate correlation among variables

Variables	Reliability	Mean (SD)	1	2	3	4
Rumination	.900	52.84 (10.67)	-
(2)Psychological distress	.756	20.37 (8.26)	.654**	-
(3) Anxiety	.753	14.43 (8.13)	.579**	.739**	-
(4) Depression	.881	12.79 (9.76)	.543**	.617**	.598**	-
(5) Inhibitory control on negative stimuli (NAP Effect)	-	9.37 (79.64)	.-075	−.149*	−.073	−.009

Notes. * p < 0.05; ** p < 0.01

7.3. Rumination to psychological distress (Path A1), inhibitory control on negative stimuli to psychological distress, and the interaction effect of inhibitory control on negative stimulus and rumination on psychological distress (Path A2)

As with the bivariate analyses, the regression coefficients for the direct effect of rumination (and covariates) on psychological distress (see. Table 4) is significant and positive. In this study, we found that inhibitory control moderated the association between rumination and psychological distress, which supported our hypothesis. As the NAP effect decreased, the relationship between rumination and psychological distress became stronger. The strength of the interaction was indicated by the explained variance rather than the b coefficient (see Figure 2). The model explained 46.19% of the variance in psychological distress (F(7,173) = 21.215, p < .001). The interaction between rumination and inhibitory control increased the variance in psychological distress explained by the model(ΔR² = 1.4%, F(1,173) = 4.45, p < .05).

Figure 2. Psychological distress scores by rumination levels, subject to inhibitory control on negative stimulus.

Table 4. Predictors of psychological distress, with covariates

	Psychological distress
Measures	b	(SE)	95% CIs (LLCI, ULCI)
Constant	21.788	(3.809)	14.269, 29.306
Rumination	.531**	(.045)	.442, .621
NAP Negative	−.013*	(.006)	−.024, −.001
Rumination x NAP Effect (IC on negative stimuli)	−.001*	(.0005)	−.0022, −.0001
Covariates
Age	−.007	(.129)	−.262, .248
Gender	−1.515	(1.215)	−3.914, .884
Education	.610	(.541)	−.459, 1.679

Notes. * p < 0.05; ** p < 0.01

7.3. Psychological distress to depression and anxiety

Regression coefficients for the mediation of psychological distress on depression and anxiety (see. Table 5) is significant and positive. The model explained 41.75% of the variance in depression (F(6,174) = 20.78, p < .001) and 56.35% of the variance in anxiety (F(6,174) = 37.43, p < .001).

Table 5. Predictors of depression and anxiety, with covariates

	Depression			Anxiety
Measures	b	(SE)	95% CIs (LLCI, ULCI)	b	(SE)	95% CIs (LLCI, ULCI)
Constant	.608	(5.087)	−9.432, 10.649	2.522	(3.665)	−4.712, 9.758
Rumination	.222**	(.073)	.078, .366	.121*	(.052)	.017, .225
Psychological Distress	.544***	(.091)	.365, .724	.622***	(.065)	.493, .752
Covariates
Age	.066	(.158)	−.246, .379	−.005	(.114)	−.231, .219
Gender	−.098	(1.493)	−3.045, 2.848	.108	(1.076)	−2.015, 2.231
Education	−.339	(.664)	−1.651, .972	−.189	(.478)	−1.134, .755

Notes. * p < 0.05; ** p < 0.01; *** p < 0.001

7.4. The direct effect of rumination on depression and anxiety (Path C’1)

As shown in Table 6, rumination was directly related to depression and anxiety, indicating that partial mediation applies.

