| name | tensor-network-emotional-memory-modeling |
| description | Tensor network modeling for order-dependent emotional memory in children (arXiv:2606.28470) |
| category | quantum-cognition |
Tensor Network Modeling for Emotional Memory
Methodology from arXiv:2606.28470 (June 2026). Quantum-inspired tensor network approach to modeling order-dependent emotional memory in children.
Core Pattern
Tensor networks capture order-dependent structure in recognition memory that standard models miss:
- Correct recall depends not just on item valence, but on valence of preceding and following items
- Classical tensor network with valence factoring achieves 77.98% accuracy
- Massive accuracy increase over standard psychological models
- Novel task protocol for exploring emotional temporal memory
Key Findings
- Order-dependence differs across events (sequences of items)
- Standard models confirm order-dependence but have low accuracy
- Standard models don't reflect how memory for emotional objects influences others
- Tensor network captures inter-item influence through valence factoring
- Task protocol enables real-world tool for emotional temporal memory analysis
Implementation Steps
- Design sequence presentation task with emotionally-valenced items
- Record recall accuracy and order errors
- Build tensor network model factoring in valence (positive/negative/neutral)
- Train on sequence recall data with order-sensitive objective
- Compare against standard psychological models and ablation controls
- Analyze how memory for one item influences recall of adjacent items
When to Use
- Modeling order-dependent cognitive phenomena
- Emotional memory research and assessment
- Quantum-inspired modeling of psychological processes
- When standard models fail to capture sequential dependencies
References
- arXiv: 2606.28470v1
- Authors: Henry Groves, Lucia F. Jackson, Barbara-Anne Robertson
