136 • Alternatives to Animal Use in Research, Testing, and Education 
that some plants, under attack by insects and 
micro-organisms, develop highly sophisticated 
defenses involving the emission of antibiotic -like 
substances and chemicals that inhibit insect diges- 
tive enzymes. Indeed, some plants can apparently 
communicate chemically with as -yet -unaffected 
neighboring plants to induce leaf-chemistry 
changes in advance of infestation (reviewed in 
ref. 164). 
USE OF NONLIVING SYSTEMS 
Inanimate chemical or physical systems are un- 
likely to prove useful in behavioral research, for 
reasons intrinsic to the nature of behavior. A dy- 
namic, emergent process, behavior functions to 
allow organisms to adapt to moment -to-moment 
changes in the environment. In a sense, all behavior 
ultimately functions to aid and abet survival and 
reproduction (51). Adaptation, survival, and repro- 
duction are not properties of nonliving systems. 
And behavior involves information processing and 
a continuous series of choices among an array of 
alternatives (140). Although chemical or physical 
systems may change in response to certain envi- 
ronmental stimuli, the nature of such changes does 
not involve decisionmaking or information -proc- 
essing. 
Behavior is a byproduct of interactions between 
sensory, neural, hormonal, genetic, and experien- 
These impressive features of the botanical world 
notwithstanding, it is unlikely that plants will make 
an important contribution to behavioral research. 
The lack of a central nervous system, and in par- 
ticular a brain, renders the plant an inappropri- 
ate model for use among the disciplines of be- 
havioral research. 
IN BEHAVIORAL RESEARCH 
tial factors. As such, it is influenced by the situa- 
tion at hand, the developmental history of the 
organism, and prior experience with similar and 
related situations . It appears inappropriate to im- 
bue inanimate chemical or physical systems with 
the capacity for experience. Devoid of such a ca- 
pacity for experience, nonliving systems are un- 
likely alternatives to using animals in behavioral 
research. 
Examples of the application of chemical or phys- 
ical systems to behavioral research are sparse. One 
involves the use of chemical reagents to mimic the 
properties of rhythmic behavioral phenomena in 
animals. Certain chemical reagents exhibit changes 
in state that oscillate periodically in a fashion sim- 
ilar to some biologically based rhythms . However, 
the chemical reactions themselves remain poorly 
understood (222). 
COMPUTER SIMULATION IN BEHAVIORAL RESEARCH 
A computer simulation is an operating model that 
depicts not only the state of a behavioral system 
at a particular point in time but also changes that 
occur in that system over time. Because dynamic 
processes are of quintessential importance in be- 
havioral research, computer simulation stands as 
a potentially useful tool for the behavioral scientist. 
In order to simulate a living system, a computer 
programmer must have information about that 
system. The more information at hand, the better 
the simulation (53). In the strictest sense, a com- 
pletely accurate simulation presupposes that every- 
thing that there is to know about the system in 
question is known. To construct a computer simu- 
lation that would fully replace the use of a live 
organism in behavioral research would require 
knowing everything about the behavior in ques- 
tion, which in turn would preclude the need for 
a computer simulation for research purposes. 
Yet, if computer simulation cannot fully replace 
living organisms, it can and does contribute to be- 
havioral research. Although the fundamental be- 
havioral qualities of adaptation, survival, and re- 
production do not pertain to computer programs, 
computer software does, for example, embody in- 
formation processing and decisionmaking. Exam- 
