The audiogram is a graph of hearing threshold against frequency. Once you can read its axes, symbols and patterns, most of clinical audiometry becomes pattern recognition. This module covers all three.

The axes

Frequency runs along the horizontal axis, low pitches on the left and high pitches on the right, usually from 250 Hz to 8000 Hz in octave steps. Hearing level in dB HL runs down the vertical axis, with 0 dB HL near the top and increasing loss downward. The downward direction is deliberate: a curve that sits lower on the chart represents poorer hearing, so “the audiogram drops off” matches the everyday sense of hearing getting worse.

Because the dB HL scale already accounts for the ear’s uneven sensitivity, a normally hearing person’s thresholds form a roughly flat line near the top of the chart at every frequency.

The symbols

Each ear and each conduction route has its own symbol, plotted at the measured threshold.

Right ear, air conduction — a circle (O), conventionally drawn in red.
Left ear, air conduction — a cross (X), conventionally drawn in blue.
Bone conduction— angled brackets: “<” for the right ear and “>” for the left, with separate symbols used when the opposite ear was masked.

The mnemonic most learners keep is simply: right is round(O) and red; left is a cross (X). Getting the ear and route right is not pedantry — the entire conductive-versus- sensorineural distinction is read from the relationship between these symbols.

TraineeWhen a tone is presented at the audiometer’s maximum output and the patient still does not respond, the threshold is recorded with a small arrow on the symbol indicating “no response” at that level. This is not the same as a threshold: it means hearing is poorer than the limit of the equipment, and it should never be connected into the threshold curve as though it were a measured point.

Degree of loss

The amount of loss is graded from the threshold, most simply by looking at the pure-tone average. The bands in common clinical use run roughly: normal hearing to about 15–25 dB HL, then mild, moderate, moderately severe, severe and profound as the thresholds increase.² The exact boundaries vary slightly between classification schemes, so the audiogram should always be described in numbers as well as in words.

Trainee The pure-tone average (PTA) summarises the loss in a single figure. The classic PTA averages the thresholds at 500, 1000 and 2000 Hz — the frequencies most important for speech. A four-frequency average that also includes 4000 Hz is often preferred for noise-exposure and medico-legal reporting, because it captures the high-frequency damage those losses cause. Both are provided as interactive calculations in the Tools module.

Type of loss: the air–bone relationship

The most important reading is not the degree of loss but its type, and that comes from comparing the air-conduction and bone-conduction curves in the same ear.

Conductive loss — bone conduction is normal, air conduction is poorer. The vertical distance between them is the air–bone gap. The problem lies in the outer or middle ear.
Sensorineural loss — air and bone conduction are equally impaired and track together, with no significant gap. The problem lies in the cochlea or the auditory nerve.
Mixed loss — both curves are abnormal and there is an air–bone gap. A conductive component sits on top of a sensorineural one.

An air–bone gap of roughly 10 dB or more is generally taken as clinically significant. Smaller gaps fall within measurement variability and should not be over-interpreted.

Configuration: the shape of the curve

The shape of the threshold curve carries diagnostic information of its own. A flat loss affects all frequencies roughly equally. A high-frequency or sloping loss spares the low pitches and worsens toward the right — the pattern of presbycusis, noise damage and ototoxicity. A low-frequencyor rising loss does the opposite and is seen in early Ménière’s disease and some conductive losses. A notchedloss dips at one frequency and recovers — most famously the 3–6 kHz notch of noise-induced hearing loss.

Clinician A disciplined way to report any audiogram is to state four things in order: the ear, the type (conductive, sensorineural or mixed), the degree (from the PTA), and the configuration(flat, sloping, rising, notched). A loss described as “left moderate sloping sensorineural” already constrains the differential considerably before any special test is run. The disease modules are organised around exactly these signatures, and the pattern-recognition trainer drills them.

What the audiogram cannot tell you

The audiogram localises a loss to the conductive or sensorineural compartment and quantifies it — but it does not, on its own, separate a cochlear sensorineural loss from a retrocochlear one, and it does not name a disease. Two patients with identical sloping sensorineural audiograms may have presbycusis and a vestibular schwannoma respectively. Closing that gap is the work of the special tests and the clinical context, which the following modules take up.