PersonaForge: Psychology-Grounded Dual-Process Architecture for Personality-Consistent Role-Playing Agents

Large Language Models excel at role-playing but struggle to maintain consistent personalities across extended multi-turn interactions.We propose PersonaForge, combining (1) a three-layer personality architecture grounded in psychological theory and(2) a dual-process generation mechanism inspired by cognitive science.We test two falsifiable claims: Claim 1 (Orthogonality): Psychology-grounded dimensions (Big Five + Defense Mechanisms) provide more orthogonal constraints than natural language descriptions, reducing long-dialogue drift.Claim 2 (Integration Necessity): High-dimensional personality constraints create "production interference" requiring a cognitive workspace (Inner Monologue) to resolve-removing it degrades performance below simpler baselines.Experiments on 88 characters demonstrate: (1) +19.4% personality consistency (PC) over the Structured-CoT baseline, with human correlation r = 0.82, (2) reduced drift over 50-turn conversations (6.3% vs. 24.8%baseline), and (3) +64.7% defense mechanism manifestation.External validation on RoleBench confirms generalization (73.2% win-rate, drift 8.4% vs. 20.4%).Selective dual-process activation achieves 96% of fullsystem performance with only 13.4% token overhead.Human evaluation confirms more authentic and psychologically coherent character behaviors.Code and data: https:// github.com/fQwQf/PersonaForge. Ablation PC∆ SA∆ DM∆ w/o Dual-Process -0.29 -0.23 -0.28 w/o Big Five -0.12 -0.03 -0.05 w/o Defense Mech.-0.03 -0.02 -0.25 w/o Speaking Style -0.02 -0.18 -0.01 w/o Dynamic State -0.04 -0.01 -0.03 Generic Structured -0.18 -0.02 -0.28 Trigger Ablations Always Dual-Process +0.03 +0