Unbounded associative learning: what is it, characteristics and how is it expressed?

Unbounded associative learning has been associated with consciousness in animals.

It has long been known that organisms learn. There are various stimuli in the environment and, when a living being is exposed to them, it internalizes information from it.

However, not all organisms learn equally, much less can they learn the same thing. A human being, a dog or a bird can learn that the sound of a bell means food, while a microbe, a sea sponge or a plant cannot.

Unbounded associative learning is an idea that emerges from a novel approach in animal cognition and aims to study the evolutionary origins of consciousness. This type of learning could be a marker of the transition from non-conscious to conscious life, a question we will explore below.

What is unbounded associative learning?

What is consciousness? What can be considered a conscious organism? At present there is some consensus in considering a conscious system as that which has the capacity to experience, to have its subjective point of perspective of the world and of its own body. This system must be able to generate an elusive property that philosophers have called phenomenal consciousness in order to be considered.

Human beings are conscious. We are animals that at some point in evolutionary history have gone from being non-conscious organisms to conscious organisms, however, the line that separates the two types of lives is much debated. That boundary must exist but finding it remains the greatest challenge of modern science and philosophy.

Yes, a volcano, a rock, a cyclone or my computer are probably not conscious of their existence (although a panpsychist would not think so), but what about a dog, a plant, a jellyfish? They are undoubtedly life forms, but are they conscious beings, and if so, to what extent? The debate is still open, but thanks to the work of Simona Ginsburg and Eva Jablonka this question seems to be approaching a solution with a new concept: unbounded associative learning.

The central postulate of unbounded associative learning is that those organisms that can perform it are living systems capable of learning about the world and themselves in an unlimited way, in the sense that they are free to interact with the environment and manifest a Wide range of responses.. This idea has been explained in a recent paper of his, "The Evolution of the Sensitive Soul: Learning and the Origins of Consciousness" from 2019.

Associative and non-associative learning.

Before we go into more depth about unbounded associative learning and its importance as a key to drawing the line between non-conscious and conscious life, let's take a brief look at the types of learning that exist. While there are many different types, learning can be grouped into two broad categories: non-associative learning and associative learning, the latter being the more sophisticated.

Non-associative learning

Non-associative learning is the type of learning common to all (or almost all) life forms, including unicellular organisms. In the case of animals, this type of learning takes place in the peripheral nervous system, although it is true that there are certain types of learning that would fall into this category and involve the activation of the central nervous system. It should also be noted that not all animals have an equivalent to our nervous system.

Non-associative learning includes habituation and sensitization.. Habituation is the phenomenon that occurs when a sensory receptor responds less frequently to a constant or repetitive stimulus. For example, habituation is what happens when we stop noticing the touch of the clothes we wear or, also, when we no longer notice the frames of our glasses on our nose and ears.

Sensitization is just the opposite of habituation. It occurs when a certain stimulus is absent for a long time, making the sensory neuron in charge of that stimulus more likely to respond when the stimulus suddenly reappears. That is, the more time that passes without receiving stimulation, the more sensitive the neuron becomes to it. Sensitization can also occur when an unexpected stimulus is given, such as noticing a mosquito bite.

Associative learning

Historically, associative learning has been divided into two types: classical or Pavlovian conditioning and operant or instrumental conditioning..

Classical conditioning is also called Pavlovian because it is this type of learning that the Russian physiologist Ivan Pavlov investigated with his famous dogs. Initially he observed that the dogs salivated when presented with food (unconditioned stimulus) so he decided to ring a bell (conditioned stimulus) every time he presented that food and see if they associated both stimuli. After several attempts he got the dogs to associate the sound of the bell with food and when they received this sound stimulus they began to salivate, even if they had no food in front of them.

Operant conditioning involves the association between an action and a reinforcing stimulus, either positive or negative. For example, if we have a rat in a cage it will be freely exploring the space until it finds a button which, incidentally, it presses. Each time it presses the button it receives a piece of food, causing the animal to associate its action with positive reinforcement. For the reinforcement to influence the behavior it must have some kind of value for the organism, whether it is attractive (p. e.g., food) or aversive (e.g., electroshock).

