Automatic Thoughts Questionnaire – Revised (ATQ-R)

1. Abstract

The Automatic Thoughts Questionnaire – Revised (ATQ-R) is an empirically validated psychometric instrument developed to assess the frequency and balance of cognition in affective psychopathology, specifically unipolar depressive episodes and cognitive vulnerabilities. Originating from the foundational work of Steven D. Hollon and Philip C. Kendall (1980) on the original 30-item Automatic Thoughts Questionnaire (ATQ), the revised 40-item edition introduced by Kendall, Howard, and Hays (1989) incorporates 10 positive self-referent items alongside the original 30 negative self-referent cognitions. This design enables clinicians and behavioral science researchers to quantify not only the absolute presence of depressogenic cognitive intrusions—such as themes of personal failure, self-reproach, hopelessness, and maladaptation—but also the presence of adaptive, protective positive cognitions, facilitating the computation of a “States of Mind” (SOM) balance ratio.

Administered via a 5-point Likert-type response scale ranging from 1 (“not at all”) to 5 (“totally”), the ATQ-R yields two overarching subscale scores: Negative Automatic Thoughts (ATQ-N; 30 items) and Positive Automatic Thoughts (ATQ-P; 10 items). Psychometric investigations consistently confirm robust internal consistency across diverse clinical and non-clinical populations, with Cronbach’s alpha values typically ranging from .94 to .97 for the negative dimension and .86 to .93 for the positive dimension. Test-retest reliability across short- to moderate-term intervals demonstrates sound temporal stability while preserving sensitivity to therapeutic change during cognitive-behavioral interventions. Construct, convergent, and discriminant validity are evidenced by strong correlations with validated measures of depression—such as the Beck Depression Inventory (BDI) and the Center for Epidemiologic Studies Depression Scale (CES-D)—as well as clear differentiation between depressed, anxious, and healthy control cohorts.

2. Keywords

Automatic Thoughts Questionnaire – Revised, ATQ-R, Cognitive Therapy, Aaron T. Beck, Negative Automatic Thoughts, Positive Automatic Thoughts, States of Mind Model, Depression Assessment, Cognitive Vulnerability, Psychometrics.

3. Authors

The conceptualization, clinical validation, and structural revision of the Automatic Thoughts Questionnaire framework stem from major figures in cognitive-behavioral therapy and clinical psychometrics:

• Philip C. Kendall, Ph.D., ABPP: Distinguished University Professor and Laura H. Carnell Professor of Psychology at Temple University. Renowned for his empirical research on cognitive-behavioral treatment modalities, internalizing disorders, and child/adult cognitive processes.
• Steven D. Hollon, Ph.D.: Gertrude Conaway Vanderbilt Professor of Psychology at Vanderbilt University. A leading researcher in the etiology, cognitive mechanics, and treatment of affective disorders, whose clinical trials laid the foundation for measuring cognitive change in depression.
• Bruce L. Howard, Ph.D.: Clinical psychologist and psychometric researcher whose work collaborated on operationalizing the States of Mind (SOM) model and expanding cognitive evaluation scales.
• Robert C. Hays, Ph.D.: Clinical researcher instrumental in investigating cognitive self-statements, affective valence, and the empirical psychometrics of self-referent speech in psychopathology.

4. Purpose

The primary purpose of the Automatic Thoughts Questionnaire – Revised (ATQ-R) is to provide an objective, reliable, and clinically sensitive evaluation of self-referent internal dialogue related to depressive symptomatology and psychological equilibrium. In cognitive models of psychopathology, surface-level conscious cognitions—known as automatic thoughts—act as direct cognitive mediators between core underlying schemas (latent deep structures) and affective, somatic, and behavioral responses. Consequently, accurately identifying and monitoring these momentary cognitions is critical for both idiographic functional analysis and nomothetic scientific investigation.

Clinically, the ATQ-R serves multiple operational functions:

• Diagnostic Screening and Severity Assessment: It assists practitioners in gauging the frequency and prominence of depressogenic cognitive intrusions that populate a patient’s internal experience. High ATQ-N scores indicate pervasive, highly active negative schemas characteristic of major depressive episodes.
• Treatment Planning and Idiographic Case Conceptualization: By examining individual item endorsements, clinicians can identify specific content themes (e.g., helplessness, perceived personal failure, social isolation, perceived burden) that require targeted cognitive restructuring, behavioural activation, or collaborative empiricism.
• Measurement of Treatment Mechanism and Outcome: Because automatic thoughts fluctuate more rapidly than entrenched core beliefs or personality traits, the ATQ-R is sensitive to therapeutic change. Administering the scale at regular intervals during Cognitive Behavioral Therapy (CBT), Pharmacotherapy, or Interpersonal Psychotherapy enables clinicians to ascertain whether reductions in clinical distress are mediated by the attenuation of negative cognitions and the restoration of positive cognitive states.
• Evaluating Cognitive Balance: Unlike the original unidimensional ATQ (Hollon & Kendall, 1980), the ATQ-R captures adaptive, resilient self-statements (e.g., “I can accomplish anything,” “I feel confident I can do anything I set my mind to”). This dual-valence tracking enables researchers to calculate cognitive balance ratios, supporting the hypothesis that psychological health is defined not merely by the absence of negative cognitions, but by an optimal balance of positive to negative internal thoughts.

5. Psychological Construct

The ATQ-R assesses self-referent automatic thoughts, which are immediate, rapid, evaluative thoughts or mental images that emerge automatically without deliberate reflective deliberation when individuals encounter environmental or internal triggers. Within the ATQ-R framework, this construct is divided into two primary dimensions comprising distinct thematic subcategories:

1. Negative Automatic Thoughts (ATQ-N, 30 Items)

The negative dimension captures the cognitive triad defined by Beck: negative views of the self, the ongoing world/experience, and the future. Factor-analytic examinations of these 30 items historically reveal four intercorrelated conceptual facets:

• Personal Maladjustment and Desire for Change: Reflects internal dissatisfaction with one’s functioning and an awareness of internal breakdown or distress. Examples include: “My life is not going the way I want it to” (Item 12), “Something has to change” (Item 35), and “I wish I were somewhere else” (Item 21).
• Negative Self-Concept and Personal Failure: Characterized by harsh self-blame, feelings of personal inadequacy, and global judgments of unworthiness. Examples include: “I’m no good” (Item 2), “I’m a failure” (Item 31), “I hate myself” (Item 23), and “I’m worthless” (Item 25).
• Giving Up and Helplessness: Reflects a perceived inability to cope, cognitive exhaustion, and a sense of surrender in the face of life demands. Examples include: “I don’t think I can go on” (Item 8), “I can’t stand this anymore” (Item 17), and “I feel so helpless” (Item 34).
• Bleak Future and Negative Expectations: Encompasses profound cognitive hopelessness, fatalistic expectations regarding outcomes, and existential despair. Examples include: “My future is bleak” (Item 38), “I’ll never make it” (Item 33), and “It’s just not worth it” (Item 39).

2. Positive Automatic Thoughts (ATQ-P, 10 Items)

The positive dimension assesses adaptive, optimistic, self-affirming, and emotionally stabilizing self-statements that counteract vulnerability and bolster psychological resilience. This construct reflects:

• Self-Efficacy and Mastery: Statements that affirm personal agency, perceived competence, and confidence in overcoming adversity. Examples include: “No matter what happens, I know I’ll make it” (Item 10), “I can accomplish anything” (Item 13), and “I feel confident I can do anything I set my mind to” (Item 24).
• Positive Self-Esteem and Affective Comfort: Cognitions reflecting self-worth, somatic and psychological calm, and subjective well-being. Examples include: “I’m proud of myself” (Item 3), “I feel fine” (Item 7), “I’m warm and comfortable” (Item 20), and “I feel very happy” (Item 28).

6. Theoretical Framework

The ATQ-R is built on Beck’s cognitive theory of depression (Beck, 1967, 1976) and subsequent models of cognitive balance, specifically the States of Mind (SOM) model proposed by Schwartz and Garamoni (1986, 1989).

