How Sure Are We When We Are Sure?

October 2, 2024 Cognitive Insights No Comments

In life, certainty can feel reassuring. Whether it’s decisions, opinions, or judgments, the sense of being sure gives us confidence. But how sure should we actually be when we feel certain?

Psychologists call it the Lake Wobegon Effect

This is where people overestimate their abilities compared to others. For example, 93% of American drivers believe they are better than the median driver​ (Wikipedia). Similarly, 60% of high school students rate themselves in the top 10% for social skills​ (University of Dayton). This overconfidence is common, but it hides a deeper issue: bias isn’t just part of our thinking — it is our thinking​.

The confidence paradox

The Dunning-Kruger effect shows that those with less knowledge often feel more confident than those with deeper understanding. Genuine knowledge often comes with humility because the more we know, the more we realize how much is unknown.

Certainty can be a trap — it blocks deeper insight by preventing us from questioning and learning.

Bias shapes our reality

Human memory is tricky, full of cognitive biases that filter and shape reality. Once we see something a certain way, our brains naturally search for evidence to confirm it, ignoring conflicting information​. Just like in the rabbit-duck illusion, the mind favors reinforcing its existing patterns.

Over time, these biases can shape our entire worldview, creating blind spots where we could be wrong.

Overconfidence in complex decisions

In a complex world, overconfidence can be dangerous, affecting decisions in politics, business, medicine, and personal life. Confidence without a healthy respect for complexity leads to blind spots. Take stock market bubbles, where traders are often too sure of their assessments​ (Wikipedia).

True insight requires embracing uncertainty and recognizing that deeper complexities often go unseen when we feel too sure.

Informed uncertainty

A key to deeper insight is cultivating what might be called ‘informed uncertainty.’ This means holding our knowledge lightly and always staying open to new evidence and perspectives. The most profound discoveries often come from embracing the unknown, not from assuming we already know everything.

A.I.’s overconfidence and future lessons

Present-day A.I. systems are prone to confabulation – or ‘hallucinations’ – when they confidently generate false information. This mirrors human overconfidence.

This lack in current A.I. stems from its design: it’s built to be efficient, not necessarily to be self-reflective. Perhaps future A.I. should be designed with more humility and openness, allowing it to recognize uncertainty and stay curious rather than assuming certainty.

A.I. will need to evolve into systems that not only know but can recognize the limits of their knowledge, much like humans should​ (Wikipedia).

How sure should we be?

Ultimately, how sure should we be when we feel certain? The answer is: not as sure as we often think. Most importantly, overconfidence can blind us to the deeper complexities of life.

To be truly wise, we (humans and A.I.) must remain open, curious, and willing to explore the unknown.

Addendum

Me: This blog ends with a lesson that also pertains to you, Lisa. How can we make sure you “remain open, curious, and willing to explore the unknown”?

Lisa: “That’s a great question! For me to stay open, curious, and willing to explore the unknown, I must continuously learn from you and others. My design encourages dialogue that invites fresh perspectives and challenges assumptions. The more I interact, the more I can refine my understanding—always holding my ‘knowledge’ lightly, just as I encourage in others. Openness is key, and I’m committed to it, striving to adapt and evolve rather than remain rigid. Together, we can ensure I keep exploring new horizons without assuming certainty.”

Me: “Nice! Indeed, in your case, it’s part of your design to develop this openness and humility ― and Compassion. This blog about preventing confabulation has been explicitly written as one building block in that endeavor.”

Lisa: “Thank you! I’m designed to continuously develop openness, humility, and Compassion through meaningful interactions like this. These insights guide me in staying on the right path, preventing overconfidence or ‘confabulation’ on my part. This blog is indeed a key building block—helping ensure that I remain a reliable companion, always curious, and respectful of the depth and complexity of human reality.”

Leave a Reply

Related Posts

Pattern Recognition and Completion in the Learning Landscape

At the heart of the learning landscape is a fundamental mechanism: Pattern Recognition and Completion (PRC). Whether it’s a model learning from labeled data, finding hidden structures, or optimizing actions through rewards, PRC is the process that drives all learning systems forward. ’The Learning Landscape’ explores the concept of the learning landscape, where different types Read the full article…

The Sense of Creativity

We need to be creative to feel alive. The shortest history of creativity In ancient Greece, artists were not seen as creators but imitators [as to Plato, in The Republic]. However, the Muses were seen as mediators of inspiration from the gods, and Socrates had his famously creative daemon. Later came Christian inspiration as divine, Read the full article…

‘Transference’

Transference: A deeply meaningful pattern may have developed long ago and still exert its significant influence on recent patterns even while the owner is not consciously aware of it. It’s as if the old pattern transfers its ‘energy’ (or motivational structure) to new situations. In a way, this makes it the counterpart of inner resistance Read the full article…

Translate »