Hierarchy of Patterns and Biological Hierarchies

By Eric Cruet

In William Gibson’s seventh novel, “Pattern Recognition”, the main character (Cayce Pollard) is a legend in the field of market research.  She is paid handsomely to recognize cultural and social patterns that corporations can turn into cash.  The truth, according to her friends, is that her sensitivity is closer to allergy, a morbid and sometimes violent reactivity to the symbols of the marketplace. Hired by Blue Ant, the world’s hippest ad agency, for the sort of high-corporate re-branding she’s known for, a more intriguing project emerges when the head of the firm asks her to determine who’s producing a mysterious series of video fragments that have gripped the imaginations of people around the world. The source of this footage, carefully concealed, has so far proven untraceable.  But what if the sense of purpose and meaning that she and others perceive in the footage is only an illusion — in other words, faulty pattern recognition? 

In Ray Kurzweil’s new book “How to Create a Mind: The Secret of Human Thought Revealed”, Kurzweil argues that the underlying principles and neural networks that are responsible for higher-order thinking are actually relatively simple, consisting of hierarchies of pattern recognition modules which make up the neocortex.  He states that many machines running current AI (Artificial Intelligence) software perform these same functions using similar principles and imitating the same neuro-structures that are present in the human brain.

Recent discoveries in neuroscience seem to confirm a subset of his Pattern Recognition Theory of the Mind or PRTM. Operating on pattern matching principles, it is hierarchical in nature for the processing of a particular input, such as the letters in a word.  It is also redundant, massively parallel to a hierarchy of concepts, for instance, when it processes the letters in the word “apple”, the differences in writing styles, the spoken word “apple”, variations in accents, different perspectives (on a tree, on a teacher’s desk), shadings, shapes, and varieties.

The human neocortex is capable of a very wide range of very complex abilities, yet the underlying structures and principles that are responsible for these abilities are very simple and straightforward, according to Dr. Kurzweil.  For example, he describes the architecture of the pattern recognition module and its operation.  Each module stores a “weight” for each input dendrite indicating how important that input is to the recognition.  But a good question would be: by what mechanism are these weights assigned?  In addition, he compares the successful recognition of a pattern by its corresponding module to the way NLP (Natural Language Processing) software encodes characteristics of time and space to recognize words with same letters but different pronunciation.  How does this work for the successful recognition of levels of attractiveness, joy, embarrassment (and the resulting blushing reflex)?

Although the book legitimately addresses brain processing functions that are similar computationally to state of the art AI mathematical modeling and learning, neuroscience and cognitive scientists need to think about the control and interface mechanisms between the neocortex and the other major brain components (thalamus, brainstem).  In closing, one key postulate from the text is the hierarchy of abstractions between the functional processing of the hierarchy of patterns by the pattern recognition modules to the biological hierarchy of cortical columns in the neocortex.

From a systems perspective, the task at hand presents itself as a large multivariate problem that probabilistically challenges whether a complete brain could ever be created to operate in the same way.  But it’s a hell of a start……

The fundamental uniformity of the neocortex (see above) was reconfirmed in a recent study using the latest in brain scanning technology (loc. 1199). The lead scientist in this study, Harvard neuroscientist and physicist Van J. Weeden, explains the findings thus: “‘using magnetic resonance imaging… what we found was that rather than being haphazardly arranged or independent pathways, we find that all of the pathways of the brain taken together fit together in a single exceedingly simple structure. They basically look like a cube. They basically run in three perpendicular directions, and in each one of those three directions the pathways are highly parallel to each other and arranged in arrays. So, instead of independent spaghettis, we see that the connectivity of the brain is, in a sense, a single coherent structure” (loc. 1212).

 

References:
Grinvald, A., & Hildesheim, R. (2004). VSDI: a new era in functional imaging of cortical dynamics. Nature Reviews Neuroscience5(11), 874-885.
Jackendoff, R. (2002). Foundations of language: Brain, meaning, grammar, evolution. Oxford University Press, USA. (2), 279-292.
Joseph, R. (2011). Neuroscience: Neuropsychology, Neuropsychiatry, Behavioral Neurology, Brain & Mind: Primer.
Kurzweil, R. (2012). How to Create a Mind: The Secret of Human Thought Revealed. Viking Adult.