Beck’s Cognitive Diathesis-Stress Model

According to Aaron T. Beck, individuals prone to depression possess latent, maladaptive cognitive schemas—deeply held structural beliefs regarding personal defectiveness, abandonment, or unlovability. When activated by congruent negative life events or stressors, these schemas bias information processing through systemic cognitive distortions (such as dichotomous thinking, catastrophizing, and selective abstraction). This schema activation generates spontaneous, stream-of-consciousness automatic thoughts. These thoughts are perceived by the individual as plausible reflections of reality rather than subjective interpretations, directly driving depressed affect, physiological lethargy, and behavioral withdrawal.

Cognitive Specificity Hypothesis

The theoretical framework of the ATQ assumes the cognitive specificity hypothesis (Beck et al., 1987; Hollon et al., 1986). This principle states that distinct forms of psychopathology are characterized by specific cognitive themes. Depression is primarily defined by thoughts of real or perceived loss, irreversible defeat, personal inadequacy, and abandonment (depressotypic cognitions). In contrast, anxiety is characterized by themes of selective threat, physical or psychological vulnerability, and underestimation of coping resources. The negative items of the ATQ-R were constructed to isolate this depressotypic cognitive footprint.

The States of Mind (SOM) Model

The critical theoretical innovation of Kendall, Howard, and Hays (1989) in revising the ATQ was grounding the instrument in Schwartz and Garamoni’s (1986) States of Mind (SOM) model. Prior to this, clinical assessment adhered largely to an asymmetry or deficit model, which assumed that mental health is simply the absence of negative cognitions. The SOM model, however, posits that psychological adaptation depends on the relative balance between positive and negative cognitive events:

$$\text{SOM Ratio} = \frac{\text{Positive Cognitions}}{\text{Positive Cognitions} + \text{Negative Cognitions}}$$

Schwartz and Garamoni identified five theoretical states of mind along this continuum:

1. Positive Monologue (SOM > .90): Hyper-positive cognitive state, characteristic of hypomania, mania, or defensive denial, which can produce poor risk assessment.
2. Dialogue of Normal Functioning (.62 ≤ SOM ≤ .67): The optimal “golden ratio” of adaptive functioning, where positive thoughts predominate to maintain optimism, while a small degree of negative thinking is retained for realistic problem-solving and error detection.
3. Conflictive Dialogue (.50 ≤ SOM ≤ .61): Mild distress or clinical ambivalence where positive and negative cognitions contend.
4. Dialogue of Uncertainty (.38 ≤ SOM ≤ .49): Moderate psychopathology where negative thoughts begin to dominate.
5. Negative Monologue (SOM < .32): Severe depressive state where negative self-referent cognitions completely overwhelm positive dialogue.

By capturing both dimensions within a single questionnaire, the ATQ-R operationalizes this functional cognitive ratio, allowing empirical evaluation of both the depletion of positive thinking and the intrusion of negative ideation.

7. Validity

The psychometric validity of the ATQ and its revised form (ATQ-R) has been established through clinical trials, analogue student designs, and cross-cultural validations.

Construct and Convergent Validity

Construct validity is evidenced by significant, robust correlations with established clinician-rated and self-report measures of depression. In early validation studies by Hollon and Kendall (1980) and Harrell and Ryon (1983), the 30 negative items of the ATQ demonstrated strong positive correlations with the Beck Depression Inventory (typically $r = .70$ to $.85$) and the Minnesota Multiphasic Personality Inventory Depression Scale (MMPI-D; $r = .65$ to $.77$). When Kendall et al. (1989) evaluated the ATQ-R, the ATQ-N subscale retained these correlations ($r > .80$ with the BDI), while the ATQ-P subscale exhibited moderate-to-strong inverse correlations with the BDI ($r = -.55$ to $-.70$) and positive correlations with measures of positive affect, self-efficacy, and self-esteem (such as the Rosenberg Self-Esteem Scale, $r = .62$ to $.74$).

