Showing papers in "Systems Research and Behavioral Science in 1977"

Social action as the control of affect

Abstract: When defining their immediate situations, people characterize themselves and others by social identities like doctor and patient, evoking sentiments that serve as guidelines for interpreting and creating events in the given situation. In particular, the assigned identities recall notions of how good, how powerful, and how lively each person is fundamentally. When events deflect feelings away from these sentiments, new events are conceived and ordinarily implemented to move feelings back toward the fundamental values. Thus, theoretically, social behavior and transient feelings form a control system with fundamental sentiments as reference signals. Interpersonal conflicts sometimes lead to events that a person cannot comprehend as sentiment confirming. This invokes redefinitions of situations, a higher order control that changes reference signals. Empirically derived formulas describing affective reactions to events have been elaborated into a mathematical model representing all of these processes. Data on a large number of social identities and behaviors have been collected to permit simulations using the model. Illustrations show that this theory of social action permits concrete, plausible analyses of social interactions, role relationships, and social reactions to deviance. Simulations of group dynamics in organizational contexts conceivably could provide guidelines for changing group functioning and for establishing social structures without historical precedent.

A theory of analytical hierarchies applied to political candidacy

Abstract: A theory of hierarchies is applied to structure complex decision problems about groups, organizations, societies or supranational systems in three steps. The problem is decomposed into a number of strata, each with several elements which may be people, variables, policies, and so on. We next analyze judgments about the interactions among elements in each stratum with respect to their impact upon elements of the immediately higher stratum of the hierarchy. A theorem on hierarchical structure permits us to recompose judgments made at each stratum to arrive at an overall preference ordering, policy choice, or solution to the decision problem. A simple, three-strata hierarchy pertaining to the choice of the most qualified candidate for the Democractic Party's Presidential nomination illustrates the theory at the level of a societal system.

A Simple Multiattribute Utility Procedure for Evaluation

Abstract: The general problem of how to determine the worth or utility of alternatives that vary on many dimensions is of great practical importance. Although the number and types of situations that require such evaluations are large, the most usual way of performing such tasks has been unaided “intuition” (or, clinical judgment); i.e., the decision maker somehow does a mental trade-off analysis between the various attributes and alternatives in order to come to an evaluation/ decision. The cognitive difficulties of performing such a feat are formidable. For example, consider a situation with ten alternatives, each varying on six attributes. The intuitive decision maker has the task of locating ten alternatives in a six dimensional indifference space and picking the one with the highest utility. In such complex situations, an accumulating body of psychological research on the decision process has shown that people will reduce task complexity by using various heuristics (e.g., Tversky, 1969; 1972; Payne, 1976). While these heuristics have the advantage of allowing a decision maker to perform a complex task, they may lead to non-optimal behavior (e.g., consistent intransitivities). Furthermore, the literature on clinical judgment (Meehl, 1954; Sawyer, 1966) has also shown that experts have great difficulty in intuitively combining information in appropriate ways.

The forward and backward processes of conflict analysis

Abstract: In this study, the forward process of planning is applied to conflict resolution, by structuring a conflict according to the levels of a conceptual hierarchy. The parties to the conflict form the first level, their objectives are at the second level, and possible solutions are at the third and final level. The elements at each level are weighted according to their relative importance by using the principal eigenvectors of a series of pairwise comparison matrices, and a weighting of the possible solutions is obtained. This forward process answers the question: given the present actors and their objectives, capabilities and policies, which outcome is the most likely to emerge? This technique is applied to a societal system, analyzing conflict in Northern Ireland. It is shown that the outcome which would in greatest measure satisfy the needs of all parties is one which gives legislative independence to Northern Ireland. This method of conflict resolution does not require long or involved calculations, and a means of further shortening the calculations is given. Unimportant branches of the hierarchy may be truncated. The use of hierarchies and the eigenvector method provides an efficient method for the evaluation of possible solutions of a conflict and provides valuable insights into the objectives of the parties and the way in which possible political structures might satisfy these objectives. This is a powerful technique to be added to the methods for conflict resolution.

Methodological problems of measurement of fuzzy concepts in the social sciences

Abstract: The paper explores the possibility of an empirical access to the membership functions of fuzzy sets representing concepts. A formalism is introduced to describe structural features of concepts. It is shown, at least for the system of concepts in the social sciences, especially those on a high level of generality, that the theory of psychological tests provides the methodology of construction of measurement tools.

Stress reactions in organizations: Syndromes, causes and consequences

Abstract: This study examines the prevalence, causes, and consequences of stress reactions among over 2000 high status members of a large organization in Canada. Five stress syndromes were identified: emotional distress, medication use, cardiovascular disturbance, gastrointestinal disturbance, and allergy respiratory disturbance. The study demonstrates the effects of sociocultural variables on selected stress responses to compare typical symptoms of French vs. English Canadians, and in demographic variables such as age and sex as reflections of stages in the life cycle. Once the sociocultural and demographic effects were isolated, the study revealed a major occupational effect in the distribution of stress syndromes. Managers compared with staff and operations people show a low prevalence of stress reactions. The explanation of this finding led to the evaluation of four possible causes: the maturity effect, the vulnerability effect, the bureaucratic effect, and the power effect. The first two causes assume the objective stressors in the environment are about the same for all occupational groups in the study and the variations in symptoms result from major personality factors held in common among the occupational groups. The second two causes assume that the variations in symptoms result from major differences in the environments of the three occupational groups. Analysis suggests that psychodynamic factors affect the tolerance for stress and the choice of syndrome. The environment, particularly the frustrations and deprivations associated with bureaucracy and the lack of power, activates the defenses against anger and rage that in turn lead to symptom formation. The study concludes with the paradox of the bureaucratic experience. Designed to minimize the uses of power in negotiating work procedures and relationships, bureaucracy requires the mobilization and uses of power to, at a minimum, reduce the risks of falling ill from frustration and anger and, at a maximum, to sense one's impact on events.

A clinical evaluation of four alternative telemedicine systems.

Abstract: A comparison was made of four alternative telecommunication systems used in delivering primary health care to remote populations. The media were color television, black and white television, still frame black and white television, and hands-free telephone. The patient population was that of a large organization in a Canadian province, a major component of a societal system. Over 1000 patients who came to a clinic seeking medical attention were examined remotely by one of the four systems. In addition, they were examined in the physical presence of a doctor at the clinic. The diagnoses, patient management programs, etc., of the clinic physician were used as the basis for comparison. We found no significant differences in diagnostic accuracy, proportion of supporting investigations requested, e.g., laboratory tests and X rays, time taken for the diagnostic consultations and the effectiveness of patient management across the four communication modes. On the contrary, in some instances it was found that behavior across the modes was significantly similar. Even patient attitudes showed only a slight preference for the more sensory rich modes of communication. It was concluded that there is little to distinguish the effectiveness of the four telecommunication modes when used for remote diagnostic consultations. Therefore, the cheapest mode is the most cost-effective. Phase four of the research program will compare the effects of the two least costly modes, still frame television and hands-free telephone, in an operational system in northern Ontario.

An analysis of voting procedures with nonranked voting

Abstract: This paper analyzes decision making by simple voting procedures in organizational or societal systems. The simple voting procedures investigated are those that do not require voters to rank candidates when one of three or more candidates is to be elected to a single office or position. The procedures are compared on the basis of their propensities to elect the simple majority candidate when it exists. Propensity is measured in terms of efficiency of the procedure, which equals 100 times the probability that the candidate elected by the voting procedure is the simple majority candidate, given the latter exists. The procedures examined use either one or two ballots. On the first or only ballot they ask each voter either to vote for not more than so many candidates (variable procedures) or to vote for exactly so many candidates (constant procedures). Depending on assumptions about voter preferences and voting behavior, either the best variable procedure or the best constant procedure can have the higher efficiency. However, in both the one-stage and two-stage contexts, the difference in efficiency between the most efficient variable procedure and the most efficient constant procedure is usually negligible from a practical standpoint. On the other hand, efficiency can vary significantly depending on the number of stages and the number of candidates to be voted for, or up to, on the first ballot.

