Daan Hulsmans

120 Chapter 5 persons (Roberts & Mroczek, 2008) and these change trends differ betweenpersons (Schwaba & Bleidorn, 2018). Given this, theory should inform research about the duration of the timeframe. Secondly, within any timeframe, we need to know at which timescale(s) the dynamics should be summarized in a network. By only estimating contemporaneous networks (i.e., the same day), we refrained from attributing temporality to lagged effects in networks. Within-day dynamics, however, remain hidden due to diary surveys not being momentary (e.g., there can be multiple impulsive moments during a day). Nevertheless, temporality manifests within a myriad of different short and long timescales, as evidenced by the autocorrelation functions in Appendix C. Not estimating lagged effects is not a solution to the non-stationarity problem per se, but at least avoids (implicitly) suggesting that dynamics manifest on one (lag-1) timescale only. Instead, the dynamics of complex systems interact across multiple timescales. Many fast and slow processes are interdependent and, in interaction with their environment, lead to the emergence of behavior (Olthof et al., 2023; Wallot & Kelty-Stephen, 2017). This interdependence across multiple timescales is not entirely new to research in personality (Hopwood et al., 2022; Wrzus & Roberts, 2017), but it is incongruent with the common practice of summarizing idiographic dynamics within a network at one or two time-lags. A theoretical account about how to understand interactions among personality processes on multiple timescales in networks is lacking. Thirdly, we need to know what the to-be-modeled personality network components are. Cramer et al. (2012) suggested to use items from personalitytrait surveys as a starting point for selecting the personality components, so we derived our variables from a personality-trait inventory (Woicik et al., 2009) that were also appropriate for a daily diary (i.e., having the potential to fluctuate from day to day). Selecting items from trait-inventories as nodes in idiographic personality network estimation is standard practice (Beck & Jackson, 2020; Beck & Jackson, 2021). According to Cramer et al. (2012), individual differences can be assessed by “allowing for individual differences in components and the strengths of the connections among them” (p. 420). Empirical science has only achieved the latter, as standardizing the set of variables across participants is common practice. That is, we compare between-person differences in edges – not components. Pressing theoretical paucities are thus is 1) which personality components to model and 2) whether these differ between individuals. The most significant theoretical implication within dynamic systems accounts lies in the necessity of explicitly incorporating if-then contingencies. People are relatively stable

RkJQdWJsaXNoZXIy MTk4NDMw