Discriminant Validity

The ATQ-R shows strong discriminant validity across different psychiatric diagnostic categories. In clinical comparisons (Hollon, Kendall, & Lumry, 1986), the negative items reliably differentiated unipolar depressed patients from non-depressed psychiatric controls (e.g., individuals with schizophrenia, generalized anxiety, or substance use disorders) and healthy controls. Depressed individuals exhibited significantly higher ATQ-N scores ($M > 95$) compared to psychiatric controls with non-depressive pathologies ($M \approx 55-65$) and normal controls ($M \approx 40-45$). Furthermore, while anxious patients occasionally score moderately on the ATQ-N due to general distress or secondary depressive symptoms, factor analytic investigations show that depressotypic automatic thoughts load on a factor distinct from anxious cognitions (such as those measured by the Anxious Self-Statements Questionnaire; ASSQ).

Predictive and Treatment Outcome Validity

The ATQ-R shows high predictive utility and sensitivity to therapeutic intervention. Numerous clinical outcome trials for major depressive disorder (MDD) demonstrate that successful pharmacotherapy and CBT lead to significant reductions in ATQ-N scores and increases in ATQ-P scores. Pre-to-post effect sizes frequently exceed Cohen’s $d = 1.20$. Crucially, research indicates that residual elevations in ATQ-N scores or chronically depressed ATQ-P scores following acute treatment predict subsequent relapse and recurrence, confirming the instrument’s clinical value for prognosis.

8. Reliability

The ATQ-R consistently demonstrates high reliability across diverse study cohorts, clinical settings, and cultural adaptations.

Internal Consistency

Internal consistency estimates for the ATQ-R are exceptionally high:

• ATQ-N (Negative Scale, 30 Items): Hollon and Kendall (1980) reported an initial Cronbach’s coefficient alpha of $.97$ in an undergraduate sample and $.96$ in a clinical validation sample (Harrell & Ryon, 1983). Kendall, Howard, and Hays (1989) replicated these findings in their revised sample, reporting $\alpha = .96$. Split-half reliability coefficients for this subscale routinely exceed $.95$.
• ATQ-P (Positive Scale, 10 Items): Internal consistency for the 10 positive items added in the revision is similarly high, with Kendall et al. (1989) reporting an alpha coefficient of $.91$. Subsequent evaluations across community, college, and clinical cohorts have yielded alpha values between $.86$ and $.93$, indicating uniform item-total correlations without redundancy.

Test-Retest Stability

Temporal stability metrics reflect both the trait-like cognitive vulnerability and state-dependent fluctuations of automatic cognitions:

• In stable, untreated cohorts measured over short intervals (e.g., 2 to 3 days), test-retest correlations remain high ($r = .85$ to $.90$).
• Across intermediate periods (e.g., 2 to 8 weeks) in non-clinical samples without active interventions, test-retest reliability stabilizes around $r = .65$ to $.80$, demonstrating reasonable cognitive stability.
• In clinical cohorts undergoing active psychotherapy, test-retest coefficients drop significantly ($r = .30$ to $.50$), confirming that the ATQ-R is sensitive to meaningful clinical change rather than assessing immutable, fixed personality traits.

9. Factor Analysis

Both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) have been conducted to evaluate the dimensionality of the ATQ-R.

Exploratory Factor Analysis (EFA)

In the original 30-item scale development by Hollon and Kendall (1980), a principal components analysis with varimax rotation yielded four distinct factors with eigenvalues greater than 1.0, accounting for 66.5% of the total variance:

• Factor 1: Personal Maladjustment and Desire for Change (items assessing personal deficiency, confusion, and the urge to escape or alter circumstances; e.g., Items 1, 9, 12, 18, 21, 22, 26, 27, 30, 35, 36, 40).
• Factor 2: Negative Self-Concept and Negative Expectations (items reflecting severe self-blame, failure, and unworthiness; e.g., Items 2, 4, 6, 11, 14, 19, 23, 25, 29, 31).
• Factor 3: Low Self-Esteem / Giving Up (items tapping exhaustion, cognitive paralysis, and an inability to continue; e.g., Items 8, 17, 39).
• Factor 4: Helplessness (items capturing external attribution of defeat and lack of agency; e.g., Items 5, 33, 34, 38).

When Kendall, Howard, and Hays (1989) examined the 40-item ATQ-R, principal axis factoring with oblique and orthogonal rotations revealed a clean, dominant two-factor higher-order structure: a comprehensive Negative Cognitions Factor (all 30 original items, with primary factor loadings typically ranging from .45 to .82) and a completely separate Positive Cognitions Factor (the 10 revised items, with factor loadings between .52 and .84), showing minimal cross-loadings (< .20).

Confirmatory Factor Analysis (CFA)

Modern structural equation modeling and CFA evaluations (e.g., Netemeyer et al., 2002; Burgess & Haaga, 1994) have tested several competing structural models: a unidimensional model, an uncorrelated two-factor model, a correlated two-factor model, and hierarchical models. The correlated two-factor model (Negative Automatic Thoughts and Positive Automatic Thoughts as distinct but moderately inversely correlated latent factors, $r \approx -.45$ to $-.60$) consistently demonstrates superior fit across clinical and community data compared to unidimensional alternatives. Standard goodness-of-fit indices commonly meet accepted psychometric benchmarks:

• Root Mean Square Error of Approximation (RMSEA) $le .058$
• Comparative Fit Index (CFI) $ge .93$
• Tucker-Lewis Index (TLI) $ge .92$
• Standardized Root Mean Square Residual (SRMR) $le .051$

10. Instrument / Measurement Tool

• Construct Assessed: Frequency of negative and positive self-referent automatic thoughts (cognitive depressive manifestations and positive coping self-statements).
• Item Count: 40 items total (30 negative items, 10 positive items).
• Administration Format: Paper-and-pencil questionnaire, digitized computer survey, or clinician-facilitated interview.
• Target Population: Adults and adolescents aged 14 and older; parallel child versions (Kazdin, 1990) exist for younger populations.
• Administration Time: Approximately 5 to 10 minutes.
• Response Scale: 5-point Likert scale assessing the frequency of each thought over the past week:
  • 1 = “not at all”
  • 2 = “somewhat”
  • 3 = “moderately”
  • 4 = “very much”
  • 5 = “totally”
• Subscale Item Composition:
  • Negative Automatic Thoughts (ATQ-N): 30 items (Items 1, 2, 4, 5, 6, 8, 9, 11, 12, 14, 15, 17, 18, 19, 21, 22, 23, 25, 26, 27, 29, 30, 31, 33, 34, 35, 36, 38, 39, 40).
  • Positive Automatic Thoughts (ATQ-P): 10 items (Items 3, 7, 10, 13, 16, 20, 24, 28, 32, 37).
• Scoring and Computational Rules:
  • ATQ-N Total Score: Sum of the 30 negative items. Theoretical range: 30 to 150. Higher scores indicate more frequent depressogenic automatic cognitions.
  • ATQ-P Total Score: Sum of the 10 positive items. Theoretical range: 10 to 50. Higher scores indicate more frequent adaptive, positive self-statements.
  • States of Mind (SOM) Ratio: Computed as:

    $$\text{SOM} = \frac{\text{ATQ-P}}{\text{ATQ-P} + \text{ATQ-N}}$$

    Alternatively, standardized frequency transformations or raw sum transformations can be used as outlined by Schwartz and Garamoni (1986).

• Clinical Interpretation Guidelines:
  • ATQ-N < 50: Non-depressed / normal range of negative automatic cognitions.
  • ATQ-N 50–70: Mild depressive cognitive symptoms / elevated negative self-statements.
  • ATQ-N 71–90: Moderate depressive cognitive intrusions.
  • ATQ-N > 90: Severe, pervasive depressogenic thinking typical of Major Depressive Disorder.
  • ATQ-P > 35: High positive cognitive balance and resilience.
  • ATQ-P < 20: Significant depletion of adaptive thoughts (positive cognitive deficit).

