REM test results...please help!

@Bimodal_user said:

I have Phonak Roger Select. Sometimes it gives strong SNR and the ambient noise is decreased, more often I didn’t see much difference.

I thought it isn’t helpful in some situations because I had to work hard. Then the battery died. Then I’ve realised how much it really helped. Yes, I had to work hard but without it, battle was just lost.

In short - turn it off and see if its helping if you haven’t done that already. I was really surprised.

About crosses…
I have one bad ear and one normal ear. First few years I wore aid only on bad ear. People talking from my bad side had to walk over to my normal side.

This year I’ve started with 2 aids, and results are awesome. Paradises 90.
The other day, shopping mall, echoey, slow day so no many people around, my husband on my bad side, wearing mask and speaking normally, we could normally converse. No mics needed.

Paradises and marvels transfer sounds between the aids in various modes. Not loud, but it’s there.

There’s also 360 mode which I’m currently looking into which should maybe be louder.

So basically I’m utilising all 4 mics.

Prior to this, fitters would use acoustic phone program and tweak it for everyday usage. Such solution is called amp cross or tricross.
So it aids bad ear AND transfers the signal to good ear (and aids it if needed).
For phonak, there’s this 360 program that should work better since you don’t have to pick side master-slave, it should detect automatically from which to which it should transfer.

Regular cross solutions just pick sounds from bad side, discarding it completely.

Check dr cliff’s video on the topic of single sided deafness / ampcross, forgot the exact title.

However if you’re totally deaf on bad side, then bicross is the term you need.

Cross only transfers the sound to the normal ear from bad side.
Bicross transfers to the good ear and helps it as well (so it’s also an aid).
Those two are devices under such name.

Ampcross solution (no device exists) is with 2 aids, aiding each ear and transferring the signal between them. Needs a fitter who knows how to make it work, and aids that aloow it.

@Blacky said:

I thought it isn’t helpful in some situations because I had to work hard. Then the battery died. Then I’ve realised how much it really helped. Yes, I had to work hard but without it, battle was just lost.

In short - turn it off and see if its helping if you haven’t done that already. I was really surprised.

Maybe you’re right. I rarely participate in social events, so I had only several occassions to use it. My experience with Roger Select is with Phonak Bolero V70-P (I haven’t transferred RogerX license to my Audeo P90-RT yet), so maybe with new HA (bought them 6th Nov) it will works better…

Yes, saying “CROS” I meaned “BiCROS”, because I am completely deaf on left side. However, AmpCROS definitely isn’t for me because I could loss e.g. AutoSense features and more. I am much interested in 360 mode. Thank you for much infos.
I expected some benefits from Binaural Voice Stream from CROS e.g. Speech in 360, Speech in very loud noise or Auto StereoZoom. This is one of many things why I have bough P90-RT version, not P70.

I am very glad you have reached so many benefits only thank to aiding second ear!

For me, I don’t hear many benefits with Paradise compared to previous aids (Venture) in situation apart streaming. Perhaps because I have it very briefly, maybe its no REM (only APD 2.0 + louder adjustments), maybe too small vent with regard to my good hearing in lows. Many things to consider!

I bought my top of the line Resound hearing aids, new on eBay for half the price of $6000 quoted by my Audi. I sent copy of my hearing test to the vendor and they were programmed by them. I then made an appointment with an Audi for a REM test. I got her name from a Dr Cliff video on YouTube. She charged me $160 to perform the REM test. So for an incredibly low price I have the latest Resound hearing aids fitted by an Audi using REM testing.

Here’s one video that explains a bit how Aurical’s OnTarget tool works: https://youtu.be/bnAeKm1ObsM10

Should REM show readings that are at or above target unless there’s a specific issue at that level (e.g., feedback)? Some of mine are below target (especially 55dB stimuli) and the lower SII percentage seems to indicate that could be improved. This was just a temporary programming, but I want to understand to get this programmed better when returning.

