İşitmeAtölyesiCanal volume is individual
The same device produces several dB more sound pressure in a narrow canal. In children the difference grows further — which is exactly why RECD is measured.
Limited on small screens.The dragging, comparison and chart interactions in this tool are built for wider screens. For the best experience, use a computer or tablet.
HEARING AIDS · INTERACTIVE TOOL
REM & RECD Workshop
Measurement in the ear, targets and verification
The gain shown in the fitting software does not tell you what reaches the eardrum: canal volume, natural resonance and venting differ from person to person. Real-ear measurement makes that difference visible. This workshop follows the order of the clinic itself: from the electroacoustic check of the device to the clinical protocol, from the measurement chain to the prescription target, from the difference a child's ear makes (RECD) to the fine-tuning that follows the measurement.
Educational · not a substitute for clinical verification
- 13 interactive modules
REAL EAR
Target match · 42%
Prescription target
Manufacturer's first fit
Corrected with REM
The first fit sits 13 dB below target at 4 kHz: the most information-rich region of speech is under-amplified.
1 Measure
2 Compare
3 Verify
Roadmap
The workshop runs in three stages
Thirteen modules follow the order of the clinic: they begin with checking the device, continue with measurement in the patient's ear, and end with target matching and fine-tuning. Click any heading to jump to a module, and use this rail to keep track of where you are.
1
Stage 1
Before the measurement
The order in the clinic: first the device itself is checked, then the ear is prepared for measurement.
0 / 3 modules
2
Stage 2
Measurement and target matching
What do the measured curves mean, which target are they judged against, and when do you stop adjusting?
0 / 5 modules
3
Stage 3
What affects the measurement, and what follows it
The discomfort limit, acoustic coupling, the volume of a child's ear, and the part of the appointment that comes after the measurement.
0 / 5 modules
- 9Dynamic range, MPO and UCLStaying below the discomfort threshold→
- 10Coupling and occlusionVent, feedback and the gain trade-off→
- 11The child's ear: RECDThe difference between a 2 cc coupler and a real ear→
- 12How is RECD measured?Four steps and simulated REM→
- 13After the measurementFine-tuning, instruction, follow-up→
Why
Right on the screen,
Right on the screen,
wrong in the ear
The manufacturer's “first fit” looks at the audiogram and proposes a gain. That proposal assumes an average ear — yet an ear canal may hold 0.8 ml or 2.2 ml. The same output produces a higher sound pressure in a small canal; and the moment you seat the earmould, you also switch off the ear's own natural gain. That is why a fit that looks right on the screen can be wrong in the ear.
The ear has a resonance of its own
The open ear naturally boosts sound by 15–20 dB around 2–3 kHz. Close the canal and that natural gain goes with it; the device has to give it back.
First fits usually miss the target
In the literature, first-fit outputs deviate markedly from the prescription target, especially at high frequencies. The only objective step that reveals the gap is real-ear measurement.
Is the device sound?
—
—
—
—
—
—
—
—
—
—
Test-box measurements (ANSI S3.22)
One question must be answered before any real-ear measurement: does the device perform to the manufacturer's specification? An electroacoustic check in a test box, using a 2 cc coupler, compares output, gain, distortion and internal noise against the manufacturer's data. The button below drops a random device into the box — some of them are faulty.
Which device goes in the box?
Ready
StageWaiting for a measurement
Stimulus—
Frequency—
How to read the measurement
A value outside tolerance means the device should go for servicing ; the problem must be resolved before moving on to REM.
A high THD usually points to a damaged receiver, while a high EIN points to a noisy microphone.
A coupler measurement does not tell you the output in the ear — for that you need RECD and a real-ear measurement.
MeasurementMeasuredToleranceManufacturerResult
OSPL90Maximum output at a 90 dB input
—dB SPL
Full-on gainHFA · average of 1000–2500 Hz
—dB
Total harmonic distortionTHD · 500–1600 Hz
—%
Equivalent input noiseEIN · the device's own noise
—dB SPL
Battery drainOperating current
—mA
Frequency response · at a 60 dB SPL input
—
Manufacturer's tolerance bandManufacturer's curveMeasured
Waiting for a measurement. Place a device in the box and press “Measure the device”.
