A sniff test to determine consciousness

Could a simple ‘sniff test’ predict recovery of severely brain-injured patients?
26 May 2020

Interview with 

Duncan Astle, Cambridge University; Helen Keyes, Anglia Ruskin University


Brain schematic


This month, local experts Duncan Astle and Helen Keyes have been delving into a few neuroscience news stories for us. This month, Duncan’s been looking at a paper about sniffing. Specifically, using sniffing in a clinical context, to assess how consciously aware someone is. Severe brain injuries can render someone in various states of conscious awareness, from a full coma at one end, towards the more conscious end of the scale. Duncan explained that it can be really difficult to determine someone’s level of consciousness, and the paper he’s been looking at sets out to use the sniff response to try and figure this out.

Duncan - Well the idea is that your sniff response might be very closely related to relatively automatic brain processes. When you sniff something really unpleasant, like some sweaty shoes or something, without perhaps even realising it, your body adjusts your breathing pattern to accommodate the fact that there's a really foul odour. Sometimes you're aware of that, but sometimes you're not aware of that at all. Your brain is doing it in the background. And so they're trying to rely upon this relatively automatic brain response as a way of getting a reliable reading out of essentially how consciously aware the person is.

Katie - Is it then that if you smell something awful, you breathe, what, less or more shallowly so you're not taking as much in?

Duncan - Yeah, exactly.

Katie - What specifically do they do to try and test this idea?

Duncan - They had 43 patients, and for every patient they would do the sniff test, so they would present different smells that were either neutral, pleasant smells like shampoo or unpleasant smells like rotten fish. And after each presentation they would look at people's breathing pattern - so the flow of air through the nose as they intake breath. Within each session they would have multiple smells that they would present. After every session, they would administer a clinical assessment that tries to evaluate how conscious someone is. So whether they are, say, in a minimally conscious state or at the other end of the scale, whether they are essentially in a coma, what's sometimes called unresponsive wakefulness syndrome. Or what used to be called vegetative state.

And what they're able to show is that if you are more at the severe end, those subjects did not differentiate the different smells. The air flowing through their nose wouldn't change depending upon whether there was a pleasant smell, an unpleasant smell, or a neutral smell. However, at the other end of the scale, the closest to being conscious, those patients would very reliably distinguish those different types of smells, so there'd be different intakes of breath depending upon the type of smell.

And actually there were 10 patients more at the severe end, who did show a differentiation of smell. And that might seem surprising, but when they tracked those 10 individuals, all of them later transitioned to the much milder, minimally conscious state. So it actually seemed to be that the sniff test was a better way of predicting their future prognosis than the standard clinical assessment.

Katie - Why is it that the sniff test might outperform other checks?

Duncan - The whole point of someone being minimally conscious at best or in a coma at worst is that it's incredibly hard to get any information from the individuals and so that's the kind of ultimate challenge. And I think the reason the sniff test works so well is that in a way it's quite a crude response. And is thus seemingly very reliable. The physiological response to some disgusting smell, which seems to be so automatic that it can actually tell you about some underlying awareness and potential prognosis, so potential for improvement, seems to be really robust.

Katie - Can we be confident that 43 people are enough to say something meaningful about this study?

Duncan - Yeah. So how many people do you need in order to know is a really good question. Ultimately, the more people you have, the better. The fact that the effect is so robust and so clear in only 43 patients suggests to me that we should have some confidence about it. Now ultimately the real acid test is going to be when you scale this up. So imagine that the sniff test became incorporated with standard clinical guidelines for assessing patients who are in this condition. Then you can start to collect data on a scale, on a kind of national scale, to see whether this really, really works in front line clinical practice, and that's ultimately going to be the kind of acid test of whether this has real value in practice.

Katie - How would you view how significant this test is?

Duncan - Really good question. So I guess there's two ways of thinking about it. One is about clinical utility. The more aware or more conscious the participant is, the better their chances of recovery and that might drive certain types of clinical decision. So that's one side of the coin. The other side is about what's it like for relatives and loved ones of these individuals? You can imagine it's very, very difficult when you are unsure how aware the person is. And so I think that potentially there are two benefits. One is to the clinical care or the clinical team and I think the other is providing more information for the relatives of these patients.

Katie - It seems like a relatively easy test to implement.

Duncan - With these assessments, they have to be standardised, right? You have to know that if you're down in Addenbrookes (hospital) and you are in a minimally conscious state or you're unresponsive, you're going to be given the same kind of tests as you'd be given anywhere else in the country, or in the world for that matter. And so the only challenge I can think of is standardising.



