1. Abstract
The Cognitive Failures Questionnaire (CFQ-B), initially developed by Donald E. Broadbent and colleagues in 1982, represents one of the most widely utilized self-report instruments designed to assess the frequency with which individuals experience everyday cognitive slips, lapses, and execution errors across the domains of memory, perception, and motor functioning. Rooted in applied cognitive psychology and human error theory, the instrument captures failures in tasks that an individual is normally capable of executing without difficulty. The standard version of the instrument comprises 25 items rated on a five-point Likert scale ranging from 0 (Never) to 4 (Very often), yielding a total composite score ranging from 0 to 100.
Although Broadbent and collaborators conceptualized the scale as measuring a unidimensional, trait-like vulnerability to mental lapses, subsequent psychometric evaluations have demonstrated multidimensional structures. Prominent structural models frequently identify three to four underlying dimensions, most notably Distractibility (Attentional Failures), Forgetfulness (Memory Lapses), Blunders (Action Execution Slips), and Spatial/Kinesthetic Disorientation. Psychometric investigations across diverse clinical, non-clinical, and occupational populations consistently report robust internal consistency, with Cronbach’s alpha coefficients typically falling between .84 and .92, and test-retest reliability estimates demonstrating temporal stability ranging from .80 to .86 across multi-month intervals.
Elevated scores on the CFQ-B have documented associations with subjective cognitive complaints, high levels of occupational stress, chronic sleep deprivation, burnout, and elevated neuroticism. Furthermore, the questionnaire serves as an indispensable tool in ergonomic investigations, industrial safety assessments, neurological monitoring (e.g., following traumatic brain injury or in early-stage neurodegenerative diseases), and pharmacological trials evaluating cognitive side effects. This comprehensive psychometric review details the theoretical background, construct taxonomy, structural factor analytic debates, validity evidence, and administrative parameters of the CFQ-B.
2. Keywords
Cognitive Failures Questionnaire, CFQ, Broadbent, everyday cognitive slips, absentmindedness, perceptual failures, action slips, human error, attention lapses, psychometrics, working memory failures, executive functioning, occupational safety, self-report cognitive assessment
3. Authors
The Cognitive Failures Questionnaire was originally conceptualized, developed, and validated by a distinguished multidisciplinary team of cognitive psychologists and occupational health researchers at the Medical Research Council (MRC) Applied Psychology Unit in Cambridge and the University of Oxford:
- Donald E. Broadbent, C.B.E., F.R.S. (1926–1993): A seminal British experimental psychologist renowned for his pioneering work in information processing theory, selective attention, and human-machine interaction. Affiliated with the Department of Experimental Psychology, University of Oxford, Oxford, United Kingdom.
- Peter F. Cooper: Research psychologist affiliated with the Department of Experimental Psychology, University of Oxford, Oxford, United Kingdom.
- P. FitzGerald: Applied psychometrician and research associate affiliated with the Department of Experimental Psychology, University of Oxford, Oxford, United Kingdom.
- Katherine R. Parkes, Ph.D.: Prominent occupational and health psychologist specializing in organizational stress, shift work, and individual vulnerability factors, affiliated with the Department of Experimental Psychology, University of Oxford, Oxford, United Kingdom.
4. Purpose
The primary purpose of the Cognitive Failures Questionnaire is to provide an ecologically valid, self-administered psychometric index of an individual’s propensity to commit minor cognitive errors during the execution of routine daily activities. While traditional laboratory-based neurocognitive assessments (such as the Stroop task, the Trail Making Test, or span tasks) isolate discrete components of executive function under structured, distraction-free conditions, they frequently demonstrate weak ecological correlations with the chaotic demands of everyday real-world performance. The CFQ was designed explicitly to bridge this methodological gap by quantifying failures that manifest when sustained attention, memory retrieval, and motor execution are challenged by real-world environments.
In clinical neuropsychology and psychiatry, the CFQ-B is deployed to evaluate self-perceived cognitive impairment in clinical disorders where objective neurological deficits may be subtle or fluctuating. It is widely utilized in evaluations of attention-deficit/hyperactivity disorder (ADHD) in adults, major depressive disorder, generalized anxiety disorder, mild cognitive impairment (MCI), chronic fatigue syndrome, fibromyalgia, and recovery trajectories following minor or moderate traumatic brain injury (TBI). In these clinical populations, the scale serves as a standardized baseline metric and an outcome measure tracking rehabilitation progress or pharmacological adverse events.
