#mcgurk

Optical illusions are pretty well-known, like a rabbit that looks like a duck (or vice versa?) or figures that appear to move or change colour when you look at black and white patterns. But there are also many fun kinds of linguistics illusions!

In this episode, Lauren and Gretchen get enthusiastic about some of our favourite bits of language that can be two things at the same time. We talk about the where the Yanny/Laurel illusion that became popular on social media a while back came from, the McGurk (and hypothetical future McGawne?) Effect, using the Stroop Test to find spies, hallucinating words from musical instruments, the Comparative Illusion (aka "More people have been to Russia than I have"), and making our own speech to song illusion to infect you with (sorry) (no but seriously) (please read the content advisory and proceed to the separate audio file at your own risk of earworm infection).

* the Stay Lingthusiastic speech to song effect audio link here.

Update: I’m not dead! I know I haven’t been posting regularly. I’m sorry. It’s down to two things really: a) I’ve been very busy with the new job and b) I’ve frankly really struggled to find any kind of inspiration lately - I suppose that’s what happens when your life is taken over by your job. And you’re an auditor.

But this week this whole Yanni/Laurel brought about a bit of a brainwave - not least because it’s done nothing but do my nut in. Literally every one of my social media feeds is infected with these words. Apart from Twitter - but only because it contains an option to mute words - but even then I’m still swamped by the overhyped, equally annoying sequel: green needle/brainstorm. 

However, as with most things I hate, I’m going to put my back into this.

A few things are going to happen in the next few minutes: we’re going to unpack the explanations behind these phenomena, and then I’m going to try to shatter your perception of the world.

The Yanni/Laurel thing has now been confirmed as an aural phenomenon: if you were to plot the frequencies present in the recording against time, much like something you’d get on Audacity or any other kind of audio-editing software, you would see that this clip is made up of a mixture of high and low frequency tones. Yanni is formed from the higher frequencies. Laurel is characterised by lower frequencies. It’s like listening to what are essentially two different tracks of music that have been overlaid. If your ear is more attuned to higher frequencies (perhaps the younger among you), or you’re the kind of animal that turns down the bass on your speakers, you’re going to hear Yanni. The vast majority of people however hear Laurel, because, well, we’re older.

Now we come to Laurel’s little sister: green-needle/brainstorm. She’s a little smarter, a tad more interesting and she was allowed to wear makeup from a younger age. What you can hear in this recording can be changed depending on simply the word you’re looking at when you hear it, which is much more than a physical phenomenon - it’s a psychological one. We know something to be true - that we’re hearing the same sound each time, but our perception of it changes. This is interesting for two reasons: firstly, on a psychological level it helps us to dissect how our brains work, and secondly, more importantly, it proves to us that objective truth is a fallacy.

Green needle/brainstorm is a slightly more evolved example of the McGurk effect, which is a widely known and studied phenomenon where your brain can interpret the same audio/visual recording as two different sounds depending on the context it’s given. This context often comes in the form of a visual cue, which is much better explained by the folks at Horizon: https://www.youtube.com/watch?v=G-lN8vWm3m0. Ultimately it comes about because of the top-down processing in our brains. What does this mean? Well, effectively our brains process a whole load of information all at once, and uses its analysis of this to work out the most probable explanation of our current circumstances to make sense of the world. For example, imagine you’re on safari. There’s not a cloud in the sky, you’ve got the sunroof down and you’re driving through a woody looking area. You hear a dense flock of startled birds swiftly fly out of the branches above you as your jeep slams through the undergrowth. They’re so close you can feel the beat of their wings in the air around you, and suddenly you feel something cold drip onto your hair and down your neck.

You’ve surely been shat on.

You look up and see a monkey peering down at you from the sunroof, drooling.

But for a second, you believed you’d been shat on, well, because you hadn’t noticed the monkey. Hey, we’re not perfect.

Additionally your analysis often relies on the outcomes of events that it’s seen before and it projects these probabilities on the current situation in order to work out what’s going on. For example, if you’ve had a horrid cough before and went to the GP, who told you it was pneumonia, the next time you get a cough you’re more likely to think it’s pneumonia again, even though that’s actually quite unlikely. The McGurk effect combines these two analytical phenomena. Most of the time you hear a hard “k” sound and seen a particular mouth shape, it’s turned out to be a word starting with that letter. But if that same exact sound is accompanied by a “g” mouth shape, your brain goes “Well, based on past experience, that word must begin with a G”.

TLDR: we can easily trick ourselves, and others based on the subset and quality of information we allow ourselves to see.

As well as being a fun illusion, like all other illusions it highlights something more insidious: we’re all primed for bias - it’s inherently how our brain deals with the mound of information it receives every single millisecond of every day. If we didn’t skip straight to conclusions we’d end up overthinking everything and ultimately not taking any action. Evolutionarily speaking, our ancestors would have died if they didn’t spring into action on hearing twigs breaking, assuming it was indicative of an imminent attack. The benefit of catching our predators pre-arrack vastly outweighed the excess energy expended on false alarms. Out of our ancestors, those who were the quickest to leap into action on hearing the quietest of sounds lived the longest. However, in modern day terms this kind of cranial processing doesn’t work as well. Sure, based on he gait of the person in front of you at Kings Cross, you might predict they’re going to take a hard swerve left to the Victoria line and you can use that information to prevent an embarrassing collision. I’m not saying this fundamental system of processing doesn’t have its merits - I’m just saying it has fewer: we don’t spend every waking moment fending off predators any more, because we’ve built infrastructure, terraformed land, driven predators out of their natural habitats and evolved societies that provide you with security against dangerous individuals in return for a cut of your income.