The paper Helen’s been looking at this month deals with the rather disconcerting theory of cognitive dissonance - the mental difficulty that arises when one’s behaviour conflicts with one’s beliefs. In order to resolve this conflict, you could change your behaviour, or your beliefs. But Helen points out that this paper is suggesting a third option. Change your memory!

Helen - So here in a series of experiments that involved quite a lot of people, over 3000 people in total, participants were made to make a series of five decisions where they got to choose how much of a pot of money they would give to an anonymous fellow participant. So they chose how much money they would keep and how much to give to the anonymous partner. And later on, unbeknownst to them, they were then asked to recall how much on average they gave to their partner. And to avoid inflation effect, so to avoid an effect where the participant was trying to just look good, the participants were financially incentivised to remember correctly. So they were told that if they remembered correctly how much on average they gave, they would receive a financial bonus. So there really was a reason to remember correctly here.

Across all of the studies that they did, they found that stingy participants remembered giving more than they actually did. So they were misremembering. And this wasn't the case for generous participants. So it wasn't a general inflation effect that everybody just remembered giving more money. If you were generous initially in what you actually gave, you remembered pretty accurately. You maintained that positive self image by actually having the good behaviour in the first place. But if you were stingy in your behaviour, you misremembered, even when you were financially incentivised to remember correctly, you remember giving more money than you did. And this was particularly true for participants who had a high standard of behaviour for themselves. So participants who really said that they believed in fairness, these participants in particular had that misremembering effect if they were stingy.

So it looks like when people think their own actions are selfish or unfair, they misremember having acted more fairly, which preserves their image of themselves as a good person.

And I think the take home here is really that the brain is rather efficient, as we know. So it adds to our evidence that the brain always uses its resources very wisely. Why bother changing your behaviour or even your attitudes when you can simply change your memory of an event. Well-played brain! And it might explain why we often remember events so differently from one another. If everyone has a tendency to remember themselves in the best light, this might explain why we really can have two different versions of events. Finally, it might prompt us to examine our actual behaviour with more scrutiny before we feel comfortable with our idea of ourselves as a good person.

Katie - It makes me feel quite disconcerted. How do we know this is motivated behaviour? Are they able to control for genuinely just misremembering something?

Helen - So we know that it wasn't a general misremembering effect because people in different groups behave so differently. So people were really accurate in remembering how much they gave if they were generous givers. And indeed the more that you saw yourself as different from your own ideal - so if you really did believe that in a certain fair distribution across society - those people had a stronger misremembering effect on people who didn't really care about fairness in the first place. There's no reason to think that those people would have worse memories. It's very likely a motivated effect in order to protect our own self image.

Katie - Do we think this is a, I don't really want to put a value judgment on it, I was going to say "good" thing, but is it a self preservation thing, or are we just deluding ourselves? It's really disconcerting to think that our memories perhaps aren't as objective as we think they are.

Helen - From a cognitive psychologist point of view, we're less interested in morals and what is good and what is bad. We're more interested in what the brain does that's efficient and that's clever. And this is really a very clever thing for the brain to do. If we do have a drive as humans to see ourselves as good, generous, fair, honest people, if that's something that we value, it's really nifty from a cognitive psychology point of view that your brain can get there, can do that really efficiently without having to do much work with regards to actually changing.

Katie - But do you think there's perhaps a wider implication to this? I'm just thinking about situations where you do rely on your memory to be objective. Like, I don't know, say if you're being interviewed by the police or something. Objectivity must be pretty important there, right?

Helen - Absolutely. I mean there's just a huge body of evidence to show that our memory is absolutely not an objective thing and that we shouldn't be relying on it. Famously eye witness testimony is terribly inaccurate. We misremember things for all sorts of reasons. I think what this paper is adding is that we don't just misremember things because we have a particular interest or focus on one aspect of what's happening. We might actually misremember things just to make ourselves feel better.

Duncan - One of the challenges is that people are not very good with nuance, right? We like to have this notion of people as good people, or bad people. We're not very comfortable with the idea that even people who are, who are good and they're not monsters, it might do some bad stuff. We've kind of straight into this realm where people are either good people or bad people and this study kind of goes with that, right? If your own actions are not sort of in keeping with the view that you have for yourself, then we just adjust the memory slightly so that it's less inconvenient. We're not so comfortable with the idea of, "well, maybe it was a bit stingy back there, but it doesn't make me think that I'm not a good person". It's all part and parcel I think of this slight dichotomy that we've developed.

Katie - It just really freaks me out. It makes me feel like maybe I am in the matrix after all.

Helen - [giggles]



Add a comment