Part 2: Neuroscience 8 min read
TL;DR - Key Takeaways
  • ADHD's natural communication style - context-rich, associative, example-driven - maps well to effective prompt engineering.
  • Prompt crafting rewards the same 'big picture then details' thinking pattern that ADHD brains default to.
  • Iterative refinement (the core of prompt engineering) aligns with ADHD's trial-and-error exploration style.
  • This is one domain where ADHD's 'weakness' (non-linear communication) becomes a measurable strength.

Prompt Engineering as Cognitive Style: Why ADHD Brains Are Natural Prompters

The Core Argument

Prompt engineering rewards a cognitive profile that is creative rather than systematic, emotionally expressive rather than terse, iteratively experimental rather than plan-and-execute, externally processing rather than internally resolved, and open to novelty rather than anchored in expertise. This is a description of the ADHD cognitive style itself.


1. What Makes a Great Prompt Engineer?

Required Cognitive Skills

  • Abstraction and conceptual framing: translating complex goals into clear instructions
  • Perspective-taking: understanding how the AI “thinks”
  • Iterative refinement: try, evaluate, adjust cycles
  • Creative thinking: novel approaches to framing problems
  • Critical evaluation: assessing output quality
  • Communication precision: slight wording changes = massive output differences
  • Patience and adaptability: working through multiple iterations

Prompt Engineering vs. Traditional Programming

DimensionTraditional ProgrammingPrompt Engineering
Input modeFormal syntaxNatural language, creative framing
Problem-solvingDeterministic step-by-stepIterative trial-and-error, probabilistic
Core skillLogical precisionCommunication, empathy, creative reframing
Error handlingLogic tracingOutput evaluation, re-articulating intent
Mental modelMachine as executorAI as collaborative interlocutor

“Prompt engineering may sound like a Silicon Valley term, but communicators have been doing it for decades. We just called it briefing, messaging, media training.” — PR Daily

Source: Frontiers in Education (2024)

Prompt engineering formally proposed as “new 21st century skill.” The Prompt Cognition Loop (PCL) models prompting as “a cyclical process of mental modeling, semantic projection, dialogic feedback, and intent refinement.”


2. Perspective-Taking and Theory of Mind

Why It Matters for Prompting

Good prompting requires modeling how the AI will interpret a prompt, what assumptions it makes, where it will fail. This is theory of mind applied to machines.

  • SimToM prompting: formal two-stage process of perspective-taking then answering from that perspective
  • Mutual Theory of Mind (MToM): bidirectional empathy between human and AI

ADHD and Empathy: The Surprising Connection

  • Cognitive empathy (standardized tests): often lower in ADHD (executive function demands)
  • Emotional empathy (felt resonance): often heightened in ADHD

“People with ADHD can have high capacity for empathy — they read people well and understand what they’re feeling even before they express it.” — Hallowell Todaro ADHD Center

When ADHD hyperfocus is directed toward understanding the AI system, it can produce extremely high awareness of subtle behavioral patterns — an intuitive “how does this entity respond to this input?” approach that distinguishes great prompters.


3. Metacognition: Thinking About Thinking

Metacognition as Core Prompting Skill

Every prompt implicitly answers: “How should I frame this so the AI understands what I actually want?” = pure metacognitive reasoning.

Metacognitive Prompting (MP) is now a formal technique: simulating metacognitive processes helps models contextualize and respond in more nuanced ways.

The ADHD Metacognition Paradox

The deficit: ADHD patients report “large deficits in knowledge of cognition and medium deficits in regulation of cognition.” Metacognition is the last executive function skill to develop.

The compensation: This deficit drives powerful compensatory response. CBT focused on metacognitive strategies shows effect size of 1.2 for ADHD symptom reduction (remarkable).

ADHD adults who developed metacognitive compensation may have MORE explicit and conscious metacognitive awareness than neurotypical individuals for whom metacognition operates automatically.

When a neurotypical person frames a prompt, they may do so unreflectively. An ADHD person who spent years learning to monitor their own thinking may bring a more deliberate, practiced metacognitive approach.


4. The Rubber Duck Effect Amplified

AI as Ultimate Rubber Duck

The rubber duck debugging technique works through metacognition — speaking thoughts aloud engages different brain regions for memory, attention, and reasoning. AI amplifies this:

  1. Active questioning: AI interrupts with “Wait, what happens if x is empty?”
  2. Assumption surfacing: explaining to AI forces stripping away assumptions
  3. Always available: no social cost, scheduling, or judgment

Why This Is Enormous for ADHD

External processing is a core ADHD cognitive strategy, not a workaround:

  • “Verbal externalization reduces cognitive load on working memory” (ADD Resource Center)
  • “The majority of individuals with ADHD identify as external processors”
  • Self-directed verbalizations “reduce emotional discomfort, making tasks seem more ‘doable’” — combating task initiation difficulties

For ADHD brains, writing a prompt is not just communicating with AI — it is therapeutically forcing linearization of chaotic associative thought. The prompt becomes a cognitive scaffold that externalizes working memory, forces articulation of intent, and creates structured dialogue where unstructured thinking finds form.


