Computational Models for Affect Dynamics

Vanhasbroeck, Niels; Ariens, Sigert; Tuerlinckx, Francis; Loossens, Tim

doi:10.1007/978-3-030-82965-0_10

Cited by 15 publications

(15 citation statements)

References 139 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The high prevalence of multimodal distributions and skewness has important consequences for statistical modeling. Currently, the most popular way to analyze emotion time series is the VAR model (e.g., Vanhasbroeck et al, 2021). However, it is well-known that the VAR model has only a single equilibrium to which it returns after perturbations drawn from a Gaussian distribution (Hamilton, 1994).…”

Section: Time Series Modelingmentioning

confidence: 99%

“…Finally, the distributional form of data has major ramifications for statistical modeling. The currently most popular statistical model for emotion time series-the vector autoregressive (VAR) model (e.g., Bringmann et al, 2013;Hamaker et al, 2015;Vanhasbroeck et al, 2021)-only fits symmetric unimodal distributions (e.g., Hamilton, 1994). If emotion measurements turn out to deviate considerably from unimodal distributions, this would mean that such models would be gravely misspecified.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Multimodality and skewness in emotion time series.

Haslbeck¹,

Ryan²,

Dablander³

2023

Emotion

View full text Add to dashboard Cite

The ability to measure emotional states in daily life using mobile devices has led to a surge of exciting new research on the temporal evolution of emotions. However, much of the potential of these data still remains untapped. In this paper, we reanalyze emotion measurements from seven openly available experience sampling methodology studies with a total of 835 individuals to systematically investigate the modality (unimodal, bimodal, and more than two modes) and skewness of within-person emotion measurements. We show that both multimodality and skewness are highly prevalent. In addition, we quantify the heterogeneity across items, individuals, and measurement designs. Our analysis reveals that multimodality is more likely in studies using an analog slider scale than in studies using a Likert scale; negatively valenced items are consistently more skewed than positive valenced items; and longer time series show a higher degree of modality in positive and a higher skew in negative items. We end by discussing the implications of our results for theorizing, measurement, and time series modeling.

show abstract

Section: Time Series Modelingmentioning

confidence: 99%

mentioning

confidence: 99%

Multimodality and skewness in emotion time series.

Haslbeck¹,

Ryan²,

Dablander³

2023

Emotion

View full text Add to dashboard Cite

show abstract

“…Examples of the second category are statistical models that capture statistical associations between different emotions over time, either by using regression models to capture the lagged correlations in emotion time series (Bringmann et al, 2016;Hamaker et al, 2015;Kuppens, 2015;Kuppens, Oravecz, et al, 2010), or by modeling the probability that experiencing one emotion will lead to the activation of other emotions (as in the Affective Ising Model; Loossens et al, 2020). These statistical models could also be interpreted as generative models (Vanhasbroeck et al, 2021), representing the hypothesis that emotions directly cause other emotions at later points in time. However, these models do not directly formalize emotion theories as described above.…”

Section: Situationmentioning

confidence: 99%

Towards a Generative Model for Emotion Dynamics

Ryan¹,

Dablander²,

Haslbeck³

2023

Preprint

View full text Add to dashboard Cite

Most theories of emotion suggest that emotions are reactions to situations we encounter in daily life. Process theories of emotion further specify a feedback loop between our environment, attention, emotions, and action that clarifies the adaptive nature of emotions. In principle, such process theories describe how emotions develop in daily life, and consequently, emotion measures collected from individuals many times a day in studies using the Experience Sampling Methodology (ESM) should be highly useful in developing these theories. However, current emotion theories are predominantly verbal theories and therefore do not make clear predictions in such data. In this paper, we take a first step towards a generative model of emotion dynamics by formalizing the link between situations and emotions, which provides us with a basic generative model of emotions in daily life. We show that this incomplete model already reproduces nine stable phenomena in emotion time series related to (temporal) associations between emotions and their distributional form. We then discuss how we can draw on existing (process) theories of emotion to extend our basic model into a complete generative model of emotion dynamics. Finally, we discuss how generative models of emotion dynamics can facilitate theory development and advance measurement, study design, and statistical analysis.

show abstract

“…The high prevalence of multimodal distributions and skewness has important consequences for statistical modelling. Currently, the most popular way to analyze emotion time series is the Vector Autoregressive (VAR) model (e.g., Vanhasbroeck et al, 2021). However, it is well-known that the VAR model has only a single equilibrium to which it returns after perturbations drawn from a Gaussian distribution (Hamilton, 1994).…”

Section: Time Series Modelingmentioning

confidence: 99%

“…Finally, the distributional form of data has major ramifications for statistical modelling. The currently most popular statistical model for emotion time series -the Vector Autoregressive (VAR) model (e.g., Bringmann et al, 2013;Hamaker et al, 2015;Vanhasbroeck et al, 2021) -only fits symmetric unimodal distributions (e.g., Hamilton, 1994). If emotion measurements turn out to deviate considerably from unimodal distributions, this would mean that such models would be gravely misspecified.…”

Section: Introductionmentioning

confidence: 99%

Multimodality and Skewness in Emotion Time Series

Haslbeck¹,

Ryan²,

Dablander³

2022

Preprint

View full text Add to dashboard Cite

The ability to measure emotional states in daily life using mobile devices has led to a surge of exciting new research on the temporal evolution of emotions. However, much of the potential of these data still remains untapped. In this paper, we re-analyze emotion measurements from seven openly available Experience Sampling Methodology (ESM) studies with a total of 835 individuals to systematically investigate the modality (unimodal, bimodal, multimodal) and skewness of within-person emotion measurements. We show that both multimodality and skewness are highly prevalent. In addition, we quantify the heterogeneity across items, individuals, and measurement designs. Our analysis reveals that multimodality is more likely in studies using an analogue slider scale than in studies using a Likert scale; negatively valenced items are consistently more skewed than positive valenced items; and longer time series show a higher degree of modality in positive and a higher skew in negative items. We end by discussing the implications of our results for theorizing, measurement, and time series modeling.

show abstract

Computational Models for Affect Dynamics

Cited by 15 publications

References 139 publications

Multimodality and skewness in emotion time series.

Multimodality and skewness in emotion time series.

Towards a Generative Model for Emotion Dynamics

Multimodality and Skewness in Emotion Time Series

Contact Info

Product

Resources

About