The study, published in BMC Psychiatry, was led by Wenjia Leng and Jie Fan of the Second Xiangya Hospital of Central South University in Changsha, together with colleagues from Hunan institutions. The researchers recruited 2,397 Chinese college students, a population often used in early validation work because of its accessibility and its elevated rates of emerging behavioral addictions such as problematic internet use. Participants completed the translated ICBC alongside established validation instruments, including the Barratt Impulsiveness Scale, the Obsessive-Compulsive Inventory-Revised, and the Young Internet Addiction Test, allowing the team to check whether the new scale behaved the way a valid measure of impulsivity and compulsivity should.

Before any modeling began, the researchers screened the original 33-item scale using item analysis and exploratory factor analysis. Two items dropped out: exercising and illicit drug use. The first likely lost its discriminating power in a context where regular exercise is common and socially encouraged rather than compulsive, and the second may reflect the comparatively low base rate of illicit drug use among Chinese students. In their place, the team added one culturally relevant item, restraining from food, a behavior that resonates with dietary restriction practices in the Chinese context. The result is a 32-item Chinese version of the checklist, subtly reshaped to fit the behaviors that actually matter in its new setting.

The heart of the study lies in its modeling strategy. Traditional confirmatory factor analysis forces every item to load on exactly one factor and assumes that items measuring the same construct are unrelated to items measuring others, an assumption that is often unrealistic for overlapping constructs like impulsivity and compulsivity. Exploratory structural equation modeling, or ESEM, relaxes that constraint, letting items cross-load on multiple factors. The researchers compared competing structures using both CFA and ESEM, and the clear winner was the most sophisticated option: a bifactor-ESEM model, which simultaneously estimates one broad general factor and several narrow specific factors, each item allowed to speak to all of them.

In the winning model, the behaviors were organized around a single general disinhibition factor flanked by two specific factors with evocative names. The first, labeled impulsive-compulsions, captures compulsive behaviors with an impulsive flavor, such as compulsive shopping or binge-like episodes. The second, compulsive-impulsions, captures impulsive behaviors tinged with compulsion, such as excessive gaming or checking. The distinction sounds like wordplay, but it reflects a genuine insight of the transdiagnostic framework: many real-world behaviors mix both qualities, and forcing them into pure categories distorts the data.

The bifactor indices then revealed how much weight each factor could actually carry. The general disinhibition factor explained 40 percent of the common variance, with high factor determinacy of 0.93 and construct reliability of 0.88, yet its omega hierarchical of 0.55 showed that it contributed only a limited share of reliable variance to total scores. The compulsive-impulsions factor emerged as the strongest specific dimension, with an omega hierarchical subscale of 0.66 and strong reliability indices. The impulsive-compulsions factor, by contrast, offered limited unique reliable information beyond the general factor, with an omega hierarchical subscale of just 0.33, prompting the authors to urge cautious interpretation of scores on that dimension.

These numbers carry a practical warning. In Western samples, the general disinhibition factor of the ICBC has appeared considerably more dominant, and the Chinese findings suggest that the balance between general and specific variance is culturally contingent. The authors conclude that the total score of the Chinese version should not be used at all, and that researchers should instead rely on factor scores from the bifactor-ESEM model, particularly the general disinhibition and compulsive-impulsions dimensions. It is a candid admission that a translated scale can be statistically sound while still diverging in structure from its original, a nuance that often gets lost when questionnaires cross borders.

The scale also passed several conventional tests of quality. Factor scores from the bifactor model showed relatively small but significant correlations with external validation measures, ranging from -0.17 to 0.46, in the directions expected for impulsivity, obsessive-compulsive symptoms, and internet addiction. Test-retest reliability, measured by intraclass correlation coefficients over a repeated administration, ranged from 0.50 to 0.59, modest but not unusual for behavioral checklists whose targets fluctuate week to week. Most importantly for future research, the scale demonstrated configural, metric, scalar, and residual invariance across gender, meaning that men and women respond to the items in comparable ways and group comparisons are statistically legitimate.

