118 Chapter 5 4. Discussion The current study explored idiographic network structures of adolescents and young adults with a mild intellectual disability or borderline intellectual functioning who completed a personality-related daily diary for 60 days. More specifically, we evaluated how variable personality networks were between individuals and within individuals over time. We found high between-person heterogeneity in network structures across the sample. Comparisons of the idiographic networks among individuals with a similar personality profile (Woicik et al., 2009) reveal similarly high levels of between-person heterogeneity. These findings are in line with the heterogeneity that is repeatedly found in other idiographic networks in various samples (e.g., Dotterer et al., 2020; Fisher et al., 2017; Reeves and Fisher, 2020). Our results further show that networks structures were not only variable between persons, but also varied within persons over time. Repeatedly estimating idiographic personality networks in a sliding 30-day window showed the structures to be variable throughout the 60-day timeline for all participants – although the degree of within-person network variability differed between persons (Table 1). This echoes the within-person network inconsistencies that Beck and Jackson (2020) found between EMA waves two years apart and even within a two-week EMA wave. How to interpret network variability over time is connected to one's conceptualization of personality. Under the theoretical assumption that the personality system is time-invariant, time-varying idiographic networks can be considered the result of unreliably estimating that ‘true’ average personality system (e.g., split-half network unreliability in Beck and Jackson, 2020; Beck and Jackson, 2021). However, dynamic systems theories expect structural variability when learning takes place and when situations significantly differ (e.g., Mischel & Shoda, 1995). The variability we find at the n = 1 level over time, visualized with dynamic network videos, showcases non-stationarity. The consequence is that averaged network estimates (i.e., edges) misrepresent the actual dynamic process. For studying individual differences this is detrimental, because it casts doubt on the validity of the summary statistics (i.e., average relations between personality components) upon which between-person comparisons are based. Yet, there are several theory-informed explanations for the nonstationarity. First, Mischel and Shoda (1995) indicate that the personality system can change due to learning or updating self-beliefs. Participants in our study received treatment, possibly inducing learning and updating selfbeliefs, which would explain the non-stationarity. Indeed, interventions are associated with changes in personality traits (Roberts et al., 2017). However,
RkJQdWJsaXNoZXIy MTk4NDMw