How to Think More Rationally: Lessons from Engineering, Cybersecurity, and Everyday Life

I wanted to write something to reflect on what I have learned and observed over the years. This article is inspired by real-life experiences as well as reflections from books such as Asking the Right Questions, The Crowd: A Study of the Popular Mind, and philosophical works that require deeper contemplation, such as The Republic.

What Is Rationality?

Before discussing how to become more rational, it is worth asking a simple question: what exactly is rationality?

In my understanding, rationality is a way of thinking and understanding the world that has been widely validated over long periods of time. It is an approach that, statistically speaking, tends to produce more benefits than drawbacks.

Of course, there are exceptions. Someone who is excessively rational may become overly cautious and reluctant to take risks, which can sometimes hinder innovation and progress. From a personal perspective, a person who treats everything as a strict calculation of gains and losses may also be perceived as lacking warmth or romance. After all, a healthy society needs both rationality and imagination, both realism and idealism.

For that reason, I believe most of us should strive to become rational individuals first, while preserving a small amount of romanticism. Rationality provides the foundation; romanticism adds color.

Rationality Through Practice

So how do we become more rational?

I believe the answer lies in practicing methods that have repeatedly proven effective. Rather than discussing abstract theories, I would like to explain this through examples from my own experience.

Consider engineering and scientific work. To solve technical problems effectively, one must first develop sufficient training and experimental experience. This is how we learn scientific thinking.

The process is simple:

Make a hypothesis.

Test it carefully.

Use evidence to determine whether the hypothesis is valid.

At the beginning, there may be multiple possible explanations. The purpose of experimentation is to determine which explanation is closest to reality. This is the essence of scientific thinking.

Most people working in science and engineering already operate this way, even if they have never explicitly studied the philosophy behind it. In practice, engineers naturally follow the principle of testing before implementation.

The Danger of Relying Solely on Experience

"Experiment first, implement later" sounds straightforward, yet it is surprisingly easy to forget.

Why?

The first reason is that we tend to trust our prior knowledge too much. We encounter a few cases, build mental models from those experiences, and then assume we understand the broader reality.

However, statistics teaches us an important lesson: the cases we have personally encountered are only tiny samples of a much larger and more complex world. They rarely provide a complete picture.

When we rely excessively on our own experience, we may choose the wrong technical direction. Months or even years later, new evidence emerges, forcing us to abandon previous assumptions and rebuild our understanding. The result is wasted effort and resources.

The second reason is time pressure.

Modern organizations often prioritize speed and efficiency. While this approach can produce short-term gains, it often leaves little room for research, experimentation, and careful validation. As a result, people rarely have the opportunity to slow down and think deeply.

There is wisdom in the old saying that careful work takes time. Slowing down allows us to see reality more clearly, while thorough experimentation helps us avoid costly mistakes before they occur.

Scientific research provides an excellent example. Researchers continuously study the latest literature in their field, building upon previous discoveries rather than reinventing the wheel. More importantly, they avoid paths that have already been proven ineffective, allowing limited resources and talent to be directed toward more promising areas. That is a healthier form of efficiency.

A Security Industry Example

The importance of experimentation is especially relevant in cybersecurity.

I once encountered a situation involving a detection system that behaved unexpectedly under a special edge case. The issue effectively bypassed the original detection logic.

At first glance, the solution appeared obvious. Several people quickly proposed actions based on intuition and prior experience.

However, as discussions continued, new examples kept emerging that challenged earlier assumptions. Every proposed solution generated additional branches of possibilities, much like an expanding decision tree. Eventually, the conversation became increasingly complicated, and we lost sight of the original goal.

At that point, I recalled an important lesson I had learned while studying machine learning: validate hypotheses through experimentation.

Instead of continuing endless debates, I proposed a series of experiments designed to test our assumptions directly. In reality, the hypothesis had only a limited number of possible outcomes. Once we framed the problem that way, the path forward became much clearer.

The experiments ultimately confirmed the correct solution, and the issue was resolved effectively.

This experience reinforced an important lesson for me: rationality in engineering does not come from imagination or confidence. It comes from evidence.

The more we rely exclusively on our past experiences, the more likely those experiences will eventually mislead us. Rational methods — especially experimentation — provide a far more reliable foundation for decision-making.

Rationality in Communication and Evaluation

Rational thinking is equally important when communicating with others.

One common example is evaluation and comparison.

Whether we are reviewing technology products or assessing solutions within a specialized industry, objective evaluation requires consistent standards.

Yet people often make a subtle mistake: they unconsciously change the evaluation criteria during the process.

This is rarely intentional. More often, it happens because those involved become immersed in the details and gradually lose sight of the original framework.

I once observed a situation involving security detection systems.

There were two fundamentally different scenarios:

• Real-time detection
• Post-event analysis

The key difference was flexibility.

Real-time detection systems require highly reliable decisions because they may block actions automatically. In such cases, the system bears much of the responsibility. False positives must be minimized.

Post-event analysis systems are different. They serve as decision-support tools, while users retain final authority. Consequently, they can tolerate a higher false-positive rate.

Initially, our evaluation treated all products as black boxes operating in a real-time detection scenario. We focused solely on their final outputs and compared them using the same standards.

However, as testing progressed, we gradually changed the criteria. Intermediate "suspicious" states produced by one system were eventually counted as final malicious detections.

This seemingly small adjustment introduced bias into the evaluation.

Why?

Because the original comparison assumed a real-time detection environment. If one system's "suspicious" category was designed primarily for post-event investigation, it would naturally tolerate a higher false-positive rate. Higher false positives often produce higher detection rates, making performance appear stronger than it actually is under equivalent conditions.

As a result, the comparison no longer reflected a fair evaluation.

The issue was not the technology itself. The issue was that the standards had changed.

How to Stay Rational When Communicating

So how can we remain rational when discussing complex issues with others?

My approach is simple.

First, resist the urge to respond immediately.

Assume that you know nothing about the topic and start from the beginning.

Give yourself time to examine the proposed standards, assumptions, and processes. Walk through the logic step by step, either mentally or on paper. Focus on maintaining consistent criteria and understanding how key metrics are actually measured.

Once you believe you have identified the core issue, avoid rushing to a conclusion. Organize your thoughts first.

Then explain your reasoning to another person who is also unfamiliar with the subject. Teaching often reveals hidden weaknesses in our own understanding. It helps expose cognitive biases and logical mistakes that we might otherwise overlook.

If two independent individuals can follow the same reasoning process and arrive at similar conclusions, that provides stronger evidence that the analysis is sound.

No one is free from bias.

I certainly am not.

That is why rationality is not a destination but a continuous practice. It requires humility, self-reflection, and a willingness to question our assumptions.

At the same time, it requires patience with ourselves and empathy toward others.

Perhaps becoming more rational ultimately means learning to think more carefully while judging less quickly.