Characteristics of unbounded associative learning

In view of all this, it is time to talk directly about unbounded associative learning. Ginsburg and Jablonka introduced in their work a new nomenclature when referring to different types of learningThey called classical conditioning "world learning" because it involves associating external stimuli (unconditioned stimulus and conditioned stimulus). As for operant conditioning, they call it "self learning" because in this case it does involve the association of an action of the individual with the reinforcement (stimulus).

These two authors postulate that for associative learning to occur to a lesser or greater extent there must be a minimally complex neurological system, a brain or something similar, and this is why this learning is not found in all animal species because not all of them meet this criterion. For example, jellyfish do not have something similar to a brain and the only learning that has been seen that they can do is non-associative.

There are some very primitive creatures that show something similar to associative learning: limited associative learning.. This mode of learning involves very simple associations that are very far from being considered conscious by the organism making them, but somehow or other approaches the line that separates non-conscious life from conscious life.

The next level of complexity is that of unbounded associative learning. This involves associations between compound stimuli coming from various sensory modalities, such as hearing, sight, taste and so on. These stimuli are integrated and elicit a wide variety of motor actions. As we progress up the phylogenetic scale, the capacity to integrate different perceptual stimuli and the behavioral range grows, becoming virtually unlimited.

And it is here where we end up talking about unlimited associative learning, which is considered to be behind consciousness and which would be would be behind consciousness and would be the key element indicating a minimum of consciousness in an animal species.. The more varied an animal's response to the same stimulus and the more it can adapt to it in various ways, the more it makes sense to think that behind its behavior there is a minimum of intentionality and understanding, something that resembles in one way or another our idea of consciousness.

• You may be interested in "Does empathy exist in the animal kingdom?"

Conscious animals

According to this idea dogs are conscious because they can associate in many ways the same stimulus and, also, can perform all kinds of behaviors in response to it. In fact, thanks to their "consciousness" we can make them learn many tricks or prevent them from misbehaving by applying negative reinforcement every time they perform an undesired behavior. On the other hand, a sea sponge, a much simpler organism, cannot associate different stimuli, which would be indicative of an absence of consciousness.

Along the same lines as what we have just discussed, both Ginsburg and Jablonka and other experts in animal cognition consider that several animal species must have a minimum of consciousness, especially those that in laboratory conditions have been shown to be capable of associating different stimuli with different stimuli.especially those that in laboratory conditions have been shown to be capable of associating different sensory stimuli. It is agreed that all vertebrates (fish, amphibians, reptiles, mammals and birds), some arthropods (e.g., hymenoptera) and very few cephalopods (mainly octopuses) have consciousness since they have shown unlimited associative learning.

Likewise, although this type of associative learning may be key to identifying a species with consciousness, it does not mean that the organism itself can be conscious. The research is still open and the idea of unlimited associative learning is too novel to claim that it is a good demarcator of the line between non-conscious living beings and conscious living beings. However, this concept has helped to clear up a little more of this debate and it seems that in the next few years it will be possible to establish more clearly to what extent a living thing is or is not conscious.

Bibliographical references:

• Birch, J., Ginsburg, S. & Jablonka, E. (2020) Unlimited Associative Learning and the origins of consciousness: a primer and some predictions. Biol Philos 35, 56. https://doi.org/10.1007/s10539-020-09772-0
• Ginsburg S, Jablonka E (2020) Consciousness as a mode of being. J Conscious Stud 27(9-10):148-162.
• Ginsburg S, Jablonka E (2019) The evolution of the sensitive soul: learning and the origins of consciousness. MIT Press, Cambridge
• Ginsburg S, Jablonka E (2010b) The evolution of associative learning: a factor in the Cambrian explosion. J Theor Biol 266:11–20. https://doi.org/10.1016/j.jtbi.2010.06.017