I consider myself an expert in Gibson's work, but I must admit that i
cannot understand the debate so far. There is an unfortunate confusion
between the works of Gibson and that of the Gestalt Psychologists. And the
responses fail to live up to the spirit of Gibson's holistic approach to
JJ Gibson and i were friends. We used to disagree all the time and argue
about it, interactions that we both enjoyed and looked forward to, ideally
over a glass of wine. (Glass? a bottle or two.)
He started life as a Gestalt psychologist but then denounced it. He
believed that you could not separate perception into domains such as
vision, audition, haptics, spatial orientation, ... They were all
intertwined. And you couldn't separate perception from action and from the
environment. And in turn, you could not separate it from people's goals.
(We agreed on all that -- our disagreements were based on the fact that I
wanted to understand the underlying brain mechanisms whereas he argued for
pure "information pickup," without any internal processing. I still don't
understand what he meant by that - and i suspect he didn't either. He
exaggerated in order to make people attend to his main points.)
We disagreed, yet when I wrote Psychology of Everyday Things (now called
"The Design of ... "), I realized that his concept of Affordance could be
powerfully applied to design.
(Gibson would probably have gotten furious for my re-interpretation of his
ideas, but that's OK. Some of my best work has come about by my
misunderstanding of the work of others. I don't care if good ideas come
about because of my misunderstandings -- as long as they are truly good
He was brilliant.
Richard started the debate by asking about Gibson.
I am currently looking into Gibson´s theory of perception after having
looked into Gestalt theory. I can see lots of examples of how Gestalt
theory can be applied to design. For Gibson the matter is less clear if one
is concerned not with physical affordances (handles, buttons, levers). The
optical examples usually listed are the converging ladder, the compressed
lines and the way proximity to the horizon affects how we perceive size.
Somehow or other it got muddled by Gestalt Theory, which as i said, Gibson
thought wrong. I agree. The Gestaltists were excellent at observation,
but truly bad at theory. The quotation is also a misreading of Gibson. He
was trying to show how the rich, interactive world of 3D moving perception,
provided powerful information to the observer. Those examples are more
like those of the gestalt theorists than of Gibson.
Gibson would demonstrate how one landed an airplane by guidance from the
optical flow of the moving convergences brought by perspective.
Example: If an object is coming at you, will it hit you? You don't need to
do a computation: just look at the symmetry. If it grows larger completely
symmetrically, it will hit you.
Richard further compounds the confusion by asking
Can anyone point to a paper or article etc where the design consequence for
3D objects are outlined? Why does an industrial designer need to worry
about converging ladders etc in the way the Gestalt Laws such as
Pragnanz and proximity matter?
These have nothing to do with Gibson: these come from the Gestaltists. I
assure you that Gibson didn't differentiate 3D from 2D. Actually, he didn't
believe in 2D -- 2D is a drawing. The real world is 3D -- 4D if you count
time, because most interesting phenomena vary with time.
Gibson's work showed how you could land an airplane by the changes in the
visual field, or catch a ball with simple cues: A ball thrown or hit high
in the sky can be caught simply by moving so that the visual angle between
you and the ball remains constant: Move toward the ball if the angle gets
lower -- move away of t gets higher. Similarly for left and right.
Gestalt laws of perception are very important to interaction design.
Industrial designers tend to be awkward at Interaction Design (I give you
Jonny Ive), so this may not matter to them.
But these laws were of little interest to Gibson. he wanted to understand
how we navigated through the world, accomplishing our goals.
Gestalt laws speak to how we divide up the visual world and categorize
things. These are critically important to interaction.
My favorite book was written by my student, Stephen Palmer. This is the
very best book on perception: it is a landmark. It covers Gestalt as well
as modern information processing approaches.
*Vision Science: Photons to Phenomenology*
Here are some of his papers. (You will need to get them via your library
-- otherwise they are expensive)
*Visual Perception of Objects*
*Modern Theories of Gestalt Perception*
PhD-Design mailing list <[log in to unmask]>
Discussion of PhD studies and related research in Design
Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design