HRTFs are usually considered as a set of linear time-invariant systems, the "TF" in HRTF means "transfer function". With that definition, HRTFs change the amplitude and phase of the existing frequencies only and they do not introduce new frequency components.
Of course, as everywhere in the acoustics, there could be some nonlinear or modulation effects caused by the pinna (or head and shoulders), however, for sound localization, there is no evidence for such effects until now. All published sound-localization studies I know use HRTFs as they are - without any additional nonlinearities or modulations.
Hope that helps,
Psychoacoustics and Experimental Audiology
Acoustics Research Institute
Austrian Academy of Sciences
Wohllebengasse 12-14, 1040 Vienna, Austria
Tel.: +43 1 51581-2511
Fax: +43 1 51581-2530
Ranjit Randhawa wrote: