The CFI measures the kind of cognitive flexibility needed to challenge and replace maladaptive thoughts: the tendency to see difficult situations as controllable and the ability to generate alternative explanations and solutions.
Dennis and Vander Wal developed it as a brief self-report complement to performance-based flexibility tests, aimed at the thinking skills targeted in cognitive-behavioral therapy; factor analysis yielded two subscales, Alternatives and Control.
In the development research, higher scores accompanied lower depressive symptoms and more adaptive, less avoidant coping, and both subscales showed strong internal consistency and stability over a follow-up interval.
Validation rests mainly on U.S. undergraduate samples; the CFI captures perceived rather than performance-based flexibility, no clinical cutoffs exist, and later work in older clinical samples cautions against relying on the total score there.
At a Glance
Items
20 (13 Alternatives, 7 Control; six items reverse-scored)
Administration time
Approximately 5-7 minutes (Dennis & Vander Wal, 2010)
Response format
7-point Likert scale, 1 = strongly disagree to 7 = strongly agree, all seven points labeled
Subscales
Alternatives and Control, plus a total score
Validated populations
Adults; developed with U.S. undergraduates (Dennis & Vander Wal, 2010); separately evaluated in clinical and nonclinical older adults (Johnco, Wuthrich, & Rapee, 2014)
License
Free for research use with attribution; copyright rests with the authors (item set published as an appendix of the development article)
Original citation
Dennis & Vander Wal (2010), Cognitive Therapy and Research
Introduction
The Cognitive Flexibility Inventory (CFI) is a 20-item self-report measure of the type of cognitive flexibility necessary to successfully challenge and replace maladaptive thoughts with more balanced and adaptive thinking. Developed by Dennis and Vander Wal (2010), it was created to measure cognitive flexibility as it matters for cognitive-behavioral therapy (CBT) and for adaptive coping with life stressors, and it was designed to be brief enough for repeated administration.
The CFI occupies a distinct position among cognitive flexibility measures. Unlike the Cognitive Flexibility Scale, which grew out of the communication-competence literature (Martin & Rubin, 1995), or performance-based neuropsychological tests such as the Wisconsin Card Sorting Test, the CFI assesses how people appraise and respond to difficult situations in everyday life (Dennis & Vander Wal, 2010). It yields a total score and two factor-analytically derived subscale scores, Alternatives and Control.
Understanding Cognitive Flexibility for Adaptive Coping
A core premise of cognitive-behavioral therapy is that depression and related difficulties are treated most effectively by breaking down automatic maladaptive cognitions and replacing them with more realistic, adaptive ones (Young, Weinberger, & Beck, 2001). Dennis and Vander Wal (2010) argue that the mechanism through which CBT achieves this change involves increasing cognitive flexibility, the ability to adjust one’s thinking in response to changing situational demands. Theoretical accounts of cognitive therapy likewise propose that the rigid, all-or-nothing thinking styles characteristic of depression help maintain the depressed state, so that increasing flexibility may reduce depressive symptoms (Moore, 1996; Teasdale, Segal, & Williams, 1995).
Consistent with this framing, the development research found that higher CFI scores were associated with lower depressive symptoms and with more adaptive, less avoidant coping (Dennis & Vander Wal, 2010).
Theoretical Foundation
Dennis and Vander Wal (2010) originally hypothesized that three aspects of cognitive flexibility are necessary for proficiency with cognitive-behavioral thought-challenging techniques: (a) the tendency to perceive difficult situations as controllable, (b) the ability to perceive multiple alternative explanations for life occurrences and human behavior, and (c) the ability to generate multiple alternative solutions to difficult situations. Exploratory factor analysis showed that aspects (b) and (c) form a single construct, yielding the CFI’s two-factor structure (Dennis & Vander Wal, 2010):
Alternatives (13 items) – the ability to perceive multiple alternative explanations for life events and human behavior, and to generate multiple alternative solutions to difficult situations
Control (7 items) – the tendency to perceive difficult situations as controllable and confidence in one’s ability to overcome difficulties
The two subscales are moderately correlated and show distinct patterns of association with coping strategies and depressive symptoms, supporting their treatment as related but separable dimensions (Dennis & Vander Wal, 2010).
🧠Key insight: The CFI operationalizes the kind of flexibility CBT is theorized to build: on this account, loosening rigid appraisals and generating alternative explanations and solutions is a central mechanism of therapeutic change in depression (Dennis & Vander Wal, 2010; Moore, 1996; Teasdale et al., 1995).
Key Features
Assessment Characteristics
20 self-report statements rated for agreement on a fully labeled 7-point scale
Two subscales, Alternatives (13 items) and Control (7 items), interleaved in the published form; six items are reverse-scored
Approximately 5-7 minutes to complete (Dennis & Vander Wal, 2010)
Developed with adult undergraduate samples; designed for repeated administration, for example across a course of therapy
Free for research use with attribution to the development article
Dimensions Assessed
Alternatives – perceiving multiple explanations for events and behavior, considering situations from several perspectives, and generating multiple solutions to problems
Control – perceiving difficult situations as controllable and maintaining confidence in one’s capacity to overcome difficulties
Versions & Adaptations
Original 20-item CFI (Dennis & Vander Wal, 2010) – the only published English form; no short form or revised version was identified during verification
Independent psychometric evaluation in clinical and nonclinical older-adult samples (Johnco, Wuthrich, & Rapee, 2014)
Validated translations exist in several languages, including Turkish, Russian, and Italian adaptations; validation quality varies by language and sample
Research Applications
Clinical psychology – CBT process research, therapy outcome assessment, and study of cognitive mechanisms of change
Stress and coping research – examining flexibility as a mediator or moderator of stress-distress relationships
Depression research – studying cognitive vulnerability and evaluating prevention programs
Health psychology and counseling – assessing flexible thinking in coping-skills and stress-management interventions
Assess your cognitive flexibility for adaptive coping across the Alternatives and Control dimensions.
Scoring and Interpretation
Response Format
Respondents are instructed: “Please use the scale below to indicate the extent to which you agree or disagree with the following statements” (Dennis & Vander Wal, 2010, Appendix 2). Each statement is rated on a 7-point Likert scale: 1 = strongly disagree, 2 = disagree, 3 = somewhat disagree, 4 = neutral, 5 = somewhat agree, 6 = agree, 7 = strongly agree.
CFI Items
The 20 items are shown below in the published administration order (Dennis & Vander Wal, 2010, Appendix 2, p. 252). Each item’s subscale is shown in brackets; items marked (R) are reverse-scored. In the standard instrument the two subscales are interleaved: the Alternatives subscale comprises items 1, 3, 5, 6, 8, 10, 12, 13, 14, 16, 18, 19, and 20, and the Control subscale comprises items 2, 4, 7, 9, 11, 15, and 17.
I am good at “sizing up” situations. [Alternatives]
I have a hard time making decisions when faced with difficult situations. [Control](R)
I consider multiple options before making a decision. [Alternatives]
When I encounter difficult situations, I feel like I am losing control. [Control](R)
I like to look at difficult situations from many different angles. [Alternatives]
I seek additional information not immediately available before attributing causes to behavior. [Alternatives]
When encountering difficult situations, I become so stressed that I can not think of a way to resolve the situation. [Control](R)
I try to think about things from another person’s point of view. [Alternatives]
I find it troublesome that there are so many different ways to deal with difficult situations. [Control](R)
I am good at putting myself in others’ shoes. [Alternatives]
When I encounter difficult situations, I just don’t know what to do. [Control](R)
It is important to look at difficult situations from many angles. [Alternatives]
When in difficult situations, I consider multiple options before deciding how to behave. [Alternatives]
I often look at a situation from different viewpoints. [Alternatives]
I am capable of overcoming the difficulties in life that I face. [Control]
I consider all the available facts and information when attributing causes to behavior. [Alternatives]
I feel I have no power to change things in difficult situations. [Control](R)
When I encounter difficult situations, I stop and try to think of several ways to resolve it. [Alternatives]
I can think of more than one way to resolve a difficult situation I’m confronted with. [Alternatives]
I consider multiple options before responding to difficult situations. [Alternatives]
Note: Items marked (R) are reverse-scored. These are items 2, 4, 7, 9, 11, and 17 (six items). Item 15 loads on the Control subscale but is not reverse-scored (Dennis & Vander Wal, 2010, Appendix 2).
Scoring Procedure
Reverse score the six items marked (R), which are items 2, 4, 7, 9, 11, and 17: reverse score = 8 – original score.
Alternatives subscale = sum of the 13 Alternatives items (items 1, 3, 5, 6, 8, 10, 12, 13, 14, 16, 18, 19, 20; range 13-91).
Control subscale = sum of the 7 Control items (items 2, 4, 7, 9, 11, 15, 17; range 7-49).
Total CFI score = sum of all 20 items (range 20-140). Higher scores indicate greater cognitive flexibility.
Interpretation and Sample Means
No published cutoff scores or interpretive bands exist for the CFI; the development article establishes none, and none are offered here. The means below are descriptive values from the development samples, not normative standards (Dennis & Vander Wal, 2010, p. 247):
Sample
N
Score
M
SD
Source
Undergraduates, Time 1
196
CFI total
102.98
13.91
Dennis & Vander Wal (2010)
Undergraduates, Time 1
196
Alternatives
67.59
9.41
Dennis & Vander Wal (2010)
Undergraduates, Time 1
196
Control
35.36
7.02
Dennis & Vander Wal (2010)
Undergraduates, Time 2 (7 weeks later)
152
CFI total
105.38
13.84
Dennis & Vander Wal (2010)
Undergraduates, Time 2 (7 weeks later)
152
Alternatives
69.41
9.40
Dennis & Vander Wal (2010)
Undergraduates, Time 2 (7 weeks later)
152
Control
35.92
6.77
Dennis & Vander Wal (2010)
In the development research, higher CFI scores were associated with more adaptive coping (problem-focused coping, seeking social support, focusing on the positive) and lower depressive symptoms, whereas lower scores were associated with maladaptive coping (wishful thinking, detachment, keeping to oneself) and higher depressive symptoms (Dennis & Vander Wal, 2010; exact coefficients in the Research Evidence section below).
Subscale-specific patterns (interpretive guidance only, not published norms or validated clinical profiles): a respondent may score high on Alternatives but low on Control, suggesting they can see multiple perspectives yet feel unable to act on them, or the reverse, feeling capable while generating few alternatives. High scores on both subscales indicate the fullest expression of the flexibility the CFI measures; low scores on both indicate the greatest cognitive rigidity.
Research Evidence and Psychometric Properties
Reliability Evidence
Internal consistency: total CFI α = .90 (Time 1) and .91 (Time 2); Alternatives α = .91 at both time points; Control α = .86 (Time 1) and .84 (Time 2) (Dennis & Vander Wal, 2010)
Test-retest reliability (7-week interval): total CFI r = .81, Alternatives r = .75, Control r = .77, all p < .001 (Dennis & Vander Wal, 2010)
Subscale intercorrelation: Alternatives and Control correlated r = .41 (Time 1) and .45 (Time 2), both p < .001, indicating related but distinct constructs (Dennis & Vander Wal, 2010)
Independent replication of reliability: internal consistency was good for the CFI in clinical and nonclinical older-adult samples (Johnco, Wuthrich, & Rapee, 2014)
Factor Structure
Exploratory factor analyses at two time points, seven weeks apart, yielded a two-factor solution accounting for 39% of the total variance in the CFI items at Time 1, with the same 20 items loading on the same two factors at both time points and factor loadings ranging from .47 to .85 (M = .64) (Dennis & Vander Wal, 2010). The structure has not been uniformly reproduced beyond the development sample: in older adults, the Alternatives and Control subscales were not significantly correlated in the clinical sample, leading the authors to caution that the CFI total score may not be valid in clinical samples (Johnco et al., 2014).
Convergent Validity
Cognitive Flexibility Scale (Martin & Rubin, 1995): CFI total r = .73 (Time 1) and .75 (Time 2); Alternatives r = .58 and .62; Control r = .65 and .66; all p < .001 (Dennis & Vander Wal, 2010)
Self-report vs performance: in older adults, the CFI and CFS converged with each other but showed little relationship with neuropsychological measures of cognitive flexibility, suggesting self-report captures a different aspect of the construct than performance testing (Johnco et al., 2014)
Relations with Depression and Coping
Correlations with the Beck Depression Inventory-II (Beck, Steer, & Brown, 1996) were negative at both time points: CFI total r = -.39 and -.35 (p < .001); Alternatives r = -.19 and -.20 (p < .01); Control r = -.50 and -.44 (p < .001), the Control subscale showing the strongest inverse association with depressive symptoms (Dennis & Vander Wal, 2010).
On the revised Ways of Coping measure (Folkman & Lazarus, 1985), CFI total scores correlated positively with adaptive coping and negatively with maladaptive coping (Dennis & Vander Wal, 2010):
Problem-Focused Coping: r = .48 (Time 1), .49 (Time 2), both p < .001
Seeking Social Support: r = .32 at both time points, p < .001
Focusing on the Positive: r = .39 (Time 1), .32 (Time 2), both p < .001
Keep to Self: r = -.33 (Time 1), -.34 (Time 2), both p < .001
Wishful Thinking: r = -.15 (Time 1), -.24 (Time 2), both p < .05
Detachment: r = -.23 (Time 1), -.30 (Time 2), both p < .001
The subscales also showed differential validity: Control was uniquely associated with lower Self-Blame (r = -.22 and -.30, p < .01) and lower Wishful Thinking (r = -.35 at both time points, p < .001), whereas Alternatives was uniquely associated with greater Seeking of Social Support (r = .37 and .39, p < .001) (Dennis & Vander Wal, 2010).
Development Sample
Study 2, Time 1: N = 196 undergraduate students at a private Midwestern university; mean age 20.20 ± 1.05 years; 75% female; 81% Caucasian (Dennis & Vander Wal, 2010)
Study 2, Time 2 (7-week follow-up): N = 152 (78% return rate); mean age 20.36 ± 0.96 years; 74% female; 81% Caucasian (Dennis & Vander Wal, 2010)
Evidence Beyond the Development Sample
Johnco, Wuthrich, and Rapee (2014) evaluated the CFI in 47 older adults with comorbid anxiety and depression and 53 nonclinical community-dwelling older adults. Internal consistency was good in all samples, and the clinical sample reported poorer cognitive flexibility than the nonclinical sample. However, the two subscales were not significantly correlated in the clinical sample, and self-report scores bore little relationship to neuropsychological measures of cognitive flexibility, so the authors recommended caution when interpreting the CFI total score in clinical populations (Johnco et al., 2014).
Usage Guidelines and Applications
Primary Applications
Measuring perceived cognitive flexibility in stress and coping research
Assessing CBT-relevant thinking patterns and studying cognitive mechanisms of change
Repeated measurement in intervention studies; the CFI was designed to allow monitoring of cognitive flexibility during therapy, though its sensitivity to clinically produced change had not been demonstrated in the development research (Dennis & Vander Wal, 2010)
Research on cognitive vulnerability to depression
Research Design Considerations
Score against your own sample: no clinical cutoffs or normative bands exist; interpret scores relative to the study sample or the development-sample means, labeled as descriptive
Use the published item order and numbering when administering and scoring; the subscales are interleaved in the standard form (Dennis & Vander Wal, 2010, Appendix 2)
Consider reporting subscales separately in clinical samples, where the total score may not be valid (Johnco et al., 2014)
Do not substitute the CFI for performance testing: self-report and neuropsychological measures of flexibility are only weakly related (Johnco et al., 2014)
Cultural Considerations
The CFI was developed and validated with a predominantly female, predominantly Caucasian U.S. undergraduate sample (Dennis & Vander Wal, 2010); generalization to other populations should be checked rather than assumed
Translations exist in several languages, but validation quality varies; consult the local validation literature before using a translated form
Limitations and Cautions
Self-report of perceived flexibility: scores reflect beliefs about one’s thinking, not measured cognitive performance, and may not mirror behavioral flexibility (Dennis & Vander Wal, 2010)
Population specificity: validation rests mainly on undergraduate students; further research in clinical populations is needed (Dennis & Vander Wal, 2010)
No clinical cutoffs: the CFI is not a diagnostic instrument and no score qualifies or excludes anyone from any category
Response bias: the items are face-valid and may be susceptible to socially desirable responding (Dennis & Vander Wal, 2010)
Total score in clinical samples: interpret cautiously; the subscales were not significantly correlated in a clinical older-adult sample (Johnco et al., 2014)
Copyright and Usage Responsibility: Check that you have the proper rights and permissions to use this assessment tool in your research. This may include purchasing appropriate licenses, obtaining permissions from authors/copyright holders, or ensuring your usage falls within fair use guidelines.
The Cognitive Flexibility Inventory was published in full as Appendix 2 of the development article; copyright rests with the authors, Dennis and Vander Wal. The instrument is not distributed under a commercial license and is available for research use with proper attribution to the original development article.
