9Objective measures → the MAP

FThe mapping problem

A MAP is the complete set of programming parameters stored in the sound processor: for every active electrode, a threshold (T) level and a comfortable (C or M) level defining the electrical dynamic range into which sound is compressed, plus the coding strategy, stimulation rate, and pulse width. Ideally every T and C is set behaviourally. In reality — at first fit, in young children, across 12–22 electrodes — full behavioural measurement of every value is slow or impossible, and that is the gap objective measures fill.[2014]

TCObjective anchors for levels

Two anchors sit at different points in the dynamic range, and they are complementary:

The ECAP threshold correlates with behavioural levels but the relationship is loose and offset between individuals; it is better at predicting the relative profile across electrodes than the absolute value on any one. The ESRT, sitting near C/M, is a stronger predictor of the comfortable ceiling. Used together they bracket the dynamic range from both ends.[2000, 2002, 1997]

Some working numbers: a Nucleus electrical dynamic range is typically about 25–60 current levels (most ~40), and the ECAP threshold usually falls above behavioural T and within roughly ±20 CL of the C-level — so a comfort level sitting far above the ECAP threshold (more than ~30–40 CL) is a flag for over-stimulation. The per-manufacturer ESRT → comfort-level calculator in Module 5 turns the comfort rule into a concrete target.[2014]

CProfile-based & scaled fitting

A key practical refinement is to use the ECAP not value-by-value but as a profile. The shape of the tNRT across the array — which electrodes need more current and which less — tends to track the shape of the behavioural T/C profile, even when the absolute offset is unknown. So a clinician can measure behavioural levels on a few electrodes, then use the ECAP profile to interpolate the rest, rather than measuring all of them.[2010]

This scaled-profile approach — anchor with a small number of behavioural measurements, shape with the objective profile — is more accurate than applying a single fixed offset to the whole ECAP profile, and it is the most clinically useful way the ECAP feeds the MAP.

A few rules govern how the ECAP threshold relates to the behavioural MAP, and they shape how you read the profile-shift above: the ECAP threshold almost always sits above behavioural T, in the upper part of the dynamic range, and is more likely to exceed C/M at faster stimulation rates; when the T and C profiles differ, the ECAP profile tends to follow the T-level shape. Because the MAP uses fast pulse trains while the ECAP uses slow single pulses, the correlation between the two is only moderate and worsens as rate rises — which is exactly why the objective profile is a scaffold to be refined behaviourally, not a finished MAP.[2013]

TCPaediatric fitting — where objective measures carry the load

In infants and young children, reliable loudness scaling is impossible and even conditioned-response audiometry is limited. Here objective measures move from supporting role to lead:

• ESRT anchors the comfortable ceiling so the child is not over-stimulated.
• ECAP thresholds / profile shape the level map across electrodes.
• eABR / cortical responses confirm the pathway conducts and matures (Modules 6–7).
• Behavioural observation then refines the MAP progressively as the child develops testable responses.

The recommended practice is a battery — no single objective measure suffices, but combined they produce a safe, audible starting MAP that behavioural refinement then optimises.[2004]

When a recipient genuinely cannot give behavioural responses, the ECAP-threshold profile itself can seed the whole MAP: measure ECAP across the array, then derive T and C by applying offsets to that profile. Because the behavioural upper profile is flatter than the ECAP profile (channel interaction smooths it), the C profile is flattened toward its mean — the refinement that made scaled-profile fitting more accurate than a single fixed offset. Set the offsets below, toggle the flattening, and watch the derived T and C profiles; the whole map is then dropped below audibility and raised in live speech while watching for a response.[2000, 2002, 2010]

Written out as a protocol, the steps are:

1. Measure the ECAP threshold on every intracochlear electrode to get the profile.
2. Set T and the upper level to approximate that profile's shape.
3. Smooth any abrupt electrode-to-electrode jumps so the profiles change gradually.
4. Offset both down by a small amount (~10 CL, or ~10% of the ECAP threshold — e.g. a 150-unit threshold → ~15-unit offset) and then drop the whole map below audibility.
5. In live-speech mode, raise levels gradually while watching for a behavioural response; set T at the minimal level that elicits one.
6. Raise the upper level slowly toward loudness normalisation, flattening the C profile as it rises and stopping at any aversive sign — even if still below where an ESRT would sit.

It applies to a defined group: young children with inconsistent responses, recipients with multiple disabilities, those with PE tubes or no measurable ESRT, and those who will not tolerate the immittance probe. This is a starting MAP for un-testable recipients only — when reliable behavioural levels or an ESRT can be obtained, they take precedence. Build from ECAP alone only when you must.[2014, 2020]

TA combined workflow

1. Impedance — confirm intact contacts; deactivate opens/shorts (Module 2).
2. ECAP — obtain thresholds/profile across the array (Modules 3–4).
3. ESRT — anchor the C/M ceiling where obtainable (Module 5).
4. Set a starting MAP — C/M from ESRT, profile shape from ECAP, T-levels conservatively below.
5. Refine behaviourally — loudness balancing, comfort checks, and (in children) observation and serial visits.

TCSetting levels at the chair

Objective measures seed and cross-check the MAP, but the levels are still finished behaviourally — and the behavioural technique is worth knowing, because it is what the objective data are integrated with. Threshold (T) levels are measured ascending (responses sit lower on descending runs, so an ascending criterion guarantees soft-sound audibility); a count-the-beeps method — the recipient reports how many bursts they heard — is robust against tinnitus confusion and false positives.[2020, 1995]

Upper levels are best set in live-speech mode: drop the whole map below audibility, then raise the upper levels globally while the recipient listens to running speech. A useful pearl — recipients often say “that's enough” beforethe optimum, and a small further increase usually improves clarity, so push slightly past the first “comfortable” report and settle where sound quality peaks. Then loudness-balance across electrodes two at a time (judging loudness, ignoring pitch, adjusting the second of each pair), and finally smooth rather than perfectly flatten the upper profile.[2020]

The whole thing is then verified behaviourally: aided sound-field thresholds (a target of roughly 25–30 dB HL in adults and 20–25 dB HL in children), the Ling six-sound test (detectable at ~15 ft within a month), and speech-perception checks. The ESRT belongs in this routine for all ages as the objective ceiling — upper levels should not exceed it.[2020, 2017]

Two modern refinements are worth noting. Fitting can be streamlined or comprehensive: on Advanced Bionics and MED-EL, T can be set to zero or estimated from the upper level with good outcomes for typical adults, reserving full per-electrode T measurement for when soft-sound audibility is actually poor (the older practice of loudness-scaling every channel for T is now reserved mainly for the rare “T-tail” — a flat span of loudness growth across the lower dynamic range). And frequency allocation should keep high-frequency audibility wide — out to ≥6 kHz in adults and ideally ≥8 kHz in children acquiring speech.[2020, 2011]

A documented pitfall, especially in children: upper levels tend to be ratcheted up a little at each visituntil, over years, they drift well above typical values for no good reason. Validate every level against the device's typical bands plus performance and the objective ceiling — do not inflate by habit (the worked over-stimulation cases in Module 10 show where that leads).[2020]

CThe limits of objective-only programming

The recurring honest caveat of this atlas applies most sharply here. Objective measures tell you the system can deliver an audible, comfortable signal; they do not guarantee speech understanding, optimal loudness balance, or the best spectral allocation. A MAP set purely on objective data is a safe starting point, not an optimised one. Where the recipient can give reliable behavioural responses, those take precedence; objective measures earn their place by getting the MAP close, fast, and safely — and by carrying programming when behaviour cannot.[2014]

TCTroubleshooting common complaints

When a recipient returns unhappy, the objective measures become a troubleshooting toolkit. Most complaints map to a recognisable set of programming adjustments — and the same anchors recur: balance loudness at the upper-stimulation level (USL) across the array, use the ESRT to sanity-check that comfortable ceiling, and sweep the array (an ECAP or loudness sweep) to find the electrode behind a poor or aversive percept. The navigator below pairs each common complaint with the adjustments to try, in order.[2014, 2014, 2012]

Two of these — facial-nerve stimulation and non-auditory percepts (tinnitus, tactile sensation) — are worked through as clinical cases in Module 10.[2023, 2023, 2000]