Links to this page open in a separate window. Close the window to return to the previous page.
Please feel free to suggest additional terms for inclusion on this page.
Amplitude (of a response): The potential difference (voltage) between two points on a waveform, typically a peak and trough, used to quantify the magnitude of the response.
Auditory Brainstem Response (ABR or BSER): The family of peaks (& troughs) recorded in the 10ms following a brief stimulus such as an acoustic click. On a simplistic basis, each peak has an anatomically distinct generator so the amplitude and latency intervals of the peaks may be used to make gross inferences on the status of the generators or their interconnections. Ironically, peaks I & II arise from (the distal and proximal portions of) the auditory nerve rather than from within the brainstem itself. The generator of peak III is thought to be the cochlear nucleus, peak IV the superior olive, peak V the lateral lemniscus though this model is an over-simplification for the later peaks, especially peaks V, VI & VII, in which multiple generators are implicated. Wave V is often employed in the assessment of hearing since it can be recorded close to the audiological threshold.
Auditory Neuropathy: An abnormality of the auditory neural pathways not evident from imaging studies leading to absent or abnormal ABR and/or later responses rather than from a defect within the organ of corti. Evidence of hair cell function (OAE or cochlea microphonic) is usually present. Patients with this disorder usually have speech discrimination problems that are much more severe than one would associate with their hearing thresholds.
Averaging: A standard means of improving the signal to noise ratio (and therefore our ability to record an identifiable response) based on the assumption that the signal (response) occurs at a fixed time (latency) after the evoking stimulus whereas the noise is random. Multiple (n) samples of activity are recorded over a fixed time period following each stimulus and are summed. The signal will sum algebraically whereas the noise, if truly random, will sum by n/√2.
Contingent Negative Variation (CNV): An "expectation" response usually evoked by a stimulus that the subject associates with a second subsequent stimulus, often of a different modality. For example paired stimuli may be an auditory tone followed by a visual flash. If the subject has been repeatedly exposed to many pairs of these stimuli, not only does the tone evoke its own auditory response, a distinct response occurs in readiness for the flash, even if it is not given.
Electrocochleography (ECochG [ECoG is not used since it refers to the electrocorticogram]): The recording of electrical events arising from within the cochlea. These include the (neurogenic) compound action potential (synonymous with peak I of the ABR), the cochlear microphonic and the summating potential, from which information regarding the function of the cochlea may be gleaned. A principal clinical application is in the diagnosis of Ménière's Disease. The small size of the ECochG when recorded using scalp electrodes requires that a special electrode is used allowing placement closer to the cochlea. Originally, trans-tympanic needle electrodes were used but this invasive technique has been largely replaced with an electrode in contact with the tympanic membrane or ear canal wall.
Epoch (time base or window): The period over which the response is recorded - the horizontal axis of an evoked response waveform. This most often starts at the instant of stimulus onset and extends to beyond the latest possible region of interest of the response. The epoch may start before stimulus onset to provide the user with a sample of non-response "baseline" activity or it may start after the stimulus to avoid recording an unwanted artefact associated with the stimulus.
Filters: The analogue or digital means of removal of unwanted noise from the data, based on its frequency content. Frequency regions beyond the spectral content of the response are attenuated whilst the frequency band of interest is not. The upper and lower boundaries of the "pass band" are usually separate high-pass (rejecting low frequencies) and low pass (rejecting high frequencies) filters.
Frequency Specificity: The extent to which a stimulus, or the response it evokes, relates to a specific or narrow region of audiometric frequencies. This is related to the effective duration of the stimulus and to the way in which it starts and stops. Unfortunately, the audiologically ideal tones used in pure tone audiometry do not evoke many of the responses used in ERA (which are onset responses) so the stimulus needs to start and stop fairly abruptly. However, this leads to a spread of energy to frequencies other than the fundamental of the stimulus (spectral splatter). Since there is a link between response latency and the maximum rise time of the stimulus, early responses such as the ABR require very brief stimuli that are inherently less frequency specific than that available for use with the late responses, such as N1-P2.
Input-Output Function: This is the relationship, expressed as a graph, between an aspect of the stimulus (e.g. intensity) and an aspect of the response (e.g. latency). Knowledge of normal and common pathological I/O functions is helpful in response interpretation (e.g. estimating threshold). Measuring an individual's I/O function can also be helpful (e.g. as an indicator of recruitment).
International ERA Study Group: The IERASG was formed in 1970 and comprises researchers, scientists, audiologists and clinicians with a common interest in ERA & OAE techniques. The group holds a 4-day symposium every two years. Go to Links to visit the IERASG web site.
Inter-Stimulus Interval (ISI): The time from the start of one stimulus to the start of the next, the reciprocal of stimulus repetition rate. Confusingly, ISI is not the duration of the silent interval between stimuli (except for stimuli of zero duration!).
Latency: The time from the instant of stimulus onset to a specific feature or peak in the response.
Masking: The process (or the sound used in the process) of temporary elevation of the hearing threshold, most often applied to the better hearing ear to inhibit the detection of sounds presented to the worse ear. Masking noise may be unfiltered (wide band or white noise) or filtered in some way (e.g. narrow band, centred around the frequency of the test tone). In some ERA tests noise is presented, together with the stimulus, to the test ear in order to improve the frequency specificity of the response or to limit the effective spectrum of the stimulus.
Middle Latency Response (MLR): As the term suggests, responses falling under this heading occur after the early latency responses of the ABR and before the long latency cortical responses, roughly in the range 15 - 70 ms. Peaks carry the labels Na, Pa, Nb, Pb etc. It is debatable whether P1 (at around 50ms) is the last of the middle latency responses or the first of the long latency responses. There is some uncertainty over the generators of the MLR but most agree that the medial geniculate nucleus and auditory cortex are implicated.
Mis-Match Negativity (MMN): If about 10% of stimuli presented to the subject is different in some way to the majority of the stimuli, in addition to the response elicited by the stimulus (e.g. N1), there is an extra negative wave (or an enhanced N1) in the range 100-300ms that is not otherwise present. The MMN may be isolated by subtracting the waveforms evoked by the "rare" and "frequent" stimuli. The MMN does not rely on the subject's attention to the "rare" stimulus, nor on different neural pathways - it can be evoked by a stimulus that is simply shorter than the "frequent". MMN is thought to be generated by the auditory cortex and frontal cortex.
P300: The stimulus paradigm of the P300 is similar to that of the MMN (see above), using an "oddball" stimulus. Unlike the MMN, if the subject is asked to attend to the "rare" stimuli or perform some task in response, an additional positive wave occurs at about 300ms. The generators of P300 are not well understood.
Plateau: The duration (normally specified in milliseconds or optionally by the number of cycles of the tone) of that part of a stimulus tone envelope that has a constant amplitude. See also notes under Tone Burst.
Repetition Rate: The rate at which stimuli are repeated during the averaging process. It is the reciprocal of inter-stimulus interval.
Rise Time: The duration (normally specified in milliseconds or optionally by the number of cycles of the tone) of that part of a stimulus tone envelope whose amplitude rises from zero to the full value. In cortical ERA the rise usually has a linear slope though a more complex transition may be used to maximise the frequency specificity of the stimulus. Traditionally, the fall time is the same as the rise time.
Sweeps: The number of times a stimulus is repeated and ensuing response is summed for the averaging process.
Temporal Integration: The process by which the loudness of a sound is reduced because of its short duration. The effect occurs only for stimulus durations less than about 100ms: the shorter the stimulus, the quieter it becomes (and the more elevated the threshold becomes) because the ear integrates the energy of the stimulus over time.
Threshold: The lowest intensity stimulus that can be detected. In most psychometric test the threshold corresponds to a 50% chance of detection. In clinical auditory evoked potentials the constraints of test time usually limit the intensities used to 10dB steps. The threshold may be taken as the minimum intensity at which a repeatable response is identified or some form of interpolation rule may be used. For example the threshold may be taken as the intensity midway between response and non-response levels, providing that the response meets certain criteria such as minimum amplitude or maximum latency.
Tone Burst: The stimulus of choice for Cortical ERA threshold estimation tests. It comprises a pure tone of the desired audiometric frequency that has a defined start (rise time), duration (plateau) and end (fall time). There is no official distinction between a tone burst and a tone pip but in practice, the term tone burst is often used when the stimulus has an overall length of more than about 20ms (and the plateau is longer than the rise/fall time), whereas a tone pip length is less than 20ms (and usually has a plateau which is the same or shorter than the rise/fall time). The International standard specifying these stimuli, IEC 645-3, confuses matters somewhat by introducing a new term: brief tones. Further, the term duration is defined as the plateau plus half the rise time plus half the fall time. There are several other concerns over the content of IEC 645-3, which is currently under review.
Vertex: The site of the active electrode when recording Cortical ERA. Also known as Cz under the International 10-20 electrode system, it is on the midline, and equidistant from the nasion (bridge of the nose) and the inion (ridge at the back of the head). For the purposes of threshold estimation, the site of the vertex electrode need not be precise since the topography of the response is gradual. Asking the patient the put their finger on the top of their head (not their crown) usually locates the vertex to within a centimetre or so. However, the use of a high forehead site leads to the unnecessary loss of response amplitude and is not appropriate.
Oto-Acoustic Emissions: First identified by David Kemp (Professor of Auditory Biophysics at University College, London) in the late 1970s, OAEs are sounds generated from within the cochlea (outer hair cells) and recorded from the ear canal. OAEs may be spontaneous or evoked by sounds, and are generally absent in cases of significant cochlear dysfunction. As such, OAEs have become popular as an efficient means of screening for hearing loss in neonates and in the investigation of cochlear function. Professor Kemp is a member of the faculty of the ERA & OAE Course.
Links to this page open in a separate window. Close the window to return to the previous page.