Showing papers in "Systems Research and Behavioral Science in 1973"

Persistence and loose coupling in living systems

Abstract: In line with current efforts to achieve a general view of living systems the degree of coupling is considered using specific examples at the levels of cell and organ, organism and group, organization and society. Loose coupling among systems, similar to Ashby's concept of independence, contributes to stability by allowing persistent behavior of the system in the face of certain inputs. It can be achieved either actively, by having a more tightly coupled subsystem which compensates for a given category of input by negative feedback, or passively, by an arrangement which allows certain variables only limited access to the system. The evolution of adaptive and maladaptive degrees of coupling in different systems is discussed.

Toward a model of interpersonal motivation in experimental games

Abstract: A geometric model of preference is described for evaluating choices in decomposed two-choice experimental games. The model characterizes a subject's (S's) choice behavior in terms of a parameter, called motivational orientation, that represents the relative weights assigned by S to his own and to the other player's outcomes. Previous definitions of motivational orientation are shown to be included in the model as particular values of the model parameter, and the characterization of decomposed games by dominance class is interpreted geometrically. Methods for measuring motivational orientation in decomposed games are discussed, and an experiment is presented demonstrating the validity and reliability of the construct.

A reparameterization of the prisoner's dilemma game

Abstract: A parameterization of the prisoner's dilemma paradigm is constructed based on recent research on the theory of risk and risk preference. The conflict is cast as intrapersonal and in terms of a risk parameter and an inducement to accept the risk, sometimes interpretable as fear and greed respectively, both acting in the context of a level for the game relative to the status quo. These correspond to the independent variables directly under the control of the experimenter. The dependent variable is the strength of the tendency to defect. These variables are acting in the context of an experimental environment which includes the benevolence or hostility of another player which is assumed to contribute to the risk variable only. Some suggestions are made for the controlled study of conjoining a hostile environment.

Success and failure as determinants of level of authoritarianism

Abstract: Various authors have argued that the more threatening an individual's environment is, the more authoritarian he is likely to be. This hypothesis was investigated in two separate experiments. In both, Ss were pretested for authoritarianism and then were exposed to one of two conditions. In the first condition, Ss were threatened by failure at two experimental tasks; in the second, they were presumably encouraged by success at these tasks. In both studies, failure increased and success decreased the Ss' levels of authoritarianism. In both studies, these changes in authoritarianism were greater for individuals who attributed their performance to internal causes. In addition, data from the second investigation indicated that these manipulations affected Ss' tendencies to conform to the judgments of an authority figure. The findings appear to support the contention that personality is an open system.

Voter concordance, simple majorities, and group decision methods

Abstract: This paper examines the effect of the degree of agreement among voters' preferences, as measured by the Kendall-Smith coefficient of concordance, on (1) the likelihood that one decision alternative will have a simple majority over every other alternative and (2) the propensity of the Borda and Copeland selection methods to give common winners. The Borda method determines the winners by maximum point total when uniformly decreasing points are awarded to the alternatives in each voter's preference order. The Copeland method determines the winners by the maximum number of other alternatives that each alternative defeats by simple majority. The general trend for (1) is an increase in the probability of a simple majority winner with an increase in voter concordance. A similar trend is observed for (2) when agreement is measured by the percent of cases in which the Borda winners are identical to the Copeland winners. However, when agreement for (2) is measured by the percent of cases in which some Borda winner is also a Copeland winner, the best agreement is obtained when there is either extreme voter disagreement or extreme voter agreement.

The contingency model of leadership effectiveness: some implications of its statistical and methodological properties.