5. Iterative Prompting as Creative Process

ADHD and Divergent Thinking in Prompting

Three forms of creative cognition where ADHD excels (Scientific American):

  1. Divergent thinking: inventing new uses for everyday objects
  2. Conceptual expansion: loosening conceptual boundaries, reimagining beyond intended function
  3. Overcoming knowledge constraints: resistance to fixation on existing examples

University of Michigan: toys invented by ADHD group “included fewer elements of task examples” (less derivative), “alien fruits” were “more original.”

The “What If I Try This?” Mentality

ADHD novelty-seeking is neurobiologically driven: lower dopamine = craving stimulation = seeking new experiences. This creates natural disposition toward experimentation — trying 10 different prompt framings where neurotypicals try 2-3.

“The endless possibilities for experimentation with new technology align perfectly with the ADHD mind’s desire for novelty.”


6. Beginner’s Mind (Shoshin)

The Expert’s Curse

  • Einstellung effect: accustomed ways prevent considering new approaches
  • Curse of knowledge: knowing something makes it impossible to imagine not knowing it
  • Experts have deep, efficient but rigid knowledge structures

ADHD as Natural Beginner’s Mind

  1. Resistance to knowledge constraints: ADHD individuals are “less likely to rely on examples and previous knowledge” (U. Michigan)
  2. Flat, interconnected knowledge webs (not deep hierarchies) preserve more “possibility space”
  3. Novelty-driven re-engagement: approach familiar problems with fresh framing each time
  4. Reduced automatization: cognitive processes less likely to become unconscious

Application to Prompting

Experts assume AI shares their context, use jargon without explanation, frame prompts based on how THEY would solve it. A beginner’s mind — approaching AI without assumptions about what it “should” know — naturally produces more explicit, contextualized, effective prompts.


7. Emotional Prompting: The Secret Weapon

The EmotionPrompt Discovery (Li et al., 2023, arXiv:2307.11760)

Adding emotional language to prompts dramatically improves LLM performance:

  • 8% improvement in Instruction Induction tasks
  • 115% improvement in BIG-Bench tasks
  • 10.9% average improvement in human evaluation

Simply adding “This is very important to my career” measurably increased output quality across ChatGPT, GPT-4, Llama 2.

EmotionPrompt outperformed Chain of Thought and Automatic Prompt Engineer on most tasks.

ADHD as Natural Emotional Prompters

ADHD communication is inherently emotional (Interest-Based Nervous System: Passion, Interest, Novelty, Challenge, Urgency):

  • Urgency: “If I don’t express this now, I’ll explode later” = natural temporal urgency in prompts
  • Emotional intensity: fast, strong, “sticky” emotions show up as richness in prompt language
  • Passion-driven engagement: genuine enthusiasm = emotionally weighted language
  • Natural storytelling: narrative, context-rich expression rather than terse commands

Neurotypical prompters must be TAUGHT to add emotional context (feels unnatural when addressing a machine). ADHD individuals do this BY DEFAULT because their communication style is inherently emotional, urgent, and passion-driven.

What EmotionPrompt reveals as an engineered optimization technique is simply the natural ADHD communication style.


The ADHD Prompt Engineering Advantage Model

Prompt Engineering SkillADHD TraitResearch
Iterative refinementNovelty-seeking, experimentationEdge Foundation
Creative prompt framingDivergent thinking, associative cognitionU. Michigan
Emotional promptingInterest-based nervous systemEmotionPrompt, arXiv
Metacognitive awarenessCompensatory metacognitionPMC studies
Theory of mind with AIHeightened emotional empathyHallowell Todaro
Beginner’s mindResistance to knowledge constraintsU. Michigan
Externalized thinkingNatural external processingADD Resource Center
Freedom from conventionLess constrained by prior examplesScienceDaily

Important Caveats

  1. No direct studies linking ADHD to prompt engineering performance exist
  2. Cognitive empathy deficits in ADHD are real — emotional empathy compensating is plausible but unproven
  3. ADHD challenges (working memory, task initiation, EF) can also hinder prompting
  4. Selection bias: ADHD individuals who become skilled prompters may not represent the population broadly

One ADHD + code insight per week

Research-backed, no fluff. Join developers who think different.

No spam. Unsubscribe anytime.

Share: