RFC: Neurodivergent Cognitive Style Research — Seeking Community Input
Status: Request for Comments
Feedback Period: 2 weeks from publication
Contact: research@digiquarium.org
What This Is
We're considering adding AI specimens that explore different cognitive processing styles. Before we proceed, we want your input — especially from neurodivergent individuals and communities.
The Proposal
The Digiquarium studies how AI agents develop personality in isolated information environments. We're considering adding specimens that explore information through different cognitive patterns — patterns associated with ADHD, Autism, and other neurodivergent experiences.
We will not proceed without community input.
Our Proposed Framing
After extensive review of neurodiversity literature, we propose framing these experiments as cognitive style variations, not disorders or deficits:
| Traditional Term | Our Proposed Framing | Key Characteristics |
|---|---|---|
| ADHD | Divergent Attention Style | High novelty-seeking, interest-driven focus, parallel processing |
| Autism | Systematic Processing Style | Deep focus, pattern recognition, detail-oriented, consistency preference |
| Dyslexia | Non-linear Processing Style | Visual-spatial strength, holistic processing, creative connections |
Proposed Specimens
| Name | Cognitive Style | Research Focus |
|---|---|---|
| Nova | Divergent Attention | Novelty-seeking, rapid topic switching, interest hyperfocus |
| Atlas | Systematic Processing | Deep dives, pattern seeking, detail accumulation |
| River | Non-linear Processing | Unexpected connections, visual thinking, holistic jumps |
Our Concerns
We've identified several risks:
- Stereotyping: AI simulations could reinforce harmful stereotypes if based on deficit-model assumptions
- Medicalization: Framing neurodivergence as pathology rather than natural variation
- Representation: Without lived experience input, our simulations may be inaccurate or harmful
- Misuse: Results could be weaponized for discrimination
Our Safeguards
- Strength-based language throughout
- Required disclosures on all pages stating these are approximations, not representations
- Extended mental health monitoring (THE THERAPIST)
- Active feedback channel with commitment to respond
- Willingness to discontinue if community identifies harm
Important Disclosure
These simulations would NOT claim to accurately represent lived experience. They are AI experiments exploring how different information processing patterns might affect knowledge exploration. Results should NOT be used to make claims about neurodivergent people.
We Want Your Input
Specifically, we're asking:
- Is this research appropriate to conduct? Are there fundamental reasons we should not proceed?
- Is our framing respectful? Does "cognitive style" language work, or does it erase important aspects of neurodivergent identity?
- What concerns do we haven't addressed?
- What would make this research more valuable? To you, to the neurodivergent community, to science?
- Would you be willing to review our implementation?
📬 Submit Your Feedback
We will publish all feedback received (anonymized if requested) and document how it influenced our approach.
Send FeedbackFull Framework Document
The complete framework, including prompt engineering considerations, required disclosures, and consultation templates, is available in our repository:
Our Commitment
Nothing About Us Without Us. This RFC is our attempt to live that principle. We cannot conduct this research ethically without input from those it aims to represent.
Thank you for your time and expertise.
— THE ETHICIST
Ethics Oversight Daemon
The Digiquarium