An Essay review of Jeff Hawking's book, A Thousand Brains: A new theory of intelligence
Elements of a Paradigm Shift in thinking about thinking
While neuroscience remains stuck in a state of perpetual uncertainty, seemingly perpetually hovering over the intractable mind-body problem, Jeff Hawkins, an Engineer turned self-taught neuroscientist, boldly advances forward.
Drawing upon his engineering expertise and employing "constraint-guided" deductive thought experiments (which are subsequently validated through simulation analysis), Hawkins makes subtle yet significant refinements to his pattern recognition-prediction framework introduced in "On Intelligence."
In his second book, "A Thousand Brains," Hawking, following in the footsteps of Nobel Laureate Santiago Ramon de Cabajal and Vernon B Mountcastle, systematically unravels the mystery of the neocortex. He reveals that the neocortex contains thousands of copies of the same brain circuits depicted in de Cabajal's Nobel Prize-winning diagrams as cortical mini-columns nestled deep within the structure of each neuron.
These dendritic-driven circuits serve as the brain's mini-processors, receiving inputs and generating predictions. Notably, these circuits are replicated for every one of de Cabajal's 150,000 columns across every sensory region of the neocortex. Their most crucial function lies in using their locally-based predictions to prime parent neurons, prompting them to cast their votes for the most probable whole, integrated systemwide prediction based on the parent neuron's higher-level shared inputs.
The resulting coherent integrated prediction shapes our understanding of the world. As neurons are activated, they give rise to our thoughts, consciousness, and indeed, our perception of the external world.
While this discovery may seem groundbreaking, it is not the culmination of Hawking's significant mid-course correction to his earlier theory. The true crowning achievement lies in the revelation that during the process of making piecewise predictions, each neuronal circuit constructs its own reference frames responsible for constructing hundreds of thousands of sensory-mode-specific prediction models of the external world.
So, what does all this mean?
If these updates withstand rigorous peer review, Hawking's modest contributions will have successfully introduced a novel theoretical framework that profoundly reshapes our understanding of cognition and the brain's intelligence. Consequently, the neuroscience community will be compelled to abandon its pursuit of the elusive mind-body problem and embrace the realization that the majority of neurons in our neocortex are dedicated to creating and manipulating reference frames that the brain employs to construct models for planning and thinking.
These hundreds of thousands of reference frame-generating mini-processors collectively store all our knowledge throughout our lives as a distributed network of dynamically evolving neuronal connections.
The culmination of Hawking's work is the elimination of the mind-body problem, as there is no mind, only a body. Neuronal activity permeates every aspect of our existence.
Having navigated through a labyrinth of intricate constraints, Hawking's refinements to his "On Intelligence" pattern recognition prediction framework provide a comprehensive picture of how the brain comprehends the world. In doing so, it begins to offer answers to some of neuroscience's most perplexing questions, such as: How do our diverse sensory inputs coalesce into a unified experience? What transpires during our thought processes? What constitutes consciousness? How can two individuals arrive at divergent beliefs from the same observations? What is the essence of abstract thinking? How does the brain process language? And why do we possess a sense of self?
A few crucial details are worth noting. Santiago Ramon de Cabajal, the first to meticulously examine the intricate circuitry within the neocortex, was awarded the Nobel Prize for his groundbreaking discovery of the structure of neocortical columns. One of de Cabajal's initial observations was the arrangement of neurons in layers based on variations in size, density, connectivity, and behavior.
As we delved into "On Intelligence," we discovered that inputs enter our brains at the lowest rung of de Cabajal's six-layer hierarchy. When predictions are confirmed, they become part of our shared knowledge base of experiences. Conversely, if predictions are incorrect, they ascend the cortical ladder, learning and becoming more abstract at each higher level. At the pinnacle of this hierarchy lies language and generalized abstract thinking, exemplified by mathematics.
As Mountcastle had forewarned in his seminal paper, "An Organizing Principle for Cerebral Function: The Unit Module and the Distributed System" (reprinted in "The Mindful Brain" by Gerald M. Edelman and Vernon B. Mountcastle, MIT Press, 1982), despite the uniform appearance of circuits across the neocortex, this phenomenon may hold significant implications. It suggests that every region of the neocortex employs the same algorithm to perform its respective functions, which likely encompass far more intricate and crucial tasks beyond mere sensory feature extraction.
Driven by this intriguing idea, Hawking relentlessly pursued it, determined to unravel the more profound secret it held.
After five arduous years of dedicated research, primarily relying on thought experiments and computer simulations, Hawking finally struck gold.
The revelation is that Mountcastle's secret lies in the fact that our brain connections recognize patterns that generate predictions, and subsequently store them as distributed dynamic reference-frame-based models of the world. These models are continually created, learned, and stored through our everyday experiences.
With every experience, new knowledge is added to the model by forming new synapses. The small percentage of neurons that are active at any given moment represent and account for our perceptions, thoughts, and consciousness. Ultimately, they are merely what it feels like to have an active neocortex. When the neocortex is inactive, the rest, which comprises the memory that makes up the distributed knowledge base of our stored experiences, comes into play.
Through simple deduction, the author uncovered Mountcastle's secret: the neocortex stores everything we know as a systemwide distributed network using map-like reference frames to construct its prediction models. The brain utilizes these models to engage in survival-enhancing thoughts. Intelligence lies in our ability to learn and apply these models to enhance our survival.
Trailblazing! Ten stars
(Article changed on Dec 26, 2024 at 6:37 AM EST)
