Anxiety is one of the more common emotions people experience and which causes them the greatest distress, affecting as many as 30% of the global population (APA, 2024). A core feature of anxiety disorders is the presence of repetitive negative thinking (RNT), which encompasses processes such as worry and rumination (Ehring and Watkins, 2008). While worry refers to a future-oriented, repetitive cognitive activity focused on anticipated threats (Borkovec, 1994), whereas rumination involves dwelling on past events or perceived failures in a passive, often self-critical cycle (Nolen-Hoeksema et al., 2008). While both processes are forms of RNT and share overlapping features, they are conceptually distinct and may operate through different cognitive mechanisms. Intrusive thoughts, although similarly distressing and often involuntary, also differ from RNT because they are typically brief and episodic rather than sustained and elaborative (Clark, 2005; Rachman, 1997). This conceptual distinction is crucial for the present study, which focuses specifically on worry given its strong links to generalized anxiety and its central role in metacognitive models of pathological thought (Wells, 2009).

Both worry and intrusive thoughts have traditionally been treated through cognitive behavioral therapy (CBT) (Masuda et al., 2004). The emergence of a metacognitive and contextual approach in recent years has led to the intro-duction of interventions that have proven to be effective where traditional CBT has fallen short (Hayes and Wilson, 1994; Hayes, 2004).

These interventions include Metacognitive Therapy (MCT) and Acceptance and Commitment Therapy (ACT), which have proven to be particularly effective in treating a series of anxiety-related issues (Sánchez Escamilla et al., 2024; Wells, 2009; Hayes et al., 1999). Worrying is often associated with uncertainty over anticipated threats and a perceived lack of control over future events (Nolen-Hoeksema et al., 2008), though other factors may also contribute in certain contexts. Given the role of worry across various diagnoses (Harvey et al., 2004; Ehring and Watkins, 2008), several therapeutic approaches have been developed to target and reduce repetitive negative thinking (RNT) (Wells, 2009; Watkins, 2016). Such thoughts may be conceptualized as experiential avoidance—defined as attempts to evade or suppress unpleasant internal experiences (emotions, thoughts, bodily sensations)—which is a key focus of ACT (Hayes et al., 1999; Wilson and Luciano, 2002).

The metacognitive theory (Wells and Matthews, 1994; Wells, 2009) is based on the principle that negative thoughts and emotions are, for the most part, transient experiences. These experiences become persistent and evolve into psychological disorders when the individual adopts a particular thinking style that disrupts self-regulation and impedes resolution of stress-related content. MCT, grounded in the Self-Regulatory Executive Function (S-REF) model (Wells and Matthews, 2016), posits that metacognitive beliefs play a critical role in the activation and maintenance of RNT. Specifically, positive beliefs about worry (e.g., “worrying helps me cope”) and negative beliefs about uncontrollability and danger (e.g., “I cannot control my worry”) contribute to a Cognitive Attentional Syndrome (CAS) characterized by persistent worry, rumination, and maladaptive coping behaviors (Wells, 2009). A meta-analysis by Normann and Morina (2018) confirmed the efficacy of MCT for both anxiety and depression. Their findings suggest that MCT not only reduces targeted symptoms but also alleviates comorbid problems, supporting its role as a transdiagnostic intervention (Sun et al., 2017). These comorbid symptoms include anxious, uncontrollable, and excessive worrying (Meyer et al., 1990) and depressive rumination (Nolen-Hoeksema and Morrow, 1991), which underpin emotional problems (Papageorgiou and Wells, 2003). Recent studies have highlighted that metacognitive beliefs about negative emotions and cognitions are central to sustaining maladaptive RNT (Xu and Tsai, 2025; Kornacka et al., 2025).

From a metacognitive perspective, rumination is sustained by beliefs about its perceived utility and controllability. Positive metacognitive beliefs (e.g., “ruminating helps me to solve problems”) motivate engagement in this process, while negative metacognitive beliefs (e.g., “my rumination is uncontrollable and harmful”) exacerbate distress and reinforce avoidance strategies (Wells and Matthews, 2016; Papageorgiou and Wells, 2003). The S-REF model conceptualizes this as part of the Cognitive Attentional Syndrome (CAS), where dysfunctional metacognitive beliefs guide attention and coping responses that maintain repetitive negative thinking.

Recently, the HExAGoN model (Tamm et al., 2024) has been proposed as a comprehensive framework for understanding repetitive negative thinking (RNT), including both worry and rumination. This model emphasizes multiple interacting pathways—such as habit formation, executive control deficits, abstract processing, goal discrepancies, and negative cognitive biases—that contribute to the onset and maintenance of RNT. Importantly, HExAGoN integrates key elements from metacognitive theories, such as the role of positive and negative beliefs about cognition (Wells and Matthews, 2016), and self-regulatory processes, such as effortful control and goal management. This aligns conceptually with MCT’s Cognitive Attentional Syndrome (CAS), where metacognitive beliefs sustain maladaptive thinking, and with ACT’s focus on psychological inflexibility, where cognitive fusion and experiential avoidance perpetuate distress. By highlighting these overlapping mechanisms, the HExAGoN model offers a useful lens for exploring potential integration between MCT and ACT approaches in targeting RNT.

