Showing posts with label cognition. Show all posts
Showing posts with label cognition. Show all posts

06 May, 2014

The flagpole of entitlements and obligations

Since the last few years, I've been interested in how the web is affecting our lives and changing the way we think. While trying to understand this, I came across the question of how our sense of "entitlement" and "obligation" affect the way we think.

The reason I came across this dilemma itself is a different question. It had to do with the wide disparities I observed in the emphasis placed on different aspects of online privacy and security, by different people.

Nevertheless, this post is not about online privacy and security, but on our sense of entitlement and obligation. Here is a theory that I've developed:

The Flagpole Model

Imagine that inside each of us is a tall flagpole. A flagpole has some element of it over the ground and some element of it buried underground.

The part of the flagpole that is overground is our sense of entitlement. It is what we think the external world owes us.

The part of the flagpole that is underground is our sense of obligation. It is what we feel we owe to the external world.

The flagpole deeply affects our strategic disposition with which we approach the world. When we are predisposed with a sense of entitlement, we tend to be aggressive, obdurate, righteous, judgmental and assertive. When we are predisposed with a sense of obligation, we tend to be empathetic, compassionate, cooperative and accommodating.

Each of us are born with a certain ratio of the flagpole above ground and the rest underground. Our innate sense of entitlement versus obligation is visible approximately when we are two years old. Take a set of two year old kids and we will probably see that the distribution of pole-lengths above ground is a Gaussian bell curve. That is, there are a small number of kids who would have an innately high sense of entitlement, and a small number of kids who would have an innately high sense of obligation. Most of the kids would be somewhere in the middle, with an almost equal sense of entitlement and obligation. They are more or less, as selfish as they are friendly.

As we grow up, our culture and system of education gently adjusts our flagpole over the years.

In some cultures, the flagpole is pulled up as people are reminded about their rights and entitlements by their culture that celebrates gumption. In some other cultures that celebrates compliance, the flagpole is pushed in and people are reminded more about their duties and obligations, rather than rights and entitlements.

Stability of a system of flagpoles

Given a society where everybody has a flagpole inside themselves, there is often a clash of entitlements. This beings us to a concept of "stability" of the society.

A society is said to be stable if a sense of entitlement by someone can be matched with a sense of obligation or duty on the part of others. In other words, the total amount of flagpole lengths that is above ground should match the total amount of flagpole lengths that is underground, for the society to be stable.

If we see a society where some folks seem to have inordinately high levels of entitlement, and the society still seems to be stable without any upheavals, it means that their high sense of entitlement is matched by several others in the society living with an inordinately high sense of duty or obligation.

If a society comprises of all people with a high sense of entitlement, it will result in conflict and clashes, till a point when some of the flagpoles are forcibly pushed underground.

Alternatively, what happens to a society where everyone lives with a high sense of obligation? On the face of it, such a society will appear to be stable too, since a sense of obligation will not seek anything from others. But such a society is not evolutionarily stable in game theoretic parlance. It means that a small set of incumbent actors with a high sense of entitlement can easily overwhelm the society. In addition, such a society is prone to a variant of the tragedy of the commons that is also well known in game theory. Since everyone in the society has a high sense of duty, selflessness and service, there is a temptation for everyone to slowly increase their own sense of entitlement over time, as it can be easily matched by a sense of obligation on the part of someone or the other. A society with an inordinate sense of obligation is also not in Nash equilibrium. As long as everybody else remains in a state of heightened sense of duty, there is a rational incentive for any given actor to give up this state and adopt a heightened sense of entitlement instead. That is one of the sources of my skepticism about the widespread practices of "bhakti" and "devotion." A society comprising of all "bhakts" may be very peaceful and empathetic, but such a society is not likely to exist for long.

Phase transitions

The sense of entitlement and the sense of obligation are like Yin and Yang. If we try to increase one of them without limit, we end up with the other. As the saying goes: "If you go too far into the East, you end up in the West." Or the Andy Grove quote: "Every successful organization contains the seeds of its own downfall."

If as a society we emphasize too much on one of the above senses, there appears a point at which there is a "phase transition" and the flagpole goes in the opposite direction of the emphasis.

If our environment emphasizes too much on a sense of duty then there comes a point when our sense of entitlement becomes close to zero. A sense of entitlement is very important for survival. Once it becomes zero, there is no mechanism for our body to convince itself to even live. But then, nature does not allow us to reach a zero sense of entitlement, without putting up a good fight. When our sense of entitlement becomes too low, our primal survival instincts surface and starts a desperate push of the flagpole upwards, resulting in rebellion, defiance and revolt.

When our sense of entitlement becomes too much, there comes a point when we start feeling hollow from within. A sense of obligation is very important for our notion of self worth. Our sense of self worth is based on how much we are needed by others. When we have lived all our lives pursuing only our self interest, there comes a point when we start feeling "soulless" and superficial living only for ourselves. This starts a desperate push of the flagpole downwards, explaining why some very wealthy people suddenly turn spiritual or go off into depression and abdication.

The entire history of the world can be seen from the lens of how this flagpole has been manipulated within ourselves and over others. Much of leadership, governance and persuasion has been about fiddling with our and others' flagpoles. 

10 September, 2013

Existential, Logical and Epistemological thinking

It is quite well known that cognition or "thinking" happens in several layers. One of the most popular theories today is by the works of Kahnemann and Tversky about Systems 1 and 2.

System 1 is our intuitive brain. It performs fast, subconscious computations and jumps to conclusions. It also embodies our emotional state, intentions, desires and dispositions in its computations. It can compare across dimensions and is innately "rational" -- that is, it is driven by self-interest and biased by who we are.

