Affective Lability (AL) refers to rapid, frequent, and intense fluctuations in expression of emotions, leading to challenges in regulating these shifts and their behavioral effects (Bandyopadhyay Prasanta et al., 2014; Leaberry et al., 2020). Referring to a type of emotional reactivity, AL can incite heightened emotional reactions among individuals when faced with life events reminiscent to past emotional experiences that had a significant impact on them (Almeida et al., 2024). Harvey et al. (1989) describe AL as the tendency to shift between various mood states similar to anger, hypomania, anxiety and depression. The prevalence of AL is found among individuals with diverse mental health disorders such as personality disorders, psychotic disorders, anxiety disorders or obsessive-compulsive disorders (Bandyopadhyay Prasanta et al., 2014; Høegh et al., 2020). In such clinical populations, AL predicts multiple debilitating outcomes, such as disease progression, increased clinical burden and impaired social functioning (Høegh et al., 2022), cognitive deficits (Lee et al., 2012), as well as suicidal ideation and attempts (Wedig et al., 2012). Beyond simply being a feature of several psychiatric disorders, AL is also a difficulty which can occur throughout the lifespan, and which was reported to be highly prevalent in the adult general population (Marwaha et al., 2013). In addition, AL observed in individuals from the general population (without a diagnosed mental disorder) was suggested to be clinically significant, and likely involved in the etiology and early, undiagnosed phases of a range of mental disorders (Broome et al., 2015). Considering that AL is closely linked to the development of mental disorders, a psychometrically sound measure is required to accurately assess the construct in both clinical practice and scientific research.

One comprehensive psychometric tool designed with the aim of evaluating changes across diverse modalities of affective functioning; assessing oscillations from one’s baseline mood (euthymic) to other affective domains, is the Affective Lability Scale (ALS) (Harvey et al., 1989). The original 54-item ALS developed by Harvey et al. (1989) consists of six subscales elaborated to tap into one’s physiological perceptions, subjective experiences and behaviors, including “depression, anxiety, elation, depression/anxiety, anger, and biphasic affect (depression/elation)” (Oliver and Simons, 2004). These subscales are generated for six types of shifts in affect, measured through three dimensions, namely perceptions of behavioral, physiological, and emotional changes associated with cognition (Harvey et al., 1989). To address the ALS’s restricted empirical support concerning its factor structure (Aas et al., 2015) and length, a shorter version of 18 items (ALS-18) was elaborated by Oliver and Simons (2004), comprising a three-factor model of AL. In fact, the initial six conceptual scales were refined through the selection of items that contributed most to each subscale’s internal consistency (Oliver and Simons, 2004). While ensuring subscales retained at least 5 items and maintained a reliability score greater than 0.75, items were excluded following an iterative elimination process, resulting in a 33-item scale (Oliver and Simons, 2004). A factor analysis was then conducted on the latter, identifying seven eigenvalues above 1.0, whereas the scree plot suggested a three-factor solution, best categorized as Anxiety/Depression, Depression/Elation, and Anger (Oliver and Simons, 2004) The scale was further refined by iteratively excluding items based on low item-total correlations maintaining a scale alpha greater than 0.80 with at least five items for each of the three scales, resulting in the 18-item ALS (Oliver and Simons, 2004). The latter scale retains a minimum of two items from the initial ALS’s six scales and shows a strong correlation with the total score of the original ALS (r = 0.94) (Oliver and Simons, 2004).

The development of the ALS and ALS-18 was initially carried out in the English language and has since been adapted and/or validated in numerous other languages including Chinese (Xu et al., 2022), Greek (Kalantzi et al., 2022), and Italian (Contardi et al., 2018). The two latter studies showed good internal reliability and consistency for the entirety of the ALS-18’s factors (Kalantzi et al., 2022; Contardi et al., 2018). In fact, a minimum value of 0.85 for Cronbach’s alpha was found for all three factors. Conversely, the Chinese study (Xu et al., 2022) yielded contrary results; failing to support the three-model factor of the initial validation of the ALS-18 through its factor analysis. Nonetheless, the ALS-18 has not yet been validated in Arabic, to the extent of our knowledge. In addition, ALS is a notable aspect of affect dysregulation on which a restricted number of research has been conducted in Arab contexts (Merhi and Kazarian, 2015). The lack of validated or adapted measures across cultures might pertain to that limit in the literature. Given that affect regulation has a culturally dependent nature (Matsumoto et al., 2008), significant differences might be found between Western and Arab samples regarding AL. The validation of the ALS-18 among an Arabic-speaking sample might give rise to the appropriateness and usefulness of the scale in an Arab setting. Moreover, a shorter version, being further concise in nature, was deemed advantageous in terms of the administration of a scale examining AL among an Arab sample. In fact, the reduction in completion time, costs, and burden on respondents within Arab countries with low to middle income are noteworthy (Fekih-Romdhane et al., 2023) as it increases response rates and data quality (Franke et al., 2013). Given the aforementioned factors, the requirement for an Arabic version of the ALS-18 adequately translated and validated becomes evident. Because AL is commonly experienced in general population adults and is potentially relevant to an array of mental health outcomes, we believe that examining the psychometric properties of the ALS-18 in such a population would be an important contribution to the body of knowledge about psychometric properties of the scale.

