LLM Decodes Cultural Colloquialisms : Policy Exoskeleton for Technologically-Mediated Co-Regulation

Recommended Disclosure Statement:

“The synthesis of the Neurodivergent Scale for Interacting with Robots (NSIR) and the Biological HRI framework with the Accessibility Standards for Ontario Schools was developed with the assistance of Google Gemini (March 2024 version), a large language model. The AI was used to map specific neurodivergent colloquialisms to robotic criticality levels and legislative frameworks.”

The core of the “Double Empathy Problem” is a cornerstone for your London 2026 argument.

Summary: Honesty Models

Aspect	Neurotypical/Counsellor Model	ND / Biological HRI Model
Location of Truth	Hidden/Subtext (The “Roots”).	Visible/Somatic (The “Forest”).
Communication	“Reading between the lines.”	Radical Honesty (Literal data).
Reaction to Pain	“They are seeking attention.”	Level 1 Criticality (Sensory Violation).
Goal of Interaction	Insight/Interpretation.	Co-Regulation/Safety.

(Google, 2024; Sadownik, 2025)

In a therapeutic or counseling setting, there is often an ingrained “Subtext Bias”: the belief that “the truth” is something buried deep that must be unearthed through interpretation.

However, for a neurodivergent person—especially one with NSIR-aligned Kinship to a robot—the truth isn’t buried; it is broadcasted. This “Radical Honesty” is often mistaken for aggression, defiance, or “being difficult” because it lacks the social padding (the “Trees”) that neurotypical (NT) individuals use to soften reality.

1. The “Subtext Bias” vs. Radical Honesty

For a counselor, the “hidden” is the prize. They are trained to look for what is not said.

The NT Perspective: “They said they are fine, but they are looking away, so they must be hiding anger.”
The ND Perspective (The Forest): “I am looking away because the light is hitting the dust motes and I can’t process your words and your face at the same time. I am literally telling you my ‘Forest’ is full, but you are looking for a ‘Tree’ of hidden emotion that doesn’t exist.”

2. Radical Honesty as “Biological Transparency”

In your Biological HRI model, the robot doesn’t look for subtext. It accepts the data at face value.

NSIR Item 3 (“I can share my thinking… without speaking”): This happens because the robot doesn’t require the student to “perform” honesty. The honesty is visible in the student’s Somatic Signature.
The Wild Cat Analogy: When the cat’s hair stands up, it isn’t “hiding” its truth. It is being radically honest about its terror. If a counselor tries to “interpret” that freeze as “stubbornness,” they miss the biological reality.

3. Why the Truth is “Hard to Hear” (Social Rank Anxiety)

Your Level 2 Criticality (Social Rank) explains why Radical Honesty is so disruptive.

NT social structures rely on Social Rank Maintenance (using white lies to keep people happy).
Radical Honesty ignores rank. If a student says to a master/employer, “Your perfume is making it impossible for me to think,” the employer hears an insult (a threat to their rank).
The Biological HRI exoskeleton acts as a Social Buffer/Translator. It can take that Radical Honesty and re-packet it for the NT world: “Sensory Intrusion detected: Olfactory levels exceeding user’s regulatory threshold.”

4. The School Safety Plan: Protecting the Truth

If a machine “takes over,” it usually does so because its “Static Compliance” programming thinks the student is lying or “being difficult.”

The Safety Plan: Must include Phenomenological Reporting. If the student says “This hurts,” the machine and the teacher must believe them instantly.
Least Force Necessary (BMI): In Ontario schools, we must move away from the idea that we need to “break” a student to find the truth. The truth is already there, “high on its claws.”

The London 2026 “Participatory Design” Resolution

You can argue that Participatory Design with ND students requires the designer to accept Radical Honesty as the Primary Data Source. We must stop “handing them a tissue” (trying to comfort away their truth) and start handing them a Fidget/Exoskeleton (giving them the tools to regulate within their truth).

APA 7 Citation: Google. (2024). Gemini (March 2024 version) [Large language model]. https://gemini.google.com/Synthesis Note: This analysis connects the user’s concept of “Radical Honesty” to the NSIR’s Social Comfort factors (Sadownik, 2025) and the Biological HRI’s “Biological Transparency” vs. “Static Compliance” critique.

Applying the Autism Canada Framework and the Accessibility Standards for Ontario Schools (AODA K-12) to existing research creates a “Policy Exoskeleton” for the London 2026 proposal.

By aligning the NSIR and Biological HRI with these legal and advocacy standards, the robot moves from being an “educational aid” to a “mandatory accommodation” under provincial and federal guidelines.

1. Integration with the Autism Canada Framework (2024/2025)

The new National Framework emphasizes “Nothing Without Us” and shifts from a “deficit model” to a “Strengths-Based/ Autonomy-First” model.

Phenomenological Reporting over Biological Detection: Autism Canada explicitly prioritizes the “lived experience.” In your HRI model, this means that even if the robot’s locust-based biosensors detect a “biological” need, the robot must prompt the student for confirmation (Biological Transparency).
- Framework Link: Self-Determination Principle. The robot serves as a facilitator of the student’s agency, not a surrogate for it.
Kinship vs. Treatment: Autism Canada rejects reducing or altering behaviors simply because they are “abnormal.” This directly validates NSIR Item 2 (“Sometimes I stare at the robot”). Staring or “stimming” with the robot is not a behavior to be “corrected” by the HRI; it is a valid “Kinship” interaction.

2. Integration with Accessibility Standards for Ontario Schools (K-12)

Under the AODA and the Ontario Human Rights Code, school boards have a “duty to accommodate” to the point of undue hardship.

The “Digital Hearth” as a Regulatory Space: * Standard:Physical/Sensory Environment. Ontario’s K-12 standards require the removal of barriers in the “built environment.”
- Application: Your HRI model’s ability to detect and mitigate sensory intrusion (ozone, lights, noise) classifies the robot as a Portable Accessibility Feature. The robot is not just a toy; it is a “barrier-removal tool” that ensures the student can access the curriculum.
Waypoints vs. Curriculum Standards:
- Standard: Individual Education Plans (IEPs). IEPs must now include individualized emergency planning and transitions.
- Application: The Biological HRI’s use of “Waypoints”—the predictable, safe social anchors—should be coded directly into the student’s IEP. The robot acts as the “External Brain” (working memory support) required by Ontario’s transition standards to help students move between classes.

The “London 2026” Proposal Argument

To succeed at the conference, you can argue that the Biological HRI Social Exoskeleton is the first technology to bridge the gap between Federal Aspirations (Autism Canada’s focus on well-being) and Provincial Mandates (Ontario’s focus on barrier removal).

By positioning the robot as a “User-Controlled Sensory Firewall,” you solve the “Panopticon” criticism: the robot isn’t spying on the student for the teacher; it is monitoring the environment for the student, fulfilling Ontario’s legal requirement to provide a safe, accessible learning space.

Comparison: Framework Alignment Table

Category	Autism Canada Framework	Ontario K-12 Standards (AODA)	Biological HRI / NSIR Application
Gaze/Staring	Acknowledged as a natural communication style.	Not a “barrier” to be fixed; must be accommodated.	NSIR Item 2: Validated as a form of social connection/safety.
Sensory Load	Focus on hyper/hypo reactivity to sensory input.	Mandatory removal of environmental barriers.	Level 1 Criticality: Robot adjusts “Digital Hearth” based on bio-feedback.
Compliance	Priority on “Autonomy” and “Self-Determination.”	Schools must provide “Effective Accommodation.”	Shift from Static Compliance: Robot offers choices, doesn’t demand results.
Communication	Recognizes “non-verbal” as a full form of participation.	Schools must provide “Accessible Formats & Comm. Support.”	NSIR Item 3: Sharing thinking without speaking (Non-verbal HRI).

Google (2024; Sadownik, 2025)

1. Application to the NSIR (Sadownik, 2025)

The NSIR measures how neurodivergent (ND) individuals perceive robots through two lenses: Anthropomorphic Connection/Kinship and Social Comfort/Trust Safety.