Abstract: The statistical and methodological adequacy of several aspects of Fiedler's (1967) contingency model of leadership effectiveness were considered. Data from several published studies were reanalyzed in order to accomplish this end. It was suggested that the use of rank order correlations instead of product moment correlations produces distortions related to sample size. LPC (least preferred coworker) was shown to be a reasonable predictor of group performance, but the model implies that LPC is unrelated to group performance under some conditions of situational favorableness. Problems with a multidimensional interpretation of LPC and a unidimensional interpretation of situational favorablness were discussed. Problems of research strategies used to test and validate the model were considered in terms of sample size, inclusion of nonsignificant results, the difficulty involved in adequately testing higher order interactions, and the inability to make between octant comparisons. It was concluded that the model is capable of directing meaningful research, but only as long as traditional research procedures designed to safeguard internal and external validity are carefully exercised.

The development of transitive structure in children's interpersonal relations

Abstract: Theories of cognitive development in children and theories of social organization in small-scale social systems when jointly considered suggest that the structure of interpersonal relations in children's groups will progress in consistent fashion from less to more transitive organization as the children become older. This social developmental hypothesis is tested by examining the organization of interpersonal sentiment preferences in 118 groups of children ranging in age from 3 to 11 years. Multiple regression techniques are used to examine the effect of age on structure. The results indicate that there is a significant relationship between age and the extent of transitive organization in the structure of the children's interpersonal preferences. Analysis of a second set of data taken from the sociometric literature provides further support for the social developmental hypothesis.

A systems philosophy of human values

Abstract: Contemporary general systems theory encompasses systems science and technology, as well as systems philosophy. The latter is the growing edge where conceptual innovations are explored and their relevance to concrete problems examined. Philosophical value theory can be reconstructed in the framework of systems philosophy by conceiving of values as expressions of various states of adaptation of the individual to his biological and sociocultural environment. Hereby the way is opened to a science of normative values, defining patterns of optimum physical and mental development in persons. The basic conceptual map underlying this program is constructed in reference to contemporary theories of value and system-environment interaction patterns.

A formal theory of actions

Abstract: This paper presents a formal theory of actions which may serve as a foundation for different domains, such as theory of organization, theory of planning, etc. The basic idea underlying the construction is that actions may be concatenated into strings, by performing them one after another, and that not all strings of actions so obtained are physically feasible. This allows us to draw an analogy between the class of admissible strings and a language: actions play the role of words in a certain vocabulary, and the admissible strings of actions play the role of sentences. One may therefore speak of a language of actions, characteristic for a given situation, and analyse it by means of mathematical linguistics. Moreover, a string of actions leads to certain outcomes. If one considers only those outcomes which are inherently connected with this string, i.e., occur whenever this string is performed, one may enrich the language of actions with a semantic structure: the well determined outcomes of a given string of actions play very much the same role as meanings of sentences. The above linguistic intuitions apply to the case of linear concatenations, where simultaneous actions are excluded; thus, the main, but not the only one, interpretation is in terms of actions of a single person. One may, however, consider also simultaneous strings of actions, for which the interpretation is in terms of team actions. In this case, some methods of mathematical linguistics are still applicable, leading to formal explication of such concepts as cooperation, blocking, etc.

A comprehensive automated psychological examination and report (CAPER)

Abstract: A program is described that generates psychological reports automatically for a battery of standard measures of intelligence and personality in adults. Problems are encountered in the programming of a battery of tests that are not present in the single test automated systems that are available. The standard measures include the Raven, MMPI, and 16PF as measures of nonverbal intelligence, psychopathology and personality traits, respectively. Newly developed measures also are included, namely, a multiple choice version of the WAIS vocabulary subtest, a questionnaire dealing with family interaction, and a sexual history questionnaire. All tests are interpreted by simple procedures that involve the printing of sentences or short paragraphs when scores fall in a given range. In the case of the MMPI/the configural systems of Gilberstadt and Duker, Marks and Seeman, and the Mayo Clinic have been included. The system provides a comprehensive clinical report that seems adequate and that requires little time for a psychology staff member.

