Biologically-Inspired Cognitive AI

The SCAN Ecosystem

Synthetic Cognitive Augmentation Network

A comprehensive AI framework that mirrors human cognition through modular agents, psychometric alignment, and neuromorphic computing—creating personalized, energy-efficient cognitive support systems.

Ben Kennedy, Atif Mohammad, Matthew Wyandt
Department of Artificial Intelligence • Capitol Technology University
Key Contributions

Research Innovations

Five foundational advances in cognitive AI architecture and personalization

Biologically-Inspired Modular Architecture

Cognitive AI system directly modeled on prefrontal cortex neuroscience, with five specialized agents (DLPFC, VMPFC, OFC, ACC, mPFC) that mirror human executive function regions.

Psychometric Cognitive Alignment (SCANAQ)

36-item psychometric assessment that profiles users across 8 cognitive dimensions, enabling agent configuration based on individual thinking styles rather than generic preferences.

Dynamic Multi-Agent Learning (SCANUE)

LangGraph-based framework for agent state management and coordination, designed to enable adaptive collaboration patterns between specialized cognitive agents.

Neuromorphic Efficiency (STAC)

Hybrid spiking-transformer architecture with projected energy reduction compared to traditional transformers, enabling edge deployment of cognitive AI systems.

Integrated Ecosystem Design

Framework design integrating cognitive architecture, psychometric personalization, adaptive learning, and neuromorphic efficiency—components that typically exist in isolation.

What is SCAN?

SCAN is not just one technology—it's a complete ecosystem of interconnected AI systems designed to augment human cognition. Think of it as a neural network that thinks like you do, learns from you, and helps you make better decisions.

Traditional AI systems are monolithic—one giant model trying to do everything. The human brain doesn't work that way. Your prefrontal cortex has specialized regions: one for planning, one for emotional regulation, one for risk assessment. SCAN mirrors this biological reality.

The ecosystem consists of four integrated components, each solving a specific challenge in building truly intelligent, personalized AI systems.

SCAN

The foundation—a modular cognitive architecture with specialized agents that work like different regions of your brain

SCANUE

The intelligent layer—adaptive agents that learn from your feedback and personalize to your needs

SCANAQ

The alignment tool—a questionnaire that maps your cognitive style to agent behaviors

STAC

The efficiency research track: converting transformers to spiking networks, with the energy savings still to be demonstrated

Deep Dive into Each Component

Foundation

SCAN

Synthetic Cognitive Augmentation Network

SCAN is the architectural foundation—a modular AI system where specialized agents work together like regions of the human prefrontal cortex. Instead of one model doing everything poorly, five focused agents excel at specific cognitive tasks.

  • DLPFC Agent: Executive planning and cognitive control
  • VMPFC Agent: Emotional regulation and risk assessment
  • OFC Agent: Reward evaluation and outcome prediction
  • ACC Agent: Conflict monitoring and error detection
  • mPFC Agent: Social cognition and perspective-taking
  • Framework: Built with CrewAI for agent orchestration
  • Models: OpenAI API for natural language processing
Intelligence

SCANUE

SCAN Using Experts (User Extensible)

SCANUE is the evolution—where SCAN agents become truly adaptive. Each agent is fine-tuned for its cognitive function, learns from human feedback, and continuously improves. It's SCAN with a memory and the ability to grow.

  • Fine-Tuned Agents: Specialized models for each PFC region
  • Framework: LangGraph for state management and agent coordination
  • HITL Integration: Human-in-the-Loop feedback for continuous learning
  • Adaptive Learning: Agents adjust based on user interactions
  • Reinforcement Learning: Optimizes decision strategies over time
  • Biometric Ready: Can integrate EEG, HRV, and other real-time signals
Alignment

SCANAQ

SCAN Alignment Questionnaire

SCANAQ is the personalization engine—a 36-item psychometric assessment that profiles your cognitive style across 8 dimensions. It tells SCANUE how to adapt its agents to match the way you think and decide.

  • 8 Psychological Scales: Risk propensity, self-efficacy, executive function, decision style, emotion regulation, stress, empathy, impulsivity
  • 36 Items: Quick assessment; items adapted from published scales
  • PFC Mapping: Scores directly inform agent parameters
  • Design-Based Research: Developed via CAUSE user surveys
  • Not a Diagnostic: A personalization tool; the composite instrument has not been reliability- or validity-tested
  • Scoring Model 2.0.0: The published scoring appendix is being corrected; the current model is in SCAN-Resources
Efficiency

STAC

Spiking Transformer Augmenting Cognition

STAC is the efficiency research track: a framework for converting pretrained transformers into spiking neural networks. With spiking off, the conversion reproduces the source model exactly. With spiking on, the operation-count projection for the current design is worse than the dense model, and no energy has been measured on neuromorphic hardware. The measured results are in the repository.

  • Hybrid Architecture: Transformers + Spiking Neural Networks
  • Two Pathways: V1 (fine-tuning) and V2 (conversion)
  • Energy: Operation-count projection only; the current design projects worse than the dense model, and no hardware power has been measured
  • SpikingJelly: Built on the SpikingJelly SNN framework
  • Loihi Target: Intel Loihi 2 is the intended platform; a constraints checker exists, hardware runs do not
  • Temporal Dynamics: Captures time-based cognitive patterns

Understanding SCANAQ: The 8 Cognitive Dimensions

First, you take the questionnaire. SCANAQ assesses your cognitive style across 8 dimensions. Then, each dimension configures a specific PFC agent. For example, your Executive Functioning score tells the DLPFC agent how much structure you want around tasks, so it can provide responses that match your natural planning style. This personalization happens for all 8 dimensions across all 5 agents.

Executive Functioning
Items 1–8. How often everyday planning, organization, and follow-through are a struggle. A higher score asks the DLPFC agent for more scaffolding: step-by-step breakdowns, reminders, and structure.
Example: "Starting a task is hard for me" (1 Never – 3 Often)
Emotion Regulation
Items 9–12. Two strategies scored separately: cognitive reappraisal and expressive suppression. Informs how the VMPFC agent frames emotional content before problem-solving.
Example: "I keep my emotions to myself" (1–7 agreement)
Impulsivity
Items 13–18. Overall impulsivity, with attentional, motor, and non-planning subtypes. A higher score suggests pause-and-plan prompts and commitment devices before decisions.
Example: "I act on the spur of the moment" (1 Rarely – 4 Almost Always)
Risk Propensity
Items 19–21. General appetite for risk. Shapes how agents present downside scenarios and how much caution they build into recommendations.
Example: "I enjoy taking risks in general" (1–5 agreement)
Decision-Making Style
Items 22–26. Rational, intuitive, dependent, avoidant, and spontaneous styles, with co-dominant styles allowed. Tells agents whether to lead with analysis, instinct, consultation, or deadlines.
Example: "I make decisions in a logical and systematic way" (1–5 agreement)
Self-Efficacy
Items 27–29. Confidence in handling difficult or unexpected situations. Calibrates how much reassurance versus autonomy agents offer.
Example: "I am confident that I could deal efficiently with unexpected events" (1–4)
Perceived Stress
Items 30–32. Recent stress load, with one reverse-keyed item. A high score asks agents to reduce load: fewer parallel asks, more sequencing.
Example: "In the last month, how often have you felt nervous and stressed?" (1–5)
Empathy and Social Cognition
Items 33–36. Empathic concern and perspective-taking, scored apart from the fantasy item. Shapes the mPFC agent's social reasoning and how it voices other people's viewpoints.
Example: "I sometimes find it difficult to see things from the other person's point of view" (1–5)
How It Works
Your scores across these 8 dimensions create a unique cognitive profile. SCANUE uses this profile to configure each PFC agent's parameters—adjusting their decision-making thresholds, emotional weighting, planning depth, and communication style to match how you naturally think and process information.

SCAN Agent Network: How PFC Regions Collaborate

DLPFC - Planning
VMPFC - Emotion
OFC - Reward
ACC - Monitoring
mPFC - Social

The Ecosystem in Action

Each component plays a specific role, and together they create a cognitive augmentation system that's both powerful and personal.

User Journey Through the Ecosystem

Follow the 8-step process from assessment to personalized AI guidance

Start Step 1 of 8 Complete
Take SCANAQ Assessment
Complete the 36-item questionnaire to profile your cognitive style across its 8 sections: executive functioning, emotion regulation, impulsivity, risk propensity, decision-making style, self-efficacy, perceived stress, and empathy.
~10 minutes

Real-World Examples

See how SCAN transforms decision-making across different domains with personalized cognitive support

Three Papers, One Vision

The SCAN ecosystem is described in three peer-reviewed book chapters: one with Springer Nature and two with IGI Global.

1

Synthetic Cognitive Augmentation Network

Focus: Foundational Architecture (Springer, SEET 2025, CCIS 2725, pp. 179–188)

Introduces the core SCAN framework—modular AI agents modeled after prefrontal cortex regions. Demonstrates biological plausibility and introduces SCANUE, SCANAQ, and STAC as future components.

Cognitive Architecture PFC Modeling Multi-Agent Systems CrewAI
Read Paper
2

Beyond Intelligence

Focus: The SCANUE Platform (IGI Global, 2025)

Operationalizes SCAN as SCANUE—fine-tuned agents for each prefrontal region, curated training datasets, and human-in-the-loop feedback for continuous realignment. Introduces SCANAQ as an upcoming instrument.

SCANUE Fine-Tuning Human-in-the-Loop Training Data
Read Paper
3

Aligned Minds, Efficient Machines

Focus: Alignment and Neuromorphic Computing (IGI Global, 2026)

Presents SCANAQ, the 36-item questionnaire that aligns agents to individual cognitive styles, and STAC, the spiking-transformer work in its V1 fine-tuning and V2 conversion forms, with hardware validation left as future work.

A correction to this chapter is in progress: a code audit found the STAC V1 spiking pathway inactive as released, and the Appendix B scoring model contained errors. The corrected scoring model is Appendix B 2.0; the full correction record is in the STAC repository.

SCANAQ Spiking Neural Networks Energy Efficiency Correction in Progress
Read Paper

What STAC Measures Today

These figures come from the STAC repository's own test suite and energy projection script (software simulation on DistilGPT-2 and SmolLM2; no neuromorphic hardware has been measured). They replace an earlier estimated comparison chart.

Quantity Measured Meaning
Conversion fidelity, spiking off 1.00× perplexity The converted model reproduces the source model's outputs.
Projected energy, spiking on Worse than the dense model Operation-count projection at the current spike coverage. Energy is coverage-bound, not sparsity-bound.
Frozen model, spiking on Collapses Post-hoc conversion needs spike-aware training to recover quality; training recovers it.
Hardware energy Not measured Intel Loihi 2 validation is planned, not done.

Where SCAN Makes a Difference

From healthcare to finance, education to edge computing—SCAN's modular, personalized approach transforms how AI supports human decision-making.

Healthcare
Clinical decision support tailored to physician cognitive styles
Finance
Risk-aware investment advising aligned with client psychology
Education
Adaptive learning systems matching student decision-making styles
Manufacturing
Real-time process optimization with edge deployment via STAC
Edge Computing
On-device AI with neuromorphic efficiency for IoT
Research
Hypothesis generation and experimental design assistance

Explore, Collaborate, Contribute

SCAN is an active research project. Whether you're a researcher, developer, or just curious about cognitive AI, there are multiple ways to engage with this work.

Read the Research

Explore the SCAN ecosystem through our published research papers and work in progress.

Beyond Intelligence Aligned Minds STAC

Explore the Code

All SCAN components are open-source. Fork, extend, and build upon the ecosystem for your own research or applications.

SCAN Core SCANUE Learning SCANAQ Assessment STAC Neuromorphic

Connect & Collaborate

Interested in collaborating, implementing SCAN in your domain, or discussing the research? Let's connect.

LinkedIn Email

Future Directions

Clinical Validation
Large-scale user studies with diverse populations to validate SCANAQ psychometrics and measure real-world cognitive augmentation impact.
Hardware Integration
Deploy STAC on Intel Loihi 2 neuromorphic chips to demonstrate real-time, energy-efficient cognitive AI on edge devices.
Multi-Modal Extensions
Extend SCAN beyond text to integrate vision, audio, and sensory modalities for richer cognitive augmentation experiences.
Domain Specialization
Develop domain-specific SCAN variants for healthcare, education, finance, and scientific research with specialized agent configurations.