They come into a world where they must struggle to survive. Over many generations, they evolve. But are they alive? Of course, one might say. But the discussion is not about amoebae, ants or alligators. Rather, it’s about computer programs. The occurrence of artificial life’ exists only within PCs and more powerful computers, but its existence in the electronic universe parallels many elements of life in the biological world. Some programs flock like birds. Others organise like bees. Some mutate swiftly from chaotic hordes to complex, stable populations in a process similar to Darwinian evolution.
As a group, artificial life programs represent the most exciting work on the edge of computer research. Study of artificial life holds promise for new ways of solving complex problems and fresh opportunities to model biology and society. Perhaps, far in the future, such research will yield the ability to blueprint living organisms. The basics behind artificial life are surprisingly simple. The programs follow a few simple rules, applying them with a speed and persistence that’s possible only inside a computer. When many such programs are run simultaneously, amazingly complex patterns can emerge. In many cases, these patterns seem like strange replications of natural behaviours.
Programs work together against common enemies and devise new ways of surviving when their environment changes. These results aren’t surprising when you consider that biological life itself consists of nothing more than variations on four simple bits of code: the four compounds that constitute DNA, the building block of all genes and therefore all life. In artificial life, computer instructions take the place of DNA code. The father of modern artificial life, research, Christopher Langton of the Los Alamos National Laboratory in the U.S., sees his work this way: ‘For us, artificial life is the study of man-made systems that exhibit behaviours characteristic of natural living systems. We’re attempting to abstract logical forms of life, not matter. We can obtain some of the same dynamics as life, albeit with different materials.’
Langton took up the study of self-replicating programs begun earlier by John von Neumann, a Hungarian mathematician whose theories contributed to the development of the programmable digital computer. An example of how programs mimic biology can be found in cellular automata (cell-like machines): structures that arise from tiny programs that each display a seemingly independent existence based on a few simple rules. Analogies between programs like this and actual life forms are inevitable. When simulated organisms cluster together, leaving rectilinear tracks on the screen, researchers call them ants’. When they do this in a three-dimensional model, they’re called ‘bees’. And perhaps the most disturbing analogy with biological life can be found in computer viruses’, self-replicating programs that display purposeful behaviour and tolerate any small physical changes in their environment.
Although some scientists regard viruses as the first programs capable of existing without the wilful cooperation of humans, the fact is that without humans to design them, they wouldn’t exist at all. Still, some of the work demonstrated at a recent gathering of the artificial life research group evoked confused excitement. ‘During five intense days’, said Langton, ‘we saw a wide variety of models of living systems, including mathematical models for the origins of life, self-reproducing automata, computer programs using the mechanisms of Darwinian evolution to produce co-adapted ecosystems, simulations of flocking birds and schooling fish, the growth and development of artificial plants, and much, much more.’
Craig Reynolds of Symbolics demonstrated his ‘boids’, computer-animated, bird-shaped creatures that flock like real birds. Reynolds programmed the ‘boids’ to follow three simple rules: they maintain a minimum distance from the nearest object; they match velocity with the nearby flock; and they fly toward the greatest concentration of the flock. The resulting flocking behaviour is shockingly real. ‘Ants’, the creation of David Jefferson and Robert Collins, also appeared. Colonies of these randomly generated creatures have developed the ability to navigate electronic mazes and search for symbols that represent food.
Independent programmer John Nagle argued that the next generation of supercomputers should challenge researchers to create ‘squirrels’, computer models with the intelligence level of a biological rodent with one gram of brain mass. Langton’s contribution, ‘Computation at the edge of chaos’, was one of the most unusual presentations. Biologists maintain that life began in a spontaneous outburst of activity that occurred when Earth’s environment reached critical thresholds of heat, atmosphere and chemical composition. A few variations on any of these variables would have altered the course of the planet into either chaos or barrenness.
Langton’s presentation was based on a computer model demonstrating similar principles. Changing a parameter in the model acts like changing the temperature of a computer-generated petri dish of single-cell creatures. When this variable passes a critical threshold, the colonies of Langton’s artificial life programs neither freeze nor evaporate but settle into recurring patterns conducive to the orderly transmission of information. ‘At one end, activity freezes; at the other end, it’s too volatile,’ notes Langton. As a result, he wonders whether ‘computation may emerge spontaneously and come to dominate the dynamics of physical systems’ much as life has. In fact, Langton speculates that life itself may have started as a chance computation on the cusp of liquid and gaseous states.
Questions 27-33
Complete the summary below. Choose NO MORE THAN THREE WORDS from the passage for each answer.
Research has shown there are possible similarities between the ‘real’ or (27)……………….and computer programs. In studying so-called (28)……………………. scientists have found that images from computer-generated graphics seem to imitate the behaviour of insects and birds, while others mutate from the disorderly into highly organised populations much like in (29)…………………Artificial life programs may allow researchers to re-evaluate models of biology and society, and may even provide the possibility to (30)…………………….living organisms. Patterns which can emerge from such programs replicate (31)…………………. Being able to adapt to environmental changes, these programs mimic (32)…………………….which, essentially, is built upon four basic codes of information. The only difference, however, is that their DNA code is replaced by (33)……………….
Questions 34-37
Look at the following artificial life programs (Questions 34 37) and the list of descriptions below. Match each artificial life program with the appropriate: description, A-H.
34. Bees
35. Boids
36. Squirrels
37. Ants
List of Descriptions
A can match the intelligence of a bird
B can group together in a rectangular form
C can cluster together moving in straight lines in 3D
D will have a gram of intelligence
E can keep a minimum distance from another object
F are able to group and form a track in 3D
G will be as intelligent as a rodent…..
H are capable of locating symbols depicting food
Questions 38-40
Choose the appropriate letter, A, B or C.
38. Researchers studying artificial life are trying to
A find how life forms impact on computer programs.
B study behavioural characteristics exhibited by man.
C identify logical forms which correspond to living matter.
39. Computer virus programs
A can change behaviour to suit their environment,
B resemble biological life closely but need human input,
C can create biological life independently of humans,
40. Researchers argue that life itself started
A on the basis of chemical variations.
B by chance, spontaneously in a particular environment.
C by a spontaneous computation of physical activity.