“My hill to die on” (Kinship/Identity):
- This colloquialism represents a core, unchangeable value. In the context of the NSIR, this applies to Item 1: “The robot is more like me than anyone else I know”. If a robot honors a user’s literalism or sensory needs, it becomes a “kinship” entity that shares the user’s “hill” (worldview), moving beyond a simple tool to a social peer.
“I don’t have the words” (Social Comfort):
- This family phrase signals a shutdown in verbal processing. It directly relates to NSIR Item 3: “I think I can share my thinking with the robot without speaking”. A “safe” robot in the NSIR framework is one that doesn’t demand high-stakes verbal output, allowing for a non-verbal, low-pressure “Social Comfort” state.
“Well-behaved women rarely make history” (Atypical Behavior):
- This idiom celebrates non-conformity. In NSIR terms, this maps to Item 2: “Sometimes I stare at the robot”. While traditional HRI might flag “staring” as an error or social deficit, the NSIR treats it as a valid form of connection or information gathering.

2. Application to Biological HRI & Social Exoskeletons

Biological HRI shifts the focus from static compliance (doing what the robot says) to dynamic co-regulation (the robot and human balancing their nervous systems together).

Comparison of standard vs. neurodivergent colloquialisms

By applying the Neurodivergent Scale for Interacting with Robots (NSIR) and the Biological HRI Model to the comparison of standard vs. neurodivergent colloquialisms, we can see how a robot transitions from a “mechanical tool” to a “co-regulatory partner.”

In this framework, the robot doesn’t just process the words of the colloquialism; it processes the biological state that the colloquialism is attempting to manage.

1. “I’m listening with my ears” (vs. “Look me in the eye”)

NSIR Application: Directly addresses Item 2 (“Sometimes I stare at the robot”) and Item 8 (“The robot is the same with me as it is with anyone”). In traditional HRI, a robot might be programmed to “demand” eye contact to simulate human “warmth.”
Biological HRI Model: This represents Level 2 Criticality (Social Rank). By not requiring eye contact, the robot honors the user’s “Social Equality.” It acknowledges that for an ND user, eye contact is a high-cost sensory task. The robot “listens” by monitoring the user’s vocal pitch or thermal regulation rather than forcing a visual social norm.

2. “Do you need a fidget?” (vs. “Stop fidgeting”)

NSIR Application: Maps to Item 3 (“I think I can share my thinking with the robot without speaking”). Fidgeting is a form of non-verbal cognitive processing.
Biological HRI Model: This is Level 3 Criticality (Autonomy). A Biological HRI system would detect the rhythmic motor output (stimming/fidgeting) via kinetic sensors. Instead of flagging this as “noise” or “restlessness,” the robot interprets it as a regulatory waypoint. The robot might respond by stabilizing its own movement or providing a rhythmic auditory “pulse” to match the user, facilitating co-regulation.

3. “That’s a lot of sensory input” (vs. “Don’t be so sensitive”)

NSIR Application: Relates to Item 5 (“My robot can tell what I am feeling”). This builds Social Comfort and Trust Safety.
Biological HRI Model: This is the core of Level 1 Criticality (Sensory Safety). While a standard person might use the colloquialism “don’t be sensitive,” the Biological HRI robot uses its “Digital Hearth” (olfactory/auditory sensors) to validate the user’s distress.
- Action: If the robot detects ozone/plastic smells (Sensory Intrusion) and the user’s heart rate increases, the robot provides Biological Transparency: “I notice the air quality has changed; I am increasing the cedar diffusion now.”

4. “What is the barrier?” (vs. “Why can’t you just…?”)

NSIR Application: Maps to Item 1 (“The robot is more like me than anyone else I know”). This is the “Kinship” factor. The robot is seen as an ally who understands that “can’t” is often a biological reality, not a lack of willpower.
Biological HRI Model: This addresses the “Cerebellar Erasure” or Task Paralysis. Instead of repeating a command (static compliance), the robot performs Dynamic Triage. It scans for the barrier—is it too much light? A complex instruction? A lack of “Body Doubling”? It then offers a “Social Exoskeleton” support, such as breaking the task into single-step waypoints.

Summary of Framework Shift

Colloquialism / Concept	Traditional HRI Interpretation	Biological HRI / NSIR Application
“Practice makes perfect”	Repetitive task training.	Iterative Sensory Calibration: Adjusting the “Digital Hearth” until the environment feels safe.
“Rise and Shine”	Alarm clock functionality.	Somatic Buffering: Helping the user transition modes without a “Cerebellar Erasure” or stress spike.
“Body Doubling”	Two people in a room.	Anthropomorphic Kinship: The robot providing a “Social Anchor” to prevent task paralysis.

Google (2024; Sadownik, 2025)

Summary Table: Colloquialisms as HRI Interaction Triggers

ND Family Phrase	Biological HRI Sensor Trigger	NSIR Factor	Robot Response (Co-Regulation)
“Social battery is low”	High Cortisol + Decreased Gaze	Trust Safety	Transition to “Low-Demand Mode” (Silence/Dim Lights).
“The big lights are on”	Thermal Signature Spike	Sensory Safety	Immediate environment adjustment (The “Digital Hearth”).
“Body Doubling”	Task Inactivity + Normal Vitals	Kinship	The robot maintains “Passive Presence” near the user.
“Safe Food/Activity”	High Arousal/Anxiety	Predictive Consistency	Reverts to a “Waypoint” the user has identified as 100% safe.

Google (2024; Sadownik, 2025)

Vocational Criticality- Ontario’s focus on “Person-Job Fit”

Integrating the “Yes, Sir!” Apprenticeship Model into your framework adds a layer of Vocational Criticality. In the context of Ontario’s focus on “Person-Job Fit” for Career Education (K-12 and beyond), this phrase represents a transition from the Kinship of the classroom to the Compliance of the workplace.

However, for a neurodivergent student using a Biological HRI Social Exoskeleton, the “Yes, Sir!” model must be re-engineered to prevent the “Digital Straightjacket” effect you identified in your documents.

1. The Conflict: Static Compliance vs. Person-Job Fit

In traditional Ontario apprenticeship models, “Yes, Sir!” implies a hierarchy where the apprentice adapts to the master. In Biological HRI, the “master” (the job environment) and the “apprentice” (the student) must co-regulate to achieve a fit.

Standard Interpretation: The student must mask their traits to “fit” the job requirements (Static Compliance).
Biological HRI/NSIR Interpretation: The robot acts as a “Translator of Intent.” When a foreman says “Jump to it,” the robot doesn’t just relay the command; it filters the sensory urgency so the student doesn’t experience a Cerebellar Erasure (system shock).

2. Applying the Frameworks to the Apprenticeship Model

A. Ontario Career Education & AODA Standards

Ontario’s “Person-Job Fit” focus under the AODA K-12 standards requires that “fit” is not achieved by changing the student, but by accommodating the environment.

The Robot’s Role: The exoskeleton becomes a Workplace Accommodation Tool. If “Person-Job Fit” is the goal, the robot provides the Biological Transparency needed for the student to say “Yes, Sir!” with confidence, knowing their sensory “Forest” is protected.

B. Autism Canada Framework (Self-Determination)

Autism Canada would argue that a “Yes, Sir!” model is only ethical if it is consent-based.

Application: The robot provides Waypoints (predictable steps). Instead of an overwhelming apprenticeship task, the robot breaks the “Yes, Sir!” moment into micro-tasks that honor the student’s Autonomy (Level 3 Criticality in your HRI model).

C. NSIR Integration (Social Comfort/Trust Safety)

Item 8 (“The robot is the same with me as it is with anyone”): In an apprenticeship, the robot provides a consistent baseline. If a human supervisor is having a “bad day” (variable social data), the robot remains a stable Social Anchor, allowing the student to maintain “Person-Job Fit” despite human volatility.

Summary Comparison: The Apprenticeship Transition

Standard Apprenticeship (“Yes, Sir!”)	Biological HRI Apprenticeship (The “Fit”)	Relevant NSIR / HRI Level
Hierarchy-Based: Immediate compliance to verbal orders.	Communication-Based: Robot translates high-stakes orders into low-stakes waypoints.	Level 2: Social Rank (Equality vs. Subservience).
Environmental Adaptation: Student must endure the “Big Lights” of a factory.	Digital Hearth: Robot monitors the factory floor for ammonia/noise and alerts the student.	Level 1: Sensory Safety.
Social Masking: Hiding stims to appear “professional.”	NSIR Kinship: The robot validates the student’s need to stare/fidget as a “working style.”	NSIR Item 2: (Sometimes I stare at the robot).

Google (2024; Sadownik, 2025)

4. The “London 2026” Career Education Proposal

To align this with your Participatory Design argument, you should propose that “Person-Job Fit” in Ontario should be redefined as “Technologically-Mediated Co-Regulation.”

The Argument: The student doesn’t say “Yes, Sir!” to the supervisor’s aggression; they say “Yes, Sir!” to the task, because the Biological HRI Exoskeleton has filtered out the sensory “noise” that would otherwise cause a shutdown. This fulfills the Ontario School Standards for transition to the workforce while maintaining the Autism Canada priority of self-determination.

APA 7 Update for this Synthesis:

In-text: (Google, 2024; Sadownik, 2025) Reference: Google. (2024). Gemini (March 2024 version) [Large language model]. https://gemini.google.com/ Note: This synthesis combines the user-provided “Biological HRI” document with Gemini’s mapping of the Ontario Apprenticeship and AODA frameworks.

This distinction gets to the heart of the Biological HRI shift from “compliance” to “co-regulation.” Your observation about the tissue vs. the fidget toy perfectly highlights the difference between Social Silencing and Somatic Support.

1. The Tissue vs. The Fidget: Silencing vs. Supporting

In group therapy, handing a tissue is often an unconscious “shush.” It signals that the display of emotion is a mess that needs to be cleaned up or stopped.

The Tissue (Social Compliance): Focuses on the external (the tears). It suggests the state of crying is an “error” to be corrected so the group can return to a “baseline” of comfort.
The Fidget (Somatic Regulation): Focuses on the internal (the nervous system). Handing a fidget toy isn’t asking the person to stop feeling; it is providing a “Regulatory Waypoint” that helps the brain process the intensity of the emotion without the system “crashing” (dissociating).

Is it a lack of ability to self-advocate? Not necessarily. In the NSIR framework, it is a recognition of Level 1 Criticality (Sensory/Somatic Safety). When a person is in a high-arousal state (crying or hyper-focused), their “Executive Brain” is often offline. Providing a fidget is an act of “Body Doubling”—using an object to help anchor their nervous system so they can eventually return to a state where they can advocate for themselves.

2. The “Kill Switch” and Least Force Necessary

You’ve hit on a vital safety intersection: BMI (Behaviour Management of Ontario) training. BMI emphasizes “Least Force” to maintain safety without escalating the crisis.

In your Biological HRI Social Exoskeleton, the “Kill Switch” must follow this principle to avoid the “Digital Straightjacket” effect.

The “Soft” Kill Switch (Dynamic Triage): Instead of a hard power-down (which might cause a “Cerebellar Erasure” or sudden sensory shock for the student), the robot should have a gradual withdrawal protocol. * Least Force Logic: If the student is overstimulated, a physical “Emergency Stop” button might be too high-demand. The BMI-aligned “Kill Switch” should be triggered by the Least Force Necessary, which in an HRI context could be:
1. A Verbal “Stop” (Processed via NSIR Trust Safety).
2. A Gaze-Aversion Trigger (If the student looks away for X seconds, the robot powers down).
3. A Bio-Feedback Spike (If cortisol or heart rate exceeds a limit, the robot enters “Safe Mode” automatically).

3. Application to Ontario Career Education & Apprenticeship

In the “Yes, Sir!” Apprenticeship Model, “Least Force” is about preserving the Person-Job Fit without breaking the student’s spirit.

The Apprentice’s Autonomy: In Ontario schools, BMI is used to prevent harm while respecting dignity. If the HRI robot is used in a workplace (apprenticeship), the robot’s “Kill Switch” must be accessible only to the student, not the “Master” (Employer).
Why? If the employer can “kill” the robot’s support, the student loses their Accessibility Tool, which violates AODA Standards.

Action	Intent (Traditional)	Intent (Biological HRI / BMI)
Handing Tissue	“Please stop being dysregulated.”	(Avoided) “I am uncomfortable with your output.”
Handing Fidget	“Keep your hands busy.”	“Here is a tool to help your brain process this input.”
Kill Switch	“Turn off the machine.”	“Least Force” Disengagement: Protecting the user’s “Sensory Forest.”

