Childhood adversity and mind wandering: the mediating role of cognitive flexibility and habitual tendencies

ABSTRACT

Background: Initial evidence proposes that exposure to childhood adversity may induce avoidance or withdrawal behaviour. However, it remains unclear whether childhood adversity results in avoidance from externally directed thinking to both deliberate and spontaneous mind wandering, i.e. intentionally or unintentionally diverting attention from ongoing task to task-independent thoughts.

Objective: To assess the associations between childhood adversity, and mind wandering, and to evaluate the mediating roles of cognitive flexibility, and habit tendencies.

Methods: A total of 601 Chinese subjects (378 females, M_age = 19.37) participated in the current study. The participants completed a series of questionnaires including demographics, childhood maltreatment, cognitive flexibility, habitual tendencies, and mind wandering.

Results: Hierarchical regression analyses showed childhood adversity, the control facet of cognitive flexibility, and the automaticity facet of habitual tendencies had significant contributions to deliberate mind wandering (β = 0.10, β = −0.40, and β = 0.06) and spontaneous mind wandering (β = 0.09, β = −0.28, and β = 0.07). Serial mediation analyses revealed that the control and automaticity partially mediated associations between childhood adversity and mind wandering (deliberate mind wandering: 95% CIs = [0.037 0.078], and spontaneous mind wandering: 95% CIs = [0.023, 0.062]).

Conclusions: The findings underscore the pivotal role of mediators in delineating the relationship between childhood adversity and mind wandering in everyday life. Interventions geared toward augmenting the control component of cognitive flexibility and regulating the automatic component of habitual tendencies show the potential to ameliorate the propensity of individuals affected by childhood adversity to disengage cognitively from the present moment.

HIGHLIGHTS

• Individuals with greater childhood adversity have higher deliberate and spontaneous mind wandering.

• Low control and high automaticity contribute to mind wandering.

• The control and automaticity partially mediate the associations between childhood adversity and mind wandering.

Antecedentes: La evidencia inicial propone que la exposición a la adversidad infantil puede inducir conductas de evitación o retraimiento. Sin embargo, aún no está claro si la adversidad infantil da como resultado la evitación del pensamiento dirigido externamente a la divagación mental tanto deliberada como espontánea, es decir, desviar intencionalmente o no la atención de la tarea en curso hacia pensamientos independientes de la tarea.

Objetivo: Evaluar las asociaciones entre la adversidad infantil y la divagación mental, y evaluar las funciones mediadoras de la flexibilidad cognitiva y las tendencias de los hábitos.

Métodos: Un total de 601 sujetos chinos (378 mujeres, M_edad = 19.37) participaron en el estudio actual. Los participantes completaron una serie de cuestionarios que incluían datos demográficos, maltrato infantil, flexibilidad cognitiva, tendencias habituales y distracciones mentales.

Resultados: Los análisis de regresión jerárquica mostraron que la adversidad infantil, la faceta de control de la flexibilidad cognitiva y la faceta de automaticidad de las tendencias habituales tuvieron contribuciones significativas a la divagación mental deliberada (β = 0.10, β = −0.40 y β = 0.06) y a la divagación mental espontánea (β = 0.09, β = −0.28 y β = 0.07). Los análisis de mediación en serie revelaron que el control y la automaticidad mediaron parcialmente las asociaciones entre la adversidad infantil y la divagación mental (divagación mental deliberada: IC del 95% = [0.037, 0.078], y divagación mental espontánea: IC del 95% = [0.023, 0.062]).

Conclusiones: Los hallazgos subrayan el papel fundamental de los mediadores a la hora de delinear la relación entre la adversidad infantil y la divagación mental en la vida cotidiana. Las intervenciones orientadas a aumentar el componente de control de la flexibilidad cognitiva y regular el componente automático de las tendencias habituales muestran el potencial de mejorar la propensión de los individuos afectados por la adversidad infantil a desconectarse cognitivamente del momento presente.