Table 6. Conditional indirect effects of rumination on depression & anxiety through psychological distress, subject to inhibitory control

	Depression & Anxiety
	b	(SE)	95% CIs (LLCI, ULCI)
Path A1 (rumination to psychological distress)	.531	(.045)	.442, .621
Path A2 (interaction of rumination to psychological distress by inhibitory control on negative stimulus)	−.0011	(.0005)	−.0022, −.0001
Path C’1 (direct path, rumination to depression)	.222	(.073)	.078, .366
Path C’1 (direct path, rumination to anxiety)	.121	(.052)	.017, .225
Path C1 (indirect path, rumination to depression)
−1 SD inhibitory control	.338	(.065)	.213, .474
M inhibitory control	.289	(.053)	.188, .397
+1 SD inhibitory control	.241	(.051)	.145, .345
Path C1 (indirect path, rumination to anxiety)
−1 SD inhibitory control	.386	(.055)	.284, .503
M inhibitory control	.331	(.043)	.248, .421
+1 SD inhibitory control	.275	(.045)	.189, .367

Notes. Confidence intervals (CIs) and standard errors (SE) are bootstrapped estimates (N = 5,000)

7.5. Conditional indirect effects (Path C1)

The indirect effects of rumination on depression and anxiety, through psychological distress (path C1), were conditional on the performance of inhibitory control on irrelevant negative stimuli, as shown by the indexes of moderated mediation for depression (b = −.0006, SE = .0003, 95% CIs = −.0013, −.0001) and anxiety (b = −.0007, SE = .0003, 95% CIs = −.0014, −.0001). Thus, it can be concluded that the mediation of psychological distress in the relationship between rumination to depression and anxiety was significantly moderated by inhibitory control. The three conditional indirect effects of rumination on depression and anxiety are summarized in Table 6.

Figure 2 shows that rumination was positively associated with psychological distress at weak, moderate, and strong levels of NAP Effect/inhibitory control performance. Interestingly, the relationship between rumination and psychological distress strengthened the NAP Effect/inhibitory control performance lowered. The indirect effect (see Table 6) shows that higher inhibitory control on irrelevant negative stimuli, rather than increases, the indirect effect of rumination via psychological distress on depression and anxiety.

8. Discussion

In this study, the mechanisms underlying the link between rumination and depression-anxiety were investigated. Psychological distress is hypothesized to be a mediator of this link. We also specifically investigated the role of the inhibition on negative stimuli in determining how strongly rumination and psychological distress are related.

One of our findings suggests that rumination is directly associated with depression and anxiety, in line with previous studies that have shown the same effect on depression (Nolen-Hoeksema et al., 2008; Watkins, 2009) and anxiety symptoms (Olatunji et al., 2013; Watkins, 2009). Thus, ruminating may increase an individual’s risk of developing depression or anxiety through the generation of persistent unpleasant feelings and thoughts, which in turn makes rumination more harmful and hinders one’s capacity for problem-solving and adaptive action (McLaughlin et al., 2007; Nolen-Hoeksema et al., 1999, 2008).

Another finding of this study also suggests that psychological distress significantly mediates the associations between rumination and symptoms of depression-anxiety. This finding provides further evidence that supports previous studies suggesting that psychological distress is an important predictor of the symptoms of depression (Cuijpers et al., 2009; Richter Levin & Xu, 2018) and anxiety (Cuijpers et al., 2009). It is noteworthy that the results point to a partial mediation of psychological distress, suggesting that, while rumination can directly predict the probability of depressive and anxious symptoms, it can also predict symptoms via psychological distress.

Our findings are also in line with Wittenborn et al. (2016) and Morris et al’s. (2010) systemic dynamic models of depression and anxiety. According to this model, ruminating may prolong psychological distress. Wittenborn et al. (2016) extensively discussed the impact of persistent activation of negative representations, leading to increased orientation and sustained attention towards negative information. This biased attention and processing of negative stimuli contributes to psychological distress, resulting in unpleasant emotions, such as sadness and biased interpretations aligned with negative mental models (De Raedt & Koster, 2010). Negative thoughts and emotions are encoded and retained in the memory, reinforcing existing negative self-referent schemas (De Lissnyder et al., 2010). Moreover, the tendency to ruminate intensifies when individuals experience negative emotions and interpretations, exacerbating psychological distress and heightening negative affect (Nolen-Hoeksema & Morrow, 1991).