Within organizational psychology, occupational ergonomics, and human factors engineering, the CFQ-B functions as an essential predictive instrument for safety-critical settings. Researchers and risk managers utilize the tool to identify personnel vulnerable to cognitive overload, situational distraction, and procedural lapses. Elevated CFQ scores consistently predict occupational injuries, industrial machinery accidents, vehicle collisions, and clinical charting errors among healthcare practitioners. Furthermore, experimental psychologists routinely employ the CFQ-B as a trait covariate to control for baseline differences in attentional control, working memory capacity, and absentmindedness in cognitive load experiments.
5. Psychological Construct
The psychological construct measured by the CFQ-B is defined as cognitive failure: an error in the execution of an intentional, goal-directed behavioral sequence on a task that the person is normally capable of executing successfully, where the failure cannot be attributed to physical disability, general intellectual subnormality, or organic sensory loss. Rather than reflecting intellectual capability, cognitive failure reflects momentary lapses in cognitive control, lapses of sustained attention, retrieval blocks, or motor schema interference.
Although originally conceptualized by Broadbent and colleagues as a single global construct, psychometric decomposition delineates four interrelated yet distinct operational dimensions:
Attentional and Perceptual Failures (Distractibility)
This subconstruct encompasses breakdowns in the sensory gating and selective filtering mechanisms responsible for maintaining task focus in the presence of competing external or internal stimuli. It manifests when an individual fails to notice salient environmental cues or loses the thread of an ongoing sensory input stream. Typical manifestations include failing to notice street signs, missing conversational points because of wandering thoughts, or looking directly at an object without consciously perceiving its presence. At an operational level, this dimension captures vulnerabilities in early-stage attentional capture and top-down executive filtering.
Memory and Information Retention Failures (Forgetfulness)
This dimension pertains to lapses in both retrospective memory (retrieving previously acquired information) and prospective memory (remembering to execute intentions at an appropriate future juncture). It encompasses working memory capacity constraints and retrieval failures under baseline cognitive load. Concrete behavioral exemplars include walking into a room and forgetting the intended objective, leaving important tasks unattended, misplacing household objects (e.g., keys, glasses), or being unable to retrieve familiar names immediately following an introduction.
Action Execution Failures (Blunders and Motor Slips)
This subconstruct reflects disruptions in procedural motor execution and the monitoring of automated action schemata. When routine behavioral routines are executed without continuous conscious supervisory control, erroneous motor programs may capture behavior. Manifestations include dropping items, accidentally bumping into people or physical obstacles, turning dials or steering wheels in the wrong direction, or executing an unintended motor sequence (e.g., pouring orange juice into a cereal bowl). This dimension directly reflects breakdowns in supervisory motor monitoring mechanisms.
Spatial and Orientation Failures
Representing a more specialized subset of cognitive slips, this facet captures disorientations within familiar spatial environments and directional processing errors. Exemplars include confusing left and right orientations when giving or receiving navigation instructions, or momentarily losing one’s bearings along routes traversed daily. It reflects transient lapses in the internal manipulation of mental maps and spatial coordinate transformation.
6. Theoretical Framework
The conceptual infrastructure of the Cognitive Failures Questionnaire rests primarily on the convergence of three foundational paradigms within cognitive science: Broadbent’s filter model of attention, James Reason’s human error taxonomy, and Donald Norman’s schema activation theory.
Donald Broadbent’s seminal Filter Model of Attention (1958) posited that the human nervous system functions as an information channel of limited capacity. In Broadbent’s framework, sensory inputs are held transiently in a sensory buffer before passing through an attentional filter that selects relevant signals based on physical characteristics, routing them to a limited-capacity processing channel. Under elevated stress, heightened arousal, or systemic fatigue, the efficiency of this filtering mechanism degrades. Cognitive failures occur when the selective filter fails to block irrelevant noise or permits transient interruptions of the primary signal channel, resulting in perceptual dropouts and task disorganization.
Complementing Broadbent’s model is James Reason’s (1984, 1990) taxonomy of human error, which draws a crucial distinction between slips, lapses, and mistakes:
- Slips: Attentional failures in which the internal plan or intention was correct, but the execution of the motor action sequence deviates from intended pathways (e.g., speech slips, motor blunders).
- Lapses: Memory failures involving internal storage or retrieval breakdowns that prevent an intended action from being initiated or completed (e.g., forgetting a scheduled meeting or losing track of an ongoing train of thought).
- Mistakes: Deficiencies in judgment, reasoning, or rule application where the plan itself is flawed, regardless of whether execution is smooth.
The CFQ-B explicitly confines its measurement scope to slips and lapses. Reason conceptualized absentminded slips as byproducts of the evolutionary optimization of human cognition: highly practiced skills become automated to conserve finite conscious attentional resources. When supervisory attention is distracted by internal preoccupations (e.g., worrying, day-dreaming) or sensory overload, the automated control loops run without error-detection oversight, culminating in action slips.
This automated-monitoring dynamic is formalized mathematically and mechanistically in Donald Norman’s (1981) Activation-Trigger-Schema (ATS) theory. Norman conceptualized human action as governed by hierarchical collections of cognitive structures termed schemata. A schema requires both sufficient activation (determined by intention and context) and an appropriate triggering condition (environmental cues) to initiate. When an individual operates under reduced attentional control, an unintended schema that shares identical trigger conditions with the intended schema can be erroneously activated—a phenomenon termed capture slips. The items comprising the CFQ-B serve as direct behavioral markers of these capture phenomena occurring within domestic, professional, and navigation spaces.
7. Validity
The psychometric validity of the CFQ-B has been substantiated across more than four decades of empirical investigation spanning experimental, clinical, and industrial contexts.
Construct and Convergent Validity
Construct validity is evidenced through robust correlations with established psychometric scales measuring affective distress, attentional control, and vulnerability to stress. Numerous studies have documented moderate to strong positive correlations between CFQ scores and measures of neuroticism on the Eysenck Personality Questionnaire (EPQ) and the NEO Personality Inventory (NEO-PI-R), with correlation coefficients consistently ranging from r = .30 to r = .48. Similarly, the CFQ exhibits substantial convergent validity with measures of trait anxiety (State-Trait Anxiety Inventory, r = .35 to .52) and perceived chronic stress (Perceived Stress Scale, r = .40 to .55).
Crucially, the CFQ correlates strongly with other self-report indices of absentmindedness and executive failure, such as the Attention-Related Cognitive Errors Scale (ARCES; r = .65 to .75) and the Everyday Memory Questionnaire (EMQ; r = .68 to .80). In neuroimaging investigations, elevated CFQ scores have demonstrated convergent associations with reduced structural volume in the prefrontal cortex and altered functional connectivity within the default mode network (DMN), supporting the construct’s biological substrate in frontoparietal attentional networks.
Criterion and Predictive Validity
The ecological and predictive validity of the CFQ is demonstrated across real-world safety domains. In a landmark occupational safety investigation by Larson et al., industrial personnel with high CFQ scores exhibited a significantly elevated incidence of workplace injuries, machinery mishaps, and near-miss occurrences over a 24-month prospective tracking window (odds ratios ranging from 1.6 to 2.4). In automotive psychology, high CFQ scores prospectively predict self-reported and state-recorded traffic violations, moving infractions, and motor vehicle collisions.
Laboratory-based predictive validity studies have demonstrated that participants scoring in the upper quartile of the CFQ show significantly elevated rates of vigilance decrements during continuous performance tests (CPT), elevated false-alarm rates on the Sustained Attention to Response Task (SART), and pronounced susceptibility to mind-wandering during reading comprehension trials.
Discriminant Validity
A central theoretical triumph of the CFQ-B is its demonstrable discriminant validity from psychometric general intelligence (Spearman’s g or IQ). Broadbent et al. (1982) demonstrated that CFQ scores do not correlate significantly with performance on standardized intelligence metrics, such as the Raven’s Progressive Matrices or the Wechsler Adult Intelligence Scale (WAIS) subtests (correlations typically hover between r = -.08 and r = .05, non-significant). This confirms that the CFQ does not evaluate intellectual capacity or declarative knowledge, but rather the operational control and execution stability of cognitive processes.
8. Reliability
The reliability of the CFQ-B has been documented across normative, clinical, and occupational cohorts worldwide, demonstrating high internal consistency and exceptional temporal stability.
Internal Consistency
In the original validation study by Broadbent et al. (1982) across occupational cohorts, the 25-item instrument demonstrated an overall internal consistency coefficient of Cronbach’s alpha = .89. Subsequent psychometric evaluations involving large, representative community samples have consistently replicated these findings:
- Wallace, Kass, and Stanny (2002) observed a Cronbach’s alpha of .91 in a sample of 458 collegiate and adult participants.