Why does this matter beyond the psychometrics laboratory? Behavioral addictions are rising concerns on campuses worldwide, from smartphone overuse to disordered eating, and clinicians increasingly want tools that measure the underlying dimension rather than each diagnosis separately. A validated Chinese ICBC opens the door to cross-cultural studies that can test whether the impulsive-compulsive spectrum is a human universal or a construct shaped by cultural context. The finding that the general factor is weaker in China than in Western samples is itself scientifically interesting, hinting that the ways disinhibition expresses itself may differ across societies, perhaps reflecting differences in how behaviors like exercise, eating, and drug use are embedded in daily life.

The study is not without limitations. It relied on a college sample, so generalization to clinical populations or older adults remains to be demonstrated, and the modest test-retest figures suggest the checklist captures behaviors that are inherently variable. The weak impulsive-compulsions factor will need attention in future revisions, possibly through item refinement or larger samples. Still, the work stands as a model of modern scale validation: a large sample, culturally informed adaptation, and a bifactor-ESEM framework that asks not just whether a scale fits, but which of its scores can be trusted. For researchers studying the tangled roots of impulsive and compulsive behavior in China, the 32-item Chinese ICBC now provides a carefully calibrated instrument, along with an honest map of its strengths and its limits.

Subject of Research: Psychometric validation of the Chinese version of the Impulsive-Compulsive Behaviours Checklist using bifactor exploratory structural equation modeling in college students

Article Title: Psychometric properties of the Chinese version of the Impulsive–Compulsive Behaviours Checklist in Chinese college students: a bifactor exploratory structural equation modeling approach

Article References: Leng, W., Liu, Q., Lei, Y., Zeng, Y., Jiang, Y., Liu, H., Zhu, X., & Fan, J. (2026). Psychometric properties of the Chinese version of the Impulsive–Compulsive Behaviours Checklist in Chinese college students: a bifactor exploratory structural equation modeling approach. BMC Psychiatry. https://doi.org/10.1186/s12888-026-08720-8

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08720-8

Keywords: impulsivity, compulsivity, disinhibition, bifactor-ESEM, psychometrics, transdiagnostic constructs, ICBC, Chinese college students, behavioral addictions, measurement invariance, cross-cultural validation, BMC Psychiatry

Frontotemporal dementia, or FTD, belongs to a spectrum of disorders called frontotemporal lobar degeneration, which primarily strikes the frontal and temporal lobes of the brain. Unlike Alzheimer’s disease, which typically announces itself through memory lapses, FTD often begins with changes in behavior, personality, and social cognition. The behavioral variant of the disease, known as bvFTD, is particularly notorious for producing stereotyped, compulsive, and ritualistic behaviors, along with a striking vulnerability to addiction-related conduct such as excessive alcohol and tobacco use. Primary progressive aphasia, which includes the nonfluent and semantic variants, erodes language abilities while leaving other functions relatively intact for longer periods.

Researchers at the Center for Neurodegenerative Diseases and the Aging Brain at the University of Bari Aldo Moro, working with the Pia Fondazione Card. Panico Hospital in Tricase, Italy, set out to determine whether patients with FTD are prone to problematic internet and social media use. Between January 2023 and December 2024, they retrospectively screened consecutive patients diagnosed with mild cognitive impairment or Alzheimer’s disease, behavioral variant FTD, and primary progressive aphasia. Diagnoses followed established consensus criteria, and patients with documented histories of overuse, compulsivity, or dependency on digital technologies were identified through clinical records and caregiver interviews.

To characterize these behaviors rigorously, the team administered a semi-structured interview grounded in the biopsychosocial model of addiction developed by Griffiths, probing six core components of behavioral addiction: salience, mood modification, tolerance, withdrawal symptoms, conflict, and relapse. Overuse was operationally defined as quantitatively excessive time online, often exceeding twelve hours per day, that disrupted daily routines. Compulsivity captured repetitive, ritualistic digital behaviors lacking any functional goal, such as constant nonfunctional scrolling or repetitive video streaming. Dependency required the presence of the six-component addiction profile. Because patients with FTD frequently lack insight into their own behavior, interviews were conducted separately with patients and caregivers.

The results were striking. Of 61 patients evaluated in the FTD group, 14, or 22.9 percent, exhibited problematic internet or social media use. Ten of these had behavioral variant FTD and four had primary progressive aphasia, with three carrying pathogenic genetic variants, two in C9orf72 and one in GRN. By contrast, in a control cohort of 354 patients with Alzheimer’s disease or mild cognitive impairment, only three patients, or 0.8 percent, showed comparable behaviors. The difference between the groups was highly significant, and the contrast with estimated general-population prevalence figures of roughly six to seven percent for problematic internet use reinforces the notion that this is a specific behavioral feature of FTD rather than a general consequence of cognitive decline.

The clinical pictures described in the series are vivid and unsettling. Excessive social media use affected six patients, producing sleep deprivation, religious obsessions, extensive dating app use that led to extramarital affairs and divorce, and hours spent browsing job portals without any genuine intention of seeking employment. Four patients engaged in problematic gaming, with sessions exceeding twelve hours a day and accompanied by social withdrawal and severely disrupted daily routines. Two patients became trapped in binge-watching cycles of repetitive video streaming, and two displayed impulsive online buying that drew them into scams and unnecessary purchases. Half of the problematic-use group showed overlapping behaviors, and social isolation emerged as a pervasive consequence across nearly every case.

Perhaps most telling were the relational dynamics that surrounded these behaviors. Nine of the fourteen patients reported preferring online interactions to face-to-face communication, deepening their withdrawal from family and social life. Six became frustrated and angry when caregivers attempted to limit smartphone access, echoing the agitation commonly seen when substance-dependent individuals are denied their drug of choice. Within the FTD group, patients exhibiting problematic internet use were significantly younger and had an earlier disease onset than those without such behaviors, though education, sex distribution, global cognition as measured by the Mini-Mental State Examination, and overall disease severity did not differ between the groups.

Subgroup analyses hinted at divergent mechanisms driving the behavior across the FTD spectrum. Among patients with behavioral variant FTD, problematic internet use was associated with a younger, early-onset phenotype, potentially reflecting a generational familiarity with digital technology or a specific behavioral trait independent of cognitive decline. Among patients with primary progressive aphasia, by contrast, problematic use correlated with greater disease severity, suggesting that in the language variants these behaviors may emerge as a nonverbal outlet, a way of compensating for the progressive loss of linguistic and social communication abilities as conversation becomes increasingly difficult.

The neurobiological plausibility of these findings is considerable. Previous research indicates that the neural circuits implicated in substance addictions, involving reward processing, impulse control, and decision-making, may also underlie problematic internet-related behaviors. These are precisely the circuits that deteriorate in FTD, particularly in the behavioral variant, where impulsivity, disinhibition, and compulsivity are hallmark features. The parallels between online addiction and more traditional forms of addiction support the hypothesis that both stem from shared neurobiological mechanisms, and the new series adds ecological validity to that framework by documenting real-world cases in a neurodegenerative population.

The authors acknowledge limitations, including the retrospective design, reliance on clinical records and caregiver report, inconsistent recording of specific applications and platforms, and the need for longitudinal research to track how these behaviors evolve as disease progresses. Even so, the societal implications are difficult to ignore. As internet use becomes ever more pervasive, patients with FTD may be especially vulnerable to maladaptive online behaviors that complicate clinical management, deepen social isolation, and increase caregiver burden. The researchers argue that as society adapts to new technologies, clinical frameworks for dementia must evolve in parallel, ensuring that emerging digital-age behavioral patterns are recognized in diagnosis and addressed in care.

Subject of Research: Problematic internet use as an emerging behavioral feature of frontotemporal dementia

Article Title: Problematic Internet Use in Frontotemporal Dementia: A Case Series

Article References: Urso, D., Volpe, G., Valguarnera, A., Vilella, D., Vitulli, A., Gnoni, V., Giugno, A., Rollo, E., Griffiths, M. D., & Logroscino, G. (2026). Problematic Internet Use in Frontotemporal Dementia: A Case Series. Annals of Clinical and Translational Neurology, 13(9), 1947-1950. https://doi.org/10.1002/acn3.70417

Image Credits: AI Generated

DOI: 10.1002/acn3.70417

Keywords: frontotemporal dementia, behavioral variant FTD, problematic internet use, behavioral addiction, primary progressive aphasia, social media, neurodegenerative disease, caregiver burden, compulsivity, digital behavior, Problematic, Internet