102 Chapter 5 contributes to more stable, high impulsivity self-assessments (Borsboom & Cramer, 2013). In the network, this is expected to be evidenced by a strong average interrelatedness between relevant personality components (Cramer et al., 2012). When an idiographic network model captures this stable personality system, these models thus can be used to study individual (differences in) personality structures. However, to successfully capture the individual’s personality structure we require theory about the timescale at which personality dynamics occur, because it is unlikely that John’s restlessness on Monday 10:00 AM will equally cause him to blurt out things at Monday 11:00 AM, Tuesday 10:00 AM or Friday 7:00 PM. Problematically, neither theory nor empirical evidence indicate which timescale should be selected to track the influence of situational change on the personality system. Moreover, personality processes unfold not at singular but at multiple timescales (Hopwood et al., 2022; Wrzus & Roberts, 2017), which further complicates the matter. Neuroticism, for example, emerges from within-day processes such as neurons firing within seconds (Read et al., 2017) or emotional changes by the minute or hour (Verduyn & Lavrijsen, 2015). Similarly, weekly or monthly processes (e.g., a depressive episode or a romantic relationship) and even processes that may fluctuate across decades (e.g., occupational status) all contribute to neurotic behavior at a certain point in time (Jeronimus, 2015). The current state of the personality as a complex system, at any given moment, self-organizes out of interactions between many processes across different timescales (Wrzus & Roberts, 2017; cf. Olthof et al., 2023; Wallot & Kelty-Stephen, 2017; Wijnants, 2014). Dynamic systems approaches to personality even suggest that the personality system and its reactivity to situational features may change within people over time as a consequence of learning and updating self-relevant beliefs (Cervone, 2005; Mischel & Shoda, 1995). Hence, massively varying idiographic personality networks (e.g., Beck & Jackson, 2021) may theoretically be expected as a consequence of either the chosen timescale, changing situations, learning processes, or a combination of all three. Idiographic network variability over time relates to stationarity: an important theoretical assumption about the processes that generate the timeseries from which networks are estimated (Molenaar, 2004). Contemporary network models assume weak stationarity (Bringmann et al., 2018), which means that the time-series used to estimate the network may be variable over time but may not change in how they vary over time. In other words, the dynamic properties of the patterns (average, lag-1 covariance) need to remain stable over time (Manuca & Savit, 1996). Consider a time-series in which
RkJQdWJsaXNoZXIy MTk4NDMw