The Normal Response

What a normal recording looks like, how the threshold is read from it, and how that recorded threshold becomes an estimated behavioural audiogram.

Amplitude grows with intensity

The single most useful fact about a normal recording is that response amplitude is related to stimulus presentation level: as the stimulus rises above the ear’s true threshold, the recorded response grows. It is precisely this orderly amplitude–intensity relationship that allows ASSR to estimate hearing threshold objectively[6].

The growth is not perfectly linear. In ears with hearing loss the response can grow unusually steeply with intensity above threshold — a manifestation of loudness recruitment, and a familiar feature of auditory evoked potential testing in general[9]. Amplitude itself is also modest: a typical response is on the order of tens of nanovolts, recorded against a residual EEG noise floor that may itself be around 40 nV in a settled adult [19]. This is why detection is statistical and why a quiet, ideally sleeping patient matters so much.

Reading the threshold

The ASSR threshold for a carrier is operationally simple: it is the lowest stimulus intensity at which the response is still statistically present. In practice the instrument steps intensity downward — for example in 5 dB steps — restarting the search at a lower level each time a response reaches significance, and stopping a channel once no significant response can be obtained[19]. The lowest level that still yielded a present response is recorded as the ASSR threshold for that carrier and ear.

Explore it

The simulator below makes the relationship concrete. Set an ear’s true behavioural threshold at each carrier and watch where the response becomes detectable; the estimated ASSR threshold is the lowest intensity with a filled (detected) marker. Switch to the polar view to see the same responses as amplitude-and-phase vectors against the noise floor.

The simulator is a teaching model with illustrative values. Real amplitude–intensity functions, normative ranges, and correction factors are device- and population-specific, as the sections below and the cited literature make clear.

The ASSR threshold is not the behavioural threshold

A measured ASSR threshold is consistently higher(poorer) than the behavioural threshold for the same ear — the recording overestimates hearing loss, especially in ears that actually hear well[18]. There are normal values for this: Van Maanen and Stapells, for instance, proposed normal multiple-ASSR threshold values across the standard carriers, against which an individual recording can be judged [18].

The size of the ASSR–behavioural gap depends on the degree of hearing loss. The gap is largest in normal and mildly impaired ears and smaller in more severe losses; one summary reported differences of roughly 0–20 dB in normal-hearing and mildly impaired ears and up to about 10 dB in moderate-to-severe loss[9]. The recruitment effect noted above is part of why severe losses give tighter estimates: the steep amplitude growth carries the response across the detection criterion at an intensity closer to the true threshold.

From ASSR threshold to estimated audiogram

Because the ASSR threshold is biased, it must be corrected before it can be read as a hearing level. Two correction strategies are in common use: subtracting a fixed, frequency-specific correction factor from the recorded threshold, or applying a regression formula that predicts behavioural threshold from the ASSR threshold [17]. Most commercial ASSR systems supply correction tables for exactly this conversion[16].

Picton and colleagues published tables of corrective values indicating that, once corrected, ASSR thresholds typically fall within about 10–15 dB of audiometric thresholds [16]. Crucially, the correction is not universal: its accuracy depends on the equipment, the carrier and modulation frequencies, recording time, residual noise, and on subject factors such as age and sleep state. The electrode montage in particular has been identified as a major source of between-study discrepancy in the ASSR–behavioural relationship. For this reason the clinician should use the correction data and references supplied by the manufacturer of the system in use rather than a generic table [16].

A limitation to keep in view

The correction works least well at the top of the audiogram. In a normal-hearing ear there is, by definition, little or no hearing loss to measure, so the relationship between the corrected ASSR estimate and the true behavioural threshold is weak in that group[19]. ASSR is therefore best understood as a tool for detecting and quantifying hearing loss and estimating its configuration — not as a precise predictor of a normal audiogram.

What a normal recording tells you

Read together, a normal four-carrier recording shows statistically present responses persisting down to low intensities at every carrier, in both ears, with estimated thresholds — after correction — consistent with normal hearing across the audiogram. Departures from that picture, and the audiometric and VEMP-style signatures of specific disorders, are the subject of the disease-focused material planned for later in the atlas.