Residual Noise is the averaged background noise. It is calculated online during the ABR recording by measuring the stability of the averaged waveform. The Residual Noise calculation used for Fmp is the same used for the calculation of Residual Noise - typically in a time window of 10 ms. The greater the stability, the less noise in the tracing. When the Residual Noise level reaches the set criteria, e.g. 40 nV for adults and 20 nV for children, the residual noise bar turns green with a checkmark.
Fmp is a statistical online analysis of the ABR recording from beginning to end. The Response Confidence is a statistical measure of a true response being present. Work by Guy Lightfoot, UK, using the Eclipse 4.6.1 has found that on infants and newborns undergoing follow up, an Fmp of 2.2 corresponds to a true response detection of 97.5%. Based on this, the UK guideline recommends using an Fmp of 2.25, which the Eclipse has always stated as being a confidence level of 97.5%.
It works on the principle of comparing response amplitude to residual noise to provide confidence level or detection rate. The underlying analysis considers the ABR recording typically in a time window of 10 ms. The Fmp ratio between the response amplitude and the residual noise is calculated. In a response situation, lower noise or larger response amplitude will drive the Fmp up (indicated by the red line and bar). In a no-response situation, the response amplitude will not rise above the residual noise, thus the Fmp value and the Response Confidence will remain low.
In summary, Residual Noise and Fmp:
Don, M. & Elberling, C. (1996). Use of quantitative measures of auditory brain-stem response peak amplitude and residual background noise in the decision to stop averaging. J. Acoust. Soc. Am., 99(1).
Elberling, C. & Don, M. (1984). Quality Estimation of averaged auditory brainstem responses. Scand Audiol., (13) 187-197