KEYWORDS:

• Mind wandering
• childhood adversity
• cognitive flexibility
• habitual tendencies
• serial mediation effect

PALABRAS CLAVE:

• Mente divagada
• adversidad infantil
• flexibilidad cognitiva
• tendencias habituales
• efecto de mediación en serie

1. Introduction

Childhood adversity refers to adverse experiences or early-life stress involving abuse and neglect for both physical and mental facets in childhood and has profound influences on the human brain and mental health resulting in problematic behaviours and mood disorders, such as suicidal behaviour, addictive behaviour, depression and anxiety (Luo et al., 2022; Teicher et al., 2016; Teicher & Samson, 2016). Recent studies have presented evidence suggesting that childhood adversity may foster habitual behaviours akin to those observed in animal studies (Ersche et al., 2017; Gordon et al., 2020; Laguna et al., 2022; Lomanowska & Kraemer, 2014; Patterson et al., 2019; Xu et al., 2022; Zhou et al., 2020). These findings propose the presence of an automatic, inflexible, and stimulus-response-based coping mechanism in response to childhood adversity may contribute to avoiding harm. However, excessive propensity for habitual behaviour is a risk factor for addiction (Everitt & Robbins, 2016; Wood & Runger, 2016), and maladaptive avoidance habits are associated with negative health outcomes (Patterson et al., 2019; Sebold et al., 2019).

In addition to the habitual behavioural avoidance, the phenomenon of diverting attention away from the external world also holds considerable intrigue. Escaping reality (e.g. games and drug) is also a coping strategy in response to a stressful environment (Chen & Chang, 2019; Everitt & Robbins, 2016; Hui et al., 2019). A study reported that individuals with childhood adversity represented more lapses in monitoring their reasoning and discourse when speaking about relevant experiences (Marcusson-Clavertz et al., 2017), which indicated the occurrence of mind wandering. Mind wandering, a mental state shifting attention from ongoing task to task-unrelated thoughts, engages in a major part of human life (Smallwood & Schooler, 2015). Nevertheless, the connection between childhood adversity and the inclination toward engaging in mind wandering, along with the potential pathways linking these variables within the non-clinical population, remains unclear. This gap impedes our ability to elucidate the origins of these associations within the clinical population. Consequently, delving into this relationship among healthy individuals will yield pertinent evidence crucial for guiding subsequent subclinical and clinical inquiries.

Due to whether intention engages in mind wandering or not, mind wandering divides into two types i.e. spontaneous and deliberate mind wandering (Seli et al., 2019). These two distinct subtypes of mind wandering are influenced by different factors. Spontaneous mind wandering is a default and unintentional state capturing a bias for cognitive resources to internal thinking, and more influenced by the amount of unused resources available and control ability (Seli et al., 2016; Smallwood & Schooler, 2015; Thomson et al., 2015). Moreover, spontaneous mind wandering is positively related to attention-deficit/hyperactivity disorder (ADHD; Seli et al., 2015), depression (Hoffmann et al., 2016), obsessive-compulsive disorder (OCD; Seli et al., 2017) and post-traumatic stress disorder symptoms (PTSD; Brosowsky et al., 2022). In contrast, deliberate mind wandering is an intentional shifting from outside world or ongoing tasks to internal thinking (Carriere et al., 2013; Seli et al., 2016; Seli et al., 2019). The level of motivation and interest is important for deliberate mind wandering (Seli et al., 2016; Smallwood & Schooler, 2015; Thomson et al., 2015). In addition, deliberate mind wandering is negatively related to PTSD symptoms (Brosowsky et al., 2022).

From the resource-control model, the extra executive resources released by automatic processing might be absorbed by mind wandering (Thomson et al., 2015). This theory proposes a relationship between the incidence of mind wandering and a gradual decline in motivation and/or effort to sustain attention on the current task. In other words, as executive resources become less inclined towards allocating themselves to the task, they are more likely to be channelled towards mind wandering (McVay & Kane, 2012; Thomson et al., 2015; Unsworth & McMillan, 2013). Habit behaviour refers to automatic responses activated by a routine context, or associated cues after a person repeatedly performs certain actions (Wood & Runger, 2016). The process of habit formation entails a shift from goal-directed behaviour towards stimulus-response association, thereby reducing the cognitive load and promoting efficient and prompt responses (Wood & Runger, 2016). Recent studies propose that the construct of habit tendencies consists of two components including automaticity and routine (Ersche et al., 2017; Ersche et al., 2019). Routines are serial action patterns that involve regularity in a familiar daily context, and automaticity are automatic behaviour patterns activated by associated cues with a lack of control and awareness. Based on the above model and evidence, it can be argued that automatic processing resulting from habituation can free up executive resources, which can then be utilized by mind wandering.

