Showing papers in "Systems Research and Behavioral Science in 1985"

An investigation into the relative manipulability of four voting systems

Abstract: This paper contains a study about decision making in groups. The author analyzes the manipulability of four voting systems: Borda, Coombs, Hare, and plurality. The Hare voting system was found to be the least manipulable; the Borda, the most manipulable. The results of this study indicate that the logical possibility of manipulation threatens the legitimacy of voting outcomes.

The future of general systems research: Obstacles, potentials, case studies†

Abstract: This paper attempts to provide an evaluative and prescriptive overview of the young field of systems science as exemplified by one of its ‘specialties’ general systems theory (GST). Subjective observation and some data on seven vital signs are presented to measure the progress of the field over the last two decades. Thirty-three specific obstacles inhibiting current research in systems science are presented. Suggestions for overcoming these obstacles are cited as a prescription for improved progress in the field. A sampling of some of the potential near-term developments that may be expected in the three rather distinct areas of research on systems isomorphies, improvement of systems methodologies, and the utility of systems applications are illustrated with mini-case studies. Throughout, there is an attempt to identify ‘key’ questions and practical mechanisms that might serve as a stimulus for research. Finally, a set of criteria defining a general theory of systems is suggested and illustrated with a case study. The paper concludes with a projection of the long-term contributions that systems science may make toward a resolution of the growing chasm between high-tech solutions and high-value needs in human systems.

A model of the eustress system for health/illness.

Abstract: The stress response system, as vital mediator of the individual's health/illness dynamics at physical, psychological, and social levels, is modeled through a systems approach. This extremely complex self-regulating system involves multiple causal factors, nonlinearities, and time delays. Key basic functions are surveillance, mobilization of resources, feedback, and purpose. The underlying physiological stress response comprises three main mechanisms: autonomic nervous system, hormonal system, and immune response system. Their afferent pathways are presented in a flow diagram, and integrated into the full system which includes the psycho-social dimensions. The main features of this overall system comprise: psycho-social dynamics, feedback and feedforward monitors, self-concept, evaluation and strategy-decision making, generalized coping resources, mobilization of response strategies, and behavioral control. Stressors, as potential producers of stress, are illustrated in the physical, psychological, and social domains. A U-shaped universal dose-response curve is helpful in understanding stressor-destressor actions. Destressors are similarly considered, with special relation to lifestyle. Finally, the concept of eustress is developed as the ideal condition toward which this complex homeostatic system works.

Euler buckling in a wheelbarrow obstacle course: A catastrophe with complex lag

Abstract: This article studies human physical work capacity under increasing load from a general systems theory perspective. There are several points of generalization between the structure of nonliving (building materials) and living systems (humans) with respect to stress, strain, and fatigue. A catastrophe model for Euler buckling was transposed and tested for human performance in a wheelbarrow obstacle course under varying loads. Subjects were 129 employees of a Midwest manufacturing plant. A cusp model was hypothesized and verified (R2 = .68, control R2 =xs .11) where vertical load was the asymmetry factor, and body balance, height, and sex-related differences all contributed to bifurcation A catastrophe model in codimension 10 was also invoked to explain memory in the system. Principal control variables were exercise habits, weight, balance, and sex-related differences (R2 = .75). The core model of human load-to-failure was concluded to be similar to that for Euler buckling: additional complexities were discovered which were attributed in part to systemic memory. Discussion points included the use of large dimension catastrophe models for problems involving complex lag effects, and the transposability of the model to the organizational systemic level.

The application of living systems theory to 41 u.s. army battalions

Abstract: This issue contains an executive summary of approximately three years of research conducted on 41 U.S. Army battalions. Included in the text is a discussion of general systems science and living systems theory. The scope/goals, methodology, findings, and summary of the two distinct phases of the research (Phase I and Phase II) constitute most of this volume. The Norms Study, conducted to obtain information from the Phase II research data to create normative tables for potential pathology determination, is also included. A brief description of the conclusions, implications, and future applications of the research studies is the closing section in this issue.

A systems model of health behavior change.

Abstract: The modeling in this paper is at the level of the individual, in relation to the subsystem of health behavior. It integrates social, environmental, psychological, and physiological factors to describe the health behavior change process. It is not intended to be quantitatively predictive of behavior change, since further development of this model is needed to determine the mathematical form of the relationships between variables The model is innovative in several ways. First, it includes health-related variables and a rich feedback structure. In contrast, attitude behavior change models often cannot explain health behavior adequately, because they do not incorporate physiological variables. Second, the model can be easily used by educators to help identify important variables for developing health education curricula. Consequently, program developers are helped in determining program intents, teaching strategies, and learning activities, and evaluative strategies which are congruent with the attitude behavior change process.

Living systems theory: A framework for management

Abstract: This article deals with living systems of all types at the level of the organization. It outlines a seven-step procedure for applying living systems theory (LST) to achieve effective management in a business enterprise. The invariant nature of LST makes it applicable to any type of organization and enhances its utility as a general theory of business. Eliminated is the need to develop separate theoretical frameworks which are dependent on organization type. The seven steps focus on identifying process structures and critical matter/energy and information flows necessary for an organization to achieve its goals.

Color matching throughout the work week: an industrial application of the swallowtail-difference equation.

Abstract: This article considers a multilevel system problem involving color perception of individual human beings, man-machine interactions, and changing social factors. Previous research on the same system found critical differences in color matching performance over a 24-hour period, modeled by a cusp catastrophe. The hypothesis now tested is whether critical differences in group performance occur in a regular weekly cycle. Subjects were 13 color matchers and 30 printers who worked rotating shifts. Data were drawn from production records for 27 pairs of multicolor jobs performed over three four-week periods. The swallowtail hypothesis was upheld for three criteria: color matching time (R2 = .55), printing press time (R2 = .55), and printing paper conserved or wasted (R2 = .83). Job length (difficulty), time period (denoting organizational dynamics taking place), and whether the job pair occurred in the Monday-Wednesday, or Thursday-Saturday part of the week (semicycle) served as control variables. Language: en

Mating kinetics in drosophila

Abstract: The kinetics of the mating behavior of a sepia-eyed strain of Drosophila melanogaster has been studied using a newly designed mating chamber. Numbers of males and females placed in the chamber for 30 minutes varied from 5 males and 10 females to 320 flies of each sex. For a given number of females the log proportion of non-mated (i.e., virgin) females decreased linearly as the number of males increased; this relationship can be expressed as (1 - v)n where v is the attrition rate of virginity and n is the number of males in the chamber. Further, the log v decreases linearly with the square-root of the number of females in the chamber. If males were to divide their courtship behaviors among two or more females under crowded conditions, v should be a function of (1/x)8 where x is the number of females available to the courting male and s is one less than the number of elements in his courtship behavior. The number of available females, x, is a function of linear distances between individuals; the number of flies in the chamber must be quadrupled in order to halve the average distance between individuals—hence, the need for the square-root scale.

Systems as clusters

Abstract: General systems theory (GST) is a relatively young interdisciplinary approach which studies interrelationships among relatively homogeneous groups of components. Numerical taxonomy (NT) is very smilar in these respects. Ironically, discourse between these complementary approaches is virtually nonexistent. The purpose of this article is to help rectify this lack of intercourse by discussing continuities and discontinuities between the two approaches. The major thrust of this article is to examine the contribution that NT can make to GST. To this end we discuss five distinctions that are emphasized by numerical taxonomists and which can also be applied to systems. These are: (1) Q- versus R-analysis; (2) form of linkage among components; (3) strength of intrasystem relationships; (4) monothetic versus polythetic groups; and (5) agglomerative versus divisive methods of group formation. All of these have been emphasized more in NT than GST, and practitioners of GST could profit by becoming familiar with them. On the other hand, GST is ahead of NT in emphasizing boundary analysis and space-time analysis, among others. In addition to these discontinuities, continuities exist in the form of common concerns emphasized by both fields. These include synchronic versus diachronic analysis, naturally occurring versus artificially constructed groups, and levels or hierarchies. These two latter factors are very important for both GST and NT. Living systems are clearly natural (as opposed to artificial) systems, and are studied at various levels (e.g., organism, group, organization, society) in GST. The concepts of natural group and group hierarchies are also stressed in NT. All of the distinctions discussed can be applied to the total system at all levels. The contributions that GST can make to NT remain to be discussed in another article.

The metatheoretical bases of social psychological theory

Abstract: The present article deals with theory concerning interpersonal interactions of systems at the levels of individual human organisms, groups, and organizations It involves all subsystems of these systems It examines a number of metatheoretical constraints that may limit the development of social psychology as an empirical science It begins by describing the sentences of one of the more classical languages used to define the structure of scientific theory, as well as criteria that are employed to evaluate these sentences It then reviews and evaluates a number of epistemological constraints that limit the development and formulation of theories For example, it considers the issue of complexity as a constraint, and briefly notes three possible ways to overcome complexity: (1) the construction of minitheories, (2) the use of systems theory, and (3) the development of ‘fuzzy models’ Finally, the paper considers, given these various constraints, whether it is indeed possible for social psychologists to formulate and test sentences in a manner more or less consistent with the positivistic goals of science

Systems simulation comparing different decision rules

Abstract: Applying the expected utility maximization rule often requires that we quantify probabilities and utilities, an undertaking which can be very complicated, time-consuming, and costly for many decision makers. For consequences of moderate to low utility—where not much is at stake–it's often much more sensible to use a heuristic decision rule This paper considers fourteen different decision rules (eight of them drawn from the election literature) using a simulation. Generally the ranking, approval, weighted ranking, and weighted approval decision rules seem to work well relative to the expected utility rule. In some cases, the rule which ignores all events but the most probable is the best These decision rules were frequently able to get 90% of the utility which would have been acquired using the expected utility rule. Thus, clearly–depending upon the expense of using the expected utility rule versus a simpler heuristic—it may, in fact, be rational for a decision maker to use a heuristic decision rule. This paper provides tables describing how our fourteen decision rules performed under various conditions.

A theory for attitude and behavior applied to an election survey

Abstract: A mathematical theory is presented for attitude as it affects decisions and behavior of individual persons, systems at the organism level. Attitude is formulated in terms of rational and emotional thought and a biological responsiveness. Behavior is formulated from attitude using probability rules and a critical attitude, which is based on standard attitudes. The application of the theory is illustrated using a US Presidential election survey. The analysis suggests voter behavior can be modeled with a cusp catastrophe equation based on influence derived from utility and importance derived from emotions.

An optimization procedure for a university

Abstract: Every organized human group has a decider subsystem. It is usually both explicit, as represented by laws and elections, and implicit, as represented by traits, taboos, and traditions. This paper proposes and illustrates the use of a new kind of democratic voting system, called dynamic value voting, as an explicit procedure for an organization to use better to reflect correct group values in making decisions. The application for a university community is discussed, with Stanford University as the example, where the explicit function is represented by group voting action on financial plans and budgets. The procedure is illustrated in detail using the TRADES financial planning system developed by Massy, Hopkins and Dickmeyer at Stanford, and the personal utility functions constructed by Dickmeyer for a sample of members of the Stanford University community. The explanation makes use of various computer programs developed by the present author, and by the Stanford team, to provide a quite detailed example of the application of the dynamic value voting procedure for this kind of organizational optimization. Although the paper deals only with the Stanford University application, the central ideas and the computer approach are applicable for any group, and should be especially useful when there is great diversity of opinion and self-interest among the group members.

On the theory of sexual reproductive kinetics with comments on some recent experimental results

Abstract: A general formalism is presented which integrates behavioral patterns into population dynamics. These behavioral patterns can be represented by a set of linear or nonlinear state equations which are said to generate the behavioral microdynamics. It is shown that under certain time-scaling assumptions a behavioral equilibrium distribution may exist which allows a considerable simplification in the representation of the behavioral microdynamic (BMD) state space. An important BMD process, sexual reproductive kinetics, is investigated in detail with reference to the specific behavioral patterns of Drosophila melanogaster. Three subsystem sexual reproductive kinetic models are investigated: (1) male mate searching behavior, (2) female spacing behavior, and (3) the response of males to female pheromone emission. The qualitative agreement between experimental results and both the general and specific aspects of Drosophila melanogaster behavioral microdynamics is discussed.

Eye-contact graphs.

Abstract: The structure of nonverbal communication expressed as eye-contact between two human beings is analyzed using graph-theoretic tools involving a theorem of Konig on bipartite graphs and various results concerning directed graphs (as in Harary). A taxonomy for possible eye-contact configurations is constructed; then a theory, formed from a sequence of theorems proved about classes of eye-contact graphs derived from the taxonomy, is interpreted to analyze possible levels of communication. This theory can apply to any living system but it is interpreted here with respect to human subsystems composed of (1) individuals with normal vision, and (2) individuals with vision disorders which lead to crossed eyes.

Infering a majority from a sample: The sawtoothed function phenomenon

Abstract: The probabilistic phenomenon presented in this paper deals with decision making in living systems at the level of the organization, society, or supranational system. Decision-making bodies often take votes even when some members are not present (given a quorum). But the probability of a sample's decision matching the population's is not an increasing function of sample size. The function is sawtoothed, and even almost complete voting bodies in some circumstances have high probability of making ‘incorrect’ decisions. These circumstances are delineated, and some of the mathematical properties of the sawtoothed function are indicated. It is shown that a quorum is a poor criterion for holding votes.