As the world battles COVID-19, the public has gotten interested in science again. Every day there are headlines about the latest research about what spreads the virus, what slows it down, and what progress is being made towards the vaccine.

It’s a good thing to have more people following scientific news, but some of them have fundamental misunderstandings about how science works, and that knowledge gap is creating problems for all of us.

One criticism that often comes up goes something like this: A few months ago the scientists said to do X, but now they say it’s more important to do Y. If they were wrong about that a few months ago, why should I trust them now? Isn’t science supposed to give answers, not create more confusion?

This is an understandable frustration. But through a simple thought experiment we can get to the bottom of why scientific inquiry produces contradictory and confusing results, but is still correct, especially over time.

For the next few minutes, imagine you’re a scientist tasked with discovering the outline of a mystery shape. Like the real world, you don’t get all the information at once. It trickles in as different observations and tests occur. All you know is that the shape is regular (not a random polygon sticking out in odd directions) and that it's one unbroken line that doesn't cross over itself.

The first piece of information you get is that this point is part of the outline. We’ll call it data point 1.

This single point of information in a sea of white doesn’t tell you much. Is it a point on a square? A circle? This is the stage of discovery when you might say: There’s something going on here, but without more research we can’t do anything with it.

Soon enough, data point 2 comes in.

This still isn't much information to go on, but we can start thinking about ways the data might connect. Maybe there's a line between these points?

Data point 3 comes in.

With this limited data, a very early hypothesis could be that the mystery shape is a triangle.

How could we test this hypothesis? We might look for points that fall on the line between 2 and 3. But when we do that we come up empty, instead we discover data point 4.

This blows up our triangle theory, but that's ok, it wasn't based on very much information and we knew it was flimsy. Perhaps it was a diamond after all.

The next round of data is in, and this time we found two points, 5 and 6. This invalidates the diamond theory but offers a tantalizing possibility.

What if our shape is the prototypical outline of a house?

The more time we spend with the house theory, the more attractive it becomes. It fits all available data, it's elegant, and something about it just feels right. But just when we were feeling sure, here comes a new piece of information, data point 7.

What's going on here? Not only does this mess up our house theory, it does it in a really confusing way. It's hard to imagine a regular shape that would include 7. You might even be tempted to throw out data point 7 in favor of the elegant house solution. But you're a good scientist and wouldn't do that!

Data points 8 and 9 roll in soon after.

It looks like the house theory is truly busted. This shape obviously has a lot more going on than we initially thought. But data points 10 and 11 give us something new to think about:

The house can't be the answer, but we're starting to see a pattern emerge. Have you noticed that all the points seem to be falling on the same vertical or horizontal spacings? It forms a grid.

Whatever shape this is, it seems to be following some rules. Can you make a prediction about another likely data point? Perhaps in the grid corner above 3? 

A test is done and eureka, there is indeed a data point there!

You're now feeling confident that whatever shape this is, it has points that fall on the grid lines. Like the house theory, there's an order and logic to it that just feels right.

That is, until data point 13 comes in. This data point was found using a new collection method that's more precise and picks up on things the earlier method missed. It sticks out like a sore thumb.

How could there be a data point there? It doesn't make sense. This shape is proving to be a much larger, more complicated problem than anyone thought.

You could continue this process, getting more and more data, generating new hypotheses, and discarding those that contradict the data. This process is tedious, time consuming, and very unglamorous, but it's the method which drives all scientific discovery. For today, we'll skip to the end and reveal what the shape was. Take one last look at the raw data and make your final guess!

And at last the answer. Scroll down to see it.

What!? But that's so tricky! How were we supposed to know it had curves? How were we supposed to know it wasn't a classic geometric shape? And yet, with patient data collection, we would inevitably arrive at the right answer. Every piece of data would fall within the line, and none would ever be off it.

In a few minutes of exploring this mystery shape, you've experienced many of the same frustrations as a hard-working scientist. At the start you had to be patient and wait for more data to come in. Then you did your best to make a hypothesis based on that incomplete data. 

Do you remember when it looked like it might be a house shape? Were you a fool to think that? No! You were finding a hypothesis that fit the best available evidence. What would be foolish would be to cling to the house theory even when the data showed it couldn't be correct.

The same holds true to the more elaborate grid theory. It felt so right, but when the data contradicted it, you had to let it go. At each step you were forced to do one of the hardest things a human ever has to do; be humble, question your assumptions, and be open to new ideas.

If you thought this shape was complicated, consider what it's like in the real world, for the global scientific community working on matters related to COVID-19.

• The data is messy and in flux. Instead of clean points on a page you're dealing with humans and the complicated social structures they live in. 

• Instead of waiting until all the data is in, scientists must get health information to the public ASAP. This is why advice changes over time. As we get a better handle on the data, the outlines of how to avoid transmission become clearer. For example, we now know that the main method of contagion is sharing microdroplet filled air with others, as you would at a family meal, office meeting, or choir practice. This is why mask-wearing has taken center stage, when in earlier months it was less emphasized. We also know how the virus affects the body in much more detail. Where it was once thought of as a respiratory disease, we now know that the virus travels through the blood and prefers to hide out in snug, tiny blood vessels (of which the lungs have many). All of this informs treatment and leads to shifting stances on which medicines work best. 

• Unlike the fixed shape, in the real world the data will shift based on how we act. For example, a country's predicted curve will change based on how it handles quarantine and testing protocols. Imagine how hard the shape would have been to figure out if every time you made your best guess it changed its outline!

• Finally, the virus itself isn't static, it's evolving in real time. The shape is alive!

We hope this thought experiment has given you empathy and patience for the scientists working day and night to beat COVID-19. Please share this article with those in your life who don't understand why the egg-heads can't seem to make their minds up. 

It's not indecision, it's the messy process of scientific inquiry taking place before our eyes. The abandoned hypotheses, blind-alleys, and wild-goose-chases are usually done in quiet labs far from public view. COVID-19 has brought it all to the town square. When we look back, we'll see it as a masterclass in how good science saves lives and is the most powerful tool humans have ever created. 

Patrick Reynolds // Kenzai Founder