ACT, in contrast, is based on the psychological flexibility model and draws from Relational Frame Theory (RFT; Hayes et al., 2001). It targets processes such as experiential avoidance and cognitive fusion—rigid patterns of responding to internal experiences—through interventions designed to foster acceptance, defusion, and values-based action (Hayes et al., 2012). Cognitive fusion, in particular, refers to the tendency to become entangled with thoughts such that they dominate behavior, while experiential avoidance involves attempts to escape or suppress unwanted internal experiences, paradoxically amplifying distress over time (Hayes et al., 1996). ACT focuses on everything involved in an individual’s learning history, and repetitive negative thinking tends to become the pre-dominant reaction, thereby reducing short-term emotional distress, but prolonging it both in duration and intensity over the long term (Côté et al., 2002).

From the perspective of ACT, while psychological problems may originate from a general absence of relational skills (e.g., in cognitive deficits), a primary source of psychopathology is how language and cognition interact with direct contingencies. These interactions produce a reduced ability to persist or shift behavior in pursuit of long-term goals. ACT proposes that psychological inflexibility arises from weak or ineffective contextual control over language processes (Ong et al., 2024).

In contexts that promote fusion, human behavior is more influenced by rigid verbal networks than by direct environmental contingencies. This leads individuals to act in ways that are inconsistent with environmental opportunities and chosen values. Justificatory contexts place excessive emphasis on the internal causes of behavior—especially private, conditioned events—while experiential control contexts aim to eliminate or manage emotional and cognitive states, often using them as the sole metric for a successful life (Addis and Jacobson, 1996; Zettle and Hayes, 1986). These contexts are intertwined and help explain why cognitive fusion supports experiential avoidance: efforts to control private experiences often increase their functional salience and reduce behavioral repertoire (Hayes et al., 1996). Unfortunately, such avoidance tends to increase the prominence of aversive experiences and limits flexibility, as many valued behaviors may evoke precisely the private events one is trying to control.

Both MCT and ACT have been explored in numerous studies (see recent reviews: Nordahl et al., 2022; Sánchez Escamilla et al., 2024). Although MCT and ACT are rooted in distinct theoretical frameworks, recent evidence suggests potential overlap in their targeting of metacognitive processes and RNT (Sánchez Escamilla et al., 2024). Both approaches emphasize shifting the individual’s relationship to their thoughts—MCT by altering metacognitive beliefs and ACT by promoting cognitive defusion and acceptance. This convergence raises the possibility of shared mechanisms that could inform integrated treatment models (Nordahl et al., 2022).

The present study aims to contribute to this integration by examining whether cognitive fusion and experiential avoidance (key ACT processes) mediate the relationship between metacognitive processes (central to MCT) and worry. This focus on worry, rather than on other forms of RNT such as rumination, reflects its theoretical and clinical relevance in generalized anxiety and its prominent role in metacognitive models. Although rumination and worry share features of perseverance cognition, worry’s future-oriented, threat-related content aligns more directly with the mechanisms targeted by MCT and ACT. Specifically, we hypothesize that broader metacognitive factors, including beliefs about the controllability, usefulness, and monitoring of thoughts, will predict worry indirectly via these mediators. Such findings would provide preliminary evidence for common processes across MCT and ACT, informing future development of transdiagnostic interventions for RNT.

This study aims to explore how key variables from metacognitive therapy and acceptance and commitment therapy can be integrated, with the aim of identifying overlapping processes that could inform future integrative approaches.

Our findings suggest that metacognitive beliefs about the uncontrollability of thoughts and positive beliefs about worrying are strongly associated with worry levels, primarily through their influence on cognitive fusion. This is consistent with metacognitive theory (Wells, 2009), which posits that negative beliefs about uncontrollability contribute to a self-perpetuating cycle of excessive worry and cognitive avoidance. Recent studies have further emphasized the central role of such beliefs in maintaining repetitive negative thinking (RNT) (Xu and Tsai, 2025; Kornacka et al., 2025). Cognitive fusion, a core ACT process, also emerged as a significant mediator, supporting previous evidence that identification with distressing thoughts amplifies emotional distress and worry (Hayes et al., 2006; Solem et al., 2011).

The results suggest that metacognitive beliefs about the uncontrollability of thoughts and beliefs about worrying have a considerable effect on worry, and carry the greatest weight, even though cognitive fusion has a highly significant impact on that relationship. According to Wells et al. (2009), uncontrollability contributes to the cycle of excessive worry by generating cognitive avoidance responses. The more one avoids worry-related thoughts, the easier it is to give them importance, and therefore trigger a cognitive fusion process (Solem et al., 2011). Cognitive fusion also mediates the relationship between positive beliefs about worrying and worry itself; in other words, believing that worry is beneficial may lead to increased focus on it, thereby intensifying distress (Wells, 2006). Although rumination was not measured in this study, future research could explore whether similar mechanisms apply across other forms of repetitive negative thinking, such as rumination.

