The role of psychological inflexibility in Beck's cognitive model of depression in a sample of undergraduates

Full Reference

Ruiz, F. J., & Odriozola-González, P. (2016). The role of psychological inflexibility in Beck's cognitive model of depression in a sample of undergraduates. Anales de Psicología, 32(2), 441-447. http://dx.doi.org/10.6018/analesps.32.2.214551

Introduction and Theoretical Background

Beck's cognitive model of depression, as formulated by Beck and colleagues (Beck, Rush, Shaw, & Emery, 1979), represents one of the most influential theoretical frameworks in contemporary clinical psychology. The model proposes that depressogenic schemas—enduring cognitive structures reflecting dysfunctional beliefs about the self, world, and future—constitute the fundamental cognitive vulnerability to depression. According to this theory, when these schemas are activated in response to stressful events, they generate the cognitive triad of negative automatic thoughts, which represent the most proximal cause of depressive symptoms.

Empirical research has provided substantial support for this mediational model across diverse populations and contexts. However, a fundamental question remains relatively unexplored: Do dysfunctional schemas directly increase the frequency and intensity of negative automatic thoughts, or do additional mediators and moderators modulate this relationship? This question carries significant theoretical and clinical importance, as it has implications for the development and implementation of preventive and therapeutic interventions.

The present study by Ruiz and Odriozola-González (2016) proposes that psychological inflexibility—a construct central to Acceptance and Commitment Therapy (ACT)—may function as both a mediator and a moderator within Beck's cognitive model. Psychological inflexibility is defined as the dominance of the discriminative functions of private events (thoughts, emotions, bodily sensations) over other sources of stimulus control, such as personal values and actual environmental contingencies. This construct is characterized by cognitive fusion (loss of perspective on the functional nature of thoughts), experiential avoidance (attempts to eliminate or escape unwanted internal experiences), and lack of clarity regarding personal values.

Hypotheses and Objectives

The study formulated two interrelated hypotheses:

1. Serial Mediation Hypothesis: Depressogenic schemas increase psychological inflexibility, which in turn amplifies negative automatic thoughts, ultimately generating depressive symptomatology. This model proposes a causal chain: DAS → AAQ-II → ATQ → BDI-II, wherein psychological inflexibility operates as an intermediate mechanism transmitting the effect of dysfunctional schemas on depressive symptoms.

2. Moderation Hypothesis: The relationship between depressogenic schemas and negative automatic thoughts is modulated by the level of psychological inflexibility. Specifically, it was predicted that dysfunctional schemas would show a significant association with negative automatic thoughts only when psychological inflexibility levels are elevated, whereas in individuals with greater psychological flexibility, this relationship would be weak or nonsignificant.

These objectives reflect an attempt to integrate theoretical perspectives from distinct therapeutic traditions: the traditional cognitive-behavioral model (Beck) and the contextual behavioral science approach of ACT.

Methodology

Participants: The study comprised 210 undergraduate students (age range: 18-45 years; M = 20.46, SD = 3.45) recruited from a university in northern Spain. The sample displayed heterogeneous disciplinary representation: 64% studied Psychology, 15% Speech Therapy, and 21% Teaching. Women constituted 84% of the sample, reflecting typical gender composition in social science and humanities university programs. Regarding mental health treatment history, 19% reported prior psychological or psychiatric treatment, 4% were currently in treatment, and 4% reported taking psychotropic medication, indicating that although a nonclinical sample, some participants had previous experience with mental health services.

Measurement Instruments: The study employed four validated instruments:

• Acceptance and Action Questionnaire-II (AAQ-II): A 7-item instrument with 7-point Likert scale measuring psychological inflexibility. Internal consistency: α = .88. The Spanish version was adapted by Ruiz et al. (2013).

• Dysfunctional Attitude Scale (DAS): A 40-item scale with 7-point Likert scale assessing dysfunctional schemas and beliefs. Internal consistency: α = .89. Spanish version by Sanz and Vázquez (1993).

• Automatic Thoughts Questionnaire (ATQ): A 30-item questionnaire with 5-point Likert scale measuring the frequency of negative automatic thoughts. Internal consistency: α = .95. Spanish version by Cano-García and Rodríguez-Franco (2002).

• Beck Depression Inventory-II (BDI-II): A 21-item scale with 0-3 scoring measuring depressive symptoms. Internal consistency: α = .86. Spanish version by Sanz, Perdigón, and Vázquez (2003).

Procedure: Participants were recruited during university classes. The study was presented as research examining variables involved in depression. All participants provided informed consent. A packet of questionnaires was administered in fixed order without monetary compensation. Participants were debriefed following completion.

Data Analysis: SPSS 22 was employed for descriptive and correlational analyses. The primary inferential statistical test was moderated mediation analysis using Hayes' (2013) PROCESS macro. For serial mediation, Model 6 of PROCESS was utilized (specifying the pathway DAS → AAQ-II → ATQ → BDI-II). For moderation, Model 1 was employed (testing effects of DAS × AAQ-II interaction on ATQ). Bootstrap resampling with 20,000 iterations was conducted with bias-corrected 95% confidence intervals (BC 95% CI). Effect sizes were reported as completely standardized indirect effects (abcs). The Johnson-Neyman technique identified the moderator transition point.

Results

Descriptive and Correlational Statistics: Depressive symptomatology showed moderate levels in the sample (BDI-II: M = 9.6, SD = 7.0). Approximately 24% of participants exhibited at least mild depression, 7.3% moderate depression, and 1% severe depression (using Sanz et al., 2003 cutoff criteria). Depressogenic schemas were relatively elevated (DAS: M = 113.4, SD = 25.9), while negative automatic thoughts displayed moderate mean levels (ATQ: M = 51.0, SD = 18.3) and psychological inflexibility exhibited low-to-moderate levels (AAQ-II: M = 20.2, SD = 7.7).

Zero-order correlations were substantial and significant (p < .001): between depressive symptoms and schemas (r = .47), depressive symptoms and automatic thoughts (r = .70), depressive symptoms and psychological inflexibility (r = .58), schemas and inflexibility (r = .59), schemas and automatic thoughts (r = .56), and automatic thoughts and psychological inflexibility (r = .65).

Serial Mediation Analysis: The serial mediation model was highly significant across all stages. First, depressogenic schemas significantly predicted psychological inflexibility (B = .173, SE = .017, p < .001; R² = .349). Second, both schemas and psychological inflexibility predicted negative automatic thoughts (DAS: B = .204, SE = .044, p < .001; AAQ-II: B = 1.215, SE = .150, p < .001; R² = .516). Third, in the complete model for depressive symptoms, when all variables were included, only negative automatic thoughts remained as a significant predictor (ATQ: B = .219, SE = .028, p < .001). Notably, the direct effect of depressogenic schemas on symptoms (B = .012, SE = .018, p = .494) was dramatically reduced from the total effect (TE = .126, p < .001), consistent with a complete mediation model.

The total indirect effect was statistically significant (.114, SE = .016, 95% BC CI [.085, .146]; abcs = .423, 95% CI [.318, .547]). This effect decomposed into three specific indirect pathways: (a) DAS → AAQ-II → BDI-II (.023, abcs = .086); (b) DAS → AAQ-II → ATQ → BDI-II (.046, abcs = .171); and (c) DAS → ATQ → BDI-II (.045, abcs = .166). The magnitudes of these indirect effects did not significantly differ from one another.

Moderation Analysis: The DAS × AAQ-II interaction on ATQ was statistically significant (B = .012, SE = .003, p < .001; R² = .498). The Johnson-Neyman technique revealed that the moderator transition point occurred at AAQ-II = 17.738 (corresponding to the 42nd percentile). Below this point, depressogenic schemas did not significantly predict negative automatic thoughts. Above this point, the prediction was significant. This interpretation suggests that dysfunctional schemas are "dangerous"—that is, translate into negative automatic thoughts—fundamentally when psychological inflexibility is elevated.

Discussion and Implications

The findings support a moderated mediation model that successfully integrates concepts from traditional cognitive therapy and ACT. Psychological inflexibility operates both as a mediator (transmitting the effect of schemas on symptoms through automatic thoughts) and as a moderator (determining the magnitude of the schema-automatic thoughts relationship).

From the perspective of Beck's cognitive model, results confirm that negative automatic thoughts represent the most important proximal mechanism linked to depressive symptomatology. However, from an ACT perspective, results suggest that psychological inflexibility is a central etiological factor that potentiates or amplifies the depressogenic effect of dysfunctional cognitive schemas.

The most important implication is that promoting psychological flexibility should constitute the primary objective in depression prevention and intervention programs. This does not mean abandoning traditional cognitive-behavioral principles but rather complementing them with strategies derived from ACT that strengthen individuals' capacity to maintain perspective on their thoughts and relate flexibly with unwanted internal experiences.

Limitations and Future Directions

The study acknowledges several important limitations: (a) exclusive reliance on self-report measures, susceptible to social desirability biases; (b) use of a nonclinical sample predominantly composed of young women, limiting generalization to clinical populations and diverse age ranges; (c) fixed questionnaire administration order, which could introduce order effects; and (d) cross-sectional design, which precludes robust causal inferences.

Longitudinal and experimental designs, as well as research with clinical samples and behavioral measures, are recommended to validate and strengthen the proposed model.

Significance and contribution

This study contributes to the field of depression research by demonstrating the integration of concepts from traditional cognitive therapy and contextual behavioral science. The findings indicate that psychological inflexibility operates both as a mediator and as a moderator in the relationship between dysfunctional cognitive schemas and depressive symptoms, suggesting that multiple mechanisms may be relevant for understanding how cognitive factors contribute to depression. This transdiagnostic perspective offers implications for designing interventions that combine traditional cognitive-behavioral strategies with procedures oriented toward increasing acceptance and psychological flexibility.

---