[-empyre-] Neuroaesthetics _ad
ben at ekran.org
Tue Sep 9 04:44:10 EST 2008
Thanks for your responses Alan. I have a few comments in-line...
ALAN DUNNING wrote:
> Hmmm, lots of questions and interesting issues here.
> The "real" world and the world of scientific truth…__
> I think that the "real" world (though that is not a term we would use
> casually) and the world of scientific truth are included in what I have
> called the virtual, symbolic and imaginary worlds. Reality in our view
> is a state that is never outside ourselves, but part of a body/world
> continuum lodged in cognitive acts. As a construction (biological,
> neurological - add your own descriptors here) reality is always
> contingent. Science has always seemed quintessentially metaphoric, full
> of images of invisible states and imagined processes as we apprehend it
> through embodied experience. Science can be seen as a kind of fiction
> that has (unusually) traction across a large number of bodies and
> worlds. Like other kinds of fiction the scientific fiction has agency.
I agree. I hope to hear other perspectives from the other guests,
especially those of the hard science ilk (if they consider themselves
> Noise and Perception…
> Kate Hayles has done so much great work in thinking about this when she
> talks about the shift toward pattern/randomness and away from
> presence/absence, in the dematerialization of the body.
> This suggests a new category of subjectivity, an embodied hybrid of
> human and machine, that privileges pattern, and its opposite noise, over
> the material and moves between the material and the immaterial. A new
> body that acknowledges the absent. This body is biologically developed
> to see pattern and grant it significance in its dynamic mental model of
> its world. Noise and signal are for me interchangeable ( a sort of
> autistic conflation of mixed messages and hyper-intensities) – its just
> depends to which you attend. Consciousness perhaps becomes a kind of
> temporal band-pass filter.
Yes, always a matter of attention or perspective, or desire (see what
you want to).
> GITM doesn't really use the machine (if you mean a computer) to make
> meaning. Our face tracker only looks for pattern, (a very simple and
> prescribed one) not meaning. Meaning is entirely dependent on the
> interpretative dance between data and receiver. What it does is filter
> the incoming data. The hybrid body/machine is the meaning maker. In my
> earlier post I said that that he images are loaded with meaning, but
> this is true only after perception. Neither the machine nor our intent
> generates any implicit meaning. Perhaps, it would be better to say the
> images are vessels/mirrors for meaning.
Could you elaborate on the "interpretative dance between data and
receiver"? This statement slightly implies that the interpretation
effects the data, rather than the sensor data initiating a pattern of
activation that sets off a dance of interpretation between elements of
previous (learned/remembered) experience?
Are not all images (or perhaps even perceptions) "vessels/mirrors for
I've been playing with a Piagetian position where life starts with
unintegrated (related to one and other) perceptions. I'm considering
meaning as what happens to link the current stimulus with the whole of
previous experience (as encoded in the neural connections). In this
sense meaning is a placement of what is seen with what has been seen.
Its clear that GITM does not have such a mechanism, but under this
proposal it is certainly possible that a machine could appreciate
meaning (albeit an infinitesimally simple way) as it own linking of
current to past experience.
I had part of this discussion with Liane Gabora who gave me this example
to counter the idea that meaning could result simply from a comparison
of sensory experiences: In a free-association a knife and a gun could
lead to one and other (and are therefore linked), even though they have
little in common from a perception point of view.
Now this is where the subtopic of emotion comes in. Perhaps the
experiential link between a knife and a gun is the stress resulting in
both. A child sees both a knife and gun as objects that cause pain.
Perhaps it is that implied (empathic) pain that causes the link between
those two objects.
To clearly return to the theme. What is neuroaesthetics but an enquiry
into the pattern of activations resulting from sensory experiences? I
think such a discussion goes to the heart of what meaning is, both for
machines modelled on our brains, and on our understanding of what our
brains are and do.
> Maria’s points are things we have considered – and maybe some will get
> implemented. But we have to date resisted these. The work demands close
> ties to EVP techniques and strategies – because they are themselves
> exactly kinds of pre-programmed events, full of bias and distinct
> deficiencies in learning. I am not so interested in generating meaning,
> or art for that matter. I don’t mean that as a provocation (though it
> probably is), but as a way for us to think differently. What interests
> us is that the desiring body is a body that irresistibly generates
> meaning. The art, if any is generated at all, is generated in the
> perceiving, and in doing so becomes the frame for the next ‘meaningful’
> event. I can imagine many versions without the machine at all (Elvis on
> the taco, the misheard whisper, nameless dread, et al) – I think of it
> only as a contextualizing sign that says, ‘let’s attend to these things,
> this way, for a bit.’
Being an artist working in a open-source and free-software context, I
would be interested in anyone developing computer models of perception
that can be used by artists exploring these issues. I don't mean current
performance based computer vision methods, but models based on current
> empyre forum
> empyre at lists.cofa.unsw.edu.au
More information about the empyre