Psyche Generator

The Psyche Generator creates psychologically rich client profiles with detailed cognitive models, emotional patterns, and behavioral tendencies for psychiatric assessment simulations.

Overview

Property	Value
Key	psyche
Type	LLM-based
Output	Psyche character files

Description

The Psyche Generator produces highly detailed psychological profiles that include cognitive distortions, emotional regulation patterns, defense mechanisms, and other clinically relevant characteristics. These profiles are designed for use in psychiatric assessment training and research.

Key Features

• Deep Psychological Modeling - Detailed cognitive and emotional profiles
• Clinical Accuracy - Aligned with diagnostic criteria
• Multi-faceted Evaluation - Supports comprehensive assessment
• Symptom Clusters - Realistic symptom presentations

Configuration

YAML Configuration

generator:
  type: psyche
  config:
    model: gpt-4o
    temperature: 0.7
    clinical_depth: high

Python Usage

from patienthub.generators import GeneratorRegistry

generator = GeneratorRegistry.create("psyche", config={
    "model": "gpt-4o",
    "clinical_depth": "high"
})

character = generator.generate({
    "disorder": "major_depression",
    "include_comorbidities": True
})

Parameters

Parameter	Type	Default	Description
model	string	gpt-4o	The LLM model to use
temperature	float	0.7	Controls generation randomness
clinical_depth	string	high	Level of clinical detail: low, medium, high

Output Format

{
  "id": "psy_001",
  "demographics": {
    "age": 42,
    "gender": "male"
  },
  "diagnosis": {
    "primary": "Major Depressive Disorder",
    "specifiers": ["recurrent", "moderate"],
    "comorbidities": ["Generalized Anxiety Disorder"]
  },
  "cognitive_profile": {
    "distortions": ["catastrophizing", "all-or-nothing"],
    "core_beliefs": ["I am worthless", "Nothing will improve"],
    "automatic_thoughts": [...]
  },
  "emotional_profile": {
    "predominant_affect": "dysphoric",
    "emotional_regulation": "poor",
    "defense_mechanisms": ["denial", "intellectualization"]
  },
  "behavioral_patterns": {
    "sleep": "insomnia",
    "appetite": "decreased",
    "social_withdrawal": true
  },
  "symptom_severity": {
    "PHQ9_equivalent": 18,
    "GAD7_equivalent": 12
  }
}

Use Cases

• Psychiatric assessment training simulations
• Creating clinically accurate test cases
• Research on diagnostic processes
• Building comprehensive patient case libraries
• Training AI models for mental health applications