Also, REM readings should not go into the grey shaded area? To avoid this, those frequencies need to be increased more to make better use of the dynamic range, even if it exceeds target? Do you just increase the lowest stimuli curve or should the higher stimuli curves also be increased so one can better differentiate loudness? You say it should be in the middle of the dynamic range all the way across, but how does one determine the adjustment levels for each stimuli level as those would then be adjusted without being fit to target? For my loss, there is a lot of dynamic range not being used and the target curves in one ear are struggling to fit my loss (crossing into the grey area) so something seems off.

As a test to see if I understand, I used the half gain rule, reduced to a one third gain rule. Is that gain value in dB HL or dB SPL?

When using X/3 to be conservative, although it looks good using the Aurical OnTarget chart and value, the db SPL levels are significantly under this computed value. Is this disparity due to the conversion from dB HL back to dB SPL? I don’t think one can just subtract the hearing threshold sensitivity values at page 1 to convert: Wayback Machine4 but when attempting that it does provide values which are closer to what Aurical is displaying. Without knowing what reference value Aurical uses to convert dB HL to dB SPL it might not be possible to compute things in an exact manner. Plus I’d have to account for any RECD for my ears?

Edit: Costco HIS says they aim to keep the numbers in the On Target table to be less than 5 (250-2k) and less than 8 (3k-6k). The graph of REM accounts for the personal ear acoustics and thus it is ok for the measurement to cross into the grey shaded area of the audiogram. This person is conservative on the 55dB stimulus to avoid causing compression of the three fitting curves. Is that a good approach? Seems it may be just underfitting softer sounds where many HOH people struggle. They also say they don’t have enough handles to get more precision, but I don’t think that’s accurate these days, with some hearing aids having up to 24 bands. I watched her adjust and she mostly grouped 2-3 bands at a time. They seem over-reliant on REM fit to guide what they are doing so that a future autofit protocol can probably do a better job.

I just realised the original comment was 4 years old. I’m very late to the conversation. You said (earlier in this thread) that the NAL prescription for this user was way off and should be disregarded as it didn’t provide enough gain in certain frequency ranges to even overcome this particular users minimal hearing thresholds. I’ve observed similar issues in my clinic when preforming REM. It is bizarre and I was trying to rationalise the clear underamplification in the prescribed gain targets by suggesting it might have something to do with preventing noise-induced hearing loss as a result of too much gain.

The REM shown in the original post is measured in REAR so it should include the effect of ear canal resonance, but gain averaging across both ears could also be a good explanation.

Old thread. Odd thread. I enjoy that the Oticon “how to REM” pdf is open on the computer in the background (it’s that uk_cookbook document)–this was 2020, they were probably double checking the autoREM setup. Blackie is right, the fit looks reasonable. Left is a bit off, but I’d guess probably attributable to open dome issues. It’s pretty standard for NL2 high frequency targets to be below threshold, particularly on a sloping loss. Here’s Gus Mueller:

Why Are Speech Mapping Targets Sometimes Below the Patient's Thresholds?28

Why speech mapping targets are sometimes below the patient's thresholds during hearing aid fittings.

And recall that usually we’re turning hearing aids up to match NL2 targets, which speaks to something about high frequency audibility on a first fit that most people are walking around with. I wonder where the OP ended up.

Speaking to earlier comments, the audiogram does appear to be entered correctly, it is indeed just converted to SPL. Note that conversational speech in SPL is about 60/65 compared to about 40/45 in HL, but it’s also not a linear converstion, differences are bigger at either end. Line 4 is MPO, which could be raised but doesn’t need to be.

Thank you! That article is perfect.

That’s basically it, the manufacturer software wants a symmetrical fitting as their directional models are based on it. If you introduce asymmetrical losses, particularly big asymmetry, the stereo summation goes out the window (on paper) but perhaps not in reality as the apparent symmetry to the user may be more balanced.

What’s a particular (perhaps moreso with AI) is the potential for asymmetric aid fittings to do different (best) responses in the presence of challenging situations. Each aid potentially going its own way to achieve the best Articulation Index, might actually work but you lose 3dB stereo summation and imho is going to sound weird as hell.