Troy, N.Y. – Since the idea of using DNA to create faster, smaller,
and more powerful computers originated in 1994, scientists have been
scrambling to develop successful ways to use genetic code for
computation. Now, new research from a professor at Rensselaer
Polytechnic Institute suggests that if we want to carry out artificial
computations, all we have to do is literally look around.
Assistant
Professor of Cognitive Science Mark Changizi has begun to develop a
technique to turn our eyes and visual system into a programmable
computer. His findings are reported in the latest issue of the journal Perception.
Harnessing
the computing power of our visual system, according to Changizi,
requires visually representing a computer program in such a way that
when an individual views the representation, the visual system
naturally carries out the computation and generates a perception.
Ideally, we would be able to glance at a complex visual stimulus
(the software program), and our visual system (the hardware) would
automatically and effortlessly generate a perception, which would
inform us of the output of the computation, Changizi said.
Changizi
has begun successfully applying his approach by developing visual
representations of digital circuits. A large and important class of
computations used in calculators, computers, phones, and most of
today’s electronic products, digital circuits are constructed from
assemblies of logic gates, and always have an output value of zero or
one.
"A digital circuit needs wire in order to transmit
signals to different parts of the circuit. The ‘wire’ in a visual
representation of a digital circuit is part of the drawing itself,
which can be perceived only in two ways," said Changizi, who created
visual stimuli to elicit perceptions of an object tilted toward (an
output of one) or away (an output of zero) from the viewer. "An input
to a digital circuit is a zero or one. Similarly, an input to a visual
version of the circuit is an unambiguous cue to the tilt at that part
of the circuit."
Changizi used simple drawings of
unambiguous boxes as inputs for his visually represented digital
circuits. The positioning and shading of each box indicates which
direction the image is tilted.

He also created visual
representations of the logic gates NOT, which flips a circuit’s state
from 0 to 1 or vice versa; OR, which outputs 1 if one or both inputs
are 1; and AND, which outputs 1 only if both inputs are 1.
"Visually
represented NOT gates flip a box’s perceived tilt as you work through a
circuit, and OR gates are designed with transparency cues so that the
elicited perception is always that the box is tilted toward you, unless
overridden," Changizi said. "The AND gate is similarly designed with
transparency cues, but contrary to the OR gate, it will always favor
the perception that it is tilted away from you."
By
perceptually walking through Changizi’s visual representation of a
digital circuit, from the inputs downward to the output, our visual
system will naturally carry out the computation so that the "output" of
the circuit is the way we perceive the final box to tilt, and thus a
one or zero.
"Not only may our visual system one day give DNA
computation a run for its money, but visual circuits have many
potential advantages for teaching logic," Changizi said. "People are
notoriously poor logical reasoners — someday visual circuits may enable
logic-poor individuals to ’see their way’ through complex logical
formulae."
Although Changizi’s visual stimuli are successful
at eliciting viewer perception, he says there are still serious
difficulties to overcome. The visual logic gates do not always transmit
the appropriate perception at the output, and it can be difficult to
perceive one’s way through these visual circuits, although Changizi
argues we may have to train our visual system to work through them,
similar to the way we need to be taught to read.
Additionally, building larger circuits will require smaller or more specialized visual circuit components.
"My
hope is that other perception and illusion experts will think of novel
visual components which serve to mimic some digital circuit component,
thereby enriching the powers of visual circuits," Changizi said.
Recent Comments