- Bridger, Johnsen, and Brasher (2013) reported alpha coefficients ranging from .88 to .92 across long-term cohorts of maritime and military personnel.
- Rast, Zimprich, Hertzog, and Hofer (2009) evaluated older adult samples (ages 60 to 91), reporting McDonald’s total omega (ω) of .90, confirming that item-scale homogeneity does not deteriorate across the aging spectrum.
Test-Retest Reliability and Temporal Stability
The temporal stability of the CFQ confirms its status as a trait-like measure of cognitive vulnerability rather than a transient state assessment. In their foundational psychometric study, Broadbent et al. (1982) administered the scale across a 21-week (approximately 5-month) interval and observed a test-retest reliability coefficient of r = .82. Further independent investigations extending over intervals ranging from 6 months to 2 years have established coefficients between r = .71 and r = .85.
Inter-rater reliability has also been confirmed through proxy-reporting paradigms. Corroborative ratings completed by spouses, partners, or work supervisors correlate moderately to strongly (r = .55 to .65) with self-reported CFQ scores, establishing that self-perceived cognitive slips correspond reliably to observable behavioral lapses visible to external observers.
9. Factor Analysis
The latent dimensionality of the Cognitive Failures Questionnaire has been the subject of extensive psychometric investigation. While Broadbent et al. (1982) originally proposed a single-factor, unidimensional structure—arguing that a singular trait vulnerability underpins diverse cognitive slips—subsequent exploratory (EFA) and confirmatory (CFA) factor analyses have elucidated nuanced multidimensional models.
The Wallace, Kass, and Stanny (2002) 4-Factor Model
One of the most widely accepted confirmatory factor structures was established by Wallace et al. (2002), who utilized both EFA and CFA on independent samples to isolate a four-factor solution comprising 21 of the original 25 items:
- Factor 1: Memory (8 items) – Captures lapses in prospective memory, retrieval of familiar semantic details, and working memory maintenance (e.g., forgetting appointments, forgetting names, misplacing objects). Standardized factor loadings range from .48 to .76.
- Factor 2: Distractibility (9 items) – Encompasses attentional capture by environmental distractions and failures in perceptual selectivity (e.g., failing to notice road signs, missing spoken words while engaged in tasks). Standardized factor loadings range from .45 to .72.
- Factor 3: Blunders (7 items) – Reflects gross action errors, physical miscoordinations, and social slips (e.g., bumping into people, dropping items, saying inappropriate things). Standardized factor loadings range from .42 to .68.
- Factor 4: Spatial Orientation (2 items) – Pertains to confusion regarding directional vectors and navigating familiar spatial layouts (e.g., confusing left and right, getting lost in known areas). Factor loadings exceed .65.
The Rast et al. (2009) Bifactor Structural Formulation
To reconcile the conflict between Broadbent’s unidimensional composite score and the empirically demonstrated multidimensional subscales, Rast, Zimprich, Hertzog, and Hofer (2009) performed advanced structural equation modeling on large cross-sectional cohorts. Their analysis demonstrated that a bifactor model provides superior fit compared to either an uncorrelated orthogonal model or a strict second-order factor model:
- Comparative Fit Index (CFI): .952
- Tucker-Lewis Index (TLI): .944
- Root Mean Square Error of Approximation (RMSEA): .041 (90% CI [.036, .046])
- Standardized Root Mean Square Residual (SRMR): .038
In this bifactor architecture, a robust general factor accounting for approximately 65% to 75% of the common variance runs across all 25 items, justifying the computation of a composite total score for screening purposes. Concurrently, distinct group factors (Attention, Memory, Motor Execution) capture residual variance, validating the clinical and diagnostic utility of profile subscale analyses.
10. Instrument / Measurement Tool
The parameters, administration requirements, and scoring algorithms of the Cognitive Failures Questionnaire are structured as follows:
- Instrument Designation: Cognitive Failures Questionnaire (CFQ; also referred to as CFQ-B in distinction to expanded or child adaptations).
- Primary Author: Donald E. Broadbent (1982).
- Assessment Type: Self-administered psychometric rating scale (also adapted for proxy/informant assessment).
- Target Population: Adolescents and adults (ages 16 to 90+ years); suitable for general population, occupational, and neuroclinical groups.
- Administration Modality: Paper-and-pencil, digital computer-based, or web-based survey interfaces.
- Administration Time: Approximately 5 to 8 minutes.
- Item Inventory: 25 standardized items formulating everyday failure scenarios.
- Response Format: 5-point Likert-type frequency scale scored as follows:
- 0 = Never
- 1 = Very rarely
- 2 = Occasionally
- 3 = Quite often
- 4 = Very often
- Scoring and Quantification Procedures:
- Composite Total Score: Calculated by summing the numerical ratings across all 25 items. Total scores range from 0 to 100. Higher scores reflect greater frequency of everyday cognitive slips and absentmindedness.
- Alternative Scoring (1 to 5): Some researchers employ a 1-to-5 scale yielding scores between 25 and 125. Standard Broadbent scoring uses the 0-to-4 metric.
- Subscale Computations: When employing multidimensional frameworks (e.g., Wallace et al., 2002), specific item subsets are summed to yield distinct indices for Memory Failures, Distractibility, Motor Blunders, and Spatial Disorientation.
- Handling Missing Data: If 2 or fewer items are missed, mean item substitution is statistically acceptable; protocols missing >2 items should be rendered invalid.
11. Permissions & Fee and Test Year
The Cognitive Failures Questionnaire was officially published in 1982 in the British Journal of Psychology by Donald E. Broadbent, Peter F. Cooper, P. FitzGerald, and Katherine R. Parkes. The underlying peer-reviewed journal article and its original item presentation fall under the copyright of the British Psychological Society (BPS) and its distribution partner, John Wiley & Sons, Ltd.
In accordance with established academic conventions, the CFQ-B is considered an open-access psychometric instrument for non-commercial academic research, pedagogical use, and independent clinical research, provided that full bibliographic citation is accorded to Broadbent et al. (1982). No licensing royalties or authorization fees are levied on individual academic researchers utilizing the standard questionnaire. However, for commercial deployments, incorporation into proprietary digital health applications, commercial clinical trials, or fee-for-service occupational screenings, researchers and organizations must obtain formal licensing permission through the copyright clearance mechanisms of the British Psychological Society / Wiley Periodicals LLC.
12. References
The following peer-reviewed literature establishes the psychometric and theoretical foundations of the Cognitive Failures Questionnaire:
- Broadbent, D. E., Cooper, P. F., FitzGerald, P., & Parkes, K. R. (1982). The Cognitive Failures Questionnaire (CFQ) and its correlates. British Journal of Psychology, 73(2), 353–370. https://doi.org/10.1111/j.2044-8295.1982.tb01817.x
- Bridger, R. S., Johnsen, S. Æ., & Brasher, K. (2013). Psychometric properties of the Cognitive Failures Questionnaire in safety-critical work. Theoretical Issues in Ergonomics Science, 14(1), 81–94. https://doi.org/10.1080/1463922X.2011.583647
- Matthews, G., Coyle, K., & Craig, A. (1990). Multiple factors of cognitive failure and their relationships with stress vulnerability. Journal of Psychopathology and Behavioral Assessment, 12(1), 49–65. https://doi.org/10.1007/BF00960459
- Merckelbach, H., Muris, P., Nijman, H., & de Jong, P. J. (1996). Self-reported cognitive failures and neuroticism, somatic complaints, and hallucinations. Personality and Individual Differences, 20(3), 407–410. https://doi.org/10.1016/0191-8869(95)00185-0
- Norman, D. A. (1981). Categorization of action slips. Psychological Review, 88(1), 1–15. https://doi.org/10.1037/0033-295X.88.1.1
- Rast, P., Zimprich, D., Hertzog, C., & Hofer, S. M. (2009). The structure of everyday cognitive failures in later adulthood: A bifactor model. European Journal of Psychological Assessment, 25(2), 110–120. https://doi.org/10.1027/1015-5759.25.2.110
- Reason, J. (1984). Lapses of attention in everyday life. In R. Parasuraman & D. R. Davies (Eds.), Varieties of Attention (pp. 515–549). Academic Press.
- Reason, J. (1990). Human Error. Cambridge University Press. https://doi.org/10.1017/CBO9781139062366
- Wallace, J. C., Kass, S. J., & Stanny, C. J. (2002). The Cognitive Failures Questionnaire revisited: Investigating the relationship between cognitive failures and accident involvement. Safety Science, 40(7-8), 679–688. https://doi.org/10.1016/S0925-7535(01)00065-9
- Wilhelm, O., Witthöft, M., & Schipolowski, S. (2010). Self-reported cognitive failures: Competence or personality? Journal of Individual Differences, 31(4), 178–191. https://doi.org/10.1027/1614-0001/a000028