Mind wandering may also be regulated by cognitive flexibility, a key part of executive control (Deng et al., 2022; Hyafil et al., 2009; Preiss, 2022; Wong et al., 2023). For instance, Deng et al. (2022) show that higher mind wandering tendencies were associated with lower cognitive flexibility, that is the ability to switch cognitive strategies to adapt to changing environments. The successful implementation of goal-directed behaviour and flexible adaptation to changes in the outcome of behavioural responses requires (among other functions) cognitive flexibility. High levels of cognitive flexibility are a protective factor that help individuals to adaptively cope with stressful life events and environment (Murphy et al., 2012). More rigid and less flexible responses have been observed in individuals following exposure to high levels of childhood adversity, such that these individuals had an impaired ability to switch from learned behavioural patterns to new action sequences (Harms et al., 2018).

Therefore, childhood adversity may represent a potential mechanism linking to mind wandering by cognitive and behavioural rigidity, based on the accumulating evidence (Laguna et al., 2022; Shan et al., 2023; Tisborn et al., 2023). In these cases, faster automatic responses and higher anxious arousal may help to avoid stressful events in the context of childhood adversity (Patterson et al., 2019). Moreover, the liberation of executive resources may foster the occurrence of mind wandering as a means to divert attention from distressing or uncomfortable realities. We hypothesize that cognitive flexibility and habit tendencies could mediate the association between childhood adversity and mind wandering via serial multiple mediation effects (Figure 1). Against this background the present study was performed to determine (1) the relationships between childhood adversity, cognitive flexibility, habit tendencies and mind wandering, (2) and the influential paths that mediate the associations between childhood adversity and mind wandering. In our serial multiple mediation analyses, we delineated cognitive flexibility and habitual tendencies as potential mediators that could explain the influence of childhood adversity on individual differences in mind wandering.

Figure 1. Hypothetical serial multiple mediation model (A) and the serial multiple mediation model in statistical diagram form with two mediators (B).

2. Method

2.1. Participants

An online platform (Chinese online survey platform, https://www.wjx.cn) was used to conduct the assessments and the collection link was deployed to campus life and study related mailing lists and online discussion groups to recruit college students. The link was distributed by members of our research team. Given that the application of a priori sample size determination in Structural Equation Modeling (SEM) is still a challenge (Wolf et al., 2013), a pragmatic approach was adopted for determining the sample size. The estimation of sample size in this study adhered to the rule of thumb, namely maintaining a ratio of 5:1 between the sample size and the number of free parameters at least (Bentler & Chou, 1987). In addition, post-hoc power analyses were conducted for each model with pwrSEM (Wang & Rhemtulla, 2021). A total of n = 689 individuals whose identity remained anonymous participated in the online assessments. Participants were required to be ≥ 18 years old and Chinese (native speakers) and free of a current or a history of a mental disorder. Participants from both sexes were included, and no restrictions with respect to employment or relationship status were imposed. As recommended by Meade and Craig (2012) an attention check item was included to identify careless participants. Additional items assessed current, or history of mental disorders and medication use to validate the exclusion criteria. To focus on general population, exclude invalid and uncompleted data, and participants who has current or historical mental disorder or usage of psychiatric medication. Finally, a sample of n = 601 (378 females) participants with M_age = 19.37 (SD = 1.31, age ranging = 18–28) was included in subsequent analyses (demographic Information see Table 1). The study procedures had full approval by the local ethics committee.

Table 1. Correlations between variables.