Subsequently, the objective of the current study was to assess the psychometric properties of an Arabic Translation of the short form of the ALS (ALS-18) in a non-clinical sample of Arabic-speaking adults from Lebanon. It is hypothesized that the scale in its Arabic version will (1) yield three-factor solution consistent with the original model, (2) show good internal reliability, (3) be invariant across sex groups, and (4) have adequate concurrent and divergent validity.

AL is a broad term employed to allude to emotion fluctuations, and is considered a predictor of various mental health disorders’ features, clinical course, and outcomes, including bipolar, personality or anxiety disorders (Bandyopadhyay Prasanta et al., 2014; Strawbridge et al., 2022). For research purposes as well as preventive and clinical practice, the availability of reliable and valid scales for AL screening is of high importance. In this vein, validating the short form of the ALS in the Arabic language was essential to better assess its practicality within the settings of Arab research. Factually, the administration of short scales requires reduced time and effort from respondents, heightening its convenience among the Arab population contending with a lack of psychometric soundness in tools for research, resources’ shortage, and financial turmoil (Maalouf et al., 2019). The investigation of the psychometric properties of the Arabic ALS-18 conducted through this present study revealed a good reliability and acceptable fit of the ALS-18 three-factor model and second-order model. The latter suggests that the three first-order factors are explained by a higher-order latent factor. Given that the three-factor model already exhibits strong correlations among factors (r > 0.90 across all items) and shows identical fit indices to the second-order model, the latter does not significantly alter the established structure of the three-factor model but rather reinforces the notion that these factors can be integrated into a higher-order construct. Both models were then retained as (1) the three-factor model would be valuable for research studying the subdimensions independently, while (2) the second-order model provides a global ALS-18 score, making it more useful in settings where a single composite score is preferred. The three-factor model in the study conducted by Doğan and Şenormanci (2025) highlighted an acceptable fit, aligning with previous studies on the ALS-18 structure (Broome et al., 2015; Xu et al., 2022; Ali et al., 2021). Findings of Contardi et al. (2018) suggest the presence of a higher-order factor, proposing the computation of a total score, as commonly reported in the literature (Aas et al., 2015; Look et al., 2010). However, the latter research (Contardi et al., 2018) was not conclusive in determining the superiority of the three-factor or hierarchical model. Correlations between the latent factors were high, suggesting substantial shared variance among the affective lability dimensions. This pattern is theoretically consistent with the conceptualization of affective lability as a higher-order construct and provides support for the plausibility of a second-order model representing a global affective lability factor. Although the χ²/df ratio was relatively elevated, this index is well known to be highly sensitive to sample size and model complexity, often leading to statistically inflated values in large samples. In contrast, the incremental fit indices (CFI and TLI), which compare the specified model to a null model and are less affected by sample size demonstrated excellent fit. Together with the low SRMR, these indices indicate that the proposed factor structure provides a robust and meaningful representation of the data despite the elevated χ²/df. Additionally, the scale showed notable concurrent validity, as well as an invariance across sex. In summary, the Arabic version of the ALS-18 is presumed suitable and convenient for screening affective lability among non-clinical Arab populations.

The internal consistency of the Arabic ALS-18 underscored in our results is deemed adequate for the scale’s three dimensions as follows: Anxiety/Depression (AD) (ω = 0.89/α = 0.89), Depression/Elation (DE) (ω = 0.93/α = 0.93), and Anger (ω = 0.90/α = 0.90). Our results are consistent with previously conducted studies investigating the psychometric properties of the 3-factor ALS-18. Among a population of undergraduates, Cronbach’s alpha coefficients obtained in the first study of the development of the ALS Short Form (Oliver and Simons, 2004) were of 0.87 for AD, 0.81 for DE, and 0.82 for Anger. An investigation of the psychometric properties carried out a few years later among individuals with personality disorders and control groups (Look et al., 2010), and within a Chinese population (Xu et al., 2022) yielded similar results, with coefficient alpha, respectively, equaling 0.82 and 0.87 for AD, 0.78 and 0.865 for DE, as well as 0.87 and 0.81 for Anger. The use of McDonald’s ω presents an advantage to the current study, in comparison with only using Cronbach’s alpha coefficient in the foregoing research, as the former coefficient’s effectiveness is considered greater than that of the latter (Crutzen, 2017).