ANSI/ASA S3.22 (and IEC 60118-7) define the conditions under which these electroacoustic characteristics are measured. The values shown are examples.
Next step
Once the device has been shown to work correctly on its own, the next step is to measure it in the ear.
Clinical protocol
REM step by step: from otoscopy to target matching
Download the protocol cardOne page · PDF · printable for the clinicThe reliability of a measurement is decided before the probe microphone ever enters the ear. If otoscopy, room setup, calibration and probe-tube placement are not done properly, the curve you obtain shows the measurement error, not the device. The nine steps below follow the order used in the clinic; click a step to see its checkpoints.
1 / 9 steps
Step 1 / 9
—
—
Checkpoints
Schematic
—
In the field
Six things that ruin a REM
If the measurement is wrong, the verification is wrong too: a curve you believe sits on target may be nothing more than the product of a measurement error. The six most common errors are summarised below — click a heading for the detail.
If the tube does not sit close enough to the eardrum, a standing-wave error appears in the high frequencies: you see a dip or a peak at 4–6 kHz that is not really there. In adults the tube is inserted to a depth of 28–31 mm measured from the inter-tragal notch, with its tip 3–5 mm from the membrane. The placement is checked against the unaided REUG curve.
If hair, a mask or the device itself covers the reference microphone, the system levels the loudspeaker incorrectly and the entire curve shifts. The patient must sit still at 0°/45° to the loudspeaker throughout. With open-fit calibration the reference microphone is disabled anyway, which makes keeping still even more critical.
In open fittings the low-frequency gain escapes; trying to “hit” the target at 250–500 Hz produces feedback and unnecessary gain. Verification in an open fitting is meaningful mainly above 1 kHz, and that is where the target should be met. The vent decision can be checked with a REOG measurement.
Looking only at 65 dB hides the compression. Measure separately at soft (50), normal (65) and loud (80) dB SPL inputs; each level has its own criterion. The device's maximum output must be verified separately with an 85–90 dB narrowband signal.
If the output ceiling exceeds the discomfort threshold, the person takes the device out — and often never puts it back in. In an ear with a narrow dynamic range, setting the MPO matters more than setting the gain; UCL should be measured frequency by frequency and the MPO kept below it.
Without loudspeaker and probe-tube calibration the whole measurement shifts — and on screen that shift looks like the behaviour of the device. Calibration is done at the start of the day, whenever the tube is changed, and separately for each ear.
Next step
Once the steps are in place, one question remains: what exactly do the measured curves tell you?
The measurement chain
REUR → REOR → REAR → REIG
The curves taken during the protocol are recordings of a single signal under three conditions: with the ear open (REUR/REUG), with the earmould or device in place but switched off (REOR/REOG), and with the device running (REAR). The differences between them reveal what the device actually adds to the ear — and the gain the device contributes on its own (REIG) falls out of that difference. Step through them in order, and change the coupling to watch the curves move.
REUR · open ear
REOR · earmould in place
REAR · device on
REIG · net gain
Step 1 / 4
REUR — Real-Ear Unaided Response
—
—
2 kHz'te
—
Coupling
—
The curves are schematic, drawn with typical values. With a sealed earmould, REOR is close to zero but not perfectly flat: small negative values reflect the insulation the mould provides. In a real measurement the values vary with the person, the probe placement, the earmould and the device.
Next step
Once you know what the curves mean, the next question is the criterion they should be judged against.
Prescription
NAL-NL2 and DSL v5 on the same audiogram
The target curve — the criterion for verification — comes from the prescription formula. Two common formulas read the same audiogram differently: NAL-NL2 aims to maximise intelligibility at a comfortable loudness, while DSL v5 aims above all to make soft sounds audible. DSL generally prescribes more gain, which is one reason it is preferred in paediatric fittings. Change the shape of the loss and see where the two targets part company.
NAL-NL2DSL v5
Loss
DSL v5 prescribes more gain than NAL-NL2
Mean difference—
Largest difference—
In paediatric fittings this extra gain is preferred for the audibility of soft sounds.
Values are shown as approximate insertion gain at a 65 dB input.
Fitting
Matching the target: bring the fit onto target by measuring
Once the target curve is set, the next step is to match the sound the device produces at the eardrum (REAR) to that target. The measurement is repeated at three input levels and the gain settings are corrected at each. The manufacturer's first fit usually misses the target — use the sliders below to bring the output onto it.
Input level
Prescription target Device output (REAR) MPO ceiling
—
A simplified representation; a real REM measures frequency-specific gain and output at soft, medium and loud inputs and keeps the MPO below the discomfort threshold. Sources: AAA/ASHA best practice; NAL-NL2, DSL v5.
Criterion
What does a good fit look like?
Target matching is not done at a single input level. The measurement is repeated at least at three levels — 65, 50 and 80 dB SPL — and each level has its own criterion. Until all three are met, gain is raised or lowered in the relevant frequency region and the measurement is repeated.
50 dB
Soft speech
The measured curve must sit above the hearing thresholds. Audibility of soft sounds is where REM-fitted devices show their clearest advantage over a first fit.
Above threshold
65 dB
Normal speech
The measured curve should run parallel to the prescription target and as close to it as possible. Deviations show up especially in the high frequencies; the software's delta values put a number on the comparison.
Parallel to target
80–90 dB
Loud input · MPO
A loud speech level is presented at 80 dB SPL, while the device's maximum output is verified separately with an 85–90 dB narrowband signal. In both cases the output must remain below the discomfort threshold (UCL). In an ear with a narrow dynamic range this criterion matters more than the gain criterion.
Below UCL
Speech mapping
What does hitting the target buy you?
The clinical value of approaching the target is measured by how much of speech rises above the hearing threshold. Below, the long-term average speech spectrum (LTASS) and its dynamic range are plotted together with the hearing thresholds in the sound-pressure domain at the eardrum. The meter on the right shows a rough index of audibility, the SII tahminini verir.
Unaided speech
Aided speech
Hearing threshold (SPL)
Audiogram
Speech level
0% = unaided · 100% = gain placed fully on target with REM
—estimated SII
—
The SII here is computed with a simplified octave-band approach (band importance weights × the audible proportion of speech); it is not a substitute for clinical SII software.
Next step
The ceiling matters as much as the gain: the measurement is read together with the discomfort limit and the acoustic coupling.
Output limit
Dynamic range, MPO and the discomfort threshold
Placing the gain on target is only half the job; the other half is delivering the sound without discomfort. The distance between the hearing threshold and the discomfort threshold (UCL) is the person's dynamic range. The narrower it gets, the more compression — and the more careful an output ceiling (MPO) gerekir.
Dynamic range—
Compression ratio required—
MPO status—
—
The dynamic range here is simplified to a single frequency (2 kHz) in dB HL; in the clinic, UCL is measured frequency by frequency and the MPO is verified with an 85–90 dB narrowband signal.
Acoustic coupling
Vent, occlusion and feedback: three faces of the same trade-off
The second great influence on the measurement is the acoustic coupling between the device and the ear — and it is what you see in the REOG. The wider the vent, the less the person's own voice booms (occlusion); but the sound escaping through that same opening both erodes the low-frequency gain and pushes the system towards the feedback limit. There is no single “correct vent” in the clinic — only a trade-off made according to the shape of the loss. Move the slider and watch the achievable gain, the occlusion effect and the maximum stable gain together.
Prescription targetAchievable gainFeedback limit (MSG)
Gain lost at 250 Hz—
Occlusion effect—
Feedback headroom—
—
- Feedback suppression typically raises the maximum stable gain by a few dB; it does not give unlimited gain.
- In an ear with good low-frequency hearing, an open fitting is preferred: there is no occlusion, and the low-frequency gain that is lost is not needed anyway.
- With a wide air-bone gap or a severe loss, a sealed earmould is required; some occlusion is unavoidable and the acclimatisation process becomes important.
The values follow typical magnitudes from the literature: a 3 mm vent costs about 20 dB of output at 250 Hz, and in an open fitting the loss extends up to 1 kHz. The occlusion effect reaches ~25 dB at 250 Hz with a closed mould, and a 2 mm vent reduces it by roughly 8–9 dB.