For most of our lives, we live by System 1. However, System 1 occasionally requires the services of "System 2" -- which is our conscious and deliberate reasoning process.

System 1 includes all activities of a perceptive nature. This includes, for example, reading a word written in a language we understand. System 2 on the other hand, is semantic in nature. It tries to understand the meaning of what is being asked for and does justice to it.

Consider the following example that shows dissonance within System 1 requiring the services of System 2. In the following set of words, say out loud whether they are written in upper case or lower case:

UPPER
LOWER
upper
LOWER
lower
UPPER
upper
lower
LOWER

We see that our System 1, first tries to just "read" what is written, rather than answer the question. System 2 then kicks in to make deliberate judgements based on what the question requires us to do. Answering this question requires a degree of "self-control" on ourselves to refrain from just reading out what we see, and to address what is being asked of us.

System 2 requires much more energy to run and is far more slower and inefficient than System 1. Also, System 2 is "lazy" (I'd like to use the term "rational") in the sense that, if System 1 already has computed an answer, System 2 would rather ratify the answer with an explanation, instead of invest resources in recomputing the solution.

The post-facto rationalizations we make to justify an impulse buying of say, a car or a camera, is an example of System 2 in action, ratifying the decisions taken by System 1, rather than computing the merits of the decision on its own.

Basically, System 2 is an "imperative engine" -- it performs logical inferences and derives conclusions from premises.

But -- and this is important -- it does its work within the confines of a mental model. The mental model comprises of the underlying premises on which inferences are made. System 1 often implicitly provides the "box" or the mental model within which System 2 performs. The post-facto ratification is an example.

This brings me to the point of this post, where I would like to propose the existence of a third layer of cognition. Rather than just calling it System 3, let me give specific names to each of these layers.

System 1 is the "existential" layer of cognition. It is an embodiment of our Person. Its thinking is driven by who we are, what we like, what we are afraid of, and such. It is oriented towards making quick, intuitive "blink"-type decisions.

System 2 is the "logical" layer. It can perform deliberate, systematic logical entailments from premises. It innately knows the rules of logic and can build an argument towards a conclusion. However, System 2 works within the confines of a mental model -- the axiomatic framework within which inferences are made. The mental model within which System 2 performs its computation, is often influenced by System 1. In that sense, System 2 is more of a "rationalizer" of the decisions of System 1 and System 1 can often "manipulate" System 2 to rationalize its leaps of intuition.

Sometimes however, we need to think beyond the confines of a mental model, and question the premises within which our System 2 is doing its thinking. This is where System 3, or the "epistemological" layer kicks in.

Consider the following question: What is 423 x 7? Answer this without using any pen or paper.

This is a typical problem that involves the operations of System 2, where we mentally calculate how to add 423 to itself 7 times.

But then, System 2 does not ask what is meant by 423, 7 and what is meant by 'x' in the question? We just assumed that their interpretations are known. Suppose I were to say that "423" and "7" are strings and 'x' is the concatenation operator, and the answer is 4237.. this constitutes a re-interpretation of the problem in a different model.

Consider the following problem:

Let's say, it is your birthday today and your friends have planned a surprise party for you. You come home in the evening after work, and your friends are all in there and yell, "Surprise!" And, you show surprise on your face. 

But suppose, you come to know of the surprise party that your friends have planned for you. When you come home and they yell "Surprise!" can you still show surprise?

The answer is yes! Because, your friends don't know that you know about their surprise.

Suppose your friends know that you know about their surprise? Can you still show surprise? 

The answer is still yes! Because, you don't know that your friends know that you know about their surprise!

This kind of a problem is characteristically different from that of "What is 423 x 7?" It requires deliberate thinking -- but deliberate thinking of a totally different kind than that used in System 2.

Epistemological thinking requires us to question our premises and perform multiple interpretations as part of its thinking.

The epistemological layer is required in many strategic situations where we need to think across different mental models. We not only have to question the basis of our own premises, but also have an understanding of what other interpretations can exist for the question at hand.

Most questions requiring us to reason about ethics and morality have this characteristic. Consider the following question:

A 10-year old argues in court that he should be allowed to be adopted by the rich childless couple in the next neighbourhood as requested by them, as they can provide him with a better life, and they also want him as their son. Besides, he argues, he cannot be held responsible for his "accident of birth" -- he did not choose his parents while being born. How should the court rule? 

This problem is complex simply because there are several models within which it can be interpreted, each giving different answers about what is the "right" thing to do. We can keep adding several additional points to this problem, making the decision sway in different directions. Suppose, we added a dimension to this problem that the current parents of the boy were living in penury and had problems of abusive behaviour. Or, perhaps that the current parents of they boy were middle class, hard working couple, who just couldn't match the wealth that the rich couple could afford. Now what?

Most strategic thinking, requires us to invoke the services of System 3, while rational thinking requires the use of System 2.

System 1 coupled with System 2 is "rational" -- in that they can both work towards fulfilling self-interest (managed by System 1) in a way that results in utility maximization (managed by System 2).

However, the activities of System 3 are synergistic and is about "enlightened self-interest."  Its aim is not to maximize utility for oneself, but to find a harmonious interpretation across several mental models to result in overall net gains in utility.

In other words: strategizing for a win is System 2, while strategizing for a win-win is System 3.

Models of Intelligence

Today, whenever we speak of AI, we mean a specific model of AI based on large language models and diffusion, both of which, undergo foundati...