In the wake of completing my AI Bootcamp course and LinkedIn’s toxicity rising to new levels in recent months I’ve been taking a digital detox and enjoying the crisp mountain air of the Austrian Alps in all their springtime magnificence.
It’s been a useful time to reflect, read and reset.
Listening to the birdsong in and around Salzburg was a reminder of just how spectacularly my brain has adapted to the input from the Cochlear Implant.
When you consider that the implant in my right sided cochlear has 22 electrodes each producing signals for a range of 10 frequencies then my brain has learned to interpret sound from just 220 inputs.
That is just over 7% of the input a healthy cochlear of 3000 hearing cells typically provides the brain.
Given that I have virtually no useful residual hearing in my left ear now then it is essentially less than 4% of the input of an individual with both their cochlear in good working order.
Over the last few weeks I’ve read a lot about neuroplasticity and the remarkable ability of our brains to change themselves.
It’s probably not surprising that one of the pioneers in this field - Michael Merzenich - was also a pioneer in Cochlear Implant technology. The work of his team at UCSF went into the creation of the Advanced Bionics implants used today.
Merzenich’s original work was developed further by Austrian engineer Erwin Hochmai who established one of the other leading brand of implants - Med-El.
The insights gained by Merzenich into brain plasticity on this early auditory input research, and subsequent experiments, helped him change the views of the entire scientific community which believed, up until the 1970s, that the brain’s structure and function was essentially fixed throughout adulthood.
As a result of multiple ground-breaking experiments Merzenich asserted that,
"If the brain map could normalize its structure in response to abnormal input, the prevailing view that we are born with a hardwired system had to be wrong. The brain had to be plastic."
Ultimately, his lifelong research resulted in Merzenich receiving the 2016 Kavli Prize in Neuroscience "for the discovery of mechanisms that allow experience and neural activity to remodel brain function."
With other neuroscientists, Merzenich invented a broad range of computer based training programmes designed to help overcome learning difficulties and disabilities and holds over 100 patents in this line of work.
The more I read about his experiments and those of other neuroscientists, since the awakening to the idea of brain plasticity in the 1970s, the more I’ve been thinking about how AI can be used to speed up and refine the neuroplasticity of our own brains.
The idea that AI is already using its own version of neuroplasticity to improve is already well documented and has been termed ‘techno-plasticity’. However, it's limitations in this aspect of AI learning and development are also being stated more and more frequently as the reason Artificial General Intelligence (AGI) is still further off than we might imagine, given all the hype around the topic.
From reading case studies of how neuroplasticity enables the brain to restructure neural pathways and rewires itself in response to new information or damage I've gained more understanding of the amazing processes that have been happening in my own brain since first being implanted five years ago.
My own experiences demonstrate that neuroplasticity can be a slow process and I know, from coming from a long line of deafened people who have also received implants, that the process is unpredictable and outcomes can vary widely for different individuals.
Looking deeper into the start-ups that are accelerating research and development into this 'perfect union' of neuroplasticity and AI I have identified several key areas of progress ...
1.Neurofeedback
Illustrated by organisations like Myndlift neurofeedback is a technique that involves training individuals to regulate their own brain activity by providing real-time feedback on their brainwaves. Through AI-driven algorithms, neurofeedback can be personalised to target specific brain areas involved in cognitive function, emotional regulation, and motor control.
By continuously monitoring brain activity, AI systems can detect subtle changes in neural patterns and provide feedback to the individual, encouraging them to alter their brain activity in a way that promotes neural reorganization. For instance, in the case of stroke rehabilitation, AI-based neurofeedback can help patients strengthen weak neural circuits, encouraging the brain to form new connections and restore lost functions.
2.Targeted Neurostimulation
An insight into the some of the research being conducted in this field comes from this fascinating paper published a couple of months back.
What was immediately apparent is the similarity of this proposed device to the Cochlear Implant processor and embedded technology that I already use day in, day out.
AI can be used to optimise neurostimulation therapies, which involve the application of electrical impulses to specific brain regions to promote plasticity and repair. Techniques such as transcranial direct current stimulation (tDCS) and deep brain stimulation (DBS) have shown promise in enhancing brain plasticity and improving cognitive and motor functions in patients with conditions such as stroke, Parkinson’s disease, and depression.
AI can play a critical role in improving the precision and effectiveness of neurostimulation by analysing brain activity in real time and adjusting stimulation parameters to achieve the desired therapeutic effect. For example, AI algorithms can identify the most effective locations for stimulation based on an individual’s unique brain activity patterns, ensuring that the stimulation is delivered to the right areas at the right intensity.
3.Cognitive Enhancement
Beyond rehabilitation, AI is also poised to play a transformative role in cognitive enhancement - something that Merzenich made great strides on largely through extensive trial and error. He proved repeatedly that brain plasticity can also be harnessed to improve cognitive performance, memory, focus, and emotional regulation. With the help of AI, brain training programs can be developed that adapt to an individual’s cognitive strengths and weaknesses, providing targeted exercises to optimize brain function.
For example, AI-powered brain training programs (such as NeuroLAT) can assess an individual’s cognitive abilities and create personalised exercises that challenge specific neural circuits. As the brain responds to these exercises, AI algorithms adjust the difficulty level and nature of the tasks to continuously promote neural reorganization and cognitive improvement.
4.Mental Health
Mental health conditions, such as depression, anxiety, and PTSD, are often linked to disrupted neural circuits and impaired brain plasticity. By harnessing AI to enhance neuroplasticity (such as the Flow Neuroscience approach), it is possible to offer new treatments for these conditions, enabling individuals to rewire their brain and overcome the neural imbalances that contribute to their mental health struggles.
For example, AI-based neurofeedback can help individuals with anxiety or depression regulate their brain activity, promoting emotional stability and reducing symptoms. Similarly, AI-driven neurostimulation can target specific brain regions involved in mood regulation, enhancing brain plasticity and improving mental health outcomes.
I've learned a lot more about AI, and particularly Machine Learning, than I ever imagined I would over the last year since returning to LinkedIn and embarking on the AI Leadership Bootcamp course at Portsmouth University.
The challenge this has presented (that would probably be described as a 'high-class problem' by a wise senior manager I worked for in the past) is how and where to focus this new found knowledge when AI has such vast scope across virtually all aspects of lives in some form or another.
When I returned to LinkedIn after a four year absence, due largely to learning to hear again with the Cochlear Implant, I expressed a desire to find a new role that offers both purpose and joy.
I'm certainly finding a sense of purpose in volunteering one day a week for Toucan Diversity.
However, the reason my much telegraphed website for 'Perfect Union' remains unfinished is I'm not yet finding a sense of 'joy' with the proposition - which is essentially Drupal and Drupal Commerce on AI steroids. At times there have been hints of 'joy' but at others too much baggage from the past undermines it.
Having this reset in the mountain air has certainly illustrated to me that the potential for AI to augment brain plasticity is vast.
With continued advancements in machine learning, neurostimulation, and brain-computer interfaces (BCIs), the possibilities for enhancing cognitive function, promoting recovery from neurological injury, and improving mental health are endless.
Exploring how AI can unlock the brain’s full potential is something that undoubtedly offers the potential for purpose and joy and seems like the absolute essence of a 'perfect union'.