The child's ear
RECD and the 2 cc coupler difference
Everything so far has assumed an adult ear. In children the conditions change: in the test box the device is coupled to a 2 cc cavity, while the real ear may hold far less. A smaller volume means a higher sound pressure for the same device output. The difference between the coupler and the real ear is called the RECD. Change the age and watch where a fit that looks on target in the coupler actually lands in the real ear.
Selected age
Adult (reference)
4 kHz'te RECD—
Difference from the coupler target—
—
- A measured RECD is safer than a predicted one: canal volume, the seat of the earmould and middle-ear effusion all change the RECD.
- In children the target is usually set with DSL v5 , and the RECD is used to carry the coupler measurement over to the real ear.
- A small canal means a higher SPL: MPObelow the discomfort threshold becomes critical.
The values are typical magnitudes close to the DSL age norms; in the clinic, RECD is measured individually.
Measuring the RECD
How is RECD measured?
The RECD is the difference between the sound pressure measured in the real ear and the sound pressure the same signal produces in a 2 cc coupler. The measurement has four steps, and one rule is critical: whichever transducer is used for the ear response must also be used for the coupler response.
1
Probe-tube calibration
The acoustic effect of the probe tube is removed from the measurement. The tip of the tube is aligned in front of the reference microphone and a calibration signal is presented.
2
Ear response
The probe tube is placed in the ear and the response is measured with an insert earphone or a button receiver. If the device itself is to be used, the same device is used throughout.
3
Coupler response
The same transducer is connected to the 2 cc coupler and the measurement is repeated. The coupler has none of the canal's resonance — and that is where the difference comes from.
4
Difference = RECD
The software computes the difference between the two curves and stores it. Every subsequent coupler measurement carries this correction: the acoustics of the ear are back in the equation.
Preferred
A measured RECD
Canal volume, the seat of the earmould, insertion depth and middle-ear status are all individual. Whenever possible, measure the RECD.
Fallback
Predicted RECD
A prescription formula (e.g. NAL-NL2) predicts an RECD from age, sex, device acoustics, insertion depth and vent size. Enter those parameters wrongly and the prediction is wrong too.
Next step
A curve on target is not the end of the fitting; getting the person to accept the sound and stay with it takes separate steps.
After the measurement
On target — but the appointment is not over
A curve that matches the target is the middle of the fitting, not the end. Accepting the sound, being able to handle the device and staying with the process each take their own steps.
Fine-tuning
First reactions such as “it's too shrill”, “my own voice bothers me” or “everything is too loud” are to be expected. Rather than cutting gain across the board, use the software's acclimatisation manager: reduce the high-frequency components temporarily and plan a gradual rise to target over time. That way the frequency shaping is preserved.
Instructing the user
Before leaving the clinic the person should be able to insert and remove the device correctly; changing or charging the battery, cleaning and care, using the buttons and programs, pairing with a phone or other devices, and the meaning of the audible alerts should all be demonstrated.
Follow-up and counselling
A follow-up appointment is scheduled; remote support is also an option. Counselling runs through the whole process: explaining the loss, involving the communication partner, aligning expectations with the benefit that is realistically available. In children the measurement is repeated more often and confirmed with sound-field testing and speech tests.
Result
From verification to validation
A real-ear measurement shows that the device has reached its target; it does not show whether the person benefits in real life. That second question is answered by validation: self-report scales and speech-in-noise tests. Verification is objective, validation is subjective — and neither replaces the other.
Self-report
IOI-HA
Short and international: hours of use, benefit, residual difficulty, satisfaction and impact on life — a general picture of benefit in seven questions.
Explore it in ODAK → Self-reportAPHAB
Compares the aided and unaided conditions in four domains: ease of communication, reverberation, background noise and aversiveness of sounds.
Explore it in ODAK →
Goal-based
COSI
Measures progress against five listening goals the person sets themselves; it is also used in the clinic to manage expectations.
Behavioural
Speech in noise
QuickSIN or matrix tests give the person's “SNR loss”: how many dB of extra SNR they need compared with a normal-hearing listener. The decision to fit a remote microphone can rest on that number.
Verification (REM) is objective and validation is subjective; neither replaces the other. Sources: AAA/ASHA best-practice guidelines; Cox (APHAB), Cox & Alexander (IOI-HA), Dillon (COSI), Killion (QuickSIN).
FAQ
Frequently asked questions
A first fit produces an estimate from the audiogram; it cannot see the ear's volume, its resonance or its leakage. Best-practice guidelines (AAA, ASHA, BSA) define probe-microphone verification as a standard step. Without REM, there is no way to say the target has been reached.
REAR is the total sound pressure at the eardrum with the device switched on. REIG = REAR − REUR: the gain the device adds on top of the open ear (insertion gain). REAG is the difference between the REAR and the level at the loudspeaker. Target curves are usually given in the REAR (SPL) or REIG (dB gain) domain.
Yes, but the gain measured at low frequencies stays small because of leakage and should not be forced onto the target. Verification in an open fitting is meaningful mainly above 1 kHz, and the feedback limit further constrains the measurement.
A small canal produces a higher SPL for the same device output — and a long REM on an awake infant is often impossible. The RECD is measured and applied to the coupler measurement so that the target can be reached in the coupler: this is “simulated REM”.
Calibration removes the acoustic effect of the probe tube and the microphone from the measurement. It should be repeated at the start of the day and whenever the tube is changed, and it is done separately for each ear. On an uncalibrated system every curve is systematically shifted — and on screen that shift looks like the behaviour of the device.
The tip of the tube should sit about 3–5 mm from the eardrum; the accuracy of the high-frequency response depends on that proximity. Average depths are measured from the inter-tragal notch: 28 mm in adult women, 30–31 mm in men; in children, 11 mm at 0–6 months, 15 mm at 6–12 months, 20 mm at 1–5 years and 25 mm above 5 years. Placement can be checked with the REUG: a marked departure from 0 dB at 6 kHz suggests the tube is too shallow.
No. Verification is objective: has the device reached the target (REM)? Validation is subjective: does the person benefit in real life (COSI, APHAB, IOI-HA)? Neither replaces the other.
Sources
What is this page based on?
- American Academy of Audiology. Guidelines for the Audiologic Management of Adult Hearing Impairment.
- American Speech-Language-Hearing Association (ASHA). Guidelines for Hearing Aid Fitting for Adults; Guidelines for the Audiologic Management of Pediatric Amplification.
- British Society of Audiology / British Academy of Audiology. Guidance on the Use of Real Ear Measurement to Verify the Fitting of Digital Signal Processing Hearing Aids.
- Keidser, G., Dillon, H., Flax, M., Ching, T., & Brewer, S. (2011). The NAL-NL2 prescription procedure. Audiology Research, 1(1), e24.
- Scollie, S., Seewald, R., Cornelisse, L., et al. (2005). The Desired Sensation Level Multistage Input/Output Algorithm. Trends in Amplification, 9(4), 159–197.
- Bagatto, M., Scollie, S. D., Seewald, R. C., Moodie, K. S., & Hoover, B. M. (2002). Real-ear-to-coupler difference predictions as a function of age for two coupling procedures. Journal of the American Academy of Audiology, 13(8), 407–415.
- Kuk, F., & Keenan, D. (2009). Comparison of vent effects between a solid earmold and a hollow earmold. Journal of the American Academy of Audiology, 20(8), 480–491.
- Dillon, H. (2012). Hearing Aids (2nd ed.). Thieme.
- Aazh, H., & Moore, B. C. J. (2007). The value of routine real ear measurement of the gain of digital hearing aids. Journal of the American Academy of Audiology, 18(8), 653–664.
- Valente, M., Oeding, K., Brockmeyer, A., Smith, S., & Kallogjeri, D. (2018). Differences in word and phoneme recognition in quiet, sentence recognition in noise, and subjective outcomes between manufacturer first-fit and hearing aids programmed to NAL-NL2 using real-ear measures. Journal of the American Academy of Audiology, 29(8), 706–721.
- Interacoustics Academy. Real-ear measurements: A step-by-step guide.
- MedRx. RECD measurement — step by step.
- ANSI/ASA S3.22-2014. Specification of Hearing Aid Characteristics.
- ANSI S3.5-1997 (R2020). Methods for Calculation of the Speech Intelligibility Index.