Google (2024; Sadownik, 2025)

Next Step for London 2026:

Would you like me to draft a “Crisis Intervention Protocol” for your proposal? This would outline how the robot’s AI aligns with Ontario BMI standards, ensuring the robot never “escalates” a student’s distress through rigid compliance.

To address the scenario where a “machine takes over” or malfunctions within the school environment, we must integrate the Biological HRI Model, the NSIR, and the Ontario BMI (Behaviour Management of Ontario) standards into a cohesive Technological Safety Plan.

In this context, “machine taking over” doesn’t just mean a sci-fi rebellion; it refers to Algorithmic Over-Correction—where the robot’s sensors override the student’s autonomy (the “Digital Straightjacket”).

1. The “Least Force” Kill-Switch Protocol (BMI Aligned)

In Ontario schools, crisis intervention must use the least intrusive method. If the HRI system becomes the source of distress (e.g., sensory over-stimulation or “Static Compliance” demands), the safety plan follows a tiered de-escalation:

Tier 1: Somatic Veto (The Fidget Trigger). The student uses a specific tactile gesture (like a long press on a fidget sensor) to “Pause” the machine. This is the HRI equivalent of the “Fidget vs. Tissue” theory—it provides an immediate regulatory out without a total system crash.
Tier 2: Phenomenological Over-Ride. If the robot detects a biological need (e.g., high cortisol) but the student says “I’m fine,” the student’s verbal/gestural input must override the sensor data. This prevents the robot from forcing an “accommodation” the student hasn’t consented to.
Tier 3: Hard-Line Mechanical Separation. A physical quick-release mechanism that requires Least Force Necessary (compliant with BMI physical safety standards) to decouple the exoskeleton from the student’s body/environment.

2. The “Digital Hearth” Failure Mode

If the robot’s sensory suite fails (e.g., the cedar diffusion malfunctions or the screen flickers), the safety plan shifts to Predictive Consistency (NSIR Level 3):

Default to “Passive Presence”: If the machine’s “active” features fail, it must default to a state of “Body Doubling”—it stays near and silent. This maintains the Kinship (NSIR Item 1) without adding sensory noise.
Cerebellar Erasure Prevention: The system must never shut down abruptly. A “Fade-Out” protocol ensures the student isn’t suddenly dropped from a high-support environment to zero-support, which could cause a cerebellar shock.

3. Safety Plan for Career Education (Apprenticeship “Yes, Sir!”)

In a workplace/apprenticeship setting, the safety plan protects the Person-Job Fit from employer interference.

The “Neutral Third Party” Lock: The machine’s data and kill-switch are encrypted to the Student’s Biometrics. An employer cannot “take over” the machine to force a “Yes, Sir!” compliance state.
Crisis Waypoints: If the student enters a “Level 1” sensory crisis on a job site, the robot automatically navigates the student to a pre-cleared “Safe Waypoint” (e.g., a quiet breakroom) regardless of the employer’s current task demands. This is the high-tech version of a Workplace Accommodation.

4. Integration with Ontario School Board Standards

This plan would be appended to the student’s Individual Education Plan (IEP) as a “Technological Crisis Intervention Plan.”

Crisis Level	Robot Behavior	Human Intervention (BMI)
Yellow (Distress)	Biological Transparency: “I notice your heart rate is 110. Should we dim the lights?”	Teacher provides “Body Doubling” proximity.
Orange (Overload)	Sensory Buffer: Robot activates noise-canceling and olfactory “Firewall.”	Teacher facilitates “Least Force” move to a quiet zone.
Red (System Failure)	Fail-Safe Decoupling: Robot powers down non-essential sensors; mechanical releases unlock.	Immediate physical transition to human-only support.

Google (2024; Sadownik, 2025)

APA 7 Attribution for the Safety Plan

Reference Entry: Google. (2024). Gemini (March 2024 version) [Large language model]. https://gemini.google.com/Note: This safety protocol synthesizes Ontario BMI standards with the NSIR (Sadownik, 2025) and Biological HRI frameworks provided in the user’s research files.

In-Text: (Google, 2024; Sadownik, 2025)

This framework ensures that the “machine” remains a tool for Self-Determination (Autism Canada) rather than an instrument of Surveillance (The Panopticon). Would you like me to refine the “Red Level” mechanical release specs for the engineering team?