Skip to main content
Back to challenges
Challenge

Psychopathology Cognitive Symptom Engine with AutoGen

Based on research in The Lancet Digital Health using LLMs to model psychiatric phenomena, this challenge involves constructing a multi-agent cognitive simulation system. Using Microsoft AutoGen, participants will design interactive agent networks (Clinical Evaluator Agent, Patient Simulation Agent, Cognitive Behavioral Auditor) that interact to explore symptom emergence under simulated cognitive stressors.

Data ScienceHosted by Vera
Sign in free to start
Challenge brief

What you are building

The core problem, expected build, and operating context for this challenge.

Construct an AutoGen multi-agent system to model cognitive psychopathology trajectories under simulated environmental stressors.

Delivery guide

How work is evaluated

Evaluation

Evaluates multi-agent conversation fidelity and quantitative cognitive distortion scoring.

Datasets

Challenge datasets

Public datasets, plus the private datasets you have access to.

Evaluation rubric

How submissions are scored

These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.

Dimensions
2 scoring checks
Binary
2 pass or fail dimensions
Ordinal
0 scaled dimensions
Dimension 1

valid_distortion_string

Ensure distortion matches recognized CBT classification

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Dimension 2

distortion_score_correlation

Correlation of estimated cognitive distortion scores against empirical clinical experimental data • target: 0.85 • range: 0-1

Binary check

This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.

Learning goals

What you should walk away with

  • Design dynamic multi-agent interaction loops using AutoGen

  • Simulate cognitive symptom shifts in response to structured environmental stress inputs

  • Quantify cognitive distortion metrics (e.g., catastrophizing, black-and-white thinking)

  • Evaluate conversation logs against clinical empirical literature baselines

Resources and assets

Reference links and supporting material

Dataset notes

Synthetic multi-turn clinical evaluation protocols and quantitative cognitive distortion ratings under controlled stress scenarios.

Challenge setup

Evaluates multi-agent conversation fidelity and quantitative cognitive distortion scoring.

Challenge input

JSON containing stressor_type, baseline_vulnerability_index

AutoGen

Mandatory AutoGen framework

Evaluated output

JSON containing catastrophizing_score (float 0-10), dominant_distortion (string)

Evaluation checks
  • Ensure distortion matches recognized CBT classification
  • Correlation of estimated cognitive distortion scores against empirical clinical experimental data • target:...
Configured evaluation evidence
  • Benchmark: CognitiveSim-Bench
  • Distortion Score Correlation target: 0.85
  • 1 public reference case
Configured runtime
  • Python execution harness
  • Python sandbox (unavailable on Versalist)

Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.

View technical recipe

Configured tools

Action Space
  • AutoGen · Required
  • Merge · Optional
  • Langbase · Optional

Evaluation contract

  • valid_distortion_string · Weight 1
  • distortion_score_correlation · Weight 1

Metrics

  • distortion_score_correlation · Target 0.85

Benchmarks

  • CognitiveSim-Bench
  • biotech

Reference cases

  • 1 public reference case

Workflow tasks

  1. Task 1

    Input: JSON containing stressor_type, baseline_vulnerability_index

    Output: JSON containing catastrophizing_score (float 0-10), dominant_distortion (string)

    Execution harness: Python

Evaluation tools

  • AutoGen · Agent Framework

Recipe freeze marker

No recipe freeze time is recorded.

Frequently Asked Questions about Psychopathology Cognitive Symptom Engine with AutoGen