CBT and Neuroticism: What the Evidence Shows
Updated: Oct 8
Author: Ukrainian Psychological Hub · Published: September 30, 2026 · Editorial Policy
Cognitive behavioral therapy can coincide with lower scores on neuroticism measures, and some studies suggest that genuine personality-trait change can occur during psychological treatment. The evidence, however, does not support the simple claim that CBT “cures neuroticism.” Neuroticism is a dimensional personality trait, not a psychiatric disorder. CBT is usually delivered to treat a clinical problem such as anxiety or depression, and a reduction in a neuroticism score can reflect several things at once: enduring trait change, reduced symptoms, less state distress at the time of testing, or some combination of these.
The broadest intervention evidence indicates that personality measures, especially emotional stability—the inverse pole of neuroticism in many Big Five models—can change during intervention. A major meta-analysis of 207 studies found an average personality-trait change of about d = .37 across interventions, with emotional stability showing the largest change. Therapy type was not strongly associated with the amount of change, so those results establish malleability more clearly than they establish a uniquely CBT-specific effect (Roberts et al., 2017).
More specific trials make the picture sharper. In major depression, cognitive therapy was associated with a substantial decline in NEO-FFI Neuroticism, but its advantage over placebo was no longer statistically significant after depression improvement was taken into account (Tang et al., 2009). In generalized anxiety disorder, both CBT and metacognitive therapy were associated with lower NEO-PI-R Neuroticism after 12 sessions, with the larger decrease occurring in the metacognitive-therapy condition (Kennair et al., 2021). In a transdiagnostic CBT approach designed to target neuroticism more directly, the Unified Protocol produced lower neuroticism scores than symptom-focused CBT in one randomized trial, although later work also shows why symptom change, measurement, treatment duration, and replication matter (Sauer-Zavala et al., 2021).
The short answer
CBT can be associated with reductions in measured neuroticism, but the strength and meaning of that reduction depend on what kind of CBT was delivered, why the person was in treatment, which personality instrument was used, when personality was measured, and whether symptom change was separated statistically from trait change.
The strongest conclusion is therefore conditional. Personality traits are more changeable than older fixed-trait views implied, and neuroticism-related measures often decrease during successful treatment. Standard disorder-focused CBT can show this pattern, but the evidence does not demonstrate that every CBT protocol directly changes neuroticism as an independent treatment target. Interventions that explicitly target broad emotional vulnerability, especially the Unified Protocol, provide more direct evidence, but that literature remains smaller than the evidence base for CBT's effects on anxiety and depressive symptoms.
For clinical interpretation, the distinction matters. Higher neuroticism is a transdiagnostic risk factor associated with many forms of internalizing distress, but a personality score does not diagnose anxiety disorder, depression, OCD, PTSD, or any other condition. A person does not need psychotherapy simply because a personality questionnaire places them high on neuroticism.
What neuroticism means in this evidence
Neuroticism is a broad personality dimension describing relatively stable individual differences in the tendency to experience negative emotions, perceive threat or stress, and react strongly to adverse events. In some Big Five frameworks the opposite pole is called emotional stability. In the BFI-2, the corresponding domain is labeled Negative Emotionality, with anxiety, depression, and emotional volatility facets (Soto & John, 2017). These labels overlap conceptually, but their scores are not interchangeable.
A trait is a probabilistic pattern across time and situations. It is not a constant emotional state. Someone high in neuroticism can feel calm on a particular day; someone lower in neuroticism can have severe anxiety during an acute crisis. Trait neuroticism is also distinct from negative affect at a specific moment, trait anxiety, psychiatric symptoms, and clinical disorders.
This is especially important in psychotherapy research because the same person may complete a personality questionnaire while severely depressed at baseline and again when symptoms have improved. Self-reports can move partly because the person is genuinely functioning differently across situations, partly because a stable pattern has begun to change, and partly because current mood affects how they answer. A lower post-treatment score is scientifically interesting, but it is not automatically proof that a durable personality trait has been rewritten.
What CBT means in this literature
CBT is a family of psychological treatments rather than one single protocol. Different forms emphasize different combinations of cognitive reappraisal, behavioral experiments, exposure, behavioral activation, problem solving, skills practice, and relapse-prevention strategies. A CBT protocol for panic disorder, a cognitive-therapy protocol for major depression, and a transdiagnostic CBT protocol for emotional disorders share a family resemblance while targeting different maintaining processes.
That variation is central to the neuroticism question. Most CBT trials are designed to reduce disorder symptoms and improve functioning. Neuroticism may be measured as a secondary outcome. Other interventions, particularly the Unified Protocol, conceptualize broad negative emotionality or neuroticism-related vulnerability as a treatment target. Evidence from these two kinds of studies answers different questions.
A symptom-focused trial can show that neuroticism scores change while a disorder is treated. A trait-targeted trial can more directly test whether changing a broader emotional disposition is part of the intervention's effect. Neither design, by itself, proves that the same effect will occur in a healthy person who simply wants a lower personality score.
What would count as real neuroticism change?
Trait change
Trait change means a reasonably enduring shift in a person's characteristic pattern of thoughts, feelings, and behavior. Strong evidence is easier to claim when change persists beyond the immediate treatment period, appears across repeated measurements, and cannot be explained entirely by concurrent symptom improvement or temporary mood.
Mean-level change
Mean-level change is a change in the average score of a treatment group. If a group starts high in neuroticism and scores lower after therapy, that is evidence that the group's mean moved. It does not mean every participant changed, nor does it tell us whether individuals kept their relative ordering.
Individual change
Individual change asks whether a particular person's score changed beyond what could plausibly result from measurement error or ordinary fluctuation. Group-level effect sizes cannot be converted directly into a prediction for an individual client. Reliable-change analyses, repeated assessments, and instrument-specific norms can help, but no universal raw-score threshold identifies “real” change across all neuroticism measures.
State fluctuation
State fluctuation is short-term movement caused by current stress, mood, sleep, illness, recent events, or the immediate treatment context. State-related movement is expected even when the underlying trait is relatively stable. Distinguishing state from trait is therefore not a semantic detail; it is one of the main methodological problems in this literature.
Symptom reduction
A person may become less depressed, less anxious, less avoidant, or less distressed while their broader personality disposition changes less. Because neuroticism measures contain content related to negative emotional experience, symptom improvement can reduce trait scores. Researchers therefore need designs that examine whether neuroticism change remains after accounting for simultaneous symptom change and whether it persists at follow-up.
The broad psychotherapy evidence: personality can change
For the broader change evidence and the dedicated transdiagnostic-treatment review, see Can Neuroticism Change? Personality Development, Therapy, and Evidence and Unified Protocol and Neuroticism: A Transdiagnostic Treatment Approach.
The most influential broad synthesis is the 2017 systematic review and meta-analysis by Roberts and colleagues. Across 207 studies that tracked personality traits during interventions, interventions were associated with meaningful changes in personality measures over an average of about 24 weeks. Emotional stability showed the largest average change, followed by extraversion. The pattern appeared in experimental and nonexperimental studies and could persist into follow-up (Roberts et al., 2017).
This finding shifted the scientific baseline. Adult personality is not infinitely plastic, but neither is it fixed in the sense that treatment cannot move trait measures. The meta-analysis also found that therapy type was not strongly related to the amount of personality change. That point prevents an overclaim: the review supports intervention-related malleability, not a conclusion that CBT is uniquely effective for lowering neuroticism.
The review also found greater personality change in some clinical groups than others, with anxiety-disorder samples changing especially strongly. That pattern is compatible with the idea that treating sustained fear, avoidance, worry, and negative emotional reactivity may alter the behaviors and experiences that personality questionnaires sample. It is still possible for symptom change and trait change to be intertwined rather than cleanly separable.
Direct evidence from cognitive therapy for depression
A particularly useful study comes from a placebo-controlled depression trial because it measured personality as well as depressive symptoms. Adults with moderate to severe major depressive disorder were randomized to paroxetine, placebo, or cognitive therapy. Personality was assessed with the NEO Five-Factor Inventory, not a generic online neuroticism quiz (Tang et al., 2009).
From intake to week 8, the cognitive-therapy group showed a neuroticism decline of about 0.54 standard deviations, while the placebo group showed a much smaller decline. At first glance, that looks like a clear CBT-related personality effect. The crucial analysis, however, asked whether the cognitive-therapy advantage remained once improvement in depression was statistically controlled.
It did not. After accounting for depression improvement, the cognitive-therapy versus placebo difference in neuroticism was no longer statistically significant. The study therefore supports two statements at once: cognitive therapy was accompanied by a substantial drop in measured neuroticism, and that drop could not be cleanly separated from the improvement in depression. Treating the first statement as proof of independent trait change would go beyond the data.
The same trial also illustrates why treatment mechanisms should not be inferred merely from correlated change. A treatment can reduce symptoms through cognitive and behavioral pathways while neuroticism scores move in parallel. Parallel movement does not establish which variable caused the other.
Evidence from CBT for generalized anxiety disorder
Kennair and colleagues examined personality change in a randomized trial of cognitive behavior therapy and metacognitive therapy for generalized anxiety disorder. Fifty-five participants completed the full 240-item NEO-PI-R before and after 12 sessions. Across treatment conditions, Neuroticism decreased, while Extraversion and Openness increased (Kennair et al., 2021).
The instrument matters. NEO-PI-R Neuroticism is a broad domain composed of six facets in that model: Anxiety, Angry Hostility, Depression, Self-Consciousness, Impulsiveness, and Vulnerability. A change on this domain is not equivalent to a change on the BFI-2 Negative Emotionality domain or the PID-5 Negative Affectivity domain, even though these constructs overlap.
In this trial, metacognitive therapy was associated with a greater reduction in Neuroticism than CBT. That result is useful precisely because it resists a simplistic “CBT lowers neuroticism” narrative. Successful psychological treatment was associated with trait-score change, but the magnitude differed by treatment, and the sample was small. The study cannot establish a universal effect for CBT across diagnoses or populations.
What group CBT studies add—and what they do not
A 2023 study used data from 156 patients receiving transdiagnostic or diagnosis-specific group CBT for major depression, social anxiety disorder, panic disorder, or agoraphobia. Decreases in PID-5 Negative Affectivity predicted lower subsequent depression and anxiety symptoms (Niemeijer et al., 2023).
This study is relevant to the neuroticism question, but the construct boundary must be preserved. PID-5 Negative Affectivity is a maladaptive personality-trait domain in the DSM-5 personality model. It is related to Big Five neuroticism, yet it is not the same instrument and should not be relabeled as a direct Big Five Neuroticism score. The study strengthens the case that broad negative-emotional traits and symptoms can change together during CBT; it does not supply an interchangeable estimate for NEO or BFI neuroticism.
The Unified Protocol: CBT designed to target broad emotional vulnerability
The Unified Protocol for Transdiagnostic Treatment of Emotional Disorders is a CBT-based treatment designed for emotional disorders across traditional diagnostic categories. It targets processes such as aversive reactions to emotions, avoidance, rigid emotion-driven behavior, and maladaptive responses to distress. In this research program, neuroticism is not merely an incidental personality score; it is conceptually close to the broad vulnerability the treatment is intended to address.
That makes Unified Protocol studies unusually informative for the question “Can CBT change neuroticism?” It also means their results should not be generalized automatically to every form of CBT. A protocol built to target transdiagnostic emotional vulnerability is not interchangeable with a disorder-specific CBT protocol whose primary target is panic, social fear, depressive cognition, or another narrower mechanism.
A randomized trial comparing the Unified Protocol with symptom-focused CBT
In a randomized trial of 223 patients with principal anxiety disorders, the Unified Protocol was compared with gold-standard symptom-focused CBT protocols and a waitlist. At week 16, the Unified Protocol group had lower measured neuroticism than the symptom-focused CBT group, with a between-group effect of d = -0.32, and lower neuroticism than the waitlist, with d = -0.43. The differences remained after simultaneous depression and anxiety fluctuations were controlled (Sauer-Zavala et al., 2021).
