Day 4, CoEDL Summer School. "Isn't that weird?"

“Just because the brain looks like a bowl of porridge doesn’t mean it’s a serial computer” (Michael Arbib, I think)

The brain is really weird. I’d love to be bothered to find a more nuanced word that captures the complexity and oddity of it, but “weird” is what I’ve been using all through summer school and weird it is. And then you get neurotypology, which attempts to make connections between one weird system (neurology) and another weird system (language) and the whole thing is completely bonkers. (In a good way.)

Like this little weirdity: stated as simply as possible, the brain makes predictions, and if something doesn’t go as predicted, you get a particular kind of "blip" or “neural signature” in the electrical activity. Due to the timing and type of the signature, it’s called an N400, but the name isn’t important - just think of it as something you can see on an EEG graph when the brain is surprised.

One way you can surprise the brain is to read a sentence where the final word is wrong for the context - for example, “She buttered her bread with socks”. In this case, “socks” is obviously the wrong word for the context, and you get an N400.

But! Again in overly simple terms, if you read that sentence and the final, wrong word is capitalised - as in,

        “She buttered her bread with SOCKS”

- then you don’t get an N400. The brain isn't surprised.

The (post-hoc) story for this is that the visual system notices the capitalisation early on before it’s even properly processed the word, and alerts the rest of the system that something weird is coming up - capitalisation spoils the surprise.

Well, that makes sense. I think.

But wait! Simply drawing attention to the incorrect word doesn't seem to be enough in itself to stop the brain being surprised; for example, you don't get this effect just by highlighting the word in a different colour. In that instance, the brain is still surprised enough to produce an N400. The fact that you can also reliably surprise the brain in these studies at all, where you might suspect that the participants are starting to figure out that “hey, some of these sentences are a bit weird”, is also weird.

What could be happening is that CAPITAL LETTERS actually encode a meaning that coloured letters don’t - in the age of computers and texting, capitalisation has started to carry the extra (emotional) meaning of shouting, and something about the visual presentation of the word having meaning - as opposed to just being visually more noticeable - might be enough to interrupt the brain's usual "I'm surprised by the rest of the meaning in this sentence" thingy.

Isn’t that weird? That’s so weird. The brain is so weird.

Moving on! Rotokas, a language spoken in Papua New Guinea, has one of the smallest phonemic inventories (meaning, number of sounds) in the world. It used to actually hold the world record for smallest phoneme inventory, and then a researcher more recently discovered that the Rotokas language, much like the language name itself, has an ‘s’ sound in it

At this point, “Linguists are not very quantitative” is I feel a relevant observation from Nicholas Evans.

Then there's this gem - apparently one day the speakers of a particular Papuan language had a meeting and decided to switch around the gender markings on all their words - so masculine words became feminine, and vice versa. The researcher who allegedly identified this somewhat surprising (N400!!) language switch “published a paper about it without a shred of evidence, produced a dictionary exemplifying it, and then died”.

That’s the last of my summer school updates - tomorrow I’m off to the first day of the Applied Linguistics conference. (Can someone explain to me how two of the biggest linguistic events in Australia happened to have a clashing day? I really, really need a timeturner.)