This repository contains the simulation code, data, and figures associated with the paper:
"The Integrative Cognitive Selection System (ICSS): A Multi-Objective Model of Decision-Making Under Internal Conflict"
Farbod Sedghi (2026)
The Integrative Cognitive Selection System (ICSS) is a computational framework that models decision-making as a multi-objective optimization process under internal conflict.
Unlike traditional models that assume a unified utility function, ICSS represents behavior as the outcome of competition among multiple cognitive subsystems, each producing its own valuation of available actions.
The core decision rule is:
S(i,t) = Σ w_k(t) · V_k(i,t) − λ(t) · C(i,t)
Where:
V_k(i,t)= value assigned by cognitive layer kw_k(t)= dynamic dominance weightC(i,t)= cross-layer conflictλ(t)= conflict sensitivity parameter
This repository provides a simple simulation illustrating:
- ✅ Dynamic dominance across cognitive layers
- ✅ Context-dependent decision reversal
- ✅ Internal conflict as a measurable cost
- ✅ Lifestyle Coherence (LC) as a system-level metric
- ✅ Non-stationary behavior under stable valuations
- ✅ Confidence bands representing uncertainty (illustrative)
The simulation produces:
- Panel A: Evolution of dominance weights over time
- Panel B: Lifestyle Coherence (LC) trajectory
- Confidence bands: Represent hypothetical variability
📌 These outputs correspond directly to Figure 3 in the paper.
icss-decision-system/
├── README.md
├── simulation.py
├── icss_data.csv
├── figures/
│ ├── icss_two_panel_confidence.png
│ ├── icss_two_panel_confidence.pdf
└── requirements.txt
Install dependencies using:
pip install -r requirements.txt
Run the following command:
python simulation.py
This will:
- Generate simulation data (
icss_data.csv) - Produce figures in the
figures/directory - Display plots
| File | Description |
|---|---|
icss_data.csv |
Simulation dataset (weights + LC) |
icss_two_panel_confidence.png |
Main figure (paper-ready) |
icss_two_panel_confidence.pdf |
Vector version for publication |
- This simulation is illustrative, not empirically calibrated
- Confidence bands are hypothetical, not statistical estimates
- The goal is to demonstrate the behavioral dynamics of the ICSS framework
All results presented in the simulation section of the paper can be reproduced using:
python simulation.py
No external datasets are required.
If you use this repository or build upon this work, please cite:
Sedghi, F. (2026).
The Integrative Cognitive Selection System (ICSS): A Multi-Objective Model of Decision-Making Under Internal Conflict.
(TODO)
This repository provides a minimal implementation. Potential extensions include:
- Stochastic simulation (randomized weight dynamics)
- Parameter estimation from behavioral data
- Multi-agent interaction models
- Integration with reinforcement learning frameworks
This project is licensed under the MIT License.
You are free to use, modify, and distribute this code with proper attribution.
Farbod Sedghi
This repository is part of ongoing work on computational models of decision-making, aiming to bridge behavioral science, cognitive systems, and engineering frameworks.