Tracing Human-like Traits in LLMs: Origins, Real-World Manifestation, and Controllability

Pengrui Han, Rafal Dariusz Kocielnik, Peiyang Song, Ramit Debnath, Dean Mobbs, Anima Anandkumar, R. Michael Alvarez

Keywords: LLMs, Personality traits, behavioral alignment, self-regulation, persona, trait manifestation

TL;DR: LLMs develop stable self-reported trait profiles through instructional alignment, yet these traits fail to predict real-world behavior.

Abstract:

Personality traits have long been studied as stable predictors of human behavior.
Recent advances in Large Language Models (LLMs) suggest similar patterns may emerge in artificial systems, with advanced LLMs displaying consistent behavioral tendencies resembling human traits like agreeableness and self-regulation.
Understanding these patterns is crucial, yet prior work primarily relied on simplified self-reports and heuristic prompting, with little behavioral validation.
In this study, we systematically characterize LLM personality across three dimensions: *(1)* the dynamic emergence and evolution of trait profiles throughout training stages; *(2)* the predictive validity of self-reported traits in behavioral tasks; and *(3)* the causal impact of targeted interventions, such as persona injection, on both self-reports and behavior.
Our findings reveal that instructional alignment (e.g., RLHF, instruction tuning) significantly stabilizes trait expression and strengthens trait correlations in ways that mirror human data.
However, these self-reported traits do not reliably predict behavior, and observed associations often diverge from human patterns.
While persona injection successfully steers self-reports in the intended direction, it exerts little or inconsistent effect on actual behavior.
By distinguishing surface-level trait expression from behavioral consistency, our findings challenge assumptions about LLM personality and underscore the need for deeper evaluation in alignment and interpretability.

Submission Number: 50