Noninteraction in an anonymous three‐person prisoner's dilemma game

Abstract: Twelve triads were run for 100 trials each in a three-person prisoner's dilemma game, where choices were made anonymously Results showed that the level of cooperation was lower than that found in previous PD games The choices made by the individual players were shown to have been made with apparently no regard for the choices of the other two players in the game

Some effects of source behavior on target's compliance to threats

Abstract: A robot player sent intermittent threat messages of 10 percent, 50 percent, or 90 percent credibility to 180 subjects, 90 of each sex, during the course of a mixed motive conflict game. The magnitude of the threatened punishment was either low or high. The simulated player was either 0 percent, 50 percent, or 100 percent accommodative in his use of coercive power on the threat-related trials. The results indicated a strong effect of source accommodation on a behavioral measure of target compliance. Supplementary analyses of the compliance data according to an expected value model of social power indicated that males complied as a direct function of the difference between the expected values for compliance and noncompliance, while females did not. Furthermore, a threat credibility by sex interaction was found on the proportion of cooperative responses made by targets on nonthreat occasions. The combined compliance and cooperation results indicated that targets' responses to threats are highly dependent upon both the words and the deeds of the threatener.

Senatorial decision making: I. The determination of structure

Abstract: A three-mode factor analysis of the U. S. Senate roll call voting was performed. The modes were: issues, time and Senator seats. The entries into the data matrix were the point biserial correlations of a Senator's votes on a particular issue in a particular time period with the popularity of the roll calls in that set. Five Senator factors were found, indicating the various voting coalitions. There were indications of the ideological, geographic and partisan structure within the Senate. There was also the formulation of the factors of time change which indicate how coalitions vary across time.

The significance of variability

Abstract: The negative attitude psychologists hold toward behavioral variability is contrasted against the more positive attitude biologists hold toward genetic variability. It is argued that variability is a necessary prerequisite to adaptation, whether genetic or behavioral, and that the sources of variability are intrinsic to the organism. One can also find constraints upon variability present in biological systems and it is proposed that too much variability is maladaptive and that variability is stabilized at some ideal level by a system of checks and balances. It is suggested that both sources of variability and constraints thereon could more profitably be investigated directly rather than continuing with the philosophy that variability constitutes noise to be eliminated.

The recall of historical events

Abstract: This investigation explores the extent to which psychological analyses allow us to reconstruct the history of individuals, their social environments, and society. The first study reports on the free recall of persons (relatives, friends, acquaintances) and compares the data with potential changes in social contingencies in the life of college subjects. In the second study, college students recollect the names of historical figures (politicians, scientists, artists). These data are compared with interpretations of the course of political intellectual history. In the third study, a report (recall) of the history of psychology by one eminent historian of the behavioral sciences, Professor Boring, is analyzed and the results compared with those of the first two studies.

A proof of May's theorem p(m, 4) = 2p(m, 3)

Abstract: A simple proof is given of Robert May's (1971) theorem which says that in an impartial culture with m voters (odd and exceeding one), the probability of no simple majority winner in a four-alternative contest is twice the probability of no simple majority winner in a three-alternative contest

Statistical significance of the lambda coefficients

Abstract: Lambda coefficients provide a measure of the strength of relationship between two nominal variables and also have proportional reduction in error interpretations. It is also possible to test the statistical significance of the lambda coefficients, but the computations are time consuming and are not generally understood by behavioral scientists. This paper presents the method. The computer program is available from the author.

The world we manage

Abstract: This article is the Presidential Address presented by Professor Beer to the Society for General Systems Research at the Annual Meeting of the American Association for the Advancement of Science, Philadelphia, December 29, 1971.

