The windscreen anecdote has been doing the rounds on social media again.
If you’re not familiar with the story, it goes a little something like this. “When I was a kid we would drive long distances for holidays and get bugs all over the windscreen. I don’t see any bugs on the windscreen anymore, therefore….” The interpretation, whether implied or stated explicitly, is that this is yet more evidence that a global insect decline is happening.
There are obvious flaws in this assumption, but the anecdote still strikes a chord with so many people, perhaps through some kind of confirmation bias. We know that biodiversity is in trouble, we know humans are having damaging effects on the environment, so it must be true, right?
Let’s unpack this anecdote again. For windscreen bug splatter to be an accurate measure of general population trends over time, a number of assumptions need to be true to rule out any possible confounding factors:
1. Human perceptions and memories are always accurate and unbiased.
2. The biomass of all insects remains constant across time and space.
3. The aerodynamics, road presence and speed of motor vehicles have remained consistent over time.
4. The width, speed limits, infrastructure and daily usage of roads have remained consistent over time.
5. The insect community found on or near any given road at any given time is representative of the broader insect community.
6. All insect species are equally likely to splatter on windscreens of passing cars.
We can quickly see that none of these assumptions are valid or possible.
Okay, but isn’t the anecdote still useful as a hypothesis? No, even then it’s flawed. To be a valid scientific hypothesis, a statement must be falsifiable. This means you must be able to design an experiment that can falsify, or disprove, it (not prove it; proof doesn’t exist in science).
So in this case, you need to design an experiment that can disprove the statement: There is less insect splatter on windscreens today compared to X decades ago. Without inventing a time machine, we can never test or falsify this hypothesis. Even if someone has photographic records from X decades ago, you would need to repeat the drive exactly (same car, same season, same weather conditions, same road corridor etc.) and run enough replicates to rigorously test the hypothesis. And even if you can repeat the drive exactly, all you can show is the difference between two points in time, which is not the same as a trend over time.
The perception of a bug-splattered windscreen is just that – a perception. What one person thinks is ‘lots of bug splats’ might be a pretty clean windscreen to the next person. Like most anecdotes, it’s also space and time specific so it’s impossible to generalise trends to other times and places.
If we want to believe perceptions and anecdotes are just as accurate as scientific evidence, what about those anecdotes that don’t fit our chosen narrative?
I see regular posts on social media sharing the opposite to the windscreen decline anecdote – that someone drove somewhere through clouds and clouds of bugs, or experienced mass bug slaughter on their windscreen or radiator grille after a long drive. These anecdotes get a small handful of likes but are largely ignored, while the opposite version goes viral.
Here’s a personal anecdote. In November 2019, in the midst of the bushfire catastrophe, I flew to Tasmania for a conference. Our arrival into Hobart was delayed by seven hours, due to severe widespread bushfire smoke on the mainland and an unseasonal storm front that passed over Tasmania that day. We had a four hour drive to our accommodation, so we ended up barrelling up the east coast of Tasmania through the night, instead of the leisurely scenic afternoon drive it was supposed to be. As soon as night fell, thick clouds of insects appeared in front of our headlights, literally thousands of them, and did not dissipate for 75 km along that dark, lonely coastal road. I have never driven through so many insects in my life. The windscreen itself was relatively clean at the end of the drive, just a regular scattering of tiny splats (our rental car was a brand new Toyota Kluger).
If we had done that drive in the afternoon, as we were supposed to, we would never have seen those insects. Why did we see so many? Was it the climatic context of that particular summer? Was it because that stretch of road remains largely undeveloped and ‘wild’? Was it the time of year? The time of night? The oceanfront location? A complete coincidence?
I didn’t stop to identify them, because we were tired and cranky and had to get to our accommodation before we were locked out. But they were unlikely to be representative of the broader insect community in the region – most likely nocturnal species associated with temperate coastal systems that can quickly reach high population density, like chironomids, mosquitoes or small moths.
My experience that night doesn’t disprove the ‘windscreen phenomenon’, any more than other people’s opposite experience ‘proves’ it. But it does highlight how many factors influence what you see on your windscreen.
Human memories are tricksy things and we can’t always rely on the accuracy of our own childhood memories. Anyone who’s lived through a family feud over a past event will recognise how even closely-related people can remember the exact same experience in completely different ways.
Your memories might also be ‘created’ by what you hear and read others ‘remembering’ – so hearing others consistently reporting a mythical past event may lead you to believe you also remember that same event.
Anecdotes are powerful storytelling tools. So does it matter if people use this ‘story’ to engage people in the bigger conservation issues? Maybe not. But it does matter if it’s presented as a ‘real’ scientifically-verified phenomenon. If we start accepting unverified anecdotes as generalised scientific evidence, will we lose the skills to know the difference between what is and what is not science?
© Manu Saunders 2021