11. Permissions & Fee and Test Year

• Publication History: The original 30-item ATQ was developed and published in 1980 by Steven D. Hollon and Philip C. Kendall. The 40-item revised edition (ATQ-R) was published in 1989 by Philip C. Kendall, Bruce L. Howard, and Robert C. Hays.
• Copyright & Intellectual Property: The copyright is held by the original authors and the respective publishers (Springer Science+Business Media / Plenum Publishing Corporation, which publishes Cognitive Therapy and Research).
• Licensing and Clinical/Research Permissions: The scale is widely considered an open-access academic measure for non-commercial research, graduate training, and clinical practice, provided the original validation papers are cited correctly. Commercial use, reproduction in published workbooks, or digital diagnostic platforms generally requires formal permission from the copyright holders or authors.
• Fees: Free of charge for academic, non-commercial clinical, and scientific research purposes.

12. References

Beck, A. T. (1967). Depression: Clinical, experimental, and theoretical aspects. Harper & Row.

Beck, A. T. (1976). Cognitive therapy and the emotional disorders. International Universities Press.

Beck, A. T., Brown, G., Steer, R. A., Eidelson, J. I., & Riskind, J. H. (1987). Differentiating anxiety and depression: A test of the cognitive content-specificity hypothesis. Journal of Abnormal Psychology, 96(3), 179–183. https://doi.org/10.1037/0021-843X.96.3.179

Burgess, E., & Haaga, D. A. F. (1994). The Positive Automatic Thoughts Questionnaire-Revised: Equivalent measures of positive thinking? Cognitive Therapy and Research, 18(1), 15–24. https://doi.org/10.1007/BF02359393

Harrell, T. H., & Ryon, N. B. (1983). Cognitive-behavioral assessment of depression: Clinical validation of the Automatic Thoughts Questionnaire. Journal of Consulting and Clinical Psychology, 51(5), 721–725. https://doi.org/10.1037/0022-006X.51.5.721

Hollon, S. D., & Kendall, P. C. (1980). Cognitive self-statements in depression: Development of an automatic thoughts questionnaire. Cognitive Therapy and Research, 4(4), 383–395. https://doi.org/10.1007/BF01173648

Hollon, S. D., Kendall, P. C., & Lumry, A. (1986). Specificity of depressotypic cognitions in clinical depression. Journal of Abnormal Psychology, 95(1), 52–59. https://doi.org/10.1037/0021-843X.95.1.52

Ingram, R. E., Kendall, P. C., Siegle, G., Guarino, J., & McLaughlin, S. C. (1995). Psychometric properties of the Positive Automatic Thoughts Questionnaire. Psychological Assessment, 7(4), 495–507. https://doi.org/10.1037/1040-3590.7.4.495

Kazdin, A. E. (1990). Evaluation of the Automatic Thoughts Questionnaire: Negative cognitive processes and depression among children. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 2(1), 73–77. https://doi.org/10.1037/1040-3590.2.1.73

Kendall, P. C., Howard, B. L., & Hays, R. C. (1989). Self-referent speech and psychopathology: The balance of positive and negative thinking. Cognitive Therapy and Research, 13(6), 583–598. https://doi.org/10.1007/BF01176046

Netemeyer, R. G., Williamson, D. A., Burton, S., Biswas, D., Jindal, S., Linda, S., & Bentz, B. G. (2002). Psychometric properties of shortened versions of the Automatic Thoughts Questionnaire. Educational and Psychological Measurement, 62(1), 111–129. https://doi.org/10.1177/0013164402062001008

Schwartz, R. M., & Garamoni, G. L. (1986). A structural model of cognitive balance: The States of Mind Model. Advances in Cognitive-Behavioral Research and Therapy, 5, 267–326. https://doi.org/10.1016/B978-0-12-010605-9.50011-8

Schwartz, R. M., & Garamoni, G. L. (1989). Cognitive balance and psychopathology: Evaluation of an information processing model of client content. Clinical Psychology Review, 9(3), 271–294. https://doi.org/10.1016/0272-7358(89)90058-4

13. Items of the Scale