Ruscio et al. (2015) supported this model by explaining the relationship between rumination, psychological distress, depression, and cognitive functions. Their research revealed that rumination prolonged the duration of negative information in an individual’s memory, thereby increasing psychological distress. Additional support for this model is povided by Ruscio et al. (2015) and Hosseinichimeh et al. (2018). These studies collectively suggest that this process enhances the likelihood of experiencing symptoms of depression and anxiety. Our findings also align with the systemic dynamic model proposed by Ruscio et al. (2015), further reinforcing the usefulness of this framework.

The present study also examined the impact of inhibitory control (IC) on negative stimuli and its effect on the association between rumination and psychological distress. Our findings suggest that stronger IC on negative stimuli reduces the strength of the link between rumination and psychological distress, whereas weaker IC strengthens this association. These results are consistent with Joormann’s (2010) research, which indicates that poor ability to inhibit irrelevant negative stimuli increases the accessibility of negative information in working memory, leading to heightened rumination and subsequent psychological distress.

Therefore, our findings underscore the significant moderating role of inhibitory control in the relationship between rumination and psychological distress. Furthermore, we provide evidence highlighting the influence of inhibitory control on negative information in moderating the strength of the association between rumination, psychological distress, and symptoms of depression and anxiety. This evidence contributes to our understanding of depression and anxiety as systemic syndromes, wherein various cognitive, affective, and behavioral components are interconnected and mutually influence one another as stated by Jeronimus (2019).

In conclusion, our findings illustrate the intricate relationship between ruminations and inhibitory control of negative stimuli, highlighting their potential to alleviate psychological distress and alleviate symptoms of anxiety and depression. These results underscore the importance of considering individual differences in inhibitory control and trait rumination in comprehending the emergence of depressive and anxiety symptoms. Future interventions should be designed to target inhibitory performance more accurately, specifically focusing on the capacity to inhibit irrelevant negative information among individuals with a higher propensity for rumination. One effective strategy for mitigating the detrimental effects of rumination is the implementation of Cognitive Bias Modification (CBM). A comprehensive review conducted by Mor and Daches (2015) reveals the efficacy of CBM in training individuals to inhibit their response to negative information. Consequently, CBM can serve as a valuable method for enhancing inhibitory control among individuals with heightened inclination to ruminate.

9. Strength and limitation

This study utilized the Negative Affective Priming (NAP) task to obtain a more objective measure of inhibitory performance. The implementation of an online data acquisition system enabled us to reach a broader audience and increase accessibility without significant resource requirements. The present study may serve as a template for future investigations, particularly those focusing on psychopathology, that utilize online cognitive tasks like the NAP Task.

While the NAP Task has been extensively employed in the field of affective processing and psychopathology in WEIRD (White, Educated, Industrialized, Rich, and Democratic) populations, its utilization in non-WEIRD populations has been limited. Our study stands among the few that has expanded the use of the NAP Task to include Indonesian participants, representing a non-WEIRD population. To our knowledge, our study is also the first attempt to examine the influence of inhibitory control on the association between rumination, psychological distress, and depression-anxiety. This highlights the significance of inhibitory control in processing negative information in the frequent co-occurrence of depression and anxiety symptoms.

This study has several limitations. Given that the population employed is non-clinical and that the sample consisted of more women than men (average age: 21–22 years), a follow-up study is necessary to test our model under more precise and controlled settings (clinical vs. control populations with more gender-balanced samples). The absence of longitudinal measurements in the current study also raises another issue. In addition, according to Ruscio et al. (2015) and Hosseinichimeh et al. (2018), mechanisms involving rumination, psychological distress, depression, and cognitive factors (i.e. Inhibitory control) are a cycle. Hence, we suggest that future studies employ a longitudinal design to better understand this cycle. Finally, many researchers believe that there is still potential to investigate the mechanisms of depression and anxiety using other cognitive tasks, such as the N-back task, which has been linked to rumination (Onraedt et al., 2014).

