The Nature Of Knowledge
In fact, most of the ideas we have about the world and our civilization we have learned so that we are who we are in good measure because of what we have learned and what we remember — Eric R. Kandel, Dec 8, 2000
Mirror
The universe is pregnant with knowledge. It got on without us for a very long time. Knowledge lay dormant in everything from quantum fluctuations to an atom, from DNA to a cell. Life tentatively flickered into existence in all its glorious forms powered by the free energy of our sun. In fits and starts and extinction events, it brought forth ancestors hardly recognizable as ours. They barely made it in a fierce competition for survival. As if by cosmic coincidence, mother nature, in all her vanity, fashioned a mirror out of a human’s conscious mind and held it up to admire her handiwork.
Wonder
This awe-inspiring story advances as humans proceed to accumulate knowledge. We are conscious! We marvel at how birds fly or bees dance or ponder how we may colonize a nearby habitable planet one day. Granted, these are all human ideas! But it could have been otherwise. A world devoid of any, much less sentient life, would be dull. For there is nobody to wonder. The existence of such planets in our cosmic neighborhood is one reason for awe.
The fact we can know and wonder is the whole point. It is made possible by our mind and its underlying apparatus. Learning and memory underpin our ability to encode and represent knowledge acquired by experience, captured by ideas, and transmitted by language. We live in the present-day knowledge economy. Over the last century, knowledge-acquisition has grown exponentially. How do organisms learn? How does learning alter behavior? Let us explore both facets.
Learning
Learning is the process of acquisition of new knowledge about the world. Nature required encoding experience for survival and proceeded to evolve nervous systems with their complexity to solve it. Behaviors are a set of actions in response to environmental stimuli that organisms enact and repeat. As experiences accumulate, they learn and retain them. The knowledge out there gets encoded in their nervous system.
Reward-based-learning is an explanatory framework that tries to explain an ancient evolutionary process roughly half a billion years in the making. The process is conserved in humans (genus: Homo) as in sea slugs (genus: Aplysia).
Life Experience
Today, a significant part of humanity resides in a civilized society. It was not the case for our primate ancestors in the African savanna. This scene may very well have occurred a million years ago in East Africa. Our cave-dwelling primate ancestors have started a controlled fire. A curious toddler gets her hand too close, only to reflexively withdraw it from the flame. Others in the group watching her have an opportunity to learn to avoid getting too close at virtually zero cost.
Any behavior is consequential, whether it is curiosity-driven or otherwise. Organisms must act despite sheer ignorance of their environment. Action underlies all behavior, even as outcomes remain uncertain. Myriad behaviors like satisfying hunger, warding off danger, or mate-seeking emerge in direct-response to stimuli. Cradle to grave, organisms acquire knowledge to adapt to changing environments. Behaviors get molded as a result.
Machines learn by algorithms and enormous data generated by humans — the domain of artificial intelligence. Outside that, the scientific study of learning in animals (humans included) has benefited enormously from insights gained by incisive introspection. In this post, let’s focus on non-associative and associative learning. Next post will cover the introspective insights.
Carrots and Sticks
Reward Drives Behavior
Learning a new behavior happens on encountering a new stimulus. External or internal stimuli trigger animals to act. The rewards (or penalties) of prior actions shape future behaviors. For instance, picking and eating a fruit constitutes a behavior. Withdrawing a finger from a hot stove or choosing not to are others. The memories of past actions get retained in the brain. Animals can choose to repeat (or abstain from repeating) learned behaviors. Reward-based learning captures this idea. It has three components.
Trigger → Behavior → Reward (Or Penalty)
Two examples and key takeaway:
- Doughnut (trigger) → eat (behavior) → feels good (reward)
- Seeing a predator (trigger) → escape (action) → survival (reward)
Modifying behavior requires altering the associated rewards (or penalties).
Surprise!
A new stimulus startles an organism, evoking that first-response. It gets sensitized (learns) by amplifying its second, third, or subsequent responses. Afterward, its response progressively peters out.