An index for the measurement of interparty competition

Abstract: This research note presents an index of interparty competition which satisfies four criteria: (1) it is a continuous variable; (2) it is a ratio scale; (3) it is linear with the concept of perceived competition, and (4) it may be generated from any array of two-party data It is maintained that these criteria establish the minimum conditions for precision measurement of political phenomena and that no other index of interparty competition accomplishes this

The computer as experimenter: New results

Abstract: In order to investigate the effect of replacing a human experimenter by a computer in a concept formation task, 150 subjects were run in three conditions One condition was run exclusively by a human experimenter, another was run exclusively by a computer, and a third was run using the experimenter solely as a teletype operator The task, designed to be difficult for the college level subject, required that all except one of the 30 possible solutions be logically eliminated before the solution was acknowledged Subjects run by a human performed at a slightly higher level than subjects run by computer This effect, which occurred only in subjects of below average intellectual ability, was attributed to an experimenter cueing effect Human administered subjects were also found to be significantly less variable regardless of intellectual level This was partially attributed to a variance dampening effect exerted by human experimenters on their subjects and which is absent from computer administered procedures One consequence of this effect is that performance in the computerized condition is more highly related to level of intellectual ability It was concluded that the effects found in this study will probably occur in any computerized procedure characterized by high task difficulty, low time pressure, and wide response latitude

A systems approach to drug user services.

Abstract: The use and abuse of drugs in America is receiving increasing public attention. Community responses to the drug problem often have been haphazard and poorly planned. This paper proposes a systems framework for studying the provision of services to drug users. The phenomenon of providing services to drug users is conceived of as a flow system with four major structural components: (1) the drug user population; (2) the nonuser population; (3) the service component; and (4) the legal component. A computer model based on this conceptualization is described, and results from the simulation of several intervention strategies are presented and compared. Discussion is centered on the costs of providing service associated with each of the strategies and on the effect of each strategy on the size of the drug user population.

Group choice with lexicographic individual orderings

Abstract: The problem of social choice when individual orderings are lexicographic is discussed in this paper. In addition to assuming that individual orderings over the social alternatives are lexicographic, the assumptions are made that the set of criteria which constitute the basis of each individual's ordering are identical for all individuals and also that the ranking of these criteria, in terms of priority, is the same for all individuals. Given these assumptions, the author establishes conditions for the existence of a socially best alternative under binary, decisive, neutral and nonnegatively responsive group decision rules, which constitute a very wide class of group decision rules. The result established turns out to be more general than the results proved earlier by Encarnacion (1969) and Taylor (1970).

An empirical examination of attitude consistency in complex cognitive structures

Abstract: Though much consideration has been given to the problem of cognitive consistency in simple, triadic structures, very little empirical work has been done with complex cognitive structures. The formulations about structural balance first outlined by Cartwright and Harary are examined and the predictions stemming from their model are applied to an empirical case study. Subjects were measured on their orientations toward four value statements. They were then shown the orientations toward these same value statements by fictitious other students toward whom they were then asked to indicate their orientations. The results indicated that the structural balance model predicted effectively for cases where maximum balance would result from a positive orientation of the Subject to his paired Other; but results failed to support the model where negative attraction was predicted. The results are discussed in relation to the theoretical difference between F Heider and T. M. Newcomb over the effects of negative attraction.

Motivation theory—A cybernetic contribution

Abstract: The signals (cybernetic signals) which operate control mechanisms have far more potential than signals which are rigidly linked to particular responses, however conditioned. They are generated internally by comparing some input with a standard. Systems so regulated are limited only by their capacity to handle information, generate standards, and resolve conflict. Psychology has been slow to credit men individually with such powers at the level at which societies manifestly exercise them. Cybernetic ways of thinking may be expected to contribute to the theory of human motivation by fostering this approach and quickening, among others, the insights that (a) regularities within social systems are created, not merely recognized, by the mutual expectations and self-expectations which they engender; (b) these expectations create and are in turn maintained by common standards of reality, by which common contexts and situations are distinguished, and common standards of obligation by which responses are judged, evoked, and inhibited; (c) the discrimination and regulation of relationships by such standards—which is a more comprehensive motivation than goal seeking—involves the management of conflicts at several levels, which this paper briefly distinguishes; (d) match signals are significant as a source of satisfaction and assurance not explained by other theories of motivation.

Decisions of timing in conflict situations of incomplete information

Abstract: Games of timing constitute a subclass of two-person, zero-sum, infinite games, where the problem facing the player is not what action to take, but rather when he should take action. The minimax solution to games of timing with incomplete information and equal accuracy functions is presented. An experimental paradigm of such games is provided by a computer-controlled two-person, zero-sum, infinite game that simulates the Western style silent duel. Ten pairs of male subjects each participated for three sessions in a silent duel experiment. Each pair played 450 duels in which both players had the same accuracy functions, but the starting number of bullets available to each player in the dyad was varied systematically. The results of this experiment are analyzed and discussed in terms of variables that relate to predictions arising from the mathematical theory of silent duels. Differences between results from the silent duel and previous research on noisy duels (Kahan & Rapoport, 1971) are discussed.

Relationship patterns in self-analytic groups: a clinical and empirical study.

Abstract: The development of an act-by-act scoring system, process analysis, is presented. An extension of Mann's member-leader system, this scheme reflects the influence of both psychoanalytic and interpersonal theorists and relies heavily on clinical inference. Several measures of interscorer agreement are reported. Two self-analytic classroom groups, 40 sessions in duration, are scored and studied in some detail. Data from the groups are combined, and separate factor analyses of member-leader and member-member interaction are presented. An intensive clinical and statistical assessment of the two factor structures is undertaken. A comparison of these findings with those of other studies of small experiential groups suggests that three bipolar dimensions characterize member-leader relations in such groups: loyalty vs. rebellion; personal involvement vs. neutrality; and a third dimension, more difficult to define, represented in this study by a dimension titled self-absorption vs. ambivalent compliance. Two bipolar dimensions emerged from the analysis of member-member interaction: expression of hostility vs. suppression of hostility and expression of affection vs. suppression of affection. Some differences between this and previous findings are noted. Finally, the strengths, potential, and shortcomings of this research strategy are discussed. The authors conclude that despite the scope, depth, and flexibility of this style of clinical and statistical investigation such research must begin to move toward simpler, more systematic, and less idiosyncratic studies.

Mathematical models of two-party negotiations

Abstract: The present paper describes a pair of mathematical models of two-party bargaining. One model is a Markov process with inputs. The other is a two-party version of the Bush-Mosteller linear operator learning model. The adequacy of the two models is examined in a computer controlled experiment. The Markov chain model fails to predict the extinction of cooperation encountered in certain experimental groups. It does adequately predict the trial of agreement. The learning model describes the proportion of cooperative responses quite accurately in all experimental groups. It is not as accurate in predicting agreement as the Markov process.

Perception of questions and variability of answers

Abstract: This paper presents the results of investigations aimed at determining whether certain evaluations of items have any value in predicting the variability of answers to these items. The method of analysis was based on the concept of minimax discrimination function. This method may have wider applicability to situations where the correlation methods fail. The following results were obtained: the stability of an item is related to negative emotional reaction to the content of the item and to negative previous experience, while the variability is connected with negative intellectual evaluation of both the question and its answer.

Kinship structures revisited

Abstract: The present work adds new dimensions of power and flexibility to an earlier formulation of a programming system that could generate and query kinship structures. The use of the associative memory, parallel processing language, AMPPL-II, and particularly its definitional capability, have proved to be of great help in programming.

A strategy for developing automated assessment systems.

Abstract: Two basic issues underlying the design of automated assessment systems are discussed and the merits of formal as opposed to nonformal, and categorical as opposed to dimensional approaches are stressed. A two-stage strategy, designed to be of use to investigators in particular settings in developing automated assessment systems, is proposed. The first or classification stage is expected to yield viable sets of categories for different populations and is based on a classification algorithm developed by Carlson. The second or identification stage is expected to permit efficient identification of new cases as members of one of the categories obtained in the first stage. It is based on a sequential and Bayesian pattern recognition technique devised by Lasker.