Another limitation is that we discovered that the potential impact of mental fatigue after filling out the questionnaires might have affected participants’ IC performance. This was shown by the number of individuals who were removed from the analysis because their RT data surpassed the cut-off set based on a prior study. Although it has been suggested that a higher percentage of dropouts in online experiments is expected (Mathot & March, 2021), we think it is possible to anticipate this problem. In the future, the IC assessment may be carried out at the beginning of the study while the participants are still motivated, or it can be carried out independently with a break of no more than 24 hours after filling out the questionnaires.

10. Implications and recommendations for future studies

Our study concludes that by improving IC performance on negative information (Maraver et al., 2016), the negative impact of rumination on mental health via psychological distress may be reduced. According to the perseverative cognition hypothesis proposed by Brosschot et al. (2006), repeated rumination heightens psychological load and prolongs psychological distress, leading ruminators to be more susceptible to the onset of depressive and anxious symptoms, which is in line with our findings. In addition, our study provides evidence supporting the systemic dynamic approach to depression (Wittenborn et al., 2016) and anxiety (Morris et al., 2010).

In relation to the implications of depression and anxiety, the findings of this study suggest the need for interventions that focus on enhancing inhibitory control (IC) performance specifically for processing negative information. A future objective is to expand these findings by conducting longitudinal studies based on the previously stated hypotheses. According to Wittenborn et al. (2016), a causal loop may provide a better understanding of the dynamics of depression and anxiety symptoms, particularly the causal effect of cognitive functions, such as inhibitory control. Therefore, it is crucial to incorporate IC in a longitudinal study as it plays a significant role in the occurrence of depression and anxiety symptoms.

Although efforts have been made to solicit a representative sample, it is important to replicate these findings using larger and more diverse samples to enhance their generalizability. Our sample consisted mostly of high school and undergraduate students, which may limit the generalizability of the results to other age groups or educational backgrounds. For example, older adults may engage in less self-focused rumination compared to younger adults, which could influence their susceptibility to depression and anxiety (Aldao et al., 2010). Furthermore, most of our participants were females, who might have had higher levels of rumination than males (Ando et al., 2020). Future research might want to include participants from different age groups, educational levels, cultural backgrounds, and socioeconomic statuse. This will help to determine the extent to which our findings can be applied to broader populations.

Open Scholarship

This article has earned the Center for Open Science badge for Open Data. The data are openly accessible at https://doi.org/10.1080/23311908.2023.2228556.

Ethics approval and consent to participate:

The Committee on Research Ethics at the Faculty of Psychology, Universitas Indonesia, has decided that the aforementioned study complies to the ethical standards in the discipline of psychology, Universitas Indonesia’s Research Ethical Code of Conduct, and the Indonesian Psychology Association’s Ethical Code of Conduct. Ethical approval was given on 10 June 2022 with 093/FPsi.Komite Etik/PDP.04.00/2022. All study participants were informed about the purpose of the study, their right to refuse participation, anonymity, and confidentiality of the information. Written informed consent was also obtained before participation in the study.

Disclosure statement

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

Data availability statement

The dataset for this study can be found in the figshare: https://doi.org/10.6084/m9.figshare.21555927 (Dataset) & https://doi.org/10.6084/m9.figshare.21947405 (List of Words for NAP).

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Notes on contributors

Kenneth Ananda Putra

Kenneth Ananda Putra is a student at the Faculty of Psychology at Universitas Indonesia, pursuing a Master's degree in Professional Psychology with a specialization in Adult Clinical Psychology. His research interests focus on cognitive process, executive function, mental disorders, and psychological intervention.

Clement Eko Prasetio

Clement Eko Prasetio is a student at the Faculty of Psychology at Universitas Indonesia, pursuing a Master's degree in Professional Psychology with a specialization in Adult Clinical Psychology. His research interests encompass emotion regulation, well-being, common mental disorders, loneliness, and cognitive inhibition.

Agnes Sianipar

Agnes Sianipar is an Assistant Professor in the Faculty of Psychology at Universitas Indonesia. Presently, she holds the position of Head of the Cognition, Affect, and Well-being Lab within the same faculty. Her research focuses on the roles of emotion, language, and cognitive control processes in psychological well-being, technology overuse, and everyday decision-making.