Finding groups using model-based cluster analysis: heterogeneous emotional self-regulatory processes and heavy...

Area of Science:

• Physiology
• Psychology
• Data Science

Background:

• Investigating population heterogeneity is crucial for understanding complex behaviors like alcohol consumption.
• Heart rate variability (HRV) is a physiological marker of autonomic nervous system function.
• Understanding reasons for drinking and their link to physiological responses can inform interventions.

Purpose of the Study:

• To apply model-based cluster analysis to identify distinct subgroups within a young adult population based on alcohol use and physiological data.
• To examine differences in reasons for drinking and autonomic responses between identified clusters.
• To explore the relationship between autonomic self-regulation and susceptibility to heavy alcohol use.

Main Methods:

• Utilized model-based cluster analysis with finite mixture multivariate normal densities.
• Clustered 36 young adults using baseline heart rate, heart rate variability (HRV), chronic alcohol use, and drinking motivations.
• Employed Bayesian information criterion for model comparison and identification of optimal cluster numbers.

Main Results:

• Identified two distinct clusters: a high alcohol risk group and a normative group.
• The high alcohol risk group exhibited higher alcohol use and endorsed disinhibition/suppression as reasons for drinking more strongly.
• High alcohol risk individuals showed significant HRV changes to positive and negative emotional cues, unlike the normative group which only responded to negative cues.

Conclusions:

• Autonomic self-regulatory difficulties may predispose individuals to heavy alcohol use.
• Alcohol use for emotional regulation is associated with altered physiological responses in at-risk individuals.
• Model-based cluster analysis is a valuable tool for uncovering population heterogeneity in health-related behaviors.