Subject: Masking release due to the pitch-based grouping? From: chen-gia tsai <tsai.cc(at)lycos.com> Date: Thu, 15 Jan 2004 08:47:02 -0500 Organization:Lycos Mail (http://www.mail.lycos.com:80)Dear list, Recently I read a paper about the correlation between the crest factor and masking. H. Gockel et al. (2003). Louder sounds can produce less forward masking: effects of component phase in complex tones, JASA 114:978-990. Gockel and colleagues synthesized complex tones composed of unresolved harmonics, which were added either in cosine phase (CPH), giving a large crest factor, or in random phase (RPH), giving a smaller crest factor. The results indicated that CPH tones are louder and produce less forward masking than RPH tones. Thus, it cannot be the case that loudness and forward masking are determined by the same peripheral processes, such as the level of excitation or neural activity. Gockel and colleagues suggested that the representation of these sounds in the auditory system may be affected by at least two processes: fast-acting cochlear compression and suppression. I am thinking about an alternative explanation of this effect. I synthesized several complex tones composed of unresolved harmonics. They have the same spectrum but various degrees of the crest factor due to different phase relationships of their harmonics. I find that both loudness and pitch strength decrease with the crest factor. Assuming that unresolved harmonics at n*F0 are better grouped as the pitch strength of F0 increases, CPH is the least effective forward masker because its components are best grouped. In my opinion the interesting effect reported in [Gockel et al. 2003] just highlights the importance of auditory scene analysis. Any comment will be helpful. Regards, Chen-Gia ______________________________________ Chen-Gia Tsai Ph.D Musicology, Humboldt University Berlin tsai.cc(at)lycos.com ____________________________________________________________ Get advanced SPAM filtering on Webmail or POP Mail ... Get Lycos Mail! http://login.mail.lycos.com/r/referral?aid=27005