Categorization and validation of handedness using latent class analysis

Area of Science:

• Neuroscience
• Psychology
• Human Behavior

Background:

• The majority of researchers agree that handedness is not a simple dichotomous (left-right) phenomenon.
• However, consensus on the precise number of hand-preference categories and classification criteria remains elusive.

Purpose of the Study:

• To investigate hand-preference categories using latent class analysis (LCA).
• To validate LCA-derived categories against external criteria, including hand demonstration tests and arbitrary cut-off points.

Main Methods:

• The Edinburgh Handedness Inventory was administered to 354 randomly selected individuals.
• Latent class analysis (LCA) was performed using LatentGOLD software to analyze the handedness data.

Main Results:

• Three distinct hand-preference clusters were identified: left-handed, right-handed, and mixed-handed.
• Subcategorization within these clusters proved non-parsimonious.
• LCA-based classification showed good agreement with hand-preference demonstration tests.

Conclusions:

• The study supports the non-bimodal nature of handedness.
• Arbitrary cut-off points for mixed-handedness can lead to misclassification.
• The developed categorization and validation framework aim to reduce reliance on arbitrary values in future handedness research.