When you’re making decisions you want to think rationally, objectively, even scientifically.

But our brains are so easily distracted and sometimes outright fooled.

Why can’t we always be rational? Here are five reasons:

1. Anecdotes

A recent study showed, no matter your education, a good anecdote will lower your ability to think clearly. What is a TED Talk but essentially one big long anecdote? Think about it.

2. Graphs! Charts! Formulas!

They all show convincingly that we’re more willing to believe something if it contains graphs charts and formulas. Put a graph plotting correlation and people tend to think it’s causation even if it contradicts the text they’re reading next to it.

3. We find it really hard to change our minds

A 2013 study had participants evaluate research on gun control. Interestingly, people better at handling quantitative information showed more bias, using their skill to either be forgiving in their interpretation or slam the findings. We’ll pretty much use all our talents to justify our pre-existing beliefs.

SEE: Artificial intelligence: A business leader’s guide (free PDF) (TechRepublic)

4. Overconfidence

A 2003 study of 308 university students found most had an inflated viewer of their ability to understand five media reports describing recent scientific research. We tend to confuse understanding the language with understanding the concepts. You know what I mean?

5. We all have agendas

IQ is not a predictor of how good you are at rational evaluation. More relevant is a mental attitude that seeks to override our natural instincts. In other words the more aware you are that you’re prone to bias the better chance you have of correcting for it.

But don’t take my word for it. Go out there and read the research for yourself. Just keep in mind where the information is coming from and account for your own biases as you do.

For more on why it’s hard to think rationally, check out this article from The British Psychological Society: 5 Reasons It’s So Hard To Think Like A Scientist.

Also see: