Subject: Re: Streams From: Bruno Repp <repp(at)ALVIN.HASKINS.YALE.EDU> Date: Mon, 21 May 2001 11:15:56 -0700Dear List, I was correct in my prediction that Al Bregman would provide cogent comments after my rash reply to an earlier message to this list. Still, it seems to me that the concept of a single stream, as discussed by Al, does not have much utility or explanatory value. It does so only when it is contrasted with at least the possibility of multiple streams. In other words, a single stream is a default condition of little theoretical interest. This is equally true for a single group. Groups become interesting only when there are (or could have been) two or more of them. Then there is the question of whether much is to be gained by talking about two streams occurring in succession. This amounts to saying that two successive acoustic events are perceived as coming from different sources or locations. There is nothing wrong with calling these events different streams, but neither does it add anything significant to saying that the perceived source or location has changed. Moreover, the term stream seems increasingly less appropriate as the temporal interval separating successive sound events increases. I doubt that events separated by 1 day, 1 hour, or even 1 minute should be considered as belonging to the same stream, even if they are perceived as coming from the same source. It seems to me that a stream, to be properly so called, should be temporally coherent, which may mean that not more than a few seconds should intervene between its elements. Consequently, the idea of hearing two different streams may become pertinent only when the streams are separated by less than a few seconds. Without any doubt, however, most or all research on stream segregation has been conducted with temporally overlapping series of (usually rapidly paced) sounds. This is why I thought that temporal overlap (including perhaps very brief separation) is an essential part of the definition of streams. Al's definition clearly is broader, but this is more of a terminological issue. Streams really come into their own only when they occur at more or less the same time. With regard to the role of attention in stream formation, this is an interesting empirical issue that certainly needs more research. --Bruno Bruno H. Repp Research Scientist Haskins Laboratories 270 Crown Street New Haven, CT 06511-6695 Tel. (203) 865-6163, ext. 236 FAX (203) 865-8963 e-mail: repp(at)haskins.yale.edu