A Network Analysis of Digital Clock Drawing for Command and Copy Conditions

Area of Science:

• Neuroscience
• Cognitive Psychology
• Gerontology

Background:

• Graphomotor and time-based variables from the digital Clock Drawing Test (dCDT) are crucial for assessing cognitive functions.
• Previous research has not quantified the strength of associations between digital clock variables during their production.
• Understanding these associations is key to refining cognitive assessment tools.

Purpose of the Study:

• To investigate and compare the network structure and interrelationships of digital clock variables produced under command versus copy conditions.
• To test the hypothesis that the production of clock features and their interrelationships differ between command and copy test conditions.
• To identify key variables influencing network connectivity in the dCDT.

Main Methods:

• Older adults (65+) performed a digital Clock Drawing Test under both command and copy conditions.
• A Bayesian hill-climbing algorithm and bootstrapping (10,000 samples) were employed to construct directed acyclic graphs (DAGs).
• DAGs were used to analyze and compare the network structure of dCDT variables for command and copy conditions.

Main Results:

• The command condition exhibited moderate associations between variables (micro|βz|=0.34) compared to the copy condition (micro|βz|=0.25).
• The copy condition network showed greater connectivity (18/18 variables) than the command condition (15/18 variables).
• Five variables significantly influenced network connectivity across both conditions, with dependencies aligning better with instruction order in the command condition.

Conclusions:

• Digitally acquired clock variables are interconnected, but their network structure varies significantly between command and copy test conditions.
• The findings highlight the importance of analyzing the production process of clock drawings, not just the final product.
• Further analysis of dCDT production dynamics can enhance the identification of normal cognitive features and aid in the early detection of neurodegenerative diseases.