[The validity of absolute "standard values" for the evaluation of the thermic vestibular test--results from electronystagmographic investigations with critical comment on the frequency nystagmogram (butterfly scheme) (author's transl)].

The findings presented here on the thermic examination of 100 healthy test persons and the statistical analysis lead up to the conclusion that the single standard deviation from the statistical mean value of a parameter (Claussen) describing the intensity of a thermic nystagmus reaction does not seem suitable to be assumed as normal range. This is true, too, if these values were achieved from a larger number of healthy persons. On application of the "butterfly scheme" (Claussen) a non-tolerable number of false positive findings resulted. The cause for this is the fact that extravestibular factors, particularly the degree of consciousness, but also the age, influence considerably the absolute value of nystagmus parameters. Furthermore, since the numerical values are not distributed normally, the statement of a standard deviation is not applicable. As an alternative, a procedure is being demonstrated which according to our knowledge was first mentioned by Jongkees. We are suggesting to base the evaluation of the thermic test on the comparison with so-called interquantile ranges of healthy persons. These are percentage ranges within the values of e.g. 80, 90 and 94% of healthy persons are found. In our opinion, in combination with the Jongkees-formula this procedure represents a more favourable criterion for the evaluation of a thermic test.