Inhibitory Control in Individuals with Clinical Levels of Depression and Anxiety Symptoms

Context and Objectives

Inhibitory control is a central executive function that allows suppression of automatic responses to reach objectives established by environmental requirements. This capacity is fundamental for limiting the influence of unwanted behaviors and thoughts, facilitating behavioral flexibility and adaptation to diverse goals. Deficits in inhibitory control can play a crucial role in emotional disorders, particularly because they can hinder the disruption of maladaptive cognitive processes such as pathological worry and depressive rumination.

Previous research suggests that individuals with depression and generalized anxiety disorder (GAD) present inhibitory control deficits compared with healthy controls. However, few studies have been conducted in Spanish-speaking countries. The objective of this study was to analyze performance on the Stroop Color and Word Test (SCWT) between groups of Colombian participants with clinical levels of depression and GAD symptoms, and a nonclinical control group.

Method

Participants: The sample consisted of 105 adult participants (68 women, 37 men, age range 18-38 years, M = 22.94, SD = 4.62) recruited through convenience sampling via the institution's social media. Thirty-seven percent were currently working. Educational levels were: 71.4% high school graduates, 9.5% technicians, 9.5% college graduates, and 9.5% postgraduates. Socioeconomic status was: 19% low, 76.2% middle, and 4.8% high.

Of the 105 participants, 27 (25.71%) showed clinical levels of depression symptoms (PHQ-9 ≥ 10) and 15 (14.29%) showed clinical levels of GAD symptoms (GAD-7 ≥ 10).

Design: An ex post facto design was implemented, comparing groups of participants classified according to their scores on screening instruments.

Intervention/Conditions: No experimental intervention was administered. Participants were classified into clinical and nonclinical groups based on their scores on the Patient Health Questionnaire-9 (PHQ-9) and the Generalized Anxiety Disorder-7 (GAD-7).

Instruments:

1. Patient Health Questionnaire-9 (PHQ-9): A 9-item Likert-type scale measuring depressive symptoms according to DSM-IV. Cutoff of 10 for depression diagnosis. The Spanish version showed adequate psychometric properties with Cronbach's alpha = .84 in this study.

2. Generalized Anxiety Disorder-7 (GAD-7): A 7-item Likert-type scale measuring GAD symptoms according to DSM-IV. Cutoff of 10 for GAD diagnosis. Showed Cronbach's alpha = .86.

3. Stroop Color and Word Test (SCWT): Consists of 3 cards, each with 5 columns and 20 rows. Participants named as many words or colors as possible in 45 seconds. The first card (Word) contains color names in black ink. The second card (Color) contains X's of different colors. These two are neutral conditions measuring processing speed. The third card (Word-Color) presents color names in non-matching colored ink, measuring inhibitory control. Stroop Interference was calculated using the formula: Interference = WC - WC', where WC' = (W × C) / (W + C).

Analysis: Bayesian analyses were conducted using JASP 0.12.2.02. Initially, equivalences between subgroups were explored using Bayesian multinomial tests (categorical variables) and independent JZS Bayesian t-tests (continuous variables). To compare SCWT performance, the informed Bayesian t-test was used, particularly appropriate for expected small to medium effect sizes, with a prior t-distribution with location 0.350, scale 0.102, and 3 degrees of freedom.

Results

Group Equivalence:

For depressed versus nondepressed comparison, Bayesian analysis showed extremely large differences in PHQ-9 scores (BF = 2.063 e+20), but no significant differences in sociodemographic characteristics (age, gender, employment status, socioeconomic status), with only anecdotal evidence for education level (BF = 1.82).

For anxious versus nonanxious comparison, extremely large differences were found in GAD-7 scores (BF = 3.13 e+19), but not in sociodemographic characteristics (BF < 1).

SCWT Results - Depressed versus Nondepressed:

On neutral conditions (processing speed), no significant differences were found:

• SCWT-Word: Depressed M = 108.37 (SD = 16.42), Nondepressed M = 109.18 (SD = 13.75), BF = 0.46, δ = 0.05 (95% CI [-0.36, 0.46])
• SCWT-Color: Depressed M = 77.56 (SD = 14.13), Nondepressed M = 77.63 (SD = 12.14), BF = 0.35, δ = 0.01 (95% CI [-0.40, 0.41])

On the incongruent condition (inhibitory control):

• SCWT-Word-Color: Depressed M = 45.63 (SD = 9.10), Nondepressed M = 50.45 (SD = 10.07), BF = 9.41, δ = 0.46 (95% CI [0.04, 0.90]) - Substantial to strong evidence

On Stroop Interference:

• Depressed M = 0.69 (SD = 7.70), Nondepressed M = 5.36 (SD = 8.44), BF = 13.70, δ = 0.50 (95% CI [0.08, 0.94]) - Strong evidence

SCWT Results - Anxious versus Nonanxious:

On neutral conditions (processing speed), significant differences were found:

• SCWT-Word: Anxious M = 99.53 (SD = 14.56), Nonanxious M = 110.54 (SD = 13.84), BF = 16.19, δ = 0.68 (95% CI [0.15, 1.24]) - Strong evidence
• SCWT-Color: Anxious M = 71.33 (SD = 11.28), Nonanxious M = 78.66 (SD = 12.57), BF = 5.98, δ = 0.50 (95% CI [-0.01, 1.04]) - Substantial evidence

On Stroop Interference (more informative when differences exist in neutral conditions):

• Anxious M = 2.14 (SD = 7.85), Nonanxious M = 4.50 (SD = 8.57), BF = 1.43, δ = 0.23 (95% CI [-0.27, 0.74]) - Only anecdotal evidence

Discussion and Conclusions

Results revealed differentiated patterns between depression and GAD. Depressed participants showed the same processing speed as healthy controls but lower scores on inhibitory control, with medium effect sizes. These findings are consistent with previous studies (Fossati et al., 2002; Javed et al., 2019; Sarosi et al., 2008) and Snyder's (2013) meta-analysis, which reported a weighted effect size of d = 0.39 for Stroop interference in depression.

In contrast, anxious participants showed clear deficits in processing speed (medium effect sizes) but not in inhibitory control when baseline differences in neutral conditions were controlled for using Stroop Interference. This contrasts with one previous study (Hallion et al., 2017) that did find inhibitory deficits in GAD, although Price and Mohlman (2007) found no differences.

The authors suggest that these differences might be attributed to expected variability among studies, cultural differences (linguistic structures, educational systems, socioeconomic status), or subtle factors such as the tendency to focus on vocal tone versus word meaning. Notably, the results partially contrast with Eysenck et al.'s (2007) attentional control theory, which predicts that worry should affect both processing speed and inhibitory control, not just the former.

Identified Limitations: (1) Inhibitory control measured only by SCWT; (2) Symptom severity variability within clinical groups not considered; (3) Sample with approximately twice as many women as men; (4) Classification of clinical participants based on questionnaires rather than clinical interview; (5) Clinical subgroup considerably smaller than nonclinical controls, particularly for GAD (n = 15).

Importance and Contribution

This study represents a significant advancement as one of the first in Spanish-speaking countries directly comparing inhibitory control between individuals with clinical depression, clinical GAD, and healthy controls using the SCWT. The findings suggest differentiated patterns of cognitive deficits: depressed individuals show deficits specifically in inhibitory control, while anxious individuals show deficits in processing speed. This has important clinical implications: psychological interventions for depression could benefit from inhibitory control training, while interventions for GAD could focus on processing speed improvements.

---