May 27, 2019 | David F. Coppedge

Brain Makes Automatic Statistical Inference

Your brain isn’t a computer, but contains a powerful one.

Scientists at the University of Cambridge say, “the brain is a really smart statistical machine: it looks for patterns and finds building blocks to construct objects.” For example, you can look at clothing on a mannequin inside a store and infer what the fabric feels like, just by judging its texture and color.

Our ability to extract distinct objects from cluttered scenes by touch or sight alone and accurately predict how they will feel based on how they look, or how they look based on how they feel, is critical to how we interact with the world.

By performing clever statistical analyses of previous experiences, the brain can immediately both identify objects without the need for clear-cut boundaries or other specialised cues, and predict unknown properties of new objects. The results are reported in the open-access journal eLife.

To accomplish this kind of inference, the brain breaks up the stream of information coming at it from the senses into chunks, the press release explains, a bit like pieces of a jigsaw puzzle. Experiments with puzzles showed that subjects correctly inferred the unknown properties from the chunks of known properties.

“These results challenge classical views on how we extract and learn about objects in our environment,” said Lengyel. “Instead, we’ve show that general-purpose statistical computations known to operate in even the youngest infants are sufficiently powerful for achieving such cognitive feats.

The paper in eLife did not use any reference to evolution, nor did the press release. Instead, it referred to something non-physical: the abstract reality of concepts.

  • The concept of objects is fundamental to cognition and is defined by a consistent set of sensory properties and physical affordances [properties that the environment offers the individual]. Although it is unknown how the abstract concept of an object emerges, most accounts assume that visual or haptic boundaries are crucial in this process. Here, we tested an alternative hypothesis that boundaries are not essential but simply reflect a more fundamental principle: consistent visual or haptic [touch-related] statistical properties.
  • The coherent organization of information across different modalities is crucial for efficiently interacting with the world and lies at the heart of the concept of what defines an object.
  • [S]tatistical learning goes beyond the learning of simple (pairwise) associations between the constituent components of objects, and has been shown to be best described as the extraction of statistically meaningful (potentially multivariate) latent ‘chunks’). Therefore, we propose that these latent chunks are the abstract representations that are built automatically during exposure and mediate the across-modality effects we observed.
  • Together, these results suggest that statistical learning is not only a domain-general mechanism but it also results in domain-general internal representations that could be the basis for the emergence of affordances and the abstraction of object concepts.

The brain, therefore, is adept at combining information and concepts. This sounds more like a top-down process of computational abilities, not a bottom-up pairwise association of sensory perceptions, as John Locke proposed for the emergence of ideas. Any bottom-up mechanical philosophy, in fact, has to presuppose the very abilities of thought that it tries to explain. That includes Darwinism.

One doesn’t have to use the silly “inner pickpocket” analogy contrived by the lead author, who said “These results suggest there is a secret, statistically savvy pickpocket in all of us.” Pickpockets might access the ability to infer properties by touch, but they simply use for evil purposes a complex ability with which all of us come endowed by our Creator.

The authors did not refer to evolution at all. How could they? A blind, aimless process like natural selection is oblivious to the world of logical inference, statistical inference, and concepts. Those require thought, even if they utilize the tools of neurons. Thought falsifies atheism. C.S. Lewis said,

Supposing there was no intelligence behind the universe, no creative mind. In that case, nobody designed my brain for the purpose of thinking. It is merely that when the atoms inside my skull happen, for physical or chemical reasons, to arrange themselves in a certain way, this gives me, as a by-product, the sensation I call thought. But, if so, how can I trust my own thinking to be true? It’s like upsetting a milk jug and hoping that the way it splashes itself will give you a map of London. But if I can’t trust my own thinking, of course I can’t trust the arguments leading to Atheism, and therefore have no reason to be an Atheist, or anything else. Unless I believe in God, I cannot believe in thought: so I can never use thought to disbelieve in God.

(Visited 489 times, 1 visits today)

Leave a Reply