Family configuration and intellectual development: A simulation

Abstract: A simulation based on the recently published confluence model is described. Within this model, family configuration variables –family size, order of birth, and length of birth intervals –play an important role in influencing intellectual development. The simulation is shown to achieve a close fit with national aggregate patterns of intelligence test scores for children of different family configurations. In addition, it is demonstrated that dissimilar aggregate profiles may be consistent with a common underlying model. The apparent differences may be due to systematic variations in the ages and birth intervals of sample members.

Stability analysis of the forward-backward process: Northern Ireland case study

Abstract: In this study, the forward and backward processes of planning are applied to conflict resolution in a way similar to that in which the forward process was applied in a previous paper by the authors (Alexander & Saaty, 1977). By repeated analysis of the policies of the parties to the conflict and possible reactions to these policies, an iterative process is established by which the stability of an outcome may be measured. It is shown that the possible outcomes of a conflict may be characterized by a set of state variables covering political, economic, social and legal factors. The composite outcome, found by use of the method, also may be described by weighting the values of the state variables so that a full description of the most probable and stable outcome is obtained. These techniques are applied to a societal system to provide a further analysis of the conflict in Northern Ireland. It is shown that legislative independence, the outcome obtained in the previous paper, is stable and cannot be upset by any of the parties to the conflict. The use of these techniques provides efficient methods for analyzing conflict and gives valuable insights into the behavior of the parties.

Comparing multiattribute evaluation process models

Abstract: An important part of the cognitive system of any individual is the process by which alternatives are evaluated. Of particular interest is the evaluation of multiattribute alternatives. Behavioral scientists have attempted (1) to describe the cognitive processes underlying such evaluations and (2) to predict the outcomes of such evaluations. This research develops four lexicographic models of the multiattribute evaluation process. The predictive accuracy of these models is tested and they are compared with four linear evaluation process models. An experiment was designed in which a decision task – evaluating salesmen for the purpose of bestowing year end bonuses –was assigned to 66 graduate students in business. Each subject provided evaluation information on four sets of five alternative salesmen. Results showed, among other things, that linear models were not superior to lexicographic models. These results raise important questions for the modeling, prediction and prescription of multiattribute evaluation processing behavior.

A computer simulation of freud's counterwill theory

Abstract: This paper reports a computer simulation at the level of the organism. Freud's counterwill theory is translated into a computer model, using Forrester's system dynamics language. Three simulation runs are described to illustrate the behavior of the system in comparison with Freud's description of patient behavior. It is shown that system dynamics methodology is an effective device to expose the complex structure of psychoanalytic theory. The experimental study of psychoanalytic theory has been predominantly directed to isolated propositions. However, a system dynamics representation of the theory may stimulate academic psychology to a more comprehensive and integrated investigation of psychoanalytic theory.

Time horizon: The moving boundary

Abstract: This paper surveys the problem of moving boundaries in physicochemical, biological, managerial, socioeconomic and cybernetic systems and in its spatial as well as temporal aspects. It is shown that a time horizon is a moving boundary which separates the foreseeable from the unforeseeable future. Its specification is required for unambiguous forecasts of short-range and long-range behavior of the system under study. In general, the entropy or uncertainty of transition across the time horizon has a paraboloid distribution. In socioeconomic systems it is especially important to distinguish between ex ante and ex post forecasts and to specify not only the errors in the variables, but also in their relations. Various computational methods used in the analysis of moving boundary problems are indicated in the literature.

A new look at the description and prescription of systems

Abstract: Present day status of general systems theory makes it appropriate to propose the notion of a kind of system which is able to generate the goal(s) for its own activity, a teleogenic system. This concept may apply to all levels and types of concrete systems. Some prerequisites for conceiving of such a system, and eventually founding a theory dealing with it, are outlined in this paper. The intimate relations that exist between system and observer are underscored. Any alleged property ascribed to reality is relative to the perspective of the observer. Description is facilitated through iterative modes of reflexion and the concomitant setting of a context. This permits distinguishing between a local and a global description. It is shown that the former tends to assume a causal aspect, whereas the latter is in conformity with a final aspect. It is necessary to distinguish between an unconcerned observer and a concerned observer or designer. The former only designs models to explore the system he observes, while the latter also builds the system he prescribes. The interplay between descriptive language and prescriptive language is considered. Any description is not only a representation of a reality, but, conversely, any reality exists for the observer/designer in the description made. In order to obtain a global view, one encompassing the meaning or significance of a description for the observer/designer, the assumption or prescription of goals is inevitable. According to the role goals assume, it is necessary to distinguish between goal directed, goal selecting, and goal-generating systems.

A competitive test of the 1(im)bargaining set, kernel, and equal share models

Abstract: An experiment was conducted to test three competing theories of payoff allocation in n-person conflicts. One hundred-eighty subjects participated in three-person groups. Subjects were decision makers in nonconstant sum games with sidepayments and nonempty core. With respect to coalition formation, the results showed a propensity for coalitions with large characteristic function values to form more frequently than coalitions with smaller values. With respect to payoffs, results indicated that the M1(im) bargaining set based on the Shapley standard of fairness predicted the observed allocations more accurately than either the kernel theory or the equal share model. The results also showed that relatively few outcomes fell in the core, a surprising observation given the theoretical prominence of this concept. Findings are discussed in the light of previous experimental research on n-person conflicts.

A model of attitude change in groups following a persuasive communication: An attempt at formalizing research findings

Abstract: From a review of the available literature related to persuasive communication, the author proposes a mathematical model of attitude change in human groups satisfying a set of premises concerning the following variables: discrepancy, source credibility, ego involvement, plausibility, distraction, and effort. This model is shown to be consistent with cognitive dissonance theory, the social judgment-involvement approach, learning theory, and to provide links with perception. After determining the proper estimating procedure, the model was extensively tested with more than 50 sets of published data. The instrument used was regression analysis. It appears to be well supported by the available data. Nevertheless, more effort is necessary to introduce other variables, including time, to incorporate the boomerang effect and to solve the problem of infinite horizon on the attitude scale. It is suggested that this model has relevance to individual as well as group behavior.

Independent utility scaling and the nash bargaining model

Abstract: A newly developed method for independent utility scaling was used to replicate Nydegger and Owen's earlier work that experimentally evaluated the Nash axioms for two-person bargaining. This was done in an attempt to devise better correspondence between abstract models and experimental applications in game theory. The results were: (1) The independently developed utility scales did apparently minimize the problem of interpersonal comparisons of utility. (2) When irrelevant alternatives are introduced into the bargaining, the Smorodinsky-Kalai monotonicity axiom is as descriptive of subject behavior as the Nash model. (3) When comparison processes are controlled, the Nash axiom on invariance under linear transformations of utility does seem to hold. This lends credence to Nydegger and Owen's contention that the failure of this axiom to hold up in early experimental studies was due primarily to players' attempts to effect interpersonal comparisons of utility. In summary, understanding of the complexity of the interpersonal bargaining setting seems to depend in part on developing more appropriate methods of making cross-systems comparisons, that is, in relating the bargaining unit as a whole to the player subsystems.

Hierarchical structure of participation in natural groups

Abstract: The meeting minutes of the ECOSOC Council of the UN, as a small natural group, were content analyzed to reveal the frequency of participation of the Council's delegates. Three different methods of coding yielded a very high correlation in ranking the order of frequency of participation among the delegates. The rank order of frequency of participation was used to test the exponential model of the distribution of group participation described by Stephan and Mishler (1952), and further formalized by Horvath (1965). The results did not fit the model statistically and the residual trend seemed to persist even after the modification of the model. It was speculated that the model may not be appropriate for a small group of relatively large size. However, for one parameter model, given the magnitude of the errors, it does not do badly at all.

Animal hierarchies and repeated comparisons

Abstract: Linear and near-linear hierarchies are well known in animal groups, but the nature of the process or processes that form such hierarchies is unknown. Chase recently demonstrated that two commonly suggested processes cannot be correct, even in theory. A simple modification of one of these processes, however, to allow repeated comparisons of strength, size and/or skill produces a plausible and biologically realistic model for linear or near-linear hierarchy formation.

Experimental: A simulation of the distribution of services to mentally deficient children

Abstract: In this study a formal analytical technique was introduced in a typical planning operation of the Quebec Ministry of Social Affairs. The mental deficiency sector of the province was chosen because important changes were imminent in both the orientation and types of services provided to children in this sector. A simulation model was constructed according to system dynamics methodology to provide an integrated description of the social impact of these changes. The immediate result of the simulation model was to focus a reevaluation by government planners on certain operational features of the system of services and the implementation schedule of the ministry's policy. In a broader context, the model demonstrated that simulation is a viable approach to planning social services. Similar social services studies have been initiated to familiarize government planners with the methodology and have been used in the preparation of precise implementation plans.

Long-term feedback and differential feedback effects in nonmetric multiple-cue probability learning

Abstract: The purpose of this research was to study the long-term effects of different types of feedback upon learning and performance of individual human organisms and the sensitivity of such subjects to changes in the validity of information in probabilistic judgment tasks. The subject's task consisted of predicting a binary event on the basis of two or three binary cues whose validity changed on each of five successive days. Subjects received one of three different types of informational feedback about their performance - outcome feedback, cue-event validity feedback, or percentage correct feedback. The results indicated, at least for tasks involving additive cues, that adaptation to new cue validities became more rapid on successive days. The different types of feedback tended to produce different response strategies. However, the different strategies had virtually no effect upon overall judgmental accuracy. Further, there was an apparent interaction between task characteristics and feedback type. This result was found to be consistent with the results of other studies utilizing differential feedback in similar tasks.

A comparison of some inductive inference models

Abstract: The psychology of discovery is explored in a laboratory setting using the inductive inference task of determining the mathematical function which fits a sample of coordinate values. This task resembles scientific inference more closely than traditional problem solving paradigms. The observed strategies in this task are seen to vary with subject experience and problem representation. Faculty members in the hard sciences with a graphical representation tend toward heuristic processing, whereas students given a simple list of points exhibit a relatively stochastic generate and test approach. The experiment is discussed in the context of related research and actual scientific discoveries. It is concluded that random error, a graphical representation, sophisticated technology, experience with similar problems, and a geometrical cognitive style should produce a heuristic discovery process while deterministic data, a list representation, unsophisticated instrumentation, lack of familiarity, and a numerical style should lead to a generate and test strategy.

Recognition of coalition by computer in a three‐person game

Abstract: An information processing model for recognition of coalitions in three-person games is developed by using a computer program, which simulates humans' move selection behavior, in Chinese checkers. The program plays as a participant against human opponents and decides if a coalition was formed after a game is over. It has two main parts: (a) a heuristic search algorithm for move selection and (b) a coalition recognition procedure which evaluates deviation patterns. The latter is defined as the difference between the move pattern of the selection algorithm and that of a human player. With this model, it is shown that humans' move selection behavior may be relatively simple even when the task environment involves human interaction and also that the moves of human players are influenced to some extent by the possible existence of coalitions. The system treated here is an information processing system, a computer program, which works as a member of a social environment and recognizes interactive behavior of other members. It models human behavior in such environments at the symbolic information processing level or the level of information processing psychology.

Embedded game analysis and international conflict control

Abstract: The use of game theoretic models for understanding international relations has been widely criticized. In this paper we suggest that the theory of embedded differential games with optimal control is a productive methodology for responding to the critics. Several noncooperative differential game models with embedded objective functions are formulated. Concepts for obtaining undominated solutions based on the game structure, the quadratic objective functions and linear differential kinematic equations are discussed. The paper concludes that models with rich empirical reference can be formulated and productively used in understanding international conflict.

Bargaining behavior when profits are unequal and losses are equal

Abstract: This study explores the effect of information about other's payoffs as well as the power of a mediator in the most difficult bargaining situation: It is clear to both parties that no agreement could benefit them equally, but failing to agree will result in an equal loss. This situation was investigated in a bilateral monopoly setting where subjects, 172 male undergraduate students, were divided randomly into pairs. One subject in each pair assumed the role of a buyer and the other that of a seller, attempting to complete a transaction involving an imaginary product. Each of the alternative prices which could be agreed upon was associated with a differential profit, and failing to agree resulted in forfeiting a deposit. The amount of information about the other's profit schedule was found not to affect the probability of reaching an agreement. It did affect the magnitude of the difference between the parties' profits, given that an agreement was reached. An interesting paradox was found concerning the mediator. His availability induced subjects to employ him. Thereafter, they tended to reject his suggestions, incurring a risk that inclined to offset the advantage he seemed to offer. Despite the possible advantages of intransigence, no party used an intransigent strategy. Hence, no significant difference was found between the proportion of agreements favoring the initiator of the bargaining process and those favoring the party having the last clear chance. The few cases ending in disagreement were all a result of an operation of a random device simulating fatigue. The results are considered to have implications for a wide range of social situations involving both mutuality and opposition of interests.

Beliefs about coalition formation in multiple resource three‐person situations

Abstract: Minimum resource theory and several different solution concepts of game theory as applied to three-person constant sum coalition situations differ in suggested outcomes. Minimum resource theory predicts the two weaker parties will form a coalition because of a payoff split according to its definition of power. Students were asked to judge coalition probability for multiple resource situations. Results showed that judges thought the stronger coalition was as likely or more likely to occur than the weaker coalition. This result indicated the possible importance of the person in the middle of a power structure and the differences between perception of and acting in a coalition. It also raises several questions concerning the interpretation of game theory solutions and minimum resource theory in the provision of testable hypotheses.

Delegation and control in organizations with varying degrees of decentralization

Abstract: The concepts of decentralization, delegation, control, participation, and responsibility are defined operationally for the analysis of hierarchically structured large organizations or service systems. These include hospitals, libraries, or governmental agencies, all of which involve the coordination of several resources or activities. Costs and benefits are examined from a client centered point of view. We take into account cost-benefit considerations for the organization or service agency and for its clients, as well as for the larger community by which the agency may be sustained. Delegation of certain classes of decisions to a particular agent or echelon is translated into providing this agent with the means to make decisions at tolerable levels of error and delay. That is, the agent must have adequate communication channels, memory facilities and other information processing capacities. The change in the utility of the decision performance of the organization after the delegation has been made is compared to the change in costs caused by providing or shifting the requisite facilities. The effects of a long-term increase in service loads upon the optimum levels of decentralization and delegation are examined. Eight dimensions of decentralization are specified and four general principles are proposed.

Response to sudden shifts in a two-nation arms race†

Abstract: This article concerns living systems at the international or supranational level. Control theoretic techniques may be used to obtain information about the behavior of formal arms race models that cannot be gained by alternative methods. A two-nation, Richardson linear reaction process arms race model is treated as a second-order linear control system in order to determine its response to a sudden parameter shift represented as a step function disturbance. In contrast to the equilibrium oriented techniques generally used in the study of arms race models, control theoretic methods allow detailed analysis of both long-term and short-term effects of transient disturbances, including the way in which the system moves to a new equilibrium. The Routh-Hurwitz criteria and root-locus plot are compared to the more traditional stability criteria. System response is analyzed in detail for both the critically damped and the over- or under-damped cases. State space solution concepts that allow treatment of many-nation systems are examined. The techniques here applied to the arms expenditures of two nations isolated from or only loosely coupled to the rest of the world are also applicable to other reaction processes, such as two-party or intergroup conflicts, and this treatment may be extended both to higher-order differential equation models and to a large class of nonlinear models. Some methods are suggested for governments to use in managing their behavior so as to bring sudden shifts in arms races under greater control.

The universe as a general system. Fourth annual ludwig von bertalanffy memorial lecture

Abstract: This is the fourth Ludwig von Bertalanffy Memorial Lecture, delivered at the 1977 Annual Meeting of the Society for General Systems Research at Denver, Colorado.

Test of a model of decision processes in an intense conflict situation

Abstract: A particularly intense form of conflict is uelative conflict It is a situation in which at most one of some set of participants may obtain an objective and in which it is possible that none will obtain their objective A mathematical model is derived from basic assumptions about the decision making process in a three-person uelative conflict situation, a truel It is proposed that the model is generalizable to re-person situations A preliminary test of the model that varied only one of the parameters revealed support for the model More extensive tests of the model are warranted In addition, the applicability of the model to internation or intergroup conflict is explored