Audiometry is the measurement of hearing. Clinical hearing tests are designed to evaluate two basic aspects of audition: sensitivity and recognition (or discrimination). Hearing sensitivity measures are estimates of the lowest level at which a person can just detect the presence of a test signal (Ward, 1964). Measures that require an identification response or judgments of sound differences are tests of auditory acuity, recognition, or discrimination. These tests provide information about a listener's ability to recognize, discriminate, or understand acoustic signals, such as speech, and usually are conducted at moderate or higher signal levels. Tests of word recognition are clinical examples of such tests (see suprathreshold speech recognition).

Hearing tests are performed for two primary purposes. One purpose is to identify hearing problems that may be caused by ear disease or damage to auditory structures. In some cases a hearing loss may indicate a medical problem, such as an ear infection. Because medical treatment of ear diseases is successful more often in early stages of the disease process, it is critical that such problems be detected and treated as soon as possible. A second purpose for hearing tests is to obtain information important for rehabilitation planning. In those cases of hearing loss for which medical treatment is not an appropriate alternative, it is important that nonmedical rehabilitative measures be considered based on the communication needs of the affected individual. Information obtained from the hearing evaluation, for example, is required to make decisions about the need for personal amplification (such as a hearing aid) or for other auditory rehabilitation services.

A basic requirement for administration of hearing tests is an acoustic system that enables control of the signals presented to the listener. An audiometer is an electronic instrument used to present controlled acoustic signals to a listener in order to test auditory function. In conventional pure-tone audiometry, the audiometer provides for the presentation of tones ranging in frequency from 125 through 8000 hertz (Hz). A hearing level (HL) dial allows the tester to control the level (in decibels, or dB) of a tone being presented to the listener. The HL control is graduated in steps of 10 and 5 dB (and sometimes smaller) and typically is adjustable over a range of 120 dB. When the HL dial is set to 0 dB, the output level of the audiometer corresponds to an average normal HL at that specific frequency. This is referred to as audiometric zero (American National Standards Institute [ANSI], 1996). This instrumental convenience accounts for the differences in absolute hearing sensitivity (in dB sound pressure level, or SPL) across frequency in persons with normal hearing. Other controls on the audiometer enable the tester to route the test signal to various receivers or transducers used to present the tones to a listener. The two most common receivers used in pure-tone audiometry are a set of earphones and a bone conduction vibrator. Earphones provide for airborne acoustic signals; a bone conduction vibrator is used to transmit vibratory energy through the skull to the inner ear (cochlea). Diagnostic audiometers also provide masking signals (typically a noise) for presentation to the nontest ear during specific audiometric tests. A masking noise is necessary for bone conduction audiometry and for cases of substantial unilateral hearing loss in which the test signal may be intense enough to be heard in the nontest ear. To rule out this possibility, the masking noise is introduced in the nontest ear. The noise elevates the threshold in that ear (masks it) and eliminates it from the test situation, to ensure that subject responses are only for the test ear.

Pure-tone audiometry consists of threshold measures for air and bone conduction signals in each ear (separately). A pure-tone threshold is the lowest level at which a person can just detect the presence of a tone. The threshold measure is statistical in nature; it is a level at which a listener responds to a criterion percentage of the signals presented. Clinically, threshold is usually defined as the lowest signal level at which the listener just detects 50% of the tones presented. Audiometric thresholds are influenced by a number of variables, including the instructions to the listener, the positioning of the earphone (or other transducer) on the head, and the psychophysical threshold measurement technique used (Dancer and Ventry, 1976; Yantis, 1994). In addition, individuals may demonstrate threshold variability related to factors such as motivation, the nature of the ear disorder, the patient's ability to comply with the test situation, and ongoing physiological changes inherent to the auditory system (Wilber, 1999).

Threshold measures obtained from pure-tone audiometry are conventionally plotted on a graph called an audiogram (Fig. 1). The audiogram enables the tester to quickly see the extent to which thresholds for a listener deviate from normal. In Figure 1, HL (dB) is plotted on the linear vertical axis and signal frequency (Hz) is indicated on the logarithmic horizontal axis. The format of the audiogram is such that one octave along the frequency axis corresponds in dimensional scale to 20 dB on the HL axis (ANSI, 1996). Recommended symbols for audiograms are also included in the legend of Figure 1 (American Speech-Language-Hearing Association [ASHA], 1990). Note that separate symbols are used to indicate bone conduction thresholds and measures made with a masking noise in the nontest ear (masked thresholds). Symbols used to record thresholds on the audiogram also may be color coded, with red indicating measures for the right ear and blue indicating results for the left ear.

Figure 1..  

Audiogram form used for recording pure-tone thresholds. Symbols shown to the right of the audiogram are those recommended by the American Speech-Language-Hearing Association (ASHA, 1990).

Specific procedures have been developed for pure-tone audiometry so that thresholds are obtained in a manner that minimizes test variability and are repeatable over time and from clinic to clinic. These procedures are based on research findings for persons with normal hearing and persons with hearing impairment, and include specifications on test frequencies, duration of test tones, step size changes in signal level, and other procedural variables (ASHA, 1978, 1997). The basic test protocol involves initially familiarizing the listener with the tones that will be heard and then determining thresholds for the tones. A bracketing technique is used whereby the tone level at a specific frequency is varied up and down and the listener indicates whether the tone is audible at each level. The tester determines the average hearing level at which the tone was heard approximately half the time over a series of presentations. This level represents the pure-tone threshold at a specific frequency and is recorded on the audiogram. Thresholds are obtained separately for air conduction using earphones and for bone conduction using a bone conduction vibrator. Air conduction tones produced by the earphone are directed down the ear canal, through the middle ear, and then to the cochlea. In bone conduction testing, however, a vibrator is used to transmit the signal through the bones of the skull to the cochlea. The vibrator is placed on the forehead or the mastoid portion of the temporal bone (behind the pinna), and thresholds are measured for the desired audiometric frequencies. All testing is performed in a sound-treated room that meets standards for the exclusion of ambient noise (ANSI, 1999).

Results of pure-tone audiometry, recorded on the audiogram, provide a description of both the degree and type of hearing loss. Because we are mainly interested in a person's ability to hear everyday speech, the customary procedure for classifying degree of hearing loss involves a computation of the pure-tone average (PTA). A person's PTA is her or his average pure-tone thresholds for the speech frequencies of 500, 1000, and 2000 Hz. Table 1 provides a degree classification of hearing loss based on average pure-tone thresholds in the better ear for these frequencies. In considering the purpose and results of audiometric tests, it is important to distinguish the terms hearing loss (or hearing impairment) and hearing handicap (or disability) (ASHA, 1997; ASHA/CED, 1998). The communicative handicap associated with a given amount of hearing loss in dB may differ considerably across individuals.

The classification of hearing loss according to type is based on comparisons of thresholds for air and bone conduction test tones. Conductive hearing losses are almost always due to abnormalities of the outer or middle ear. In such disorders, an interruption or blockage of sound conduction to the cochlea accounts for the hearing loss. Wax in the ear canal or breaks in the ossicular chain (middle ear bones) are examples of conditions that restrict the flow of sound from the outer ear and middle ear to the cochlea and result in a conductive hearing loss. The primary audiometric sign of a conductive hearing loss is an air–bone gap. An air–bone gap is present when hearing sensitivity by bone conduction is significantly better than by air conduction. This can occur in cases of outer and middle ear disorders because the path of sound transmission for bone conduction is primarily through the bones of the skull directly to the inner ear, essentially bypassing the affected outer and middle ear structures. An audiogram for a conductive hearing loss is shown in Figure 2A. Note that although the air conduction thresholds are elevated by 50–60 dB, bone conduction thresholds are within normal limits (0 dB HL). The primary effect of a conductive disorder is to reduce the level of sound reaching the inner ear. The impairment is primarily a loss in hearing sensitivity, not in speech understanding. If the sound level can be increased, the person will be able to hear and understand speech.

Figure 2..  

Example audiograms for conductive (2a), sensorineural (2b), and mixed (2c) types of hearing loss. Further description is provided in the text.

Hearing loss resulting from damage or disease to any portion of the inner ear or neural auditory pathways is classified as sensorineural. Because the problem lies in the inner ear or neural pathways (or both), there will be an equal hearing loss for both air-and bone-conducted signals. An audiogram for a patient with a sensorineural hearing loss is shown as Figure 2B. Notice that the bone conduction thresholds are the same as the air conduction thresholds—there is no air–bone gap. In contrast to conductive hearing loss, sensorineural hearing loss typically involves both a loss in hearing sensitivity and a reduced ability to understand speech. Even when speech is made louder, the person will still have some difficulty in understanding.

A mixed hearing loss is a combination of both conductive and sensorineural losses. Figure 2C is an example audiogram for a mixed hearing loss in both ears. Note that air conduction thresholds are poorer than normal, averaging about 60 dB HL, and bone conduction thresholds are also poorer than normal, averaging 35 dB HL. There is an air–bone gap of approximately 25 dB, suggestive of some conductive hearing loss. In addition, there is a loss by bone conduction, indicating some abnormality of the inner ear and/or auditory nerve.

The case of a mixed hearing loss underscores the fact that both sensorineural and conductive disorders can exist simultaneously in the same patient. A person with a significant sensorineural hearing loss, for example, can still experience an ear infection or other pathology that may result in a conductive component in addition to the sensorineural loss. Finally, it should be understood that many patients with significant disorders or pathologies of the auditory system may have no significant hearing loss (in dB). Patients with disorders of the central auditory nervous system, for example, may demonstrate normal pure-tone thresholds. Normal hearing sensitivity does not always indicate a normal auditory system (Wiley, 1988).