Uncovering general, shared, and unique temporal patterns in ambulatory assessment data.

Area of Science:

• Psychological Science
• Quantitative Psychology
• Computational Neuroscience

Background:

• Intensive longitudinal data (ILD) are increasingly used in psychology to study individual temporal dynamics.
• Existing analytical methods often assume homogeneity across individuals, limiting their application to ILD.
• Subgrouping Group Iterative Multiple Model Estimation (S-GIMME) offers a novel approach to model individual-level temporal processes.

Purpose of the Study:

• To evaluate the performance of the S-GIMME algorithm with ambulatory assessment data, specifically daily diary data.
• To investigate the impact of smaller variable sets, fewer time points, and weaker autoregressive effects on S-GIMME's performance.
• To demonstrate the utility of S-GIMME in identifying general, shared, and unique temporal processes in psychological research.

Main Methods:

• The study evaluated the Subgrouping Group Iterative Multiple Model Estimation (S-GIMME) algorithm.
• Simulations were conducted across conditions typical of daily diary data (e.g., fewer variables, time points, smaller effects).
• An empirical example applied S-GIMME to intensive longitudinal data from individuals with borderline personality disorder (BPD).

Main Results:

• S-GIMME demonstrated the importance of autoregressive effects in accurately recovering data-generating connections and directions.
• The algorithm proved effective even with shorter data lengths commonly found in daily diary studies.
• The empirical example successfully identified distinct temporal processes within the BPD sample.

Conclusions:

• S-GIMME is a valuable tool for analyzing intensive longitudinal data in psychological research, accommodating common data characteristics.
• The method advances the understanding of individual-level temporal processes, moving beyond homogeneity assumptions.
• S-GIMME holds significant potential for novel insights into human behavior and mental health through the analysis of ILD.