Hi Steve,
Thanks for your reply, and for your welcome to the TinnitusTalk forum. I've been a keen follower of the forum since I first got tinnitus myself back in April 2012. To discuss all of your points most effectively, I've put a quick numbered summary immediately below. But to properly support my arguments, I've then inserted proper full responses into your original posting below. (I've also numbered the most important inserted replies, matching with the numbers in the summary below.)
Summary:
(1) Most people who try the HushTinnitus online preview state in their online response that they get a masking response and/or a residual inhibition response from the preview sound. This fits with independent research. But like any system addressing the hugely varied condition of tinnitus, it doesn't work for everyone. (I've been totally open about this in what I've published online.) Steve, I have a plausible explanation for why it didn't work in your particular case.
(2) Regarding safety: nowhere on the website does the text ask you to turn the volume level up until it masks your tinnitus. In two places (including during the preview sound), the website reminds you to keep volume levels so they are "not uncomfortably loud".
(3) The website generates custom sounds for each user. These are frequency-tuned to the estimated frequencies of the user's tinnitus, using a correlation identified by several independent research studies (detailed below).
(4) As an additional aspect of safety, the online software includes a number of design and logic safety features, which have been extensively tested on end users. These provide a range of additional checks, balances and methods to help keep sound levels in a safe range.
My full replies:
Hi Clyde and welcome,
Reading my post back the language is a little harsh sounding, so apologies for that. We usually debate things in this forum and my comments were written too quickly and in a condemning style.
Thanks for that, Steve, that's appreciated. I'm a big fan of open and reasoned debate and peer review, and I welcome having a reasoned discussion on all these matters, in a mutually respectful style.
I was referring to the sound generated when I went through your process. It is a personal opinion only, but for me the type of sound generated didn't inhibit my tinnitus in any way - and I can get RI effects through certain sounds.
(1) Sure. As I'm sure we've both read, research has shown that tinnitus is extremely varied from person-to-person: in terms of the types of tinnitus sounds that people hear, the pitch ranges of their tinnitus, the loudness of their tinnitus, and of course the way that different people cope (practically and emotionally) with the condition day-to-day. Add to that the findings of research specifically into residual inhibition and masking, and the picture becomes even more varied: two people with the same tinnitus sound characteristics can have completely different levels of residual inhibition response (in terms of the depth and duration of their tinnitus reduction) or distinct differences in their masking response. With this enormous variation between different people, I think it's fair to say that no one system can work for everybody.
That's my main thinking behind offering a pre-purchase preview on the website (and also for providing a post-purchase "for any reason" refund guarantee). I feel it's essential that people can try it out, to see if it works for them – with their particular tinnitus and responses.
For those who are interested in published numbers from general tinnitus research (like how many people show a positive residual inhibition response, how many show a positive masking response, and so on) I've summarized the findings of the independent research in my review document (please see the link in my previous posting). Please see page 6 onwards for the numbers for residual inhibition. The headline figures… The different studies (which cover a wide variety of different test sound types) found that between 52% and 88% of people had at least some residual inhibition response. (In the "52%" study the experimenters played the sound quieter than in the other studies, which probably explains the lower response rate.) Regarding masking, 94% of people achieved masking responses with pitch-tuned sound in one study (see reference [56] of the review). (On the HushTinnitus website, as you'd probably expect we see similar figures to these, looking at the responses people give after listening to the pre-purchase preview.)
So, when an individual person tells me "it didn't work for me", I have to say that it doesn't surprise me. Given the huge person-to-person variation in tinnitus, and in responses to stimulus sounds, we'd always expect a minority proportion (unfortunately) to not have a tinnitus reduction response. Of course, there is a good proportion of people (consistent with the studies) who do show a positive response.
But Steve, from what you've told me about your own tinnitus below, I believe I have a plausible explanation for why the pre-purchase preview didn't work in your particular case, with your particular tinnitus. You've given me enough information to point the finger to a very specific matter, specific to the description of your tinnitus. Please see later for my thoughts on that.
I turned the sound up louder though and if I had put it to a level where it masked my tinnitus it would have been high enough to cause hearing damage over a sustained period of listening. On headphones / earbuds we know the issue on perception of loudness and actual loudness can be damaging, so I think there is potential.
(2) OK, there are a number of points here. Regarding the text on the website:
· Firstly, as already discussed, the text reminds people in two different places that, like any other sound, volume should be kept so that it is "not uncomfortably loud".
· Secondly, you say "I turned the sound up louder though and if I had put it to a level where it masked my tinnitus it would have been high enough to cause hearing damage over a sustained period of listening." Regarding the website text, the main point here is that *nowhere* does the website text ask you to turn the volume level up until it masks your tinnitus. In fact, on the screen where the preview is playing, first there is the volume restraint reminder, then the text simply says, "As it plays, please note whether your tinnitus is reduced." That's totally different. The user has been reminded (for a second time) about volume restraint, and is now simply being asked to make a yes/no observation.
In addition to the guidance in the website text, there are other features in the software designed for safety (whether the user is using headphones or speakers) – which I discuss more below.
For me it generated a high frequency sound, quite shrill. This made my hearing dull after a short time, the effect you get when you have had a long day of mixing, listening to music or similar, or when you hear a sound that sends your ears into a protective state.
General masking, which you have referenced above, uses a broadband sound that covers all frequencies, easy to mask a particular tone, whereas the sound I heard was in a specific range, not so easy to mask specific tones unless it is designed in. If a person has 500Hz tinnitus for example your sound would have been very loud before it drowned it out, I have a low rumble which wasn't even close to being masked at a higher volume by your sound.
I believe that when you are using a particular frequency and telling the user to mask their tinnitus with it that you should identify their tone, preferably their hearing profile also. This could be done via a program, though it would need to be relatively sophisticated as correctly identifying your own frequency is difficult for most.
(3)…
· The HushTinnitus System software does indeed use a strategy to estimate the frequency of the tone or tones of each person's tinnitus. (Please see page 24 of the review, paragraph 2 onwards.) That's because a large number of different studies have found that sounds, for causing residual inhibition or masking, are most effective when they are tuned to the tinnitus frequencies. (All the research references regarding this are on page 24.)
· Another relationship found by several different research teams is that the frequency range(s) of the tinnitus generally matches the frequency range(s) of regions of hearing loss. (Recent references [60], [64], and earlier references [2], [4], [5] and [9] in the review.) I've used this relationship to estimate the tinnitus frequency regions (detailed further in the next point).
· Steve, you say that "…correctly identifying your own frequency is difficult for most". I totally agree. Several studies have found that many patients have difficulty reliably finding the pitch of their tinnitus in the "traditional" way, i.e. using an external pure tone to decide "is it higher, is it lower?" (References [23], [40], [64], [6], [43], [60].) There are also "likeness" matching tests ([43], [60], [64]), which report better test-retest reliability. However, they are somewhat lengthy to complete, and many people don't have patience with lengthy online tests. So, in the pursuit of both reliability and shorter test times, the HushTinnitus System uses the relationship in the previous bullet-point.
· The website test makes a set of measurements to get estimates of some particular features of the user's hearing profile. I'd like to stress (as I also do in the review document) that the test is not trying to take a full standards-calibrated audiogram. That's not the point of it. Rather, the point is to estimate the location of certain *large features* (tens of decibels variation) in the user's hearing profile that are typically associated with frequency regions of both hearing loss and tinnitus. If you're interested in the exact details of what the HushTinnitus test does, please read pages 24 and 25 of the review (it would make this posting far too long to repeat it all here).
· The word "typically" is important here. The HushTinnitus test investigates hearing-loss features that cover the large majority of people with tinnitus: the upper audiometric edge frequency (which varies greatly from person-to-person, and is therefore specifically estimated with a frequency-sweep test), and any hearing loss at the commonly-identified notch frequencies, which (in order of prevalence, according to later studies) are 6, 4, and 2 kHz. Across all the tinnitus studies in the review, this covers the large majority of people with tinnitus.
· But no one system works for everybody, and by publishing a detailed description on the web (of both what the system does, and why), I've tried to be totally clear about what the system does and doesn't cater for. For example, the system doesn't currently cater for people with Meniere's disease (which accounts for about 10% of tinnitus cases). Also, Steve, if you have tinnitus down at around 500 Hz (which would generally be expected to accompany at least some hearing loss there too), that's quite rare (if you don't have Meniere's) according to the studies, and I'm afraid the HushTinnitus software doesn't cater for that. Even 2 kHz notches are relatively rare, and (except for Meniere's) anything lower is even less common, according to the studies.
