 Okay, so today's topic is clinical masking. Now, when we've covered audiometry, we've covered examples where we place sounds in one ear and they could be air conduction or they could be bone conduction. And we've assumed that when we deliver sounds to, let's say the right ear, if I play an air conduction sound to the right ear, it's the right ear that detects it. And if I put a bone conductor on the mastoid behind the right ear, then we've assumed pretty much that it's the right ear that detects it. Similarly, when we swap over to the left, we think that we're testing the left ear. But there are a number of reasons why this isn't always the case. There are cases where when we stimulate one ear or we attempt to stimulate one ear, we're actually stimulating both ears simultaneously, either at the same level or at a higher or lower level. And it makes it a little bit ambiguous about which thresholds we're actually measuring. So today we're talking about clinical masking, which outlines the reasons why this, what we call cross hearing can occur, the transfer of the signal between the ears. But it also will go into detail about how we go about resolving that ambiguity and making sure that the ear that we think is testing, that we're testing is the actual ear that we're examining. And that process is called masking. Okay, so just to refresh your memory about the mechanisms of bone conduction. So this is going way back to the first lecture. We mentioned that when we put a bone conductor behind the ear and stimulate, there are a number of different mechanisms, all of which are happening at the same time. So the three mechanisms of bone conduction we mentioned were the distortional mechanism. So we pass vibrations into the skull that vibrates the skull bones and causes the walls of the cochlea to distort slightly. And that creates sound borne vibration in the inner ear fluids, which is detected as sound by the cochlea. So it's the first mechanism of bone conduction. The second one is what's called the inertial mechanism. And that is the vibration of the skull and the temporal bone causes with the sound presented by the bone conductor causes it to move. But we've got the oval window where the stapes foot plate is attached. And the stapes foot plate is loose in that area and it's got a bit of mass. And the fact that it has mass means that it will lag behind any other movement. So if you imagine the stapes is sitting there in space in the middle ear cavity and we've got the cochlea that we're vibrating with the bone conducted sound, that vibration moves backwards and forwards and creates a relative movement of the stapes foot plate and the oval window which imparts sound borne vibration again into the cochlea fluids. So that is the inertial mechanism. So that's two ways we've got now of getting sound into the cochlea with bone conduction. The third one is called the osteotempanic mechanism. And that is reliant on the fact that we have the bony portion of the ear canal when we vibrate the bones of the skull that vibration actually causes the air inside the ear canal to have sound in it. It's basically passing that sound into the air and the ear canal. And just like any air, just like any sound in the air, it travels towards the tympanic membrane, vibrates the eardrum, the middle ear structures and gets passed into the cochlea just like it would be if it were air conduction. You can sense that right now if you were to put your fingers on your ears and talk, you would hear your own voice quite loudly. And part of that is a result of the sound into your ear canal. One thing you'll also notice is that if you block your ears, that bone conducted sound starts to sound louder. So if I were to block my ear, I am actually stopping some of that sound escaping. So that's the other really important thing is that we have sound energy entering the ear canal, some of it goes towards the tympanic membrane, but some of it also heads out. So we've got sound escaping out of our ears whenever we have bone conduction taking place. And we'll talk about what happens if we stop that sound from escaping a little bit later on. So everyone clear on that? They're the concepts of bone conduction. So when we apply a bone conductor behind the ear, we're actually stimulating both ears at once. This is something we haven't mentioned up until this point. So vibration from the bone conductor not only stimulates the ear that we're trying to test, and we call that the test ear. We're gonna abbreviate it to TE. So we're not only testing the test ear, but the sound also crosses through the entire skull and makes its way to the contralateral ear, the ear on the other side that we're not testing. We call that the non-test ear or the NTE. And it's a very efficient process. When we vibrate the skull with the bone conductor, it's incredibly efficient. And the sound level that we read on one side of the ear is exactly the same as what we get on the other side. So the vibration essentially reaches both cochleas at the same level. So what that means is if I put a bone conductor on you and play a sound at 10 dBHL to the test ear, then I'm actually also stimulating the non-test ear, in this case the left ear, also at 10 dBHL. So that's a bit of ambiguity that we've got going on there. And we call that cross hearing. So imagine we've got a situation like you can see here where we've already done some air conduction results and we've received the following thresholds. So the right ear threshold, as you know, represented by the red circle is at 40 dBHL. And the left ear threshold is at 10 dBHL. That's shown by the blue cross. So we've got these results that we've achieved with headphones by air conduction testing. And now we go to do bone conduction. So we place a bone conductor behind the right ear and I present the sound. And what I record is a response at 10 dBHL, and I write that down on my audiogram using the symbol for unmasked bone for the right ear, which is a bracket. When you think about the side of the audiogram that it should go on, if you're facing the person, so imagine you're facing the client and you are putting your hand to their right ear, for example, that's the shape that you would get, like so, and that's the symbol that we get on the audiogram there. So that triangle is indicating it's a response from the right ear, but is that actually what it means? Can we be confident that that's what it means? So there's two options here. One is with these results we could have, these results here could indicate that we've got a conductive hearing loss on that right ear. How do we know it's conductive? Well, there's an ear bone gap. So if we count the gap between the ear conduction and the bone conduction in the right ear, you can see it's 10, 20, 30 dB ear bone gap. So yes, it's probably conductive. Or is it? Could it be that when we stimulated the right ear, it wasn't the right ear that was responding, it was the left ear that was responding? What that means is that the left ear could have excellent bone conduction thresholds and no hearing loss whatsoever. It's threshold is at 10. And what that means is that actually we don't know what the bone conduction threshold is for that right ear. It could be 10, in which case it's a conductive loss. It could be 20, it could be 30, or it could be right there next to that ear conduction threshold at 40, in which case it's a mild sensory neural hearing loss. At this stage, we don't actually know with this data that we've got here, we don't know which of those two possibilities it is. And not only, I mean, so we've spoken about that cross-hearing, we've spoken about that happening with bone conduction because of the really great efficiency in transferring that information from one side of the head to the other. But it actually happens with the air conduction too. Cross-hearing isn't limited just to bone conductors. So if I put super oral headphones on, that sends sound waves down the ear canal towards the eardrum. Yeah, we know that. And so that's air conduction, purely air conduction. Great. But the other thing that happens is it spreads it over quite a large area of the head. It's that vibration is hitting a reasonably large surface area. And that vibration passes in through the cartilage and those soft tissues and enters the skull. So the surrounding tissue around the ear canal and the skull also receive that vibration there. It's almost as if you've got a bit of a bone conductor on there. And not only that, it's not always a perfect fit to the side of the head. And we get sound that leaks out around the cushion of the transducer. So of those two mechanisms, the direct passage into the tissues and the bone is the most, is the dominant one. So that's quite efficient. But we also have sound that leaks out and goes across to the other side that we're not testing to the non-test ear and they can just hear that in the environment. You might notice if you're ever testing someone who has quite a severe or profound hearing impairment and you're in the booth with them, you're presenting sounds through the audiometer. They're sitting over there and you can hear the sounds that are leaking out of their transducers because they're up so loud. And often they're not aware that those sounds are being presented. So two mechanisms there with their conduction. We get sound entering the skull just like a bone conductor but we also get leakage around the outside. All right, so I mentioned that the bone conduction mechanism is super, super efficient at getting that vibration to the other side but that the speaker on the side of the head, it still gets across there but it's less efficient. So if I place a speaker on the other side of the head, we do get a passage of that sound but look at this, it's sort of faded out a bit as we've got across to that contralateral ear. And with the bone conductor, we said if I present at 10 in the test ear, it's gonna be at 10 in the non-test ear. That's not the case with super aural headphones. So in this case, what we see is that the sound that reaches the non-test ear is reduced in level by about 40 decibels. So a 40 decibel difference between the ears. And we call that interaural attenuation. So attenuate means to get smaller, interaural means between the ears. So as the sound passes between the ears, it gets smaller on the far side. What does that mean? Well, if I'm presenting, let's say an 80 dB tone to the test ear, which is the bright ear in this case, then the test ear would receive it at 80 but it would pass through the skulls and be attenuated and would arrive at the non-test ear at a level of 40 dB HL. So we are reducing the level by 40 dB as we cross from one side to the other. 80 minus 40 gives us that 40. So is that a problem? Can you think of that as being, could that possibly be an issue in the accuracy of our testing? Well, the answer is yes, it most certainly can. So imagine we've already done air conduction testing and when the person came in, they said, oh, my right ear isn't as good as my left ear. My left ear is my better ear. So of course, when we do air conduction testing, we usually start with the better ear. So here we go, we've got some thresholds and we've tested their left ear and we've got a result at one kilohertz. We've got a result at 35 dB HL. And now we go to test their right ear, which is their poorer ear. We present at a reasonably loud sound because they were at 35 in the left ear. You might start up there or a bit above there. We get louder and louder. And when we get up to 80, that's when they push their button. So using the ascending method, we make the sound louder and louder and louder till we get to 80. And that's the point at which they push their button. So that's where they've heard it. We go, ah, 80, okay. So we make our mark saying that that is their threshold at 80. So they've said that their right ear is poorer than their left ear and we get these results and we say, yes, your right ear is poorer than your left ear. It's at 80, that's quite bad. Can we trust that? Is it really their right ear that has detected this sound? Or could it be the left? Is there a possibility it could be the left? And if so, how do we tell? Mm, it's a bit of a conundrum. So remember we said that the mode of that vibration passing through the skull is via bone conduction. It enters the bone and gets across to that non-test ear. It's not really the ear conduction threshold we're interested in in that non-test ear. It's the bone conduction threshold because it's gotten across there. And in this case, we've got the ear conduction at 35. Their bone conduction threshold in that non-test ear can't be any worse than 35 usually. So it can't be 45 or 50. That doesn't make any sense at all. It has to be either the same as that 35 or better than it. In which case, it's entirely likely that the sound has crossed over at a level that is now audible in that contralateral ear. So is it? Let's have a look. We've spoken about super oral headphones, the big ones that sit on the side, but what about insert earphones? We've mentioned them before as another transducer that we use for air conduction. Sometimes we prefer to use it. I'll talk about why that isn't a moment or it'll become clear in a moment. So insert earphones. Let's say we pop an insert into the ear canal and we squirt in some sound. We play some sound into that ear canal. You can see there that the amount of contact that the transducer has with all of the tissue and the bone is greatly reduced. It's a much smaller signal, much closer to the ear canal. And we've actually sealed the ear canal by placing the insert earphone in there, which means we're not actually presenting as much energy. It's hitting the eardrum at the same level, but we're not wasting a lot of energy going into those tissues that are spreading across to the other side of the head. So because of this, it means that they don't send anywhere near as much auditory energy into the skull as the super oral headphones do. And because we've got that seal in the ear canal, they also don't leak as much air coming out. So the advantage of this is that 40 dB crossover that we've got from one side to the other is reduced in terms of the amount of energy that comes across is greatly reduced. So if I place sound into the test ear, it's actually reduced by up to 75 dB to getting across to the non-test ear. Now we don't use that figure of 75 in a clinical sense because we're always aiming to be conservative when we're doing this. We don't want to think there's no cross hearing when in fact there is. So what we do is we choose conservative values for these inter-oral attenuations to almost to say what's the worst case scenario and plan for that. The reason we do that is that everyone is different. Everyone's skulls are different and the amount of tissue and fat and things that they've got around their head differs. And so we have a different degree of energy transfer between people just in the same way that people have different thresholds. People's inter-oral attenuations are different. So we have a range of those. So we've got to choose a value that is going to guarantee or allow us to plan for the fact to account for any cross hearing that may be taking place. So when we're doing insert earphone testing, we assume a couple of different values. If we're testing at one kilohertz or a frequency lower than one kilohertz, we assume that the inter-oral attenuation is about 60 dB. Again, that's a really conservative value. And if it's above one kilohertz, then it's at least 50 dB inter-oral attenuation. So again, that's less efficient than the super-orals at spreading energy, which is great. What does that mean for our testing now? So in that example with the super-orals, we thought, well, hang on, yes, they've got 35 in the left ear and they're 80 in the right ear. What if that sound was as a result of the left ear picking it up because of that 40 dB inter-oral attenuation? Well, in this case, because our crossover is much greater than that, and we're at one kilohertz, so we can assume 60 dB inter-oral attenuation, that 80 that we're presenting to the right, when it crossed over to the left, that would be down at 20 and would therefore be inaudible to that left ear, if we, again, if we assume that that left ear threshold is purely sensory neural, that would be inaudible to that left ear. So that gives us more confidence that the result that we've recorded here for the right ear is, in fact, the real result. There's no much less possibility that it's due to cross-hearing. So let's look at that. We've got 80 dB in the test ear, but here in the non-test ear, it's reached that left ear, the non-test ear at 20. So that's below the threshold and therefore inaudible. So we can be confident. So although there is crossover, it's not above threshold, so therefore it's not cross-hearing. Okay, so what about now? What about if their left ear threshold isn't at 35, it's down at zero? Can we still be confident? Well, the sound we're presenting with our inserts in the right ear at 80 is now reaching at 20, the non-test ear, which is definitely above threshold. So the stimulus would reach that non-test ear and so therefore we can't be confident that we've achieved that result. So changing the transducers will help. Obviously bone conduction is the worst, but we need to use bone conduction if we're testing bone conduction. When we're using air conduction and super oral headphones, we have a 40 dB inter-oral attenuation which gives us more problems with cross-hearing, but switching to insert earphones reduces that problem, it doesn't solve it entirely. So we need some other strategy to get around those issues where changing the transducers hasn't really solved our problem entirely. So how do we prevent cross-hearing is the question that we have to ask ourselves. And so far we've seen that for bone conduction there is no inter-oral attenuation. You play it at 20 in one ear, it's 20 in the other ear. So how do we tell which cochlear is responding? We know that for air conduction with super orals we've got a 40 dB inter-oral attenuation and for inserts, depending on the frequency we've got between 50 and 60 dB conservatively inter-oral attenuation. But what happens if the gap between the air conduction threshold in our test ear and what we think is the bone conduction threshold in the non-test ear, what if that gap is larger than the inter-oral attenuation? Like that last example, what do we do? We need some way of ensuring that we're only stimulating the test ear. We need some way of keeping that non-test ear busy, sort of disguising what's coming to it so that it can't really detect the pure tones that we're presenting, make it unable to detect those stimuli. And the technique that we do to do this is called masking. So masking is where one sound prevents the ear from hearing another sound. So it's the process whereby the detection threshold for one sound, which we will call the probe, it's like the test stimulus if you like, that threshold is increased, it's made worse, by the presence of another sound, the masker. So we're experiencing masking all the time in our everyday life. If you're in a noisy environment and you're trying to hear someone speak, the background noise will prevent you from hearing that because of masking. That's what's going on there. And the way masking works typically is that if you're testing or presenting at one particular frequency, of all the frequencies contained within a masker, it's the ones near that the frequency of interest that are gonna have the most effect. And we can see that on something like this. This is a tuning curve, which we spoke about in the cochlear function lecture around cochlear mechanics. This is saying that an inner hair cell at a particular part of