‘Numerical trap’. A new look at outcome representation in studies on choice behaviour

In almost all experimental studies on choice behaviour the consequences of choices are limited to denumerable goods (usually money) and represented in a numerical way. It is argued that a different way of representing consequences would cause differences in social orientations. In an exploratory study (1) the kind of consequences (money, satisfaction with money and satisfaction with the situation) and (2) the way of representing consequences (numbers, rectangles, and faces) were varied. No differences were found between the three kinds of consequences represented by rectangles. Representation of money by numbers did not differ from representation of money by rectangles. However, subjects in the condition in which satisfaction with money was represented by faces were more cooperative and altruistic and less equality-oriented than subjects in the corresponding condition in which satisfaction with money was represented by rectangles. Additional correlational analyses showed differences between ways of representing consequences. It was concluded that both for theory and for future research it is important to consider how consequences are represented.     

On two types of conditional reflex: General laws of association

Both types of conditioning are based on the general laws of associations-connections between the centers involved. Whereas the experimental procedures of classical conditioning expose mainly the conditioned stimulus (CS)-unconditioned stimulus (US) connection, those of instrumental conditioning expose the conditioned stimulus (CS)-response (R) connection. Thus, the main differences between the two types of conditioning are those associated with the different centers involved in each, not the associative-connective laws themselves.     

Data compression in machine learning applied to natural language

In this paper, we investigate the possibility of applying machine learning methods to data derived from the area of natural language and show how rules, induced by machine learning, are changed after the original data are compressed by grouping together entries, attributes, and attribute values. Also shown is how excessive compression of input data may affect the accuracy of induced rules.     

Logics preserving degrees of truth

The paper introduces a concept of logic applied to a formalization of the so-called inferences preserving degrees of truth. Semantical and syntactical characterizations of three kinds of logics preserving degrees of truth are provided. The other approach than in [3] and [9] to the problem of expressing that a sentence is less true than a sentence is presented.     

The emergence of personality: Dynamic foundations of individual variation

We conceptualize personality and individual variation from the perspective of dynamical systems. People’s thoughts, feelings, and predispositions for action are inherently dynamic, displaying constant change due to internal mechanisms and external forces, but over time the flow of thought and action converges on a narrow range of states—a fixed-point attractor—that provides cognitive, affective, and behavioral stability. An attractor for personal dynamics develops through two mechanisms: the synchronization of individuals’ internal states in social interaction, and the self-organization of thoughts and feelings with respect to a higher-order property (e.g., goal, self-concept). We present formal models of both processes and instantiate each in computer simulations. Discussion centers on the implications of interpersonal synchronization and self-organization dynamics for issues in personality psychology, including shared vs. non-shared environmental influences on personality development, the expression of personality in social interaction, personal stability vs. change, personal vs. situational causation, and the emergence of self-concept.