This is stronger evidence for a trait-targeting interpretation than a simple pre-post decline because the trial included an active CBT comparison and adjusted for concurrent symptom change. It suggests that neuroticism change may be more robust when the broader trait-like vulnerability is explicitly targeted.
The effect should still be interpreted at the scale it was observed. A d of roughly one third of a standard deviation is a group-level difference, not a promise that an individual will move a certain number of raw-score points. The study also involved people with anxiety disorders seeking treatment, not a general-population sample selected only for high neuroticism.
A larger effectiveness trial gives a more restrained picture
A Spanish public mental-health trial randomized 488 participants with emotional disorders to group Unified Protocol treatment or treatment as usual consisting of individual, disorder-specific CBT. Personality, depression, anxiety, affect, and quality of life were tracked through treatment and follow-up. Both conditions improved across the major outcomes, and improvements were maintained at follow-up (Osma et al., 2022).
This larger pragmatic study matters because it shows that trait-related improvement is not confined to one tightly controlled laboratory comparison. It also makes the comparative claim more modest. The study's abstract highlights larger improvements for the Unified Protocol on depression, anxiety, and quality of life, while personality improved in both conditions. The evidence therefore supports change during treatment more confidently than it supports a large, general superiority of one CBT form for neuroticism.
State change can appear before trait change
One of the most informative recent lines of work measures neuroticism repeatedly during the Unified Protocol rather than only before and after treatment. In a 38-person clinical study, neuroticism showed state-level change by session 6 and trait-level change by session 12. Within-person neuroticism reductions were bidirectionally related to anxiety change and predicted later session-to-session reductions in depression (Stumpp et al., 2024).
The value of this study is conceptual as much as numerical. Early movement on a personality measure can reflect temporary state change; a longer pattern of repeated change is more consistent with a trait-level shift. The sample was small, so the timing should not be treated as a universal six-session versus twelve-session rule.
A newer randomized study of 70 adults with emotional disorders examined six versus twelve Unified Protocol sessions and different module sequences. Neuroticism change did not differ by treatment-duration condition over the first six sessions, but participants assigned to the full twelve-session treatment showed greater reductions from sessions 7 through 12. When depression and anxiety symptoms were controlled, significant neuroticism reductions were absent in the first six sessions and emerged during sessions 7 through 12 in the full-treatment condition (Stumpp et al., 2026).
That pattern strengthens the case for separating rapid symptom-linked movement from more durable trait change. It also supports a practical scientific lesson: if personality is the outcome of interest, measuring it only immediately after a brief intervention may be insufficient.
Why symptom improvement and neuroticism change are so difficult to separate
Neuroticism and internalizing symptoms share substantial psychological content. A person who is depressed may endorse items about sadness, discouragement, vulnerability, or negative emotionality more strongly. A person whose anxiety improves may report less worry, threat sensitivity, tension, and emotional volatility. Those answers can lower a trait score even if the person's deeper cross-situational disposition has changed less.
The reverse direction is also plausible. If therapy changes habitual avoidance, threat appraisal, emotion-driven behavior, and expectations about coping, those repeated changes could eventually alter the broader personality pattern that a neuroticism measure is designed to capture. The relationship can therefore be reciprocal rather than a contest between “symptoms” and “personality.”
Research on neuroticism and emotion regulation helps explain why. Neuroticism is associated at the group level with differences in how people respond to negative emotion, but regulation is context-sensitive and trainable. Repeated changes in appraisal, avoidance, exposure, and behavioral response may change the experiences that maintain both symptoms and trait-like patterns.
How CBT might influence neuroticism
Changing threat appraisal
Many CBT protocols help people test predictions about danger, rejection, failure, uncertainty, or loss of control. If a person repeatedly discovers that anticipated outcomes are less catastrophic or more manageable than expected, future situations may evoke less intense negative anticipation. Over time, that could affect trait-relevant patterns rather than only a single symptom.