· So, in your case Steve, the system probably tuned your sound to around your upper audiometric edge frequency (if you have no other notches at the common notch frequencies). This is where it would be most effective as a masking or residual inhibition sound for *most* people, given that measured profile. (But, to briefly touch back onto safety, even though it's not effective for your particular tinnitus it's still totally safe for you try it – if you're following the on-screen guidance, and/or the same everyday approach you would use to setting any volume level safely.)
· All this goes back to why I have made a pre-purchase preview available in the first place. It gives an upfront opportunity for everybody to sample whether or not the system works for them.
Properly overseen for me would be a program to make sure the listener has the right tone, that it doesn't need to be played too loud (using the baseline data from the initial sound test you could analyse the difference in volume during a Q&A with the listener and roughly check they don't need to have it too loud when listening), checks and balances, or alternatively direct intervention. The latter would be difficult given the product and delivery method.
(4) During the online test there are other checks, balances and methods in place regarding loudness – not just for safety, but also for comfort during the test. For example, the online check includes a series of "threshold" tests to find the quietest tones (in terms of relative amplitude) that people can hear at a number of frequencies. All of these tests progress from extremely quiet (too quiet to hear) to only-just-audible – at which point the user clicks to finish that stage of the test. This keeps these test stages extremely quiet and comfortable for the user.
The same is true of the frequency-sweep tests. These sweep down, from the high frequency end, so it's totally imperceptible to start with. (Its amplitude also keeps to a quiet level during the sweep.) So again, the user only just hears it quietly, and then clicks to proceed.
I've put a lot of consideration into the relative sound level of the preview sound compared to the earlier sounds, and I've done extensive user testing on this. From that testing, I've derived figures for the range of typical sound pressure levels that people experience when the preview sound starts. I've designed the system to place those typical sound pressure levels in a comfortable and safe range (while still being in the required range for effectiveness for most users). The definition of "safe range" was taken from independent publications on audiologically safe levels (e.g. regarding audiological testing and hearing aid setting).
There's a lot more detail involved the considerations here, but I'm trying to keep this post as short as I can while still covering the important points. For example, consistency tests are performed by the software to make sure that certain user responses "make sense" (compared to the requested action) in terms of the relative levels inferred by the responses. This adds an extra "checks and balances" level of safety.
When the preview starts playing, the design of the website wording and sound levels is such that the large majority of users end up with a level that is both comfortable and at the right level to be effective. However, of course different users do different things (and can have different reactions to the same sound pressure level), and for that reason the words at that stage encourage people to change the volume (up or down) if needed.
Steve, you mention direct intervention. From talking to a lot of users, I think most people don't want that. They want freedom over what device to use (be it an MP3 player, an iPhone/iPad, a laptop, a CD player, or whatever) – because then they can use the tracks where and how they choose. All these devices have volume controls, of course. And I don't think users want their volume control "nailed down" to a particular setting. So, it's really then down to having the right guidance wording on the website and in the User Guide – which I feel I have already put in place.
I don't believe we need hand holding, but I do believe that somebody desperate for help has an implicit trust that you have all checks in place.
Steve
I have gone to considerable lengths to try to make sure that all reasonable checks and guidance are fully in place. (In addition to all the discussion above, please see the first page of the User Guide, on the HushTinnitus website, linked on the About page.) This is in no small part motivated by my own personal experiences: excessive sound levels (from a car music system) were exactly what caused my own tinnitus. (My "tinnitus story" is on the HushTinnitus website, again linked on the About page.)
Taking the widest overview on all of this, the "safe use of volume levels" is a general subject, as I see it – much more general that just HushTinnitus. In the tinnitus world, it's general to all tinnitus hardware devices that allow the user to adjust the volume (and there are many), and of course all MP3s, CDs and audio apps aimed at people with tinnitus. Having had this discussion, I don't really see a convincing argument that there is a problem with HushTinnitus regarding safe use of volume levels. All system designers need to thoroughly consider safety, and I would argue strongly that I have fully and comprehensively considered the various aspects involved, and addressed them.
To finish, here's a quick summary of HushTinnitus safety aspects (in addition to the user's common sense, of course):
· Repeated text on the website guiding users to maintain volume restraint, as they would with any sound.
· A fully analyzed and extensively user-tested design of sound levels throughout the online test.
· An extensive review of published research on residual inhibition and masking (published online), including a section dedicated to safety.
· Clear guidance in the User Guide, right on page one.
Member
) The sounds of the HushTinnitus system are based entirely on that review of independent research findings.
