Dyslexia affects at least one in five people. This academic problem often affects social relations, self-esteem and mental health, leading to lost potential and unfulfilled dreams. And that’s on top of the cost: Federal spending on special education in the United States reached $15.5 billion in 2023.
3 Fast Facts About Learning Disorders
- An estimated 10 percent of people living in the United States are diagnosed with a learning disorder during their life.
- Dyslexia (difficulty reading) accounts for 80 percent of learning disorders.
- Learning disorder is a medical term; learning disability is a legal term.
Source: Cleveland Clinic
Dyslexia is a language-based disorder involving the linguistic system in the brain. This linguistic system is massively complex, comparable to a large computer operating system with millions of lines of code. A person has difficulty reading, or dyslexia, when the linguistic system in the brain is not running efficiently.
The inefficiencies can be anywhere in this highly complex system. And different parts of the system may be operating at different levels of efficiency. Based on my decades of fieldwork in clinical linguistics, individual differences within the dyslexic population are immense. That is why dyslexia needs a computational solution to resolve it.
What Is Linguistic Capability?
Linguistic ability is about combinations: combining sounds into words, words into sentences, sentences into texts. In using language, we go through billions of such combinations every day. When there are processing inefficiencies in a person’s brain, the intervention expert must be able to locate them precisely within these billions of data points as a first step in resolving them.
Neuroscience cannot yet get to this level of precision with brain imaging. In fact, neuroscientists concede that they may never be able to identify dyslexia traits from individual brain scans — ever.
Unlike neuroscientists, educators need not concern themselves with the physical structure of the brain. After all, teachers only need to ensure that the student’s language output is appropriate. Thus the concern in education is with the functional structure of the brain: How can we manipulate the input to yield the desired output? We need autonomous AI to manipulate this input-output function.
How Autonomous AI Helps With Dyslexia
An autonomous AI system makes decisions independently without human input. An autonomous AI system for dyslexia is already in the market. This dyslexia intervention program uses a game interface.
When a person plays a game, certain targeted brain processes are activated for analysis. Critical bits of information about these brain processes are contained in the user’s game responses. Which processes ran efficiently and which ones did not? At what level of efficiency? This autonomous AI system delivers online games continuously to a person with dyslexia, using previous gameplay data to build the next game.
A dyslexia intervention program requires autonomous AI because it has to overcome three major obstacles: Complexity, speed and capacity.
Complexity
The complexity issue is explained above. A human expert simply cannot manage the complexity of dyslexia at the level of brain processing, to diagnose it accurately and resolve it fully.
Speed
The brain has to process language super-fast for us to use language. Many language processes occur in hundreds of milliseconds. How can a human expert capture them exactly to make an accurate diagnosis? And design therapy that nudges a patient’s processing by milliseconds?
Capacity
A human expert cannot easily recall a patient’s exact outputs on a past day so as to cross-reference them with her outputs today in real time during gameplay. But a computer expert system can cross-reference these data points and more, within the user database and across other databases to make the next decision instantly.
Because of the sheer speed of language processing, an autonomous AI expert system is essential for dyslexia. When a student sits at the computer to start the intervention, the expert system has to create a game, analyze the user’s game responses, cross-reference them with relevant databases, build the next game and the next one in real time in one continuous loop — without pause.
Benefits of Autonomous AI for Dyslexia
Autonomous AI overcomes long-standing problems in this field, including chronic neurological disorders, lack of access to services and exorbitant costs.
Finding and treating chronic neurological disorders
Through this new, non-invasive capability to get to brain processes in individual brains, we can now locate problems at this level of granularity to correct them. Thus, previously chronic neurological disorders like dyslexia can now be treated as transient conditions to be resolved. The other learning disabilities, dysgraphia (writing) and dyscalculia (math), can be treated similarly with autonomous AI.
From my fieldwork experience, math difficulties are often due to language processing inefficiencies as well. When students have difficulty processing math teachers’ verbal instruction in primary grades, they may lack the math foundation necessary for more advanced math in later grades.
IMPROVED access to services
There are more than 13 million children with dyslexia in this country, yet barely half of them get special services at school. Without AI, dyslexia diagnosis presently requires highly trained, certified specialists working one-on-one using primarily pen-and-paper test batteries. That is, without AI, dyslexia diagnosis is a non-scalable, inexact science. Consequently, many students with dyslexia are never diagnosed.
Similar problems occur with dyslexia intervention. It presently requires highly trained, certified dyslexia/reading specialists to work one-one-one or in small groups of three to five students. These intervention methods just help students cope but do not correct their reading difficulty. These students still need extra support year after year, which is why the other half of students with dyslexia are locked out of special education.
MAKING TREATMENT LESS EXPENSIVE
Labor-intensive methods are costly, especially when they are not effective. Special services have to be given year after year to the same group of struggling learners who still cannot read on grade level despite this costly support. Training teachers to administer dyslexia screening and intervention is too expensive for many schools.
These traditional methods cost more than $10,000 to $20,000 for each student a year in most states. Because these compensatory methods are still needed for the rest of a student’s school years, their total cost can be as high as $200,000 per pupil. In contrast, a scalable AI program costs only 1 to 10 percent of this amount, when the dyslexia is corrected in one to two years. These huge savings benefit all students, when districts no longer have to make wrenching decisions over budget cuts.
If you know someone with dyslexia, you probably did not expect that the dyslexia problem would be solved in your lifetime. But the future is already here — in the form of young adults who were once diagnosed with dyslexia yet are now functioning like their typical peers after using autonomous AI.
In the next few years, if not months, you will encounter more examples of siloed AI, custom-built by domain experts for a specific purpose, and witness the magic of their solutions.