So we find ourselves in conflict. We have brains that are used to using whatever information is conveniently available and pre-existing knowledge to judge, but vastly reduced the need for that judgement. We’ve also reduced the benefits of this judgement - if anything it’s often frowned upon. We’ve developed a new term for unnecessary judgement: prejudice. And we often think we’re well aware of our own prejudices and can therefore escape them - but I’m here to tell you that the vast majority of us can’t. Take this for example:

Try to memorise these words: Adventure, curious, sun, brave, clean, friendly, ocean, white, fruit, learn, free, wholesome, holiday, talented.

Now read this: Alan is making plans for his gap year. He wants to visit the South America but is struggling to fit that in with his plans to take part in a motorcross rally. He missed it the year before because he broke his leg in the practice round. He needs to find his passport, which he lost on his last trip back from Bali and hopes his friend accidentally picked up. He also wants to visit India and needs to find time to move into his flat in Camden before he starts at his London uni.

What do you think of Alan?

What would you have thought had you memorised these words instead: Jealous, green, selfish, cocaine, petty, reckless, red, corrupt, idiot, lad, careless, clown, rude.

Go back and read the paragraph again - see what you think.

He might have seemed a bit of a gap yah wanker that time, methinks.

This is something called priming, which is an extension of the broken thinking we discussed earlier. It’s exactly how advertising works - we can’t help but associate things together when they’re close together, either spatially or temporally. You judged Alan because those lists of words made you linger on different sets of details in the narrative each time. If you start to form an opinion, you’re more likely to see details that reinforce them.

So what’s my point? I’ve just shown you that this kind of thinking is inescapable: you knew where this article was going and yet you likely painted a picture of two different Alans. I’ve told you that our brains are hard-wired for bias. That our perception of the world is inherently, inescapably warped. That we all have our blind spots. That we can convince ourselves of anything depending on what details we choose to notice. And that our choices of details are rooted in past experience. The logical conclusion of this is that as we get older, we get more biased. Something happens, we learn from it, maybe even form a slight opinion, we stumble across varied details in the subsequent hours, days, weeks of our lives, and out of these details our brains are primed to pick out those that are familiar, opinions and beliefs are justified and strengthened, our filter for details gets narrower, our opinion gets stronger, our blinkers come down even more, so on and so forth. Incidentally it’s eerily similar to how evolution works.

We’re built from bias.

This means that in order to even be able to grasp at objective truth, you have to work. Really work. Hard. And I think this is something that is totally overlooked in our current political climate. We all think that facts are facts - they’re not, simply by virtue of being beheld by us. We, these flawed, inherently biased networks of synapses in cages of bone and bags of skin. But we need to guard against this. No man is an island, and as a society we need to believe in the concept of objective truth, even if we accept we’ll never achieve it. If we don’t, we lose our baseline for discussion, leading to a society which is unable to sort opinion from fact: one in which radical, absurd and harmful ideas could propagate at the same speed as those more closely aligned with common sense, driven by whimsy. Truth is the tare weight for any battle of wits - without it, there could be no consensus.

So if we must believe in an objective truth, but can only ever see it through a glass, darkly, so to speak, how can we polish the lens?

This brings us full circle to audit, my bread and butter, and perhaps why the question of truth is at the front of my mind. Audit is fully preoccupied with objectivity and truth - firms drop clients and lose money because of it all the time. This is because our job is to take the draft financial statements a company prepares before they’re published and ensure that the figures in them haven’t just been made up, or tweaked. We need to assess whether the numbers show an adequately “true and fair” view of what’s happened to that company during the year. As with everything, we can never be 100% certain of the truth, or fairness of accounts, so we test the numbers to a reasonable level of assurance.

Believe it or not, there are a couple of aspects that are quite interesting about it:

Firstly, sampling. Much like biologists attempting to study animals in a large habitat, the feat of fully auditing every single transaction a company makes during the year is nigh on impossible. Instead we choose a representative sample of transactions and look at those in more detail to work out if they were recorded correctly. We’re always terrified of choosing the wrong number of transactions - if we audit too few, we might miss one large one which was fraudulent or recorded wrongly - one typo could change an overall profit to a loss. If I wasn’t thorough enough, I could lose my job over that.

Secondly, we rely heavily on the people running the audited company to tell us what happened during the year. If for example they failed to tell us that they underwent a huge merger, we might audit them against the wrong set of financial standards. We might think it’s all fine by those standards - but that’s a false positive. We used the wrong measure of truth, because we didn’t have all of the facts.

So why did I bother to tell you all this?

Because auditors measure truth for a living, and you might learn something from the highly discussed and regulated procedures we use day in, day out. The next time you find yourself judging something - anything, for that matter, however small - ask yourself these questions:

How much detail can I subtract from the situation before I change my view of it?

Is there a detail or perspective I’m missing because I’m being primed by my prior beliefs, assumptions or experiences?

If you find the threshold for Q1 and an example for Q2, you’ll be much closer to the truth than you were before.

You never know, you might end up finding truth in the most unlikely of places, and applying measured skepticism can lead to some of the most - sometimes surprising - eye-opening revelations. Those “MY LIFE HAS BEEN A LIE” moments. Never be afraid of disagreeing with your past opinions - it’s a sign of learning.

🔍 ¿Qué es el Efecto McGurk? · #CU53.COM!