Why we make bad decisions and what influences them

Have you ever wondered why we make poor estimates, or why we judge situations differently when they involve us and when they involve someone else? In computing terms, our brains have certain bugs or features, depending on whether the reader works in quality assurance or development. One interesting feature of the brain that affects our decisions and judgment is what we call cognitive biases.

Let's start with a simple question.

A bat and a ball cost 1.10 Czech crowns together. The bat costs 1 crown more than the ball. How much does the ball cost?

If you answered 0.10, you got it wrong (the correct answer is 0.05), but don't worry about it. The author of Thinking Fast and Slow, Nobel laureate Daniel Kahneman, posed the same question to students at prestigious American universities such as Harvard, Princeton and MIT, and more than 50% got it wrong. If you answered 0.10 without giving it much thought, you fell for anchoring. Anchoring is a cognitive bias that involves giving weight to the first piece of information when making a decision. If I tell you Václav Havel died past the age of sixty and then ask how old you think he was, you'll guess somewhere around sixty. Maybe sixty-three or sixty-five. The first part anchored you there.

According to Kahneman, our brains use two systems to make decisions. He calls them System 1 and System 2. Each has different characteristics and we use them in different ways. System 1 gives very quick answers, which can be wrong. We use it unconsciously for what you could call “thinking without thinking.” Its main advantage is that it responds quickly to the signals we send it. We use System 2 to solve more complex problems, and its results are much more accurate. Its answers take time, though, and running it costs something. We need both systems to make decisions, and we use them depending on the stimulus and the situation. When we face a decision about a new problem, System 2 kicks in. If we have faced the problem before, System 1 guides the decision.

Most decision-making errors come from System 1, which is subject to cognitive biases. What other biases or preconceptions influence our decisions? I picked a few from Pragmatic Thinking and Learning: Refactor Your Wetware and a few others that caught my attention.

Self-serving bias
The tendency to credit ourselves for success and attribute failure to external factors. If I do well on an exam, I credit my intelligence or the hours I spent studying. If I fail, I blame the teacher's bias, their inability to explain the material, or bad luck in the question I drew. This mechanism protects and reinforces our pride or self-esteem.
Need for closure
The desire to eliminate uncertainty or doubt about a problem. It leads us to try to settle every uncertainty and doubt.
Confirmation bias
The tendency to favor information that supports our own hypotheses and opinions. It applies to how we gather, retain and interpret information. Poor decisions influenced by this bias have been identified in fields such as military affairs and politics.
Mere-exposure effect
The tendency to prefer things simply because we are familiar with them. In the 1960s Robert Zajonc ran an experiment in which he first showed two groups of people various nonsensical Chinese characters. He then told them that each character had a positive or negative connotation and asked them to say which it had. People tended to assign positive connotations to characters they had seen before and negative ones to the others.
False memory
Mixing up events or details as they occurred with the way we remember them. How many times have you and your friends told the same story about something you went through together, only for it to sound like a completely different story? Everyone had a different version even though you all experienced the same thing.
Planning fallacy
The tendency to underestimate how long a task will take, regardless of our experience with similar tasks. It applies to tasks in any field. We underestimate because we fixate on an optimistic scenario or simply want to be done by that time.
Gambler’s fallacy
The tendency to think probabilities change because of what has already happened, even though they stay the same. If I toss a coin five times and it comes up heads five times, I tend to think it has to come up tails the sixth time. The probability is still the same.
Hindsight bias
The tendency to see events that have already happened as more likely than they seemed at first. It is sometimes called the “knew-it-all-along” effect. It probably comes from a distortion in the information stored in memory, which leads to a mistaken judgment.
Fundamental attribution error
We judge individual behavior more by personality or character and overlook the situation and context in which it occurred. It shows up especially when we judge other people's behavior. When judging our own behavior, we tend to credit ourselves for the outcome and give less weight to outside influences.

I've listed only a few, but even this small sample shows how much our decisions can be distorted without our realizing it. One way to counter these distortions is to engage System 2. Each cognitive bias can be addressed differently, but we have to be aware of it. For example, we can start by learning to work with some uncertainty (need for closure). That lets us avoid making certain decisions too soon. If we're choosing a programming language, tool or methodology and settle on one option, it helps to find someone who will take the opposing position and challenge us (confirmation bias). If something fails, we can look at our own role and the wider context as well as external influences.

If the subject interests you, I recommend the following articles and presentations in addition to the sources already linked.

• Moneyball and the Science of Building Great Teams, an article I drew on heavily.
• A visual study guide to cognitive biases