Major Depressive Disorders are characterized by persistent depressive mood, anhedonia, diminished interests, impaired cognitive abilities (e.g., difficulty concentrating), and autonomic nervous dysfunction (e.g., sleep disorders). These disorders significantly contribute to disability and mortality, primarily through suicide (Abdallah et al., 2018; Otte et al., 2016). According to data from the World Health Organization, Major Depressive Disorders currently rank as the third-largest disease burden globally and are projected to become the primary cause of disease burden by 2030, profoundly impacting the well-being of individuals and families. Moreover, the COVID-19 pandemic has played a substantial role in the increased prevalence of depressive disorders (Chang et al., 2022). Recognizing and intervening early is crucial in preventing the progression of mild depressive symptoms to clinical disorders and addressing other mental health issues.

The Electroencephalogram (EEG) has emerged as a potentially objective tool, surpassing traditional self-reporting and clinical assessment, for detecting and diagnosing preclinical individuals with depressive symptoms (Schumann et al., 2012). Various analytical methods have revealed abnormal neural activity in these individuals. For example, lower alpha and beta power spectra were observed in preclinical individuals with depressive symptoms compared to healthy controls (Lee et al., 2018). Alpha asymmetry in the left and right frontal lobes was a robust indicator for predicting depression (Nusslock et al., 2011; Stewaret and Allen, 2018). In the population without clinical diagnosis, individuals with high depressive symptoms exhibited lower alpha-connection asymmetry compared to those with low symptoms (Imperatori et al., 2019). Additionally, recent studies utilizing microstate analysis and omega complexity have identified altered EEG spatiotemporal complexities in preclinical individuals with depressive symptoms (Qin et al., 2022; Zhao et al., 2022).

Temporal correlations of neuronal oscillations present a promising avenue for detecting early depressive symptoms. Spontaneous neuronal oscillations exhibit intricate temporal structures and are interconnected across thousands of oscillation periods, referred to as long-range temporal correlations (LRTC) (Gaertner et al., 2017). These LRTC patterns, typically assessed through detrended fluctuation analysis (DFA), represent a distinctive feature of ongoing neuronal oscillations in healthy individuals (Linkenkaer-Hansen et al., 2005). They reflect the adaptability of neuronal assemblies, maintaining a delicate balance between stability and flexibility (Huang et al., 2022). Previous studies have substantiated the reliability (Nikulin and Brismar, 2004) and widespread presence of LRTC across various age groups (Berthouze et al., 2010).

The abnormal LRTC of neural oscillations has been observed in clinical depressive patients, although the direction of change – whether increased or reduced – has not been consistent. Some studies reported an inverse association between LRTC and depressed severity, with clinical patients exhibiting decreased or absent LRTC in theta and broadband oscillations (0.5-30 Hz) compared to healthy controls (Hou et al., 2017; Linkenkaer-Hansen et al., 2005; Wang et al., 2016). However, there is also evidence supporting a positive relationship between LRTC exponents and depressive intensity, suggesting that factors such as rumination and psychomotor retardation may contribute to the persistence of LRTC in clinical depressive patients (Lee et al., 2007). This positive association is further supported by the findings that the increased LRTC in clinical patients can be mitigated by mindfulness training or stress reduction training (Gaertner et al., 2017).

Additionally, the resting state hypothesis of depression (RSHD) hypothesis suggests that changes in spontaneous neural activity in clinical depressive patients may lead to abnormal switches between resting and task behavioral states (Gupta et al., 2021). Although supported by limited evidence, this hypothesis suggests a potential link between disrupted resting-state neural dynamics, along with the abnormal rest-task switches, and the manifestation of depressive symptoms (Duncan et al., 2020; Gupta et al., 2021). Importantly, recent studies have begun to explore the relationship between maladaptive negative emotional regulation strategies, a precursor to clinical depression, and LRTC exponents. In preclinical depressed individuals, intricate associations between LRTC exponents and maladaptive strategies like rumination and thought suppression were observed (Bornas et al., 2015; Bornas et al., 2013). These findings imply that aberrant LRTC may precede the onset of clinical depression, potentially impacting the ability to switch between different behavioral states. However, the temporal correlations of neural oscillations in preclinical depressed individuals and the interaction with behavioral states (resting versus task states) remain unclear. This study explored the relationships between temporal correlations of neural oscillations and preclinical depressed symptoms during resting and task states, along with exploring maladaptive negative emotion regulation strategies. Additionally, this study measured the short-range temporal correlations of neural oscillations through cumulative distributions of Lifetimes and Waitingtimes of neuronal oscillations, a metric observed in disorders of consciousness and mild spastic paralysis (Gao et al., 2017; Wei et al., 2023).

A naturalistic task – action video gaming – was employed as task stimuli in this study for two reasons. First, this choice was motivated by the widespread use of naturalistic tasks, such as narratives, movies, virtual reality, and video games, in researching various cognitive domains, including attention, emotion, social cognition, memory, and language (Bottenhorn et al., 2018; Jaaskelainena et al., 2021; Lenormand and Piolino, 2022; Sonkusare et al., 2019; Zhang et al., 2021). Second, action video gaming, characterized by dynamic and diverse stimuli, provides an ecologically valid setting with complex challenges, including visual and auditory elements (Bavelier and Green, 2019). The gaming environment featured moving enemies, expansive battlefields, diverse background scenery, and continuous skill acquisition, creating unpredictable and challenging situations. Participants collaboratively engage in inter-team battles, requiring various cognitive abilities such as perception, visuospatial cognition, attention, working memory, and executive control. This mirrors real-world scenarios, enhancing the study's ecological validity by simulating complex and rapidly changing situations, both sensorimotorly and strategically.

This study specifically selected action video gaming experts as participants for three reasons. First, significant associations between depressive symptoms and action video gaming experience existed (Jiang et al., 2023; Limone et al., 2023; Mun, 2023; Ostinelli et al., 2021). Thus, focusing on action video gaming experts provides an effective and efficient approach to identifying a population with preclinical depressive risk factors. Second, experts were recruited to control for confounders related to video game experience and game proficiency. Irrmischer et al. (2018) reported reduced LRTC from eye-closed resting to focused attention meditation for meditation practitioners, whereas increased LRTC was observed for meditation-naïve participants (Irrmischer et al., 2018a). This suggests that long-term experience in a specific domain may influence LRTC. Third, action video gaming experts were familiar with the task stimuli employed in this study – action video gaming. Consequently, from a logistical perspective, they required no prior training to perform the designated task. Additionally, this selection criterion was implemented to mitigate the impact of novel experiences on the results.

Building on the findings of previous studies and recognizing the marked variations between different behavioral states (Irrmischer et al., 2018a; Irrmischer et al., 2018b), our hypothesis posits that substantial alterations in the temporal correlations of neural oscillations, spanning all frequency bands, would be evident from resting to task states. Particularly, alpha oscillations (8-13 Hz) are implicated in inhibiting task-irrelevant information, a process impaired in clinically depressed patients (Klimesch, 2012; LeMoult and Gotlib, 2019). Previous studies have demonstrated the critical role of alpha oscillations in predicting preclinical depressed populations (Imperatori et al., 2019; Lee et al., 2018; Nusslock et al., 2011; Stewaret and Allen, 2018). Consequently, we hypothesize the presence of significant correlations between temporal correlations of neural oscillations and depressive symptoms in the alpha band during resting state, influencing the transitions from resting to task states (Duncan et al., 2020; Gupta et al., 2021). Furthermore, even though individuals with depressed moods and maladaptive emotion regulation strategies may find themselves immersed in situations with unrestricted attention, their performance during focused tasks is anticipated to parallel that of healthy controls (LeMoult and Gotlib, 2019). Thus, we anticipate no significant correlations between temporal correlations of neural oscillations and depressive symptoms during the task state. Given the close correlations between maladaptive negative emotion regulation strategies and depressive symptoms (Young et al., 2019), we expect to replicate similar results concerning maladaptive strategies.