	Mean (SD)	1	2	3	4	5	6	7	8
1. Age	19.37 (1.31)	–
2. CTQ	40.66 (12.94)	−0.05	–
3. Alternatives	45.96 (8.69)	0.14***	−0.06	–
4. Control	22.81 (4.74)	0.11**	−0.31***	0.22***	–
5. Routine	53.45 (10.41)	−0.01	0.15***	0.15***	−0.31***	–
6. Automaticity	33.48 (9.01)	−0.11**	0.14***	0.04	−0.34***	0.58***	–
7. MW-D	12.49 (6.03)	−0.06	0.33***	−0.06	−0.44***	0.27***	0.28***	-
8. MW-S	12.84 (6.22)	−0.12**	0.29***	0.00	−0.32***	0.22***	0.24***	0.75***	-

Note. CTQ: Childhood Trauma Questionnaire; MW-D: Mind Wandering: Deliberate scale; MW-S: Mind Wandering: Spontaneous scale, *p < .05, **p < .01, ***p < .001.

2.2. Questionnaires

In line with the aims of the present study all participants were administered the Childhood Trauma Questionnaire (CTQ; Bernstein et al., 2003), the Cognitive Flexibility Inventory (CFI; Dennis & Vander Wal, 2010), the Creature of Habit Scale (COHS; Ersche et al., 2017) and the Mind Wandering: Deliberate (MW-D) and Mind Wandering: Spontaneous (MW-S) scales (Carriere et al., 2013). The CTQ is a validated measure to assess the level of exposure to childhood adversity experiences. The CTQ includes 28 self-report items spanning five kinds of adverse experiences (emotional abuse, physical abuse, sexual abuse, emotional neglect, physical neglect) by means of five-point Likert scales ranging from 1 (never true) to 5 (very often true). In line with previous studies, we used the CTQ total score (Ersche et al., 2017; Liu et al., 2021). Higher scores reflect higher exposure to childhood adverse events. The validated Chinese CTQ version was used in the present study (Zhao et al., 2005).

The previously validated Chinese version of CFI was used to assess the cognitive flexibility (Dennis & Vander Wal, 2010; Wang et al., 2016), which requires participants to rate 20 self-report terms on a five-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree). The CFI includes two subscales to measure cognitive flexibility: control subscale, (1) the tendency to perceive difficult situations as controllable, and alternatives subscale measures, alternatives subscale, (2) the ability to perceive multiple alternative explanations for life occurrences and human behaviour and generate multiple alternative solutions to difficult situations. Higher scores indicate higher cognitive flexibility with respect to alternatives or control.

To measure the habitual tendencies, the previously validated Chinese version of the COHS was employed in the present investigation (Ersche et al., 2017; Zhou et al., 2020). The COHS includes 27 items categorized into two subscales automaticity (11 items) and routine (16 items). Specifically, automaticity encompasses automatic responses triggered by specific contexts, whereas the routine subscale encompasses a series of familiar action patterns that involve regularity and are likely to be implemented on a daily basis (Ersche et al., 2017). Participants state their level of agreement for each of the 27 items on a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), with higher scores indicating higher levels of automaticity or routine behaviour, respectively.

Deliberate and spontaneous mind wandering were assessed by the validated Chinese version of the Mind Wandering: Deliberate (MW-D) and Mind Wandering: Spontaneous (MW-S) scales (Carriere et al., 2013; Gao et al., 2018). Each scale has 4 items with a seven-point Likert scale ranging from 1 (rarely) to 7 (a lot). The high scores of each scale indicate a greater tendency to engage in mind wandering deliberately or spontaneously in everyday life.

Reliability of the coefficients for all scales was high enough to provide further support for satisfactory measurement precision of all scales, i.e. CTQ (Cronbach’s α = 0.86, McDonald’s ω = 0.93), CFI (alternatives: Cronbach’s α = 0.94, McDonald’s ω = 0.94, control: Cronbach’s α = 0.86, McDonald’s ω = 0.87), COHS (routine: Cronbach’s α = 0.84, McDonald’s ω = 0.84, automaticity: Cronbach’s α = 0.84, McDonald’s ω = 0.85), MW-D (Cronbach’s α = 0.89, McDonald’s ω = 0.89), and MW-S (Cronbach’s α = 0.88, McDonald’s ω = 0.88).