Blasé
Already known stimulus may elicit a habituated (learned) response that it can recall from experience. Unless the reward (or one that’s perceived) gets altered, it does not evoke any response. If and when it does, it is reflexive and effortless (innate).
Before sensitization, its behavioral repertoire for that stimulus is empty. After habituation, it’s replete. In both cases, associating past stimulus to the one at hand is rendered unnecessary. These forms of learning are non-associative.
Conditioning
Associating a specific behavior with a stimulus or learning the relationship between two is not unusual. Pavlov’s dog learned to associate the ringing of a bell with food. The mere ringing of a bell triggered salivation, even though there was no food eventually. The aroma of freshly brewed coffee gets some of us going each morning. Ordinary sights or sounds may bring back fond or longing memories of events past. Consider the following scenario.
Giving-in to peer pressure (stimulus), a teenage kid may take up smoking. Once initiated, her behavior gets reinforced over thousands of iterations throughout her life. Decades later, it may get re-purposed as a stress-reliever, except now, she is in a habitual rut, even trying hard to quit.
The anatomy of this behavior is part psychology and part physiology. Initially, peer-pressure may provoke it. The anticipation of a reward (cool-kid/stress-relief) facilitates and justifies it. Repetition hardens it, causing addiction. Nicotine stimulates the limbic system for dopamine release, and its lack ensures withdrawals. Eventually, even fleeting thoughts (of good-old-days or job-related stress) can trigger cravings. The addictive loop of the trigger → behavior → reward ossifies over time.
The behavior is an example of operant conditioning. It involves modifying behavior by reinforcement (nicotine triggering dopamine reward). We saw the molecular basis of memory and learning previously (link below).
Understanding the molecular basis of learning presents fascinating insights into animal behavior.
Behavior is Conservative
Imagine being able to correlate the psychology of behavior to physiological changes occurring in the brain. Eric Kandel struck this veritable gold-mine in the 1960s. Many have followed in his footsteps since. Isolating less than a hundred nerve cells of the sea slug (Aplysia Californica) from its abdominal ganglion, he studied a set of behaviors in its rudimentary repertoire. The fact that these nerve cells are gigantic couldn’t have hurt.
Its so-called gill and siphon withdrawal reflex is one such behavior that causes it to retract (behavior) them both when the critter is disturbed (stimulus). One can isolate these cells in a lab and watch the nerve cells respond. One can even alter the behavior facilitating or deterring the desired response for a sustained period. The molecular story (the strengthening of synaptic gap mediated by neurotransmitters) occurring as the animal gets habituated or sensitized to its changing environment can be written in detail.
So What?
After all, Aplysia is a simple invertebrate that forages for food and doubles up as a male or a female to reproduce. What can one make of a human brain with highly evolved and specialized structures for specific functions, driving complex behavior?
Recall the recurring theme: if things ain’t broke, they don’t get a makeover. It is the same old tape. It applies to humans, mice, fruit flies, and sea slugs, among the myriad other creatures on this planet.
Humans share many behavioral patterns with simple animals.The evolution of learning and behavior is conservative.
The How never needed an upgrade. The What and Where got upgraded. The enlargement (doubling) of cranial capacity and the emergence of anatomical structures like the hippocampus, the amygdala, cerebellum make us who we are — conscious, sentient animals that are only just-human. Indeed, it is an icing on the cake. But deep-down, we are animals that share behavioral traits with other humble brethren.
The fact that we get to wonder, go to the moon, or colonize space one day are damn good reasons to celebrate, to be curious, and take extra good care of our biodiversity.
This is the end of part three in the series. Parts one is here:
References
- Operant Conditioning of Gill Withdrawal in Aplysia, The Journal of Neuroscience, March 2006, Robert D. Hawkins, Gregory A. Clark, Eric R. Kandel
- Learning and Memory, Molecular Cell Biology, 4th Edition.
- Craving to Quit: psychological models and neurobiological mechanisms of mindfulness training as treatment for addictions, Judson A. Brewer, Hani M. Elwafi, Jake H. Davis
©️ Dr. VK, 2021. All Rights Reserved.
