Q&A With the Expert: John Pumford, Au.D., Director of Audiology


On this edition of our Q&A with the Expert series, we’re discussing the topic of RECD tips and tricks with special guest John Pumford, Au.D., the Director of Audiology and Education at Audioscan. Read on to learn more about RECD from an expert who knows it best!

1. John, you have authored numerous publications that focus on hearing instrument fittings, verification, and probe microphone measurements. Why are you so passionate about this area of research?

In large part,this interest reflects my experiences working in both research and clinical audiology roles where I came to understand that hearing aid benefit is often more dependent on how a product is fitted than on the technology itself. You can add all the bells and whistles you like to a product, but if you don’t get the basic acoustic foundation set appropriately, it is all for nothing. 

While hearing aid programming software has come a long way in helping clinicians configure this acoustic foundation at first-fit, the reality is the acoustic transforms used in these calculations are typically based on average data. While there are patients out there with average ear canals, I still haven’t found a way to just look at someone and know if they are ‘average’.  

With probe microphone (or real-ear) measurements, there is no need for guessing and the clinician can access the measurement data they need to ensure that the device is delivering what is required to the eardrum of the patient. Real-ear measurements are best practice for ensuring the optimal outcome for all hearing aid users, whether children or adults, and the real-ear to coupler difference (or RECD) measurement is a valuable piece of that equation. 

2. You mentioned RECD. How is RECD used in the hearing aid verification process?

First, RECD is used to convertaudiograms (typically those from insert earphones),from dB HL to dB SPL so we can accurately figure out the patient’s thresholds at the eardrum in approaches that use SPLograms – like in Speechmap. 

The second purpose of the RECD, which is often more top of mind, is to conduct what’s called “simulated real-ear measurements” in the test box. In essence, if we know the acoustic difference between a person’s ear canal and the measurement coupler – which the RECD provides us – we can add the RECD to any coupler / test box measure to help predict on-ear performance.   

3. For someone who might be new to completing RECD, what should the clinician be mindful of to make sure the measurement is accurate?

First things first, you want to get the coupler measurement right. You need to select themanufacturer’s recommended coupler and then attach the RECD transducer that generates the test signal while avoiding any slit leaks. Not selecting the proper coupler is a key error to avoid. Often verification systems will provide guidance during the procedure to ensure that this stage is done properly. 

The next step is to complete the on-ear measurement, which first requires placing the probe tube at the proper depth in the ear canal relative to the eardrum. Again, some products provide methods for guiding this probe tube insertion process. Once the probe tube is in place, the clinician wants to consider methods for minimizing any slit leaks that might occur when the RECD transducer and earpiece is inserted into the ear canal.

4. Any guidance regarding how we might know if we've done things correctly?

Well, it probably goes without saying that you can see some variability in the final RECD values across patients- if they all looked exactly the same there probably wouldn’t be any point in measuring it in the first place. That said, there are some trends you should look for. For instance, large deviations from the RECD age-appropriate averages of say 10dB or more in the low frequencies might suggest that slit leaks were present in your measurement. 

To double-check, the clinician would want to reinsert the foam tip more deeply and/or make sure that any vent on a personal earmold is plugged from the medial end as the RECD is not meant to capture vent effects. However, deviations from the averages in the lows can be entirely acceptable and accurate, particularlywith patients havinga surgical ear, TM perf or PE tubes. 

Significant negative deviations from the age-appropriate averagesin the high frequencies may be due to improper probe tube placement where the medial end of the tube is too far away from the eardrum.However, some roll-off in the high frequency RECDs may occur with earmold RECDs, as the tubing in the earmold will roll off the highs.

Again, it is entirely possible that deviations from the RECD age-appropriate averages are accurate, given the variability in ear canals across patients. When in doubt, you may want to simply remeasure the real-ear portion of the RECD, beginning with confirming probe tube placement, reinserting the earpiece and following the steps I previously mentioned. If the result remains consistent, you can feel confident your RECD measurementis valid.

5. Once an accurate RECD is obtained, is it possible to simulate real-ear measurements in the test box. How important is it to select the type of hearing aid (BTE, ITE, Canal, etc.)?

Very important. You are touching on another important component of coupler-based simulated REM, and one aspect of that is the need to include the microphone location effects that would be present on the patient’s ear when the hearing aids are worn. Essentially what I’m referring to is the boost in a sound field signal caused by the influence of the head and outer ear as that signal travels from the source to the hearing aid mic. Obviously in test box measures, there is no ‘head’ or ‘outer ear canal’ present in the box – that would create a whole other set of problems - so we need to build that boost into the test signal. 

Different hearing aids have different mic locations, and therefore a different influence of these external anatomical factors on the signal. For instance, BTEs, with mics over the pinna, don’t benefit from the enhancement of sound caused by the pinna, concha bowl, etc.CICs on the other hand get an additional benefit from these anatomical factors as sound is gathered and boosted by the pinna, etc. If we want to be accurate in our simulation, we should therefore choose the proper hearing aid type from the Speechmap device type dropdown. 

6. It wouldn't be a thorough discussion without some mention of the creation of the ANSI standard for real-ear measurement (S3.46-2013). Why was the HA-1 Coupler specified for RECD measurement in the ANSI standard?

In essence, research has found that in some situations, the RECD value obtained using an HA-2 coupler could vary on the same ear depending on the signal transducer used to deliver the signal. In a nutshell the standard pointed out that the influence of sound source impedance on RECD measurements is taken care of when any tubing or simulated earmold in the coupler component of the RECD is removed – meaning the use of an HA-1 coupler reference. Now, while some systems still use an HA-2 for the RECD measurement, the equipment will apply an HA2-to-HA-1 conversion factor (or math) to get the RECD specified to an HA-1 reference and ensure a consistent RECD measurement.

7. There is a new 0.4 coupler in town. Can you tell us about that?

Yes, that is thenew coupler being used in the Verifit2 – 0.4 cc coupler – which results in what’s termed the wRECD (or wideband RECD). The concept is exactly the same as the standard RECD, the only difference being the volume of the coupler. So, while RECD assumes a 2cc coupler volume, wRECD assumes a 0.4cc coupler. The 0.4 cc coupler was used in this case to allow us to accurately measure high frequency information beyond the standard 8 kHz possible with a 2cc coupler. In essence the 0.4 cc volume allows measurements to get above the mic noise floor in the high frequencies and addressesmeasurement artifacts in the high frequencies given the smaller volume. And should there be a need to get this wRECD specified in an HA-1 reference – say for use in fitting software or other measurement devices – the Verifit2 does the math and provides the equivalent HA1 referenced RECD in a data table.

8. In closing, what do you feel is the most exciting update in the area of real-ear measurement?

Tough question, as there are so many exciting advancements happening in real-ear measurement systems today. If I had to choose just one, I suppose I couldn’t go wrong by pointing to the ongoing developments around autoREMfits. VerifitLINK represents Audioscan’s version of this feature which links the programming software and the Verifit2to enable automatic adjustment of hearing aid settings to targets based on actual real-ear measurement data. The process streamlines best practice verification from both a time management and accuracy perspective, simplifying the whole process. 

Research on our approach with a couple different hearing aid manufacturers has shown it be accurate, reliable and easy-to-use while offering time savings relative to traditional manual methods of verification to targets1,2.I am optimistic that autoREMfits such as VerifitLINK will encourage all clinicians to make real-ear measurements a standard part of all their hearing aid fittings.


Folkeard P, Pumford J, Abbasalipour P, Willis N, Scollie S. A comparison of automated real-ear and traditional hearing aid fitting methods. Hearing Review. 2018;25(11):28-32.

Pumford J, Mueller HG. Using autoREMfit for hearing aid fitting and verification: Evidence of accuracy and reliability. Hearing Review. 2020;27(8):24-27