Reducing avoidance
Avoidance can preserve fear and prevent corrective learning. Exposure and behavioral experiments create opportunities to learn that distress can be tolerated and that feared outcomes are not inevitable. A repeated shift from avoidance toward approach behavior can change both disorder-specific symptoms and broader beliefs about one's ability to cope.
Changing responses to emotion
CBT-based transdiagnostic treatments often teach people to observe emotions, reduce secondary reactions to them, and choose behaviors that are less controlled by immediate distress. This is especially relevant to stress reactivity and recovery, which are closely related to neuroticism but remain distinct processes.
Building repeated experiences of effective coping
Personality change is more plausible when new behavior is repeated across situations and sustained over time. A single insight or a temporary mood improvement is unlikely to define a trait shift. Repeated successful coping can gradually change expectations, habits, and self-perception, creating a pathway by which therapy-related learning might consolidate into personality change.
These mechanisms are plausible and partly supported by process research, but they should not be presented as a proven universal chain from CBT technique to lower neuroticism. Different protocols use different mechanisms, and the neuroticism literature is still smaller than the symptom-outcome literature.
CBT does not need to target neuroticism to be clinically useful
For the broader therapy-level evidence beyond CBT-specific studies, see Therapy for High Neuroticism: What Treatment Can and Cannot Change.
The clinical goal of CBT is usually not to produce an ideal personality profile. It is to reduce distress, improve functioning, help a person pursue valued activities, and address the mechanisms maintaining a specific problem. A client can benefit greatly even if a personality score barely changes.
The Tang depression trial is a good example. Cognitive therapy is an established treatment approach for depression, yet the study did not show an independent neuroticism effect after depression improvement was accounted for. That does not make the therapy unsuccessful; it means the personality outcome should be interpreted in relation to the symptom outcome rather than treated as the only meaningful marker of change.
Similarly, people with anxiety or depression should be assessed and treated according to their actual symptoms, impairment, preferences, diagnosis when applicable, and clinical context. A neuroticism score can add personality information, but it does not replace clinical assessment.
Can CBT cure neuroticism?
“Cure” is the wrong model for a personality trait. Neuroticism is not an illness that a person either has or does not have. It is a continuous dimension found across the population. People differ in where they fall on that dimension, and those differences can shift over development and in response to life experiences or interventions.
A more accurate question is whether CBT can produce a lasting reduction in neuroticism-related personality scores or in the patterns those scores represent. The evidence says this can happen for some people and in some treatment contexts. It does not show that CBT eliminates neuroticism, guarantees a particular amount of change, or should be prescribed solely to make a personality score lower.
The 2026 systematic review of nonclinical neuroticism interventions also underscores how heterogeneous the intervention literature has become. Across 99 articles, interventions and proposed mechanisms varied substantially, reinforcing the need to identify which strategies change which components of neuroticism, for whom, and for how long (Wright et al., 2026).
Does high neuroticism mean someone should get CBT?
No. High neuroticism by itself is not a diagnosis and is not, by itself, an indication for psychotherapy. Many people high in neuroticism function well, maintain relationships, work effectively, and do not meet criteria for a mental disorder. The trait can be associated with vulnerability, but vulnerability is not the same thing as illness.
CBT may be worth considering when a person has persistent distress, avoidance, worry, depressed mood, impairment, or another problem for which a CBT approach is clinically appropriate, or when the person has a clear change goal that therapy can address. The treatment target should be the person's difficulties and goals, not an abstract demand to become “less neurotic.”
Measurement determines what a study can claim
NEO-PI-R
The NEO-PI-R is a 240-item inventory that measures five broad personality domains and six facets within each domain. Its Neuroticism domain is the measure used in the generalized-anxiety study by Kennair and colleagues. Facet-level information can reveal whether broad score change is distributed across anxiety, depression, vulnerability, self-consciousness, impulsiveness, and angry hostility or concentrated in only part of the domain.
NEO-FFI
The NEO Five-Factor Inventory is a shorter measure of the five broad domains. The Tang depression trial used the NEO-FFI. A change of a certain number of NEO-FFI points cannot be translated automatically into an equivalent NEO-PI-R, BFI-2, IPIP, or Eysenck score.
