Modeling Psychological Attributes in Psychology – An Epistemological Discussion: Network Analysis vs. Latent Variables

Abstract

Network Analysis is considered as a new method that challenges Latent Variable models in inferring psychological attributes. With Network Analysis, psychological attributes are derived from a complex system of components without the need to call on any latent variables. But the ontological status of psychological attributes is not adequately defined with Network Analysis, because a psychological attribute is both a complex system and a property emerging from this complex system. The aim of this article is to reappraise the legitimacy of latent variable models by engaging in an ontological and epistemological discussion on psychological attributes. Psychological attributes relate to the mental equilibrium of individuals embedded in their social interactions, as robust attractors within complex dynamic processes with emergent properties, distinct from physical entities located in precise areas of the brain. Latent variables thus possess legitimacy, because the emergent properties can be conceptualized and analyzed on the sole basis of their manifestations, without exploring the upstream complex system. However, in opposition with the usual Latent Variable models, this article is in favor of the integration of a dynamic system of manifestations. Latent Variables models and Network Analysis thus appear as complementary approaches. New approaches combining Latent Network Models and Network Residuals are certainly a promising new way to infer psychological attributes, placing psychological attributes in an inter-subjective dynamic approach. Pragmatism-realism appears as the epistemological framework required if we are to use latent variables as representations of psychological attributes.

Network analysis of specific psychopathology and psychiatric symptoms in patients with eating disorders

“…In addition, it should be noted that modeling psychological features and psychiatric symptoms with network analysis should not be seen as opposed to classical latent component analysis, yet the two methods should be seen as complementary(Guyon, Falissard, & Kop, ), providing useful information from two different a priori assumptions whose ontological and epistemological validity remains equally justified. Network analysis is a proper tool to look into how symptoms reciprocally interact within or across mental conditions and how they can be influenced by external events.…”

“…toms measured with EDI in inpatients treated for an ED (Olatunji et al, 2018). (Guyon, Falissard, & Kop, 2017), providing useful information from two different a priori assumptions whose ontological and epistemological validity remains equally justified. This work should be considered in view of its limitations.…”

Are Schools Alienating Digitally Engaged Students? Longitudinal Relations between Digital Engagement and School Engagement

“…Partial correlations are presented, so the edges in the latent partial correlation network can be interpreted similarly as regression path coefficients, as they are controlled for gender as well as each other, but without assuming any direction of effects. This type of modelling allows for a powerful measurement error corrected modelling and visualization of contemporaneous relations between latent variables when the direction of effects cannot be inferred from the data (Guyon, Falissard, & Kop, 2017).…”

On the Equivalency of Factor and Network Loadings

“…Although factor and network loadings provide similar information when the data generating model is a factor model, it is the selection and representation of the model that implies a specific causal interpretation (Borsboom, 2006;Bringmann & Eronen, 2018). It's plausible that a factor actually represents an emergent property of a collection of variables while it's equally plausible that a collection of variables in a network represent a common cause (Guyon, Falissard, & Kop, 2017;Kruis & Maris, 2016;van der Maas et al, 2006). For factor models, their representation implies that a latent variable causes the relations between variables; for network models, their representation implies that the variables co-occur through causal relations and that dimensions emerge from these relations (Bringmann & Eronen, 2018;Christensen et al, in press;Marsman et al, 2018).…”