Proper Attribution: When using or referencing this inventory, cite the original development:
Dennis, J. P., & Vander Wal, J. S. (2010). The Cognitive Flexibility Inventory: Instrument development and estimates of reliability and validity. Cognitive Therapy and Research, 34(3), 241-253. https://doi.org/10.1007/s10608-009-9276-4
Dennis, J. P., & Vander Wal, J. S. (2010). The Cognitive Flexibility Inventory: Instrument development and estimates of reliability and validity. Cognitive Therapy and Research, 34(3), 241-253. https://doi.org/10.1007/s10608-009-9276-4
Moore, R. G. (1996). It’s the thought that counts: The role of intentions and meta-awareness in cognitive therapy. Journal of Cognitive Psychotherapy: An International Quarterly, 10(4), 255-269. https://doi.org/10.1891/0889-8391.10.4.255
Teasdale, J. D., Segal, Z. V., & Williams, J. M. G. (1995). How does cognitive therapy prevent depressive relapse and why should attentional control (mindfulness) training help? Behaviour Research and Therapy, 33(1), 25-39. https://doi.org/10.1016/0005-7967(94)E0011-7
Young, J. E., Weinberger, A. D., & Beck, A. T. (2001). Cognitive therapy for depression. In D. H. Barlow (Ed.), Clinical handbook of psychological disorders: A step-by-step treatment manual (3rd ed., pp. 264-308). New York: Guilford Press. (no DOI)
Validation Research:
Johnco, C., Wuthrich, V. M., & Rapee, R. M. (2014). Reliability and validity of two self-report measures of cognitive flexibility. Psychological Assessment, 26(4), 1381-1387. https://doi.org/10.1037/a0038009
Measures Used in Validation:
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Beck Depression Inventory-Second Edition manual. San Antonio, TX: The Psychological Corporation. (no DOI)
Folkman, S., & Lazarus, R. S. (1985). If it changes it must be a process: Study of emotion and coping during three stages of a college examination. Journal of Personality and Social Psychology, 48(1), 150-170. https://doi.org/10.1037/0022-3514.48.1.150
Armadillos at a decision point — representing the ability to adapt thinking, consider alternatives, and shift perspectives measured by the CFI (Cognitive Flexibility Inventory)
Frequently Asked Questions
What does the CFI measure?
The Cognitive Flexibility Inventory (CFI) measures the type of cognitive flexibility necessary for challenging and replacing maladaptive thoughts with more balanced, adaptive thinking (Dennis & Vander Wal, 2010). It has two subscales: Alternatives, the ability to perceive multiple explanations for events and to generate multiple solutions to difficult situations, and Control, the tendency to perceive difficult situations as controllable and to feel confident about overcoming difficulties.
How long does the CFI take to complete?
Approximately 5-7 minutes, per the development article (Dennis & Vander Wal, 2010). With 20 statements rated on a 7-point agreement scale, it is brief enough for repeated administration in research and intervention monitoring.
Is the CFI free to use?
Yes, for research use with proper attribution. The full item set was published as an appendix of Dennis and Vander Wal (2010) in Cognitive Therapy and Research; copyright rests with the authors and the instrument is not distributed under a commercial license. Cite the original article whenever you use the inventory.
How is the CFI scored?
Reverse score the six reverse-keyed items (items 2, 4, 7, 9, 11, and 17) as 8 minus the original score. The two subscales are interleaved in the published form: the Alternatives subscale is the sum of its 13 items (items 1, 3, 5, 6, 8, 10, 12, 13, 14, 16, 18, 19, 20; range 13-91), the Control subscale is the sum of its 7 items (items 2, 4, 7, 9, 11, 15, 17; range 7-49), and the total CFI score is the sum of all 20 items (range 20-140). Higher scores indicate greater cognitive flexibility.
How does the CFI differ from the Cognitive Flexibility Scale (CFS)?
The CFI targets the flexibility relevant to CBT and adaptive coping with stressors, whereas the CFS of Martin and Rubin (1995) grew out of the communication-competence literature. In the development study the two measures correlated r = .73-.75, indicating related but distinct constructs (Dennis & Vander Wal, 2010).
How reliable is the CFI?
Internal consistency ranged from α = .84 to .91 across subscales and time points in the development samples, and 7-week test-retest correlations were r = .81 for the total score, .75 for Alternatives, and .77 for Control (Dennis & Vander Wal, 2010). Internal consistency was also good in an independent study of clinical and nonclinical older adults (Johnco, Wuthrich, & Rapee, 2014).
Can the CFI be used with clinical populations?
With caution. It was developed with undergraduate students, and no clinical cutoffs exist. In older adults with comorbid anxiety and depression, the clinical group reported poorer flexibility than a nonclinical group, but the two subscales were not significantly correlated in the clinical sample, so Johnco, Wuthrich, and Rapee (2014) cautioned that the total score may not be valid there; reporting the subscales separately is prudent.
