Some properties of the output error method

The output error identification method is studied in various respects. The stationary points of the associated loss function are investigated. Sufficient conditions for a unique local minimum are given. The loss functions can be minimized using a quasilinearization algorithm. Such an algorithm will give good local convergence. It is, however, shown that global convergence does not always occur. The output error method is also compared with some other estimation methods from the accuracy point of view. It is proved that a prediction error method will give better accuracy. An instrumental variable technique may give better or worse accuracy depending on the actual noise correlation.     

A theoretical analysis of recursive identification methods

In this paper five different recursive identification methods will be analyzed and compared, namely recursive versions of the least squares method, the instrumental variable method, the generalized least squares method, the extended least squares method and the maximum likelihood method. They are shown to be similar in structure and need of computer storage and time. Making use of recently developed theory for asymptotic analysis of recursive stochastic algorithms, these methods are examined from a theoretical viewpoint. Possible convergence points and their global and local convergence properties are studied. The theoretical analysis is illustrated and supplemented by simulations.     

Individual versus group goal conflict.

"An experiment was performed in which a conflict was produced between individual and group goal attainment. An overt situation was compared to a covert situation, and a high individualistic… with a low individualistic motivation condition… . [The] finding that individuals… feel freer to change secret… than public decisions is discussed." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Team versus individual training, training task fidelity, and task organization effects on transfer performance by three-man teams.

Transfer performance of teams was measured in a simulated radar-controlled aerial intercept task. Superior performance occurred after training on an independently organized task (as compared to that after training which required verbal interaction among controllers), and superior performance occurred in an independently organized transfer task. However, these 2 variables interacted such that performance on an interaction condition of the transfer task was equivalent to that on an independently organized task if prior training was under the independent task organization. Training task fidelity influenced performance only on the interaction transfer task, with superior performance following a high-fidelity training situation in which controllers could acquire the same skills to be required in transfer for communication to interceptor pilots. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Complete Convergence of Competitive Neural Networks with Different Time Scales

This paper studies the complete convergence of a class of neural networks with different time scales under the assumption that the activation functions are unsaturated piecewise linear functions. Under this assumption, there are multiple equilibrium points in the neural network. Traditional methods cannot be used in this neural network. Complete convergence is proved by constructing an energy-like function. Simulations are employed to illustrate the theory.     

A study of the additivity of variables affecting social interaction.

"Groups in which all three variables were predicted to affect interaction in the same direction (i.e., either in favor of or opposed to shift in judgment) showed the greatest consistency in response… . Those Ss for whom two variables pointed to convergence and one opposed it showed more social influence than the Ss for whom two variables opposed convergence. However, this difference was not significant… . Thus prediction of individual behavior in a social interaction situation is possible only when the factors affecting interaction point in the same direction." (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Individual differences in mental activity at sleep onset.

Reports of sleep-onset and nocturnal (EEG Stage 1, REM period) mentation, CPIs, and thematic fantasy responses were collected from 32 young-adult Ss. In confirmation of recent findings by Foulkes and Vogel, dreamlike mental activity was found to occur with fairly substantial frequency at sleep onset, albeit with wide individual differences. Patterning of personality correlates with hypnagogic and nocturnal dream recall suggests that the former varies directly with waking ego strength and adaptive flexibility, while the latter varies directly with indications of waking maladaptive symptomatology. A psychodynamic interpretation of these results is offered. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Individual differences in dream recall.

Ss who typically fail to remember dreams at home (nonreporters) and Ss who frequently remember dreams (reporters) slept in the laboratory for 4 nights each. Gradual or abrupt awakenings were made at each EEG Stage-I REM (dream) period. Although nonreporters and reporters did not differ in REM-period frequency or EEG patterns during sleep, nonreporters did report dreaming less frequently following REM-period awakenings. Ss showed self-consistency in frequency of dream reporting and in type of failure to report. Some nonreporters typically failed to remember any content; others typically said they were awake and thinking. Comparisons among reporters and sub-groups of nonreporters for eye-movement frequency, arousal threshold, and dