ALR / Cortical Evoked Response Audiometry
What are auditory late responses (ALRs)?
Auditory Late Responses (ALRs) are longer latency components that for most part are generated in higher regions of the auditory CNS, including the auditory cortex.
What is cortical evoked response audiometry (CERA)?
Cortical evoked response audiometry (CERA) refers to the technique of measuring ALRs for the purpose of assessing hearing ability. The responses are typically measured using surface electrodes placed on the scalp of an individual.
Why measure ALRs?
CERA is traditionally used to help determine the degree of hearing loss in adult populations. Compared to traditional auditory brainstem responses (ABRs), or behavioral audiometry, ALRs demonstrate audibility at the cortex without the need for the listener to play an active role in the procedure. This is a key advantage in various medico-legal scenarios or in cases where the individual is unable or unwilling to provide accurate behavioral responses to a sound.
Other advantages include the relative robustness of the ALRs to myogenic activity, the high frequency specificity of the tonal stimuli due to their longer duration, and is much closer to the behavioral audiometric pure tone compared to traditional ABR octave wide stimuli, and the ability to calibrate the stimuli according to international standards (BS EN ISO 389 as for pure tone audiometers) and not the standard 389-6 used for Tone burst/Tone pips & Chirps .
The key ALR waveforms observed in CERA are the P1, N1, and P2 responses. The ALR latency typically ranges from 50 – 300 ms and N1-P2 amplitude ranges from 0-20 µV. (see Figure 1, showing P1, N1 and P2 responses to a 2 kHz toneburst stimuli, used to provide a typical threshold recording from the left ear. In this case, hearing sensitivity is shown to be within the normal audiometric range.
How to measure ALRs
Patient preparation
Patient arousal and attention state has a significant effect on the amplitudes of the ALR. The ALR waveform changes as a person becomes drowsy or falls asleep. When a patient is asleep the N1 amplitude is smaller and the P2 amplitude is larger. However, when the subject is listening for a change or paying close attention to stimuli the N1 increases in amplitude and in subjects who are not attending to the stimuli, the N1 can become difficult to measure at low intensities (Näätänan and Picton, 1987). The response also habituates quickly, an affect which is more apparent nearer to the threshold of the listener, so it is important to limit the number of stimulus presentations within each ‘run’. Typically, between 15 and 20 presentations might be made per run and a number of successive runs are then merged into a grand average in order to reveal the ALR. The patient is typically instructed to sit quietly during the procedure, maintaining passive attention for example by reading or watching a close-caption movie with the sound muted. It is not advised to perform ALR and under sedation (Crowley & Colrain, 2004).
Electrode placement
It is possible to obtain ALR with a standard 2-channel electrode montage, with an active vertex electrode referenced to either right or left mastoid. However, since the ALR has generators orientated towards the frontocentral regions of the scalp then response strength may be greater when recorded from a point on the midline that is slightly forward of the vertex position.
Setting up the Eclipse
The Eclipse comes with a pre-programmed protocol for ALR testing (license), and is ready for immediate use. Protocols can be created or modified easily to fit your clinic needs. Consult your Eclipse Additional Information to learn how to create or modify a protocol.
Protocol settings
In CERA, ALRs should be measured using tone burst (250 Hz – 4 kHz) stimuli at intensity levels between 0 and 100 dB HL to establish threshold
Summary of parameters for ALR
| P1, N1, P2 | |
| Patient state | Awake and quiet |
| Eyes | Open |
| Condition | Attend or ignore conditions |
| Types of stimuli | Tone burst, speech vowels or consontant-vowel combinations |
| Inter-onset interval | 1 to 2 seconds |
| Stimulus duration | 50 to 80 ms (including 10 to 15 ms onset/offset ramps) |
| Presentation | Typically insert or supra-aural headphones |
| Intensity | Starting at 60 to 80 dB HL |
| Reference electrode | Right/left mastoid (optimally linked electrodes using the jumper cable) |
| Filtering | 1 to 30 Hz |
| Analysis time | Pre-stimuli: -100 ms; post-stimuli: 700 ms or more |
| Sweeps | 50 to 300 (comprised of 10 to 20 sweep sub averages) |
| Waveform reproducibility | Set within the latency range of 30 to 270 ms |
| Measurements in adults | P1, N1, P2 |
| Measurements in children | P1, N200-250 |
| Measurements in infants | Replicable components |
| Measures | Baseline to peak amplitude, peak latency (use latency window from mean data) |
| Response presence determined by | 1) Replicable components 2) Response amplitude should be larger than or equal to 2.5 times the residual noise 3) Residual noise amplitude should be less than or equal to 1.5 µV (as determined by the average difference between replicates) |
Interpretation of the ALR result
Typically the ALR threshold recording is started at 60dB HL and increased or decreased by 20dB based on the response. 5 or 10 dB steps are typically used when close to threshold.
It is recommended that the amplitude of the response is taken as the peak-to-peak amplitude between N1 and P2, and a lower value of 2.5 µV is recommended before a response can be clearly identified. The Waveform Reproducibility function can also be used to indicate the response reliability. This function provides a measure of the correlation between curves held in the A and B buffer when there is a response present. The residual noise in the trace can be estimated by measuring the ‘average gap’ between replicates (for example, between rarefaction and condensation responses displayed in the A-B buffer).
See Figure 2, which shows the A and B curves separate for each trace. Over the duration of the epoch (-150 ms to + 750 ms) the average gap between traces can be observed to be below 1 µV, and the Waveform Reproducibility for the highlighted curve is shown to be high (73%) indicating reliable response.
Electrophysiological threshold estimation and correction factors
An ALR threshold at 20dBHL at 2kHz would be considered within the range of normal hearing. Applying a typical correction factor would estimate the behavioral threshold to be 13.5dBHL at 2 kHz. For masking the ALR, please follow the pure tone audiometry guidance’s.
Lowest level response >5uV: interpolate
Lowest level response <5uV: is threshold
ALR threshold behavioral correction factors to estimated hearing thresholds [2]:
| 500 Hz | 1000 Hz | 2000 Hz | 4000 Hz | |
| Stimuli (dB HL) | 50 | 60 | 65 | 65 |
| Mean correction (dB) | -6.5 | -6.5 | -6.5 | -6.5 |
| dB eHL | 43.5 | 53.5 | 58.5 | 58.5 |
Reporting
Choose the Report Icon 
When complete, choose Save and Exit.
References
[1] Crowley, K.E. & Colrain, I.M. (2004) A review of the evidence for P2 being an independent component process: age, sleep and modality. Clin Neurophysiol.(115(4) 732-44.
[2] Näätänen, R. & Picton, T. (1987) The N1 wave of the human electric and magnetic response to sound: A review and an analysis of the component structure. Soc. For Psychophysiological Research, Inc. 24 (4) 375-425.
[3] Liverpool UK (2014). Available from www.corticalera.com.