the cochlear responds very well to one particular frequency. In this case, this is in the guinea pig cochlear, this is a hair cell in what we call the 20 kilohertz region of the cochlear. So when we play a sound at 20 kilohertz, it's those inner hair cells that are detecting it. So they're the inner hair cells with the job of detecting 20 kilohertz. But if I play some noise, which is centered around that 20 kilohertz, if I play noise in this area, around 20K, it doesn't have to be very loud to prevent you from hearing a pure tone that's reasonably quiet. If I present noise that is centered around that frequency, that will be very effective at masking your ability to hear that frequency. The ear isn't really affected by frequencies that are higher than the frequency you're trying to listen to. So here at 20K, if I present, and remember guinea pigs can hear up to 40 or so. If I present something at 30, it's not affecting your ability to hear 20 kilohertz at all. The ear doesn't really care. But if I go to lower frequencies, if I make it loud enough, lower frequencies do affect your ability to hear higher ones. We call that the upward spread of masking. We don't need to learn or know too much about that for today. The only thing that is really important is that if we're listening to a particular frequency, if we play noise around that frequency, that will interfere with that part of the cochlear's ability to detect that really quiet pure tone. Okay. So we do that when we're doing pure tone audiometry. So you remember our audiometer presents pure tones that have a very particular frequency characteristic. They're sinusoids. And so when we say I'm presenting four kilohertz, I know that it's a sine wave at four kilohertz. Similarly at one kilohertz, it's a pure tone. It's a sine wave at 1,000 Hertz. And so that's the part of the cochlear that I'm stimulating. So if I want to stop your ear from hearing that quiet pure tone, what I can do is present masking noise that is centered around that frequency. And our audiometer has a button that allows us to do that. So here you can see, this is part of the front panel of an audiometer. We can see the transducers we've chosen are earphones. So they're gonna be playing through earphones, not bone conduction, not speakers, not inserts or not special high frequency ones, but just the standard super orals. It's going in the left ear. What are we presenting? Are we playing a tone, pure tone stimulus? No, in this case, what we're gonna do, oh, and we're not talking to them through a microphone. We're not playing tracks off a CD or a computer. What we're presenting to them is noise. And the type of noise that we're presenting, it's not speech noise or white noise. It's narrow band noise. And this is the spectrum of narrow band noise. This is what it looks like. I've color coded them here based on different frequencies. And I'm gonna play them to you now so that you can get a feel for what they sound like. So this is 250 Hertz. I'll play 500. Sounds like a really windy day. Here's one kilo Hertz. That's an even windier day. Two kilo Hertz, 4K. Very unpleasant. And now 8K. So you can hear that the frequency characteristics of all of those different noises were quite different. You can see that, whereas a pure tone is a straight line at a particular frequency, these are relatively wired. In fact, we talk about the bandwidth as being a proportion of that center frequency. So it's a constant band width that's somewhere between 25 to 35% of that mask of frequency. So you notice though the characteristics of those sounds. When we're doing pure tone audiometry, we're playing a pure tone in the test ear, the one that we're really interested in. We wanna keep that non-test ear busy. And so that what I just played you are the sounds that we present to that non-test ear. If I'm playing one kilo Hertz in the test ear, I play one kilo Hertz narrow band noise in the non-test ear. What you say to the person who you're testing, the client, when you're putting on the headphones and you're about to do some masking, which is what we're doing here, is you say, now you're gonna hear some, like a rushing noise or a windy noise in your left ear. But what I want you to do is just ignore that. And I just want you to concentrate on hearing the beeps in the right ear. So just listen for the beeps. Don't worry about the masking noise, just listen for the beeps. And they sort of nod and then you play that mask up. They get used to it. So they're hearing, let's say it's 8K, you hear, they're hearing that. And then while that's being presented, you then go ahead and play those pure tones. In this case, it's 8 kilo Hertz and they push the button when they hear the pure tones. When you do that, and if you do it appropriately, you know that you're keeping that non-test ear busy and that you're presenting to the ear you think you are, which is the test ear. Now have a look at these spectra. They're all roughly the same shape as a percentage of, because it's a logarithmic axis, they've all got the same bandwidth. But notice that the levels are slightly different. They're differing between different frequencies. We don't actually care about the absolute physical level that those sounds are at. We do care about how we calibrate it though. But the way we calibrate it is in terms of what's called effective masking level. And that isn't to do with sound pressure level or anything like that. It's stated in terms of the amount of masking noise we need to lift someone's threshold to a certain value. So we call that DBEM, decibels of effective masking. And that effectively tells us the DBHL, that I, so we're on the audiogram, that a mask will shift someone's threshold too if we present that noise. This only applies to if the mask noise is actually above their threshold. If it's inaudible, it doesn't have an effect. So if their non-test ear threshold is zero in this side, if we play 50 dB of effective masking in that side, it now, it's basically lifted the threshold of that non-test ear up to 50 DBHL, which means that a sound would have to reach it at 50 or above in order for them to detect it with that ear for it to be audible. Similarly, if their threshold wasn't at zero, if it was at 30, and we present 50 dB of effective masking, their threshold still comes up to 50. So regardless of what your threshold is, it will rise to that level of effective masking. Does that make sense? All right, but if their threshold was 80 and we add 50 dB of effective masking, that masking noise is inaudible and won't change what their non-test ear threshold is. So that