2.3. Data analysis

Statistical analyses were conducted using RStudio (https://www.rstudio.com), a professional software for data science based on R and R packages (https://www.r-project.org). To avoid a common method bias, leading to common variance due to the measurement method or the instrument – rather than the actual predispositions of individuals that the measurements are thought to examine – we performed Harman's one-factor test on all items of the questionnaires in the current study (Podsakoff et al., 2003). The first single factor accounted for 17.522% variance, arguing against the presence of significant common method bias. Subsequently, Pearson's correlation coefficients were computed to assess the relationships among variables, encompassing age, CTQ, alternatives, control, routine, automaticity, and MW-D/S. To determine the contributions of each variable and identify the most important variables for future mediation analyses, we conducted separate hierarchical regression analyses. In the first step, age and sex were entered as control variables. In the second step, childhood adversity was entered as a core variable. In the third step, cognitive flexibility was entered as a core variable. In the fourth step, habitual tendencies were entered as a core variable. By entering the variables step-by-step, we were able to determine which variables were significant based on the significance of the variance and the significance of the beta coefficients. To confirm no multicollinearity in the model, the variance inflation factor (VIF) was calculated for each predictor variable. Finally, based on a prior hypotheses and hierarchical regression analysis, we assessed the serial multiple mediation model (Figure 1(B)) to determine the mediation effects of cognitive flexibility and habitual tendencies on the association between childhood adversity and deliberate and spontaneous mind wandering by routines implemented in the lavaan package (https://www.lavaan.ugent.be). 95% confidence internals (CIs) based on 5000 bootstrapping, and Sobel test were used to test the significance of indirect effects (Preacher & Hayes, 2008). For all analyses, p values of less than 0.05 were considered statistically significant. Post-hoc power analyses were performed with sample size N = 601, alpha level = 0.05, and 5000 simulations.

3. Results

3.1. Correlation coefficients of scales

The results indicate that CTQ scores exhibited a significant negative correlation with the control subscale of CFI (r = −0.31, p < .001), revealing that individuals who experienced fewer adverse childhood events demonstrated a greater feeling of control. Conversely, CTQ scores demonstrated significant positive correlations with COHS's Routine (r = 0.15, p < .001) and Automaticity (r = 0.14, p < .001) subscales, implying that individuals with more adverse childhood experiences tended to exhibit more habitual behaviours. Furthermore, CTQ scores were significantly positively associated with MW-D/S scales (MW-D, r = 0.33, p < .001; MW-S, r = 0.29, p < .001), suggesting that greater adverse childhood experiences were associated with increased instances of mind wandering. Age exhibited significant correlations with specific measures, including both subscales of CFI (alternatives, r = 0.14, p < .001; control, r = 0.11, p < .01), signifying that older age was associated with increased cognitive flexibility. Furthermore, age demonstrated a significant negative correlation with the automaticity subscale of COHS (r = −0.11, p < .01), suggesting that older age was linked to a greater inclination towards automatic habitual tendencies. Additionally, age exhibited a significant negative correlation with MW-S (r = −0.12, p < .01), indicating that older age was associated with an increased propensity for spontaneous mind wandering. In addition, CFI’s alternatives subscale was not significantly correlated with COHS’s automaticity subscale, MW-D and MW-S (details are provided in Table 1).

3.2. The contributions of childhood adversity, cognitive flexibility and habitual behaviours on deliberate and spontaneous mind wandering

Regarding to the contribution of age, sex, childhood adversity, cognitive flexibility, and habitual behaviours on the prediction of deliberate and spontaneous mind wandering, two hierarchical regression analyses were respectively performed (Tables 2 and 3). The VIF values for age (1.05), sex (1.13), CTQ (1.18), control (1.40), alternatives (1.14), automaticity (1.59) and routine (1.62) demonstrated a level of multicollinearity below 5, suggesting a low presence of intercorrelation.

Table 2. Hierarchical regression analysis on MW-D considering age, sex, CTQ, CFI, and COHS.

	β (SE)	T	R^2	Δ R^2
Step 1
Age	−0.22 (0.19)	−1.17	0.006
Sex	0.69 (0.51)	1.34
Step2
Age	−0.11 (0.18)	−0.61	0.122	0.116
Sex	1.45 (0.49)	2.96**
CTQ	0.16 (0.02)	8.87***
Step 3
Age	−0.03 (0.17)	−0.18	0.234	0.112
Sex	0.43 (0.48)	0.90
CTQ	0.10 (0.02)	5.67***
Control	−0.47 (0.05)	−9.33***
Alternatives	0.03 (0.03)	1.16
Step 4
Age	0.01 (0.17)	0.03	0.253	0.019
Sex	0.38 (0.47)	0.81
CTQ	0.10 (0.02)	5.46***
Control	−0.40 (0.05)	−7.43***
Alternatives	0.01 (0.03)	0.33
Automaticity	0.06 (0.03)	2.01*
Routine	0.04 (0.03)	1.82

Note. CTQ: Childhood Trauma Questionnaire; MW-D: Mind Wandering: Deliberate scale, * p < .05, ** p < .01, *** p < .001.

Table 3. Hierarchical regression analysis on MW-S considering age, sex, CTQ, CFI, and COHS.

	β (SE)	T	R^2	Δ R^2
Step 1
Age	−0.57 (0.20)	−2.90**	0.014
Sex	−0.15 (0.53)	−0.28
Step2
Age	−0.47 (0.19)	−2.49*	0.094	0.08
Sex	0.51 (0.51)	0.99
CTQ	0.14 (0.02)	7.29***
Step 3
Age	−0.44 (0.18)	−2.39*	0.155	0.061
Sex	−0.16 (0.52)	−0.32
CTQ	0.10 (0.02)	4.94***
Control	−0.35 (0.05)	−6.43***
Alternatives	0.06 (0.03)	2.16*
Step 4
Age	−0.40 (0.18)	−2.17*	0.171	0.016
Sex	−0.22 (0.51)	−0.44
CTQ	0.09 (0.02)	4.76***
Control	−0.28 (0.06)	−4.89***
Alternatives	0.04 (0.03)	1.47
Automaticity	0.07 (0.03)	2.19*
Routine	0.03 (0.03)	1.09

Note. CTQ: Childhood Trauma Questionnaire; MW-S: Mind Wandering: Spontaneous scale, *p < .05, **p < .01, ***p < .001.

For deliberate mind wandering, age and sex were included in step 1 as control variables, accounting for percentage of variance of 0.6%, F(2,598) = 1.84, p = .16. In step 2, exposure to childhood adversity increased the percentage of variance explained by 11.6%, F(1,597) = 78.69, p < .001. In step 3, both control and alternatives were included and increased the percentage of variance explained by 11.2%, F(2,595) = 43.59, p < .001. Finally automaticity and routine accounted for 1.9%, F(2,593) = 7.63, p < .001. However, regarding individual predictors, results indicated that only greater childhood adversity (β = 0.10), lower control (β = −0.40) and higher automaticity (β = 0.06) were significantly associated with higher deliberate mind wandering.

For spontaneous mind wandering, age and sex were included in step 1 as control variables, accounting for percentage of variance of 1.4%, F(2,598) = 4.20, p < .05. In step 2, exposure to childhood adversity increased the percentage of variance explained by 8%, F(1,597) = 53.12, p < .001. In step 3, both control and alternatives were included and increased the percentage of variance explained by 6.1%, F(2,595) = 21.25, p < .001. Finally automaticity and routine accounted for 1.6%, F(2,593) = 5.81, p < .01. However, regarding individual predictors, results indicated that being younger (β = −0.40), greater childhood adversity (β = 0.09), lower control (β = −0.28) and higher automaticity (β = 0.07) were significantly associated with higher spontaneous mind wandering.

3.3. Serial mediation effects of cognitive flexibility and habitual behaviours on associations between childhood adversity and mind wandering

In conjunction with previous discoveries (Zhou et al., 2020) and the aforementioned outcomes derived from hierarchical regression analyses, the consecutive multiple mediation models exclusively centre on elucidating the mediation effects of the control component in cognitive flexibility and the automaticity component in habitual behaviours.

The serial multiple mediation model used in this study revealed significant indirect effects of control (95% CIs = [0.031 0.066], p < .001, Table 4), and both control and automaticity (95% CIs = [0.002 0.012], p = .008) on the association between childhood adversity and deliberate mind wandering. However, not all indirect paths showed significant effects. Both confidence interval and significance indicated the path from childhood adversity to automaticity to deliberate mind wandering was not significant (95% CIs = [−0.003 0.009], p = .385). Furthermore, there was a significant direct effect in the model (Figure 2(A)), suggesting a partial mediation effect. In addition, a similar pattern for the spontaneous mind wandering was observed, suggesting significant indirect effects of control (95% CIs = [0.016 0.050], p < .001, Table 4), and both control and automaticity (95% CIs = [0.002 0.013], p = .012) on the association between childhood adversity and spontaneous mind wandering. The path from childhood adversity to automaticity to spontaneous mind wandering was also not significant (95% CIs = [−0.004 0.010], p = .385). Moreover, the significant direct effect reflected a partial mediation model (Figure 2(B)).

