In our previous article, What’s Missing in the AI Literacy Framework: Oversights and Opportunities we examined a critical oversight in the AI Literacy Framework. The Framework, and others like it, is a comprehensive overview of what students should be able to do in an AI-world. It does not ask whether they are cognitively equipped to do it.
This article explores that missing foundation in depth: the role of cognitive capacity in learning, decision-making, and AI literacy—and why it cannot be assumed, nor directly taught, in the ways current frameworks often imply.
The AI Literacy Framework is a thoughtful and timely response to the growing need for students to understand and engage with artificial intelligence. It offers a structured vision of AI literacy across domains such as:
Yet across these domains, there is a recurring cognitive assumption: that students already possess—or can readily be taught—the mental abilities required to perform these tasks. The Framework asks students to:
These are complex cognitive demands. But the Framework does not address the underlying mental architecture—the cognitive functions—that make these tasks possible.
It is increasingly common in education to see expectations like "apply knowledge," "think critically," or "monitor your learning" built into curriculum goals and assessments. Teachers are expected to "foster metacognition" or "teach critical thinking" as if these are instructional techniques.
But neuroscience tells us something different: these higher-order processes are not skills to be taught. They are outputs of cognitive functions—such as attention, reasoning, and cognitive flexibility—which vary widely between students.
Telling a student to “focus harder” or “analyze more deeply” without first understanding the status of their cognitive capacity is akin to asking someone to lift a weight without knowing whether they’ve built the muscle to do so.
At the heart of every learning experience—every decision made, every concept grasped, every question asked—is the brain. It is not one factor among many. It is the determining force that governs how we take in, organize, and respond to the world around us.
The brain’s cognitive architecture shapes not just what we learn, but how we learn: with what degree of ease or struggle, at what depth, and with what level of independence. From the ability to hold and manipulate information in our minds, to the capacity to sustain attention, shift perspectives, detect patterns, and reason through complexity—these processes are the product of intricate, interdependent systems within the brain.
And those systems do not operate uniformly across all learners. Each individual possesses a unique cognitive profile—a distinctive configuration of strengths and inefficiencies that affects their capacity to engage with learning environments. For some students, these variations mean learning feels intuitive and fluid. For others, the path is marked by friction, fatigue, and frustration—not because of lack of effort or motivation, but because the necessary brain functions have not fully developed or are under strain.
Nowhere is this more visible than in students with neurodiverse profiles—those with ADHD, dyslexia, dyscalculia, auditory processing difficulties, and other learning challenges. These students are routinely expected to perform tasks that require reflection, inference, or ethical reasoning—without adequate recognition of whether the brain systems required to do so are currently in place.
Drawing from the Arrowsmith framework, the following cognitive functions are particularly relevant to AI literacy, 21st-century skills, and lifelong learning:
| Cognitive Function | What It Enables |
| Symbol Relations | Supports reasoning, synergizing of ideas. Allows switching between perspectives, adapting to new information or feedback from AI systems |
| Symbolic Thinking | Sustains focus across time, resists distraction, supports deep engagement with AI-generated material |
| Predicative Speech | Underpins coherent argumentation, hypothesis generation, and rational critique of AI performance, also helps follow multi-step AI processes, coding sequences, or logic-based instructions |
| Non-Verbal Thinking | Assists with interpreting social-emotional dimensions of human-AI interaction and collaborative tools |
| Memory for Info + Instructions | Holds information, essential for comparing AI outputs without personal knowledge or context |
These functions cannot be meaningfully improved through instructional strategies alone.
When education systems assume these cognitive functions are universally available or easily taught, it creates unintended consequences—especially for neurodiverse and cognitively vulnerable learners:
These impacts are subtle but significant—and they deepen inequities over time.
Many competencies in the AI Literacy Framework—reflection, critique, discernment—are framed as teachable outcomes. But these are not simply instructional challenges. They are cognitive ones.
And it places pressure on educators to deliver outcomes that actually depend on their students’ cognitive readiness.
Teachers can model strategies and offer tools. Teachers can model strategies and offer tools. But they cannot build the brain’s infrastructure—such as attention span, reasoning, or mental flexibility—through pedagogy alone. They cannot differentiate instruction for students whose reasoning systems are not yet ready to perform AI-related tasks. Nor can they close capacity gaps through content delivery or classroom management.
To ask them to do so is not just unrealistic—it’s unfair.
This is not a critique of pedagogy—it is a call for frameworks, including the AI Literacy Framework, to acknowledge and address the cognitive prerequisites of performance.
For students with cognitive profiles marked by underdevelopment in essential functions (like to article about LD), exposure to AI in the classroom without adequate scaffolding can have unintended consequences:
These risks are rarely visible in the moment, but they compound over time. This is especially critical when implementation timelines stretch across years, leaving vulnerable students unsupported during key developmental windows.
A Shift in Thinking: From Skill Acquisition to Capacity Building
If education systems are to take AI literacy seriously, they must also take cognitive infrastructure seriously.
That means moving away from the idea that complex cognitive outcomes can be taught like content. It requires recognizing that the ability to reason, critique, reflect, and adapt are not taught—they are enabled by cognitive systems that must be assessed, supported, and in some cases, actively developed.
In the final article in this series, we will explore how neuroplasticity-based programs—like Arrowsmith—are helping schools strengthen the very cognitive functions that frameworks assume to be present. These interventions offer a scalable way to prepare students' brains to engage not just with AI—but with learning itself.
True AI-literate education doesn’t begin with access to tools.
It begins with readiness to think.