Moreover, our findings underscore an invariance across sex at the metric, scalar and configural levels among an Arabic-speaking sample, highlighting that its factorial structure do not vary between sex in the stated population (Fekih-Romdhane et al., 2023). To the extent of our knowledge, the measurement invariance across sex is scarce in the literature. Indeed, few studies have examined this characteristic, one of which was conducted by Harvey et al. (1989) on the ALS initial 54-item version. The later researchers highlighted differences among sex solely for the depression scale. Similarly, in their research within a Portuguese sample, Almeida et al. found notably higher scores among women than men on the total and subscales’ scores (Almeida et al., 2024). Contardi et al. (2018) contend that the ALS-18’s scores show no association with sex. The present study showed significantly higher scores on AD, DE, and Anger among female respondents in comparison to male respondents. The distinctions in emotional experiences among males and females have indeed been underscored through precedent research, particularly concerning the frequency of negative emotions (Almeida et al., 2024). In fact, Gunnlaugsson et al. (2011) postulates that women have a predisposition to heightened levels of AL. This may explain the increased susceptibility of females to diverse mental health disorders relating to emotions and its fluctuations (Staugaard and Berntsen, 2021; Bradley et al., 2001).

Findings of the present study highlighted a significant association between greater scores on all three ALS-18 dimensions (i.e., depression-anxiety, depression-elation, and anger) and higher depression, anxiety, and stress, indicating that individuals with increased AL would tend to experience high levels of psychological distress. In fact, greater ALS-18 total scores, ALS-18 depression-anxiety, depression-elation, and anger with DASS-8 Depression, Anxiety, and Stress indicate significant and moderate positive associations. These associations suggest that individuals scoring higher on affective lability would tend to report greater scores on the latter variables. Results of the ALS short form validated by Xu et al. (2022) denoted a positive correlation between affective lability, and anxiety and depression. Particularly, stress resistance and adaptability were shown to be positively impacted by the role depression-elation plays (Xu et al., 2022). In their study focused on the psychometric properties of the ALS-18. Contardi et al. (2018) highlighted a significant association between the scale’s general factors and subdimensions, and corresponding instruments assessing difficulties in emotion regulation and depression. These results would also suggest the role of a risk factor that affective lability may play for depression, anxiety, and stress (Xu et al., 2022). Despite the contrasting results from Oliver and Simons (Marwaha et al., 2013) offering no further clarifications, our findings are supported by numerous previous studies (Xu et al., 2022; Contardi et al., 2018; Weibel et al., 2019; Li, 2023). In fact, the association observed can be uncovered by the high comorbidity existing between depression and anxiety (Pollack, 2005). Furthermore, a link between depression and the ALS-18 factor of anger was established, supporting the idea that anger is a prominent feature of depression and significantly contributes to its development (Busch, 2009). Moreover, stress is identified among affective lability, depression, and anxiety (Li et al., 2021).

Additionally, an association between higher ALS depression-anxiety, ALS depression-elation and ALS anger, and lower self-esteem was denoted. Self-esteem has become central in psychology, defined as “the individual’s positive or negative attitude toward the self as a totality” (Rosenberg et al., 1995). It has been shown that poor self-esteem correlates with various mental disorders with affect and emotion regulation emerging as crucial factors (Mouatsou and Koutra, 2023). In other words, self-esteem is associated with the two extensively studied emotion regulation strategies: individuals employing reappraisal to manage emotions tend to exhibit higher self-esteem (Mouatsou and Koutra, 2023). Furthermore, fluctuations in self-esteem and mood are significant characteristics observed in clinical conditions like depression (Kashdan et al., 2006). Individuals with low self-esteem often interpret their daily events as less positive and more influential on their emotional states (Campbell et al., 1991). Our findings showed that, although the correlation between self-esteem and AF was statistically significant (likely due to the large sample size), the magnitude of the correlations was small (in the range of 0.03–0.15). This suggests that self-esteem and AL are not similar constructs, thus attesting for the divergent validity of the scale.

The current study sought to investigate the psychometric properties of the ALS-18 within an Arabic-speaking population, providing evidence of scale’s adequacy for assessing affective lability in Arab settings, particularly among a sample of Arabic-speaking adults residents and citizens of Lebanon. Additional cross-cultural validations of the ALS-18 in various cultural and religious contexts across Arab countries is suggested. Providing scholars access to the Arabic version of the scale may expand study prospects in this topic across various Arab cultural backgrounds and contexts.