Alan B. Bond - US grants

Animal Behavior Program

Non-technical Abstract



Title: Predator cognition and prey evolution: The virtual ecology approach

PI: Bond, Alan

Proposal #: 9974669


The evolution of color patterns in animals has been studied for over a century, but many basic questions are still poorly understood. Why do camouflaged species often occur in multiple, distinctive forms? Why are distasteful animals usually brightly colored? There is little direct evidence bearing on these issues, because it is difficult to conduct evolutionary experiments in the laboratory. The proposed research will explore the evolution of color patterns in prey species using blue jays trained to search computer displays for the presence of digital "moths." The moths exist in an artificial population maintained in computer memory, a "virtual ecology," with a genetic system derived from the genetics of real moths. They breed and evolve through successive generations based on whether jays detect them when they are displayed on screen. Changes in their color patterns over time mimic many of the evolved features of color patterns in real insects.

Predator-prey systems are important both biologically and practically. The proposed research will advance understanding of the complex dynamics involved in predator-prey interactions over evolutionary time. It will also test the validity of virtual ecology for studying evolutionary dynamics, potentially adding a powerful new technique to the scientific study of organic evolution.

Virtual Ecology: Experimental Tests of Evolution in Predator-Prey Systems

Alan B. Bond

Some of the most profound issues in biological evolution concern the selective effects of interactions between organisms, particularly the influence of predators on the color patterning of their prey. The proposed research will study the effects of predation using a 'virtual ecology', in which predators hunt moth-like virtual prey whose appearance is controlled by a computer-based genetic algorithm. When the prey are detected on the computer screen, they are considered to have been eaten. Only surviving prey can breed to produce the next generation, so the actions of the predator, over time, influence the appearance of the virtual prey dynamically, causing it to evolve into new forms. This procedure will be used to test hypotheses that relate prey appearance to patterns of predation. For example, most palatable prey are also cryptically colored, difficult to detect, and highly variable in appearance. In contrast, prey that are poisonous or distasteful tend to be conspicuously colored and to vary little in appearance. By establishing populations of virtual moths that vary in coloration and palatability, several hypotheses that have been proposed to account for differences in appearance and variability of the prey will be rigorously tested. The proposed research will advance our understanding of the complex dynamics involved in predator-prey interactions over evolutionary time. It will also establish the validity of virtual ecology for studying complex evolutionary dynamics, potentially adding powerful new techniques to the scientific study of ecological relationships among organisms.


--------------------------------------------------------
Michael D. Greenfield
Program Director, Animal Behavior
Div. Integrative Biology & Neuroscience
National Science Foundation
4201 Wilson Blvd., Rm 685
Arlington, VA 22230 U.S.A.
703-292-8421

DESCRIPTION (provided by applicant): Knowledge of the social environment, the relationships among ourselves and other individuals, is crucial for humans and other animals that live in complex, structured groups. Social cognition includes capabilities, such as tracking relationships among individuals and transitive reasoning, that are integral components of human intelligence and that are strongly affected by abnormal development or neural disease. Given their importance, the fundamental mechanisms of social cognition are surprisingly poorly understood. The research proposed here takes advantage of a robust model system, using operant procedures and carefully structured paired interactions to explore detailed cognitive mechanisms in a highly social animal. Emphasis will be placed on understanding how stimulus hierarchies are organized in the mind, how dyadic relationships are tracked, and how such mechanisms operate when applied to the task of determining relationships among members of a social group. By addressing such issues in a naturalistic context and making explicit comparisons to the cognitive capabilities of less social species, the proposed approach has the potential to yield a novel, unifying perspective on the processes underlying social cognition. These studies are apt to provide critical evidence to help us differentiate between the social complexity hypothesis and the forging hypothesis concerning the evolution of complex cognition.

DESCRIPTION (provided by applicant): The proposed research seeks to develop an operant test for implicit episodic memory, which is memory for unique, individually experienced events. A primary criterion of episodic memory is the binding together of information about What happened, Where it happened, and When it occurred (WWW memory) into an integrated memory trace. Experimental studies of episodic memory commonly use designs that depend on verbal reports of conscious memories in human subjects, and as a result, episodic memory is often considered an integral function of human consciousness. It is possible, however, that the formation of at least some episodic memories requires neither conscious recall nor verbal ability, that animals, pre-verbal children, or cognitively compromised adults may be able to make use of implicit memories of specific past occurrences that fit the primary diagnostic features of episodic memory. This possibility has important theoretical and practical implications. If subjects with cognitive deficits retain some unconscious memories of past events, this could modify our view of their cognitive capabilities of these populations and potentially lead to new forms of assessment in normal and at-risk populations. A determined search for criteria that do not depend upon verbal report is needed. One potentially valuable approach to this problem would be to develop such tests with nonhuman animal models. We propose to explore a range of novel experimental procedures employing operant techniques with California scrub jays, which have demonstrated impressive 'episodic-like'memory in a naturalistic context. The proposed experiments will test for integrated event memory, using designs that are more amenable to precise experimental control. PUBLIC HEALTH RELEVANCE: If any of these studies are successful in demonstrating episodic memory in scrub jays, the methodology can be readily applied to human subjects, providing significant benefits for the diagnosis and treatment of neuropsychiatric disorders, such as autism and schizophrenia. Development of a useable animal model of episodic memory will also permit further investigations of the underlying neurobiology, using pharmacological or neuroanatomical techniques that cannot be employed in human studies.