In turn, cognitive fusion is directly related to the acceptance of thoughts and emotions - a central process in ACT. The inability to accept unwanted thoughts seems to reinforce cognitive fusion, which leads to an increase in levels of worry. This relationship between fusion and acceptance highlights ACT’s therapeutic value, which seeks to dismantle cognitive fusion by promoting experiential acceptance (Hayes et al., 2012). Moreover, this research shows that acceptance acts as a crucial mediator between cognitive fusion and worrying, which supports the hypothesis that the development of psychological flexibility through acceptance may alleviate worry and its associated emotional distress (Ruiz, 2010).

The mediation model also reveals that the weight of acceptance and action is significant in the prediction of worrying, which is consistent with other studies (Rood et al., 2012). Beliefs about the uncontrollability of worry may represent an important pathway to worry through their association with cognitive fusion, whereas acceptance appears to exert a smaller direct effect in this model (Wald and Mellenbergh, 2018). However, given that acceptance and cognitive fusion show similar zero-order correlations with worry, future research should examine whether their respective roles differ depending on contextual and population-specific factors.

Taken together, these findings tentatively suggest that future research could focus on evaluating whether interventions that help individuals accept the uncontrollability of thoughts are effective in reducing repetitive negative thinking.

This study presents several limitations that should be considered when interpreting the results. First, the cross-sectional and non-experimental design precludes any inference of causality among the variables examined. Although the tested mediation model is supported by theoretical frameworks, the directionality of the observed associations remains uncertain. To strengthen causal interpretations, future research should employ longitudinal or experimental designs that allow for the examination of temporal dynamics and causal pathways.

Second, the generalizability of the results may be limited by the characteristics of the sample. The sample was predominantly composed of female participants, which may restrict the applicability of the findings to the general population or to male sub-groups. Moreover, participants were drawn from a non-clinical population, and no formal diagnostic criteria were applied, which may further limit the extent to which these results can be extended to clinical settings or individuals with diagnosed anxiety or depressive disorders. Future research should aim to include more diverse and representative samples in terms of gender, age, and clinical status to enhance the external validity of the findings.

In short, the present findings highlight the central role of metacognitive beliefs—particularly those concerning the uncontrollability of thoughts—in sustaining worry. This effect appears to be primarily mediated by cognitive fusion, with acceptance contributing as a secondary mediator. These results align with prior evidence emphasizing the interplay between metacognitive processes and psychological flexibility (Hayes et al., 2006; Wells, 2009; Tamm et al., 2024).

From a clinical perspective, these insights suggest that interventions targeting both maladaptive metacognitive beliefs and experiential avoidance may offer complementary benefits. For instance, modifying beliefs about the uncontrollability of thoughts through detached mindfulness (MCT) (Mansueto et al., 2024; Wunder and Jones, 2023; Nordahl et al., 2022) could be combined with ACT strategies that promote acceptance and cognitive defusion (Cookson et al., 2020), which emphasizes the importance of fostering acceptance to reduce experiential avoidance and disengagement from unhelpful thought patterns (Hayes et al., 2006; Tamm et al., 2024). However, it is important to interpret these implications cautiously. Given the cross-sectional and exploratory nature of this study, causal relationships cannot be established. Further longitudinal and experimental research is needed to confirm these mechanisms and evaluate their relevance for integrative therapeutic approaches.

Future studies should also consider using more diverse and clinically representative samples, as the present study involved a predominantly female, non-clinical population. Moreover, replication in larger samples and across different cultural contexts would strengthen the generalizability of these findings. Designing experimental paradigms that assess worry responses in real-time (e.g., via ecological momentary assessment or laboratory-based induction tasks) could help clarify how metacognitive beliefs and cognitive fusion interact dynamically during episodes of repetitive negative thinking. Finally, some of the indirect effects observed in the model were associated with wide confidence intervals, suggesting reduced precision in estimating the magnitude of these effects. This limitation should be taken into account when interpreting the strength and reliability of specific pathways in the model.

Finally, future directions should also explore the integration of Metacognitive Therapy (MCT) and Acceptance and Commitment Therapy (ACT) techniques in ap-plied settings. Experimental studies testing the efficacy of interventions that combine strategies from both models—such as cognitive defusion exercises from ACT and attention training or detached mindfulness from MCT—could provide valuable in-sights into their synergistic effects. In addition, designing experimental studies that employ computer-based tasks to simulate worry-inducing scenarios could help assess how metacognitive beliefs and cognitive fusion processes influence worry responses in real time. These paradigms could offer an ecologically valid method to examine the immediate effects of interventions targeting fusion or metacognitive beliefs under con-trolled conditions. Furthermore, longitudinal models could be used to track the evolution of cognitive fusion, metacognitive beliefs, and acceptance over time, and their role in the onset and maintenance of pathological worry, allowing for a more comprehensive understanding of these transdiagnostic processes.