I’m struggling to understand how one could talk about the maximum temporal window independently of discussion of the dimensions of
‘interrelatedness’. Your example of a sequence of different pitches (though you don’t mention timbre/harmonic content, or envelope structure) presumes differences in the frequency domain, but similarity in other respects, is that right?
So your findings might, for argument’s sake, vary according to the degree of interrelatedness (or at least, the degree of ‘similarity’
along several dimensions).
Similarly, if similarity were very high (i.e. identical – a loop) – the maximum temporal window might actually be the maximum possible
under any circumstances, speculatively.
In that context (‘looping’), my students, in an exercise to synthesise 10 minute spatial soundscapes (in ambisonics) often try to get
away with looping some ambient background material of, say 2 mins. After repeated listening of the whole piece, by the third time, the loops are very noticeable, to me (and much to the students’ surprise). But that’s because particular features are identical,
the listener is accustomed to hearing loops (even expecting them) and probably various other factors.
Whether this sheds any light on interrelated-but-not-identical, I’ve no idea. But it’s not very compatible with the ‘tape recorder’
metaphor, and is more akin to an example of Cocktail Party Effect.
Dr. Peter Lennox
Senior Lecturer in Perception
College of Arts
University of Derby, UK
t: 01332 593155
From: AUDITORY - Research in Auditory Perception [mailto:AUDITORY@xxxxxxxxxxxxxxx]
On Behalf Of Dario Sanfilippo
Sent: 01 February 2017 14:40
Subject: Maximum temporal window for pattern recognition and the perception of redundancy
I would like to ask you a couple of questions and I will be very grateful if you could help me. It would be great to be pointed out to specific publications, thank you so much
in advance for that.
The first question is on the maximum temporal window for the recognition of patterns in long-term audio events. Generally speaking, what is the largest (temporal) distance between
audio events so that we can still process such events as interrelated? As an example, let's assume to have a sequence of different pitches equally spaced in time; what is the largest possible distance between them to still be able to perceive a melody?
The second question is on the perception of redundancy in relatively complex (i.e., dynamical equilibrium) long-term audio events. As an example, think of the sound of the sea:
its internal structure is never the same although it keeps a strong identity from a global point of view. I believe that this is highly dependent on both the cultural background of the listener as well as the degree of complexity of the audio event itself,
but is there any study which tries to relate the perception of redundancy (i.e., the moment in which the listener's attention drops) with a specific temporal window? And is there a connection or similarity between this window and the one described in my first
Thank you and best wishes.
The University of Derby has a published policy regarding email and reserves the right to monitor email traffic.
If you believe this was sent to you in error, please reply to the sender and let them know.