BFI-2 Negative Emotionality
The BFI-2 labels the relevant Big Five domain Negative Emotionality rather than Neuroticism and measures three facets: anxiety, depression, and emotional volatility (Soto & John, 2017). This is a modern operationalization of the same broad Big Five territory, but instrument-specific labels, item content, scoring, reliability, and norms still matter.
PID-5 Negative Affectivity
PID-5 Negative Affectivity is part of a maladaptive personality-trait model used in psychopathology research. It overlaps with neuroticism but is not simply another name for a Big Five score. The 2023 group-CBT study therefore provides related trait-change evidence, not a direct estimate of change on a Big Five Neuroticism instrument.
Raw scores and percentiles
No raw score or percentile has a universal meaning across instruments. Interpretation depends on the questionnaire, scoring direction, normative sample, age and population where relevant, and whether the score is a domain, facet, or maladaptive-trait measure. A “high score” before therapy and a “lower score” afterward should be interpreted within the same validated scoring system.
How large is the effect?
There is no single CBT effect size for neuroticism. The literature combines different populations, disorders, treatment formats, measures, durations, and comparison groups. Broad intervention meta-analysis gives an average across many intervention types; specific CBT trials answer narrower questions.
In the Sauer-Zavala randomized trial, the Unified Protocol versus symptom-focused CBT difference at week 16 was d = -0.32. That is a modest group-level difference, not a universal individual outcome. In Tang's depression trial, cognitive therapy showed a sizable pre-post decline in NEO-FFI Neuroticism, but the independent advantage over placebo disappeared after depression improvement was controlled. Kennair's GAD trial found neuroticism reductions in both CBT and metacognitive therapy, with greater reduction in the latter condition.
Those findings are more informative together than any single number. They suggest that neuroticism can move during CBT, but the magnitude depends on the clinical context and on whether the treatment is explicitly designed to target broad emotional vulnerability.
How long would personality change take?
The evidence does not support a universal session count. Personality change can begin quickly on self-report measures, yet early changes may contain a larger state component. The repeated-measure Unified Protocol research is useful because it suggests that state-level movement can appear before stronger evidence of trait-level change.
In the 2026 Stumpp study, controlling for anxiety and depressive symptoms weakened the evidence for neuroticism change during the first six sessions, while significant reductions appeared during sessions 7 through 12 in the full-treatment condition. That finding is consistent with the idea that repeated practice and consolidation may matter. It should not be turned into a rule that “twelve sessions change personality” for everyone.
Are changes durable after CBT ends?
Durability is one of the key criteria for interpreting personality change. The Roberts meta-analysis found that intervention-related personality changes could persist at follow-up across the broader literature. Some Unified Protocol effectiveness studies have also followed participants after treatment and found maintenance of broader outcome improvements. Still, long-term evidence specifically isolating CBT-caused neuroticism change from symptom change is much thinner than the evidence for short-term pre-post movement.
A strong future study would measure neuroticism repeatedly before treatment, during treatment, immediately afterward, and months or years later; use reliable trait instruments; assess symptoms at the same time points; include an active comparison condition; and test whether individual change generalizes across situations and informants. Much of the existing literature meets only part of that ideal.
What the evidence does not show
The evidence does not show that high neuroticism is a disease. It does not show that every person high in neuroticism needs psychotherapy. It does not show that any one CBT technique directly lowers a Big Five trait in every population. It does not show that a lower post-treatment questionnaire score is always enduring trait change. It does not show that symptom-focused CBT is ineffective if neuroticism barely moves.
It also does not justify converting group averages into personalized predictions. A trial may show that a treatment group changed more than a comparison group, while individual trajectories vary widely. Some people may show large reductions, others little movement, and some may fluctuate in both directions.
Finally, the evidence does not establish that lower neuroticism is always the only desirable outcome. Sensitivity to threat, emotional responsiveness, vigilance, and concern can be useful in some contexts. Therapy is better evaluated by whether it reduces unnecessary suffering and expands flexible functioning than by whether it pushes every personality score toward one population average.
