The Affective Ising Model: A computational account of human affect dynamics

Abstract

The human affect system is responsible for producing the positive and negative feelings that color and guide our lives. At the same time, when disrupted, its workings lie at the basis of the occurrence of mood disorder. Understanding the functioning and dynamics of the affect system is therefore crucial to understand the feelings that people experience on a daily basis, their dynamics across time, and how they can become dysregulated in mood disorder. In this paper, a nonlinear stochastic model for the dynamics of positive and negative affect is proposed called the Affective Ising Model (AIM). It incorporates principles of statistical mechanics, is inspired by neurophysiological and behavioral evidence about auto-excitation and mutual inhibition of the positive and negative affect dimensions, and is intended to better explain empirical phenomena such as skewness, multimodality, and non-linear relations of positive and negative affect. The AIM is applied to two large experience sampling studies on the occurrence of positive and negative affect in daily life in both normality and mood disorder. It is examined to what extent the model is able to reproduce the aforementioned non-Gaussian features observed in the data, using two sightly different continuous-time vector autoregressive (VAR) models as benchmarks. The predictive performance of the models is also compared by means of leave-one-out cross-validation. The results indicate that the AIM is better at reproducing non-Gaussian features while their performance is comparable for strictly Gaussian features. The predictive performance of the AIM is also shown to be better for the majority of the affect time series. The potential and limitations of the AIM as a computational model approximating the workings of the human affect system are discussed.

Efficient estimation of bounded gradient-drift diffusion models for affect on CPU and GPU

“…By virtue of the interactions, affect states that would rarely be experienced because of the entropy can become more natural. For a more elaborate description of the parameters and their interpretation, see Loossens et al (2020). Equation 1 describes how the affect system updates in an infinitesimal interval dt.…”

“…Continuous-time stochastic modeling is becoming increasingly popular in the field of affect research (de Haan-Rietdijk et al, 2017;Hamaker et al, 2015;Oud, 2002Oud, , 2010Voelkle & Oud, 2013). Continuous-time models, like the Ornstein-Uhlenbeck (OU) model (de Haan-Rietdijk et al, 2017;Hamaker et al, 2015;Oud, 2002Oud, , 2010Voelkle & Oud, 2013) and the Affective Ising Model (AIM; Loossens et al, 2020) have several advantages over discrete-time models, such as the popular vector autoregressive model (Bos et al, 2012;Bringmann et al, 2013;Lodewyckx et al, 2011;Pe et al, 2015;Snippe et al, 2015;Wichers, 2014;Zheng et al, 2013); for instance, affect processes are generally conceived as continuously unfolding across time. A continuous-time description is more conform with this idea.…”

“…Another aspect of affect dynamics that is brought more frequently into the spotlight is its non-linear nature. Evidence has been found for heavily skewed (Crawford & Henry, 2004 ; Merz et al, 2013 ; Yik et al, 1999 ), V-shaped (Kuppens et al, 2013 ; Mattek et al, 2017 ) and even multimodal data distributions (Loossens et al, 2020 ). There is a particular interest in non-linear dynamics because of possible relationships with chaotic behavior and catastrophe theory (Van der Maas & Molenaar, 1992 ).…”

“…Recently, the Affective Ising Model (AIM) was introduced, which is a non-linear, continuous-time, stochastic model for affect (Loossens et al, 2020 ). The model was applied to two-dimensional affect data characterised by the positive (PA) and negative affect (NA) dimensions.…”

“…Furthermore, it is open and free to use. First, we will briefly discuss the AIM as introduced by Loossens et al, ( 2020 ). Next, we will discuss the numerical algorithm that is used in the software package to compute the likelihood of the model.…”