Constructing personality maps, mapping personality constructs: multidimensional scaling recovers the big five factors...

Area of Science:

• Psychology
• Linguistics
• Cognitive Science

Background:

• The lexicon of personality traits is a rich source for understanding human perception.
• Previous research has explored trait similarity using various methods, but convergence between semantic and usage-based approaches needs further investigation.

Purpose of the Study:

• To investigate the underlying structure of similarities within the personality-trait lexicon.
• To compare similarity structures derived from semantic judgments and actual word usage (covariance).
• To determine if the lexicon's structure reflects trait perception or personality structure.

Main Methods:

• Analysis of a lexicon of 60 trait adjectives.
• Collection of semantic similarity judgments.
• Measurement of trait-attribution covariance in actual use.
• Application of multidimensional scaling (MDS) for structural representation.
• Utilizing canonical correlation and individual-differences MDS for cross-validation and interpretation.

Main Results:

• Substantial convergence was found between semantic similarity and trait-attribution covariance.
• Both data types revealed a similar underlying structure when analyzing subsets of adjectives.
• Individual-differences MDS facilitated the interpretation of shared dimensions across datasets.

Conclusions:

• The structure of the personality-trait lexicon is primarily related to trait perception.
• The observed structure is robust across different measurement methods (semantic vs. covariance).
• The findings suggest that the organization of trait language reflects cognitive processes of perception rather than the objective structure of personality.