won't have an effect. All right, but that effective masking level scheme is as a calibration method is really intuitive and helpful and it's really useful in terms of deciding how much masking to apply. When we are deciding how much masking to apply, there's obviously a range. Masking can be too quiet to the point where it's not masking at all. It's not affecting the non-test ear's ability to hear a sound. It could be just right or it could be far too loud. And we've got this range of masking that you can see here. So the minimum masking level is where we've made it just loud enough where it's starting to have an effect at preventing the ear, the non-test ear from detecting the sound that we're presenting to the test ear when we take into account how much of it passes over to that non-test ear. So that's the minimum masking level that's required to give us some confidence that we're testing the test ear and not the non-test ear. Anything less than that is under masking and then we're still in that ambiguous case there where we still don't know. If we're under masking, we don't know if it's the test ear or the non-test ear. We don't know that until we achieve that minimum masking level. The maximum we can apply, although we should apply, is the amount of masking that is enough that we are still sure that we're only testing the test ear, but it's not too loud. Now, what do we mean by too loud? Well, we'll talk about that in a moment, but just as the pure tone, our stimulus can pass from the test ear to the non-test ear. If we make our masking noise too loud in the non-test ear, that can pass back to the test ear and prevent it from hearing the test tones. And that's obviously not what we wanna do. We don't wanna mess with the threshold of the test ear. We only wanna mess with the non-test ear threshold. So if we've got too much masking, we call that over masking. Not only that, it's actually pretty uncomfortable. So you notice those sounds weren't exactly pleasant to listen to. Sometimes when we're presenting masking, we need to, I'll demonstrate the procedure, but we need to increase the masking level, increase the masking level, increase the masking level, and you see the client kind of screw their face up because it's quite unpleasant. We sort of reassure them it won't be there for very long. You try and do your procedure as quick as you can, get accurate results, and then you're able to take away the masking and there's often a bit of a sigh of relief there, particularly if you're doing a lot of masking across an audiogram. So that area between the minimum masking level and the maximum masking level, we call the plateau. So we call it the plateau because it looks like, here's a, like imagine we're going up a hill and here's a flat bit, here's a plateau on that curve. So what you see here is the masking noise that we're applying in the non-test ear. And here we're seeing what, when we're doing the testing, what presentation level we're finding the person's threshold at in the test ear. So you can imagine they've demonstrated some really low testing threshold in their test ear, but we're not convinced that it's actually the test ear. We might think it's the non-test ear receiving that. So what we do is we present, let's say, 20 dB of masking level in the non-test ear and that lifts the threshold in the test ear. They were down at some lower number, now they're up at 60. And we increase that masking noise to 30 and we sort of have gone up 10 dB. Their threshold goes up 10 dB in the test ear. So, ah, we're actually showing that me changing the masking in the non-test ear is shifting the, what we're measuring for their test ear, which means it was actually the non-test ear responding. We go up high, we go up to 40 and by the time we get to 40, this test ear has stopped increasing. So we go up another 10 dB and our threshold stays dead still, it stays where it is. So that means, ah, I'm messing with the non-test ear, but that's not changing the threshold. The test ear threshold is staying the same. We've reached this plateau. So that means I'm confident that now I'm testing the test ear. So if I go up 10 dB and then another 10 dB, that's what we call a 20 dB plateau. So if I get a 20 dB plateau, I go, yep, okay, I know that what I've just measured there is the test ear threshold. Not the non-test ear. I can say, yes, I've effectively masked that threshold. I've got a proper result there. And I can move on to my next frequency that I'm trying to test. So that's sort of this technique of using a plateau to decide on your masking was developed by Hood. It's called the Hood technique. And it's one of the masking techniques that we use. We're not gonna go into those techniques in a huge amount of detail until you do the m-word and then we'll teach you all about it then. Okay, so that's the point in which we accept that threshold and move on. All right, so the masking procedures, which as I said, are beyond the scope of this lecture, all follow the same sort of pattern. So firstly, we've got a transducer that we're applying sounds to our test ear with. So that could be bone conduction. Let's say it's the right ear we're looking at. It could be bone conductor applied to the right mastoid. It could be a super oral headphone or an insert earphone if we're doing air conduction testing on that right ear. So firstly, we determine what the threshold is on that. What we think is the test ear. And then we use our information about the non-test ear and what we know about inter-oral attenuation to decide, oh, is there any chance that it could be cross-hearing that's taking place? So we've got to make that decision. What's the inter-oral attenuation? What is the predicted sound level that we'd be reaching the bone conduction thresholds on that non-test ear? Notice the difference in the inter-oral attenuation sort of signified by the shading here. There's no, it doesn't fade out at all for bone conduction. Fades out by about 40 dB for the super orals and by 70 dB for the insert earphones. Then we apply some initial level of masking. We increase that level of masking to the point where we reach that plateau and then we can be sure that the result that we've got is the result we're interested in from that test ear. And then when we've done that, we can accept that threshold as being masked and we can move on and we can take that away and just move on to the next threshold and repeat the process all over again for the next frequency. I'll show you one example. This is for bone conduction. Let's think about the results we've got so far. So here you can see a bone