Figure 2. Serial multiple mediation models and mediation effects of control and automaticity on the relationship between childhood adversity (CTQ) and (A) deliberate mind wandering or (B) spontaneous mind wandering. Estimated coefficients are presented with standard errors in parentheses. c’: direct effect, c: total effect, * p < .05, ** p < .01, *** p < .001. CTQ: Childhood Trauma Questionnaire, MW-D: deliberate mind wandering scale, MW-S: spontaneous mind wandering scale.

Table 4. Indirect effects of serial multiple mediation models.

Paths	Estimate	SE	Lower CI	Upper CI	z	p	power
CTQ->Control->MW-D	.047	.009	.031	.066	5.298	<.001	1.00
CTQ-> Automaticity->MW-D	.003	.003	−.003	.009	.869	.385	.08
CTQ->Control->Automaticity->MW-D	.006	.002	.002	.012	2.649	.008	.94
Total indirect effect	.056	.011	.037	.078	5.306	<.001	1.00
CTQ->Control->MW-S	.031	.009	.016	.050	3.678	<.001	1.00
CTQ-> Automaticity->MW-S	.003	.003	−.004	.010	.869	.385	.08
CTQ->Control->Automaticity->MW-S	.007	.003	.002	.013	2.527	.012	.94
Total indirect effect	.041	.010	.023	.062	4.131	<.001	1.00

Note. CTQ: Childhood Trauma Questionnaire; MW-D: Mind Wandering: Deliberate scale; MW-S: Mind Wandering: Spontaneous scale; SE: standard error; Lower CI: lower-level confidence interval; Upper CI: upper-level confidence interval.

Summarizing, the level of cognitive flexibility, specifically the control facet, mediated the positive association between childhood adversity and mind wandering, while the automaticity facet of habitual behaviour partially mediated the association between control and mind wandering. In addition, automaticity was not the mediator between childhood adversity and mind wandering. These findings highlight the importance of the control facet of cognitive flexibility linking childhood adversity and mind wandering. In additional analyses including age and sex as confounding variables, the findings from both models were consistent with the main analyses.

4. Discussion

The objective of this study is to investigate the effect of childhood adversity on the human mind, particularly with regard to mind wandering, as well as the underlying pathways that mediate this relationship. This research provides evidence for potential pathways by which the control component of cognitive flexibility and the automaticity component of habitual tendencies sequentially mediate the associations between childhood adversity and both subtypes of mind wandering, i.e. deliberate and spontaneous mind wandering. It was also observed that the control component of cognitive flexibility exhibits a negative association with mind wandering in adulthood, thus underscoring its role as a mitigating factor. These findings suggest that childhood adversity may have a lasting impact on both deliberate and spontaneous mind wandering, partially mediated by a diminished ability to perceive self-control over behaviours and the environment and increased automatic processing. This study reveals the effect of childhood adversity on mind wandering, which may be related to resilience or psychopathological development.

The present study confirms the contributions of childhood adversity, cognitive flexibility and habit tendencies on deliberate and spontaneous mind wandering in the non-clinical population, respectively. The hierarchical regression analyses reveal that only childhood adversity, automatic behavioural patterns, and the control facet of cognitive flexibility contributed to the prediction of deliberate and spontaneous mind wandering. These findings confirmed a priori hypothesis that higher exposure to childhood adversity is associated with increased mind wandering in daily life, and are in line with the resource-control model (Thomson et al., 2015). Our research provides preliminary evidence supporting a link between childhood adversity and mind wandering in the non-clinical population. These findings suggest an adaptive cognitive response by individuals who have experienced childhood adversity to avoid painful environments. It is noteworthy that mind wandering exhibited a positive correlation with the control facet of cognitive flexibility, which pertains to the capacity to exercise self-control over behaviours and surroundings, rather than the ability to generate and perceive multiple solutions (Dennis & Vander Wal, 2010). It suggests that if individual do not perceive or believe in their ability to control the environment, their cognitive resources allocation and even intention tend to mind wandering. Together with the formation of automatic responses both deliberate and spontaneous mind wandering increased.

Our sequential multiple mediation models unveil that the connections between childhood adversity and mind wandering, encompassing both deliberate and spontaneous types, were partly mediated by the cognitive flexibility's control facet and automatic responses within habitual behaviours. The pathways from childhood adversity to the development of automatic behavioural patterns align with a growing body of evidence indicating that chronic stress increases an individual's vulnerability to the emergence of dysregulated habits (Enoch, 2011; Uhart & Wand, 2009). Cognitive flexibility represents an important ability to adaptively cope with stressful life events (Murphy et al., 2012). A previous study found that the control ability totally mediated the association between childhood adversity and automaticity (Zhou et al., 2020). The level of perceiving the controllable environment may be related to the intentional control of individual behaviour. However, childhood adversity increases the uncertainty and uncontrollability of the early environment, then reduces the intention in behavioural responses to external stimuli. In addition, the finite cognitive resources in humans establish an inherent equilibrium between various internal and external cognitive tasks or processes. In other words, there is a limitation on the number of tasks that can be effectively supported simultaneously (Esterman & Rothlein, 2019; Thomson et al., 2015). The cognitive resource allocation, therefore, is critical for adjusting weights among different cognitive tasks, even internal and external cognitive processes, including mind wandering.

In conclusion, compromised cognitive flexibility, particularly in its control facet, can lead to the allocation of cognitive resources and the shift of attentional control from the external to the internal realm. These potential mechanisms provide an explanation for the observed findings that both deliberate and spontaneous mind wandering are influenced by childhood adversity, with mediation occurring through the control facet of cognitive flexibility and the automaticity of habitual tendencies. The level of control can affect the degree to which an individual is driven by the stimulus-response mode, thereby affecting the occurrence of mind wandering in a chain way. While the current study primarily examines the non-clinical population, it nonetheless offers valuable clinical insights into preventive strategies for depression and anxiety in adulthood. This is because both deliberate and spontaneous mind wandering have been found to be associated with psychiatric disorders (Brosowsky et al., 2022; Hoffmann et al., 2016; 2018; Seli et al., 2015). In summary, we believe that based on the current findings, there is value in targeting individual control or reducing rigid behaviour.

The present findings should be interpreted within the context of certain limitations. Firstly, the study utilized a cross-sectional design, making it challenging to establish strict causal relationships. To obtain more robust evidence regarding the lasting effect of childhood adversity, a longitudinal design within a developmental cohort study would be beneficial. Furthermore, although the current sample had a restricted age range, we observed a negative correlation between age and spontaneous mind wandering, but not deliberate mind wandering. To accurately estimate the impact of age, it is essential to replicate these findings in larger cohorts from both the general population and clinical settings, encompassing a wider age range. While our study offers insights into the ramifications of childhood adversity, it is imperative to emphasize the necessity for additional validation of pertinent variables within clinical cohorts. In the future, the amalgamation of models across health and clinical populations will substantively enhance our comprehension of the clinical ramifications associated with cognitive flexibility, habitual tendencies, and mind wandering.

In this study, we utilized hierarchical regression analysis and serial multiple mediation analysis to explore the influences of adverse childhood experiences, cognitive flexibility, and habitual tendencies on deliberate and spontaneous mind wandering. Moreover, we examined whether the relationship between childhood adversity and mind wandering in adulthood is mediated by cognitive flexibility and habitual tendencies. Consistent with our hypothesis, the control facet of cognitive flexibility and the sequential involvement of automatic behavioural patterns served as mediators in the association between childhood adversity and both deliberate and spontaneous mind wandering.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability

The study related data can be obtained from OSF (https://osf.io/kwser).

Additional information

Funding

This work was supported by National Natural Science Foundation of China [grant number 32300905]; Natural Science Foundation of Sichuan Province [grant number 2023NSFSC1185]; Sichuan Normal University [grant number XJ20220113].