Clinical interpretation: trait, symptoms, and diagnosis must stay separate
A high neuroticism score is a personality assessment result. Anxiety and depressed mood can be symptoms. A screening questionnaire estimates the likelihood or severity of a problem. A psychiatric diagnosis is a clinical classification based on a defined set of criteria, context, duration, impairment, and differential assessment. These levels are related but they are not interchangeable.
Neuroticism can function as a broad vulnerability factor across multiple disorders, which helps explain why transdiagnostic treatments are scientifically interesting. Yet treatment should still be matched to the actual clinical presentation. Someone with panic attacks, major depression, OCD, trauma-related symptoms, or another condition needs disorder-relevant assessment rather than a treatment plan inferred from a personality score.
What a person in CBT can realistically look for
For someone already in CBT, it is reasonable to notice whether emotional reactions become less intense, whether recovery after stress becomes faster, whether avoidance decreases, whether catastrophic interpretations weaken, and whether coping becomes more flexible. These are meaningful changes even before anyone asks whether the Big Five trait itself has moved.
If personality change is an explicit goal, repeated use of the same validated instrument can provide one source of information, but it should be interpreted alongside functioning, behavior, symptoms, and life context. Repeating an online test every few days is more likely to capture noise and state fluctuation than stable trait change.
A therapist and client can also separate goals that are actionable from labels that are too broad. “I want to recover faster after criticism,” “I want to stop avoiding uncertainty,” or “I want to handle worry without losing hours to it” are more useful treatment targets than “I want to erase neuroticism.”
The current scientific bottom line
CBT can be accompanied by lower neuroticism scores, and the broader intervention literature supports the conclusion that adult personality traits can change. Direct CBT evidence is strongest when interpreted study by study rather than collapsed into a slogan.
In depression, cognitive therapy lowered measured neuroticism, but the independent effect was difficult to separate from symptom improvement. In generalized anxiety disorder, CBT was associated with lower NEO-PI-R Neuroticism, while metacognitive therapy produced the larger reduction. In the Unified Protocol literature, a CBT-based transdiagnostic intervention explicitly targeting neuroticism-related vulnerability has produced reductions beyond symptom-focused CBT in one trial, while larger and more recent studies emphasize modest effects, treatment duration, and the distinction between state and trait change.
The most defensible conclusion is that neuroticism is modifiable, CBT can contribute to that change in some contexts, and the scientific meaning of a lower score depends on measurement, symptoms, duration, comparison condition, and follow-up. Neuroticism remains a personality trait rather than a diagnosis, and clinical success should not be reduced to whether a personality score moves.
Frequently Asked Questions
Can CBT reduce neuroticism?
Yes, neuroticism scores can decrease during CBT, and several clinical studies have observed this pattern. The size and interpretation of the change vary by disorder, protocol, measure, and whether symptom improvement is controlled.
Does CBT permanently change personality?
Psychological interventions can produce personality-trait changes that persist beyond treatment, but permanence cannot be assumed from a pre-post score difference. Long-term follow-up is needed, and individual trajectories vary.
Is neuroticism a disorder that CBT treats?
No. Neuroticism is a dimensional personality trait. CBT treats clinical problems and behavioral or cognitive patterns; some CBT approaches also target broad processes closely related to neuroticism.
Is the Unified Protocol the same as ordinary CBT?
It is a CBT-based transdiagnostic treatment, but it is a specific protocol designed for emotional disorders and broad underlying processes. Its neuroticism findings should not be generalized automatically to every CBT protocol.
If my neuroticism score drops after therapy, does that prove my trait changed?
No. The drop may reflect genuine trait change, symptom improvement, temporary state change, measurement error, or a combination. Repeated measurement and follow-up make trait-change interpretations stronger.
Which neuroticism test should be used before and after CBT?
Use a validated personality instrument appropriate to the purpose and population, and use the same scoring system at both time points. NEO-PI-R, NEO-FFI, BFI-2 Negative Emotionality, IPIP measures, and Eysenck scales are not numerically interchangeable.
Does high neuroticism mean I need therapy?
No. A high trait score alone is not a diagnosis or treatment indication. Therapy is considered in relation to distress, functioning, symptoms, goals, and clinical assessment.
Can CBT make someone emotionally stable?
CBT can help people change patterns of appraisal, avoidance, coping, and emotional response, and emotional-stability measures may improve. It cannot guarantee a specific personality profile, and emotional stability exists on a continuum rather than as an all-or-none state.
Related Articles
References
Kennair, L. E. O., Solem, S., Hagen, R., Havnen, A., Nysæter, T. E., & Hjemdal, O. (2021). Change in personality traits and facets (Revised NEO Personality Inventory) following metacognitive therapy or cognitive behaviour therapy for generalized anxiety disorder: Results from a randomized controlled trial. Clinical Psychology & Psychotherapy, 28(4), 872–881. https://doi.org/10.1002/cpp.2541
Niemeijer, M., Reinholt, N., Poulsen, S., Bach, B., Christensen, A. B., Eskildsen, A., Hvenegaard, M., Arendt, M., & Arnfred, S. (2023). Trait and symptom change in group cognitive behaviour therapy for anxiety and depression. Clinical Psychology & Psychotherapy, 30(5), 1058–1070. https://doi.org/10.1002/cpp.2857
Osma, J., Peris-Baquero, Ó., Suso-Ribera, C., Farchione, T. J., & Barlow, D. H. (2022). Effectiveness of the Unified Protocol for transdiagnostic treatment of emotional disorders in group format in Spain: Results from a randomized controlled trial with 6-months follow-up. Psychotherapy Research, 32(3), 329–342. https://doi.org/10.1080/10503307.2021.1939190
Roberts, B. W., Luo, J., Briley, D. A., Chow, P. I., Su, R., & Hill, P. L. (2017). A systematic review of personality trait change through intervention. Psychological Bulletin, 143(2), 117–141. https://doi.org/10.1037/bul0000088
Sauer-Zavala, S., Fournier, J. C., Steele, S. J., Woods, B. K., Wang, M., Farchione, T. J., & Barlow, D. H. (2021). Does the unified protocol really change neuroticism? Results from a randomized trial. Psychological Medicine, 51(14), 2378–2387. https://doi.org/10.1017/S0033291720000975
Soto, C. J., & John, O. P. (2017). The next Big Five Inventory (BFI-2): Developing and assessing a hierarchical model with 15 facets to enhance bandwidth, fidelity, and predictive power. Journal of Personality and Social Psychology, 113(1), 117–143. https://doi.org/10.1037/pspp0000096
Stumpp, N. E., Southward, M. W., & Sauer-Zavala, S. (2024). Assessing theories of state and trait change in neuroticism and symptom improvement in the Unified Protocol. Behavior Therapy, 55(1), 93–105. https://doi.org/10.1016/j.beth.2023.05.005
Stumpp, N. E., Southward, M. W., Hines, A., Kushner, M. L., Schäuffele, C., Fruhbauerova, M., Bridges, E., Rahman, A., & Sauer-Zavala, S. (2026). When and under what circumstances does neuroticism change in the Unified Protocol? Psychotherapy Research, 36(4), 785–796. https://doi.org/10.1080/10503307.2025.2521779
Tang, T. Z., DeRubeis, R. J., Hollon, S. D., Amsterdam, J., Shelton, R., & Schalet, B. (2009). Personality change during depression treatment: A placebo-controlled trial. Archives of General Psychiatry, 66(12), 1322–1330. https://doi.org/10.1001/archgenpsychiatry.2009.166
Wright, A. J., Haehner, P., Hopwood, C. J., & Bleidorn, W. (2026). A systematic review and taxonomy of neuroticism interventions for the general public. Personality Science. https://doi.org/10.1177/27000710261421531
Zemestani, M., Ommati, P., Rezaei, F., & Gallagher, M. W. (2022). Changes in neuroticism-related constructs over the Unified Protocol for Transdiagnostic Treatment of Emotional Disorders in patients on an optimal dose of SSRI. Personality Disorders: Theory, Research, and Treatment, 13(2), 171–181. https://doi.org/10.1037/per0000482