conductor on the right ear and when we've done ear conduction, which we did before we did bone conduction, in the left ear, we got a threshold at zero. So zero dB HL threshold in the left ear. In the poor ear, in the right ear, we've got a threshold of 30. And so we're thinking, oh, okay, so there's 30 dB in that side. Right, great, so that's the ear conduction threshold. Now I present bone conduction on the right ear, on the left ear and I get zero and I do it on the right ear and I also get zero. So just based on this audiogram here, what pattern does that say to you? What does it say? Does it say that right ear is sensory neural or does it say it's conductive? Well, on the basis of just looking at this, you would say it's conductive if you didn't know about masking and didn't realize that these bone conduction symbols were unmasked bone. You would look at that ear bone gap and you would say it's a 30 dB ear bone gap, it's a conductive hearing loss. All right, so we know that there is no inter-oral attenuation for bone. So when I'm stimulating the left ear, sorry, the right ear at zero, it's reaching the left ear at zero and if the left ear is good, that's the one that will be detecting the stimulus. So what we have to do is mask, we know that. So we put a transducer on that non-test ear, it could be a super oral, it could be inserts, whatever you like. And we present some masking noise. Now, usually we start at what the bone conduction threshold is in the non-test ear plus a bit of what we call safety. So it's a say, a 10 dB safety factor just to get us above that point where we know we've started to mask that non-test ear. So here you can see 10 dB of masking applied to the non-test ear. I've indicated that on the audiogram by a gray arrow. We don't indicate that on the audiogram usually because it's not something we record. We only record the end results on the audiogram but I've put a marker there just to illustrate where we're at in this masking process. So I present 10 dB of masking to the non-test ear and I'm presenting bone conduction tones to the right ear. Now, if I am actually stimulating the left ear, if it's the non-test ear that's responding, that 10 dB of masking, effective masking, should elevate my right bone conduction threshold by that degree of effective masking, 10 dB. And so I present it zero, they don't respond. I present at five, they don't respond and I present at 10 and they do respond. So that 10 dB of effective masking has shifted that test ear threshold up by 10 dB. Okay, so I need to go further. I increase my masking noise by another 10 dB. What should that do to the test ear threshold if it's not a real threshold? The answer is that should also shift up by 10. So here we go, we've moved up by five, moved up by another five. So we've gone 10 for 10 so far. We're on that under masking part of the plateau where 10 dB of mask gives us another 10 dB of threshold shift. So it's an indicator that we haven't reached a plateau yet. So I now boost up to 30 dB of effective masking. No response at 25, but I do get a response at 30. So we've hit a point now. We don't know if we're at a threshold if we've hit a plateau yet. So I need to increase my masking even further. I go up to 40 and I test at 30 and they do respond. So there we go. So I've stopped moving. I'm now onto the flat bit of the plateau and I've got 10 dB of plateau at this point. I go up another 10 dB. This noise is quite loud now in that non-test here. Test the right ear bone conduction again. What do I get? I'm still at 30. So that is a clear sign that I've now got 20 dB of plateau. I am increasing the masking noise in the non-test here but it's not affecting what I'm measuring with my bone conductor in the test ear, the right ear. And so because of that, because I've got that 20 dB of plateau, I can accept it and say, yes, that is a masked threshold. I changed the symbol from being an open triangle to being that closed triangle that you see here. So if I'm looking at this, I go, I can say, well, that's a 30 dB sensory neural loss in the right ear there. That left ear threshold for bone conduction is still unmasked. Do I need to go ahead and mask that? The answer is no. What else could it be? It has to be the left ear. We don't have a third ear that's even better than those two. So there's no need for me to go ahead and mask that left ear bone conduction threshold because I'm very confident that it's not the right ear responding to that. So that's great. That's good enough for us and we can go on from there. Does that make sense? Very good. So that's bone conduction masking. One thing I, so let me just go back. Masking is actually a lot more complicated than that and there's lots of things we've got to take into account. And one of the things we have to take into account is that mask that we applied to the ear does something really interesting. And that means what it actually does is can improve. Let's say I put that super oral headphone on the non-test ear and I'm applying bone conduction to the test ear. Whatever we had as the sensitivity of the non-test ear to bone conduction might actually improve when I put the headphone on that ear even without playing masking noise. And the reason for that is when I spoke about the osteotempanic mechanism of bone conduction and I said that sound goes towards the eardrum but also heads out, if I stuck my finger in my ear that prevents that sound energy from escaping and makes it more intense in the ear canal. And so the level that passes through the eardrum is actually boosted up. So I've made that bone conduction osteotempanic mechanism more efficient and that could shift our threshold. So that makes the bone conduction thresholds appear better than they are. But occlusion also happens when the ear canal is blocked with transducers like a super oral or the inserts here. So we still get that boost. So that's called the occlusion effect. You can, if you all want to take a moment now you can do a little demo where you go and then put your fingers in your ears and you go... Your nose gets much, much louder when you've got your fingers in your ears because the energy that was escaping is now unable to and it's bigger when it's heading back into your middle ear and into your cochlear. Okay, so that's the occlusion effect. How do we avoid that? Well, if we're doing bone conduction basically make sure that when we're in the test ear we don't have an insert earphone in that test ear and if we're using super oral headphones to apply masking in the non-test ear so here's the non-test ear and I've blocked it with the headphones that I'm providing masking with. The other bit of my headphone instead of sticking it here where it's going to interfere with the bone conduction threshold in my test ear I can just shift it off the head. So people will kind of laugh at that. It's a bit funny when you're sitting there with an earphone sticking on the side of your head. I think you've done it wrong but what we're trying to do is not occlude that ear to allow that energy out so that the bone conduction threshold we're measuring is still accurate. We haven't altered it just by occluding that ear canal. So even though that speaker isn't delivering sound to the test ear it would be occluding the ear canal and it would make our bone conduction stimulus louder and that would give us the wrong threshold in our test ear. Here's some examples, here's some pictures where we've got the super oral headphone on the left ear presenting masking noise that you can see here but on the ear that we're actually testing with bone conduction we've left that ear open. Here again we've left that ear open because that's the ear that we're testing with the bone conductor. All right, one sign when you've done an audiogram one sign that you should look out for that oh hang on maybe there's been some cross hearing happening is if we get something called a shadow curve a shadow curve. So let's imagine a situation where the client comes in and they've actually got a profound sense of this renewal hearing impairment in their left ear. So let's say they're not aware of it and you sit them down and you're doing audiometry you test their right ear and you get these sets of thresholds here you've got the right ear air conduction and you've got the right ear bone conduction indicated by the triangles and then we go and test the left ear and this is the pattern of left ear results that we've got. So here you can see I've got air conduction thresholds at 35 up to 5K 500 Hertz sorry 45 at one and 2K and 50 at four and 8K. One thing you may notice is that if I take away that air conduction on the right ear what you'll notice is that there's a constant degree of offset between the right ear bone conduction threshold and the left ear air conduction threshold. That should raise alarm bells to you. So here you can see in both cases I've got the bone conduction threshold here and the left ear air conduction thresholds are all separated by a constant number of decibels 10, 20, 30, 40 decibels. Hmm, I've been testing with super orals and I've got results in my test ear that is 40 dB above the bone conduction threshold in the non-test ear. That's a clear sign that cross hearing has occurred. And if I put back those air conduction thresholds what you can see is that if the air conduction even if the air conduction was over here it's not the air conduction threshold that we're getting the air bone gap with it's the same distance from the bone conduction thresholds in that non-test ear. So it's the test ear bone conduction thresholds that dictate the shape of the shadow curve not the air conduction thresholds. But if you see a shadow curve like this and this will become important in our next topic this is a clear sign that cross hearing has taken place. So when you do your master of audiology and you actually need to do pure tone masking this is a situation where you pull out a heap of rules to decide when do I mask for air conduction. So what I've given you in this lecture is just like a feel for the situations where you need to mask but we are a bit more quantitative when we come to do that in actual clinical audiometry. So what I'm gonna do now is run through quickly some of the rules and the rationales for doing that. You don't need to learn these particular bits for the exam in terms of writing out these rule formulae but I need you to know the underlying rationale. So these formulae should help you understand what that rationale is. So masking rules, yes it does. No, masking rules. Number one, when do we mask for air conduction? Apparently, sorry, what we do is we mask the non-test ear when the gap between our, so this is the non-test ear. We mask the non-test ear when the gap between the unmasked air conduction thresholds and the apparent bone conduction thresholds is at least as big as our conservative value for that interaural attenuation. So remember we always choose the most conservative value for the interaural attenuation just so we're not getting cross hearing by mistake. So the rule is we mask the non-test ear if the air conduction in the test ear minus the bone conduction in the non-test ear is greater than or equal to the interaural attenuation. So almost if you think about as an airborne gap but across the head, if that airborne gap across the head is bigger than the interaural attenuation or the same size as it, then you go ah, it could be cross hearing so we need to mask. What if we haven't measured the bone conduction yet? So we normally do air conduction left and air conduction right first before we do the bone conduction. So what do we do in those cases? Well, we basically need to estimate what the bone conduction thresholds would be in that non-test ear. What's the probable side of lesion? So if it's, that you've got clues from that from your timponometry and that sort of thing. If it's probably conductive, then you can assume a value again conservatively, assume that the bone conduction in that non-test ear is at zero and decide accordingly based on that. If you think it's probably sensory neural, so you've got type A temps and their history doesn't indicate any middle ear disorder, then what we assume is that the bone conduction in the non-test ear is at least as good. It can't be any worse than the air conduction in that non-test ear. So that is we know that the bone conduction thresholds usually won't be worse than air conduction threshold if it's purely sensory neural. So we can just use that instead and make our masking decisions on the basis of that. So in those cases, if we haven't done bone conduction yet, we mask the non-test ear if the air conduction thresholds in the test ear and the non-test ear are different by an amount that is equal to or greater than the inter-oral attenuation for our transducers. So 40 dB if it's super orals or 50 or 60 if it's insert earphones. All right. When do we mask for bone conduction? For bone conduction, we mask the non-test ear when the test ear air bone gap using our unmasked bone conduction thresholds is 15 dB or more. So if we've got an air bone gap in the test ear, so we've got a test ear air conduction threshold and the air bone gap to the unmasked bone conduction threshold is 15 dB or more then we have to go, I think we need to mask that. That's the first criterion. The second criterion is that this test ear air conduction threshold has to have some sort of hearing loss. If we're in the normal hearing range, if we're at 10 or 15 or something in the test ear, then there's no point masking because we've got usually about five dB leeway with thresholds in any direction just in terms of test retest reliability and your test ear one could be down by five dB and your non-test ear up by five dB and that would be totally reasonable. So we only do the masking when the test ear air conduction threshold is at 20 or above, then there's a need to mask. So if the test ear air conduction threshold is 20 or above, so showing some sort of hearing loss and we've got that 15 dB air bone gap to our unmasked bone thresholds, then we mask. So that is stating it as a rule with symbols and stuff. We mask the non-test ear, if the air bone gap in the test ear is greater than or equal to 15 dB, where the air bone gap we mean, we're using the unmasked bone conduction thresholds and we've got some sort of loss in the air conduction that's greater than or equal to 20 dB HL. So that 15 dB threshold, that the air bone gap of 15 dB is what's most commonly used. Some clinics might decide to use 10 dB rather than 15 but because as I said that can happen by chance with one being up and one being down, you could be wasting valuable clinical time doing masking in all of those situations. So we tend to use 15. In some cases you really, really do absolutely need to know the masked bone conduction thresholds for the ears. You can't say, oh well there's no loss, so don't worry about it. If it's a surgical situation or you're doing research where you do need to know that, then by all means go ahead and use masking because you will get the correct answer. It just takes time that isn't worth doing under normal clinical circumstances but under surgical research situations it's definitely worth doing. All right, so we've spoken about cross hearing for ear conduction and bone conduction. For pure tones, what about speech audiometry? Can we get crossover of speech from one side to the other when we're doing speech audiometry? And of course the answer is yes. If we're playing sound into one ear, it can vibrate the bones and end up in the contralateral ear. So the values that we use are similar to the ones that we use for the pure tone into oral attenuation values. But what we take into account is the fact that a pure tone is at one frequency but speech is a broadband stimulus. So for super orals it's 40 dB at every frequency so we go with 40 dB for speech. For insert earphones it's 50 at some frequencies but 60 at other frequencies. So because we've been conservative we choose the worst of those two values. So we say, actually no, we go for 60 dB across the board 50 to 60 dB for the inserts. So the rule in this case is very similar to the ear conduction masking rule. We've masked the non-test ear for ear conduction, pure tones, we said we mask it if the ear conduction in the test ear and the bone conduction in the non-test ear is greater than or equal to the interoral attenuation. For speech it's the same thing. Instead of the ear conduction threshold for the test ear we think about what level of speech are we putting in. So we talk about the presentation level. We could be putting speech in at 50. And because speech is broadband we say 50 in this ear is the presentation level. What is the best bone conduction threshold on that non-test ear? No matter what the frequency is what's the best one? What's the most sensitive frequency there? We take that because that means that at least some of that energy is gonna cross over and be received by that bunch of hair cells. So to clarify that we mask if the non-test ear if the presentation level in our test ear is different from the best case scenario bone conduction in the non-test ear by an amount that is equal to or greater than the ear bone gap. Sorry the interoral attenuation for that transducer. If for some reason you've done your speech audiometry before you've done your pure tone thresholds then what we do is we mask if the difference in the half peak levels or the SRTs is at least as big as the interoral attenuation of the transducer you are using. So if you do your speech in one ear, speech in the other ear and you notice that the two curves are separated by an amount that's greater than or equal to the interoral attenuation for the transducers then that's a sign maybe again we're detecting it with the non-test ear and we need to mask that. All right so we mask if the difference between the SRTs between the ears is greater than or equal to the interoral attenuation. Okay the last very last thing I want to talk about is something called the masking dilemma. So remember when we spoke about not enough masking under masking, the masking plateau and then over masking there are situations where we have extreme differences between the two ears and it was Norton in 1960 and Studebaker shortly after that who were the first to describe what's now referred to as the masking dilemma. And that means that because of the differences between the ears the moment we apply enough masking to the non-test ear that's actually already too much masking and it's gonna cross over to the test ear and muck with our threshold on the test ear. So here's the illustration here we've got our pure tone is crossing from the test ear to the non-test ear and we're trying to mask that but the level of masking that we've got to provide means that that mask a noise is so big that it's crossing back and altering the threshold of our test ear which is obviously not what we want. So in those cases as soon as we start masking we're already over masking so that's a problem. It occurs when we've got a significant loss in the non-test ear and a conductive loss in the test ear. So that combination is what makes us get the masking level up so high and having to provide enough to get it across there. If you don't recognise the fact that you've got a masking dilemma you're gonna get weird results. It's gonna be difficult to interpret and you'll get the wrong answer. So you need to be able to spot it and the way you address it is by switching transducers. So you can best get over that by using inserts instead of super oral headphones and not only does that decrease the situations where you need to mask but it also decreases the level of oral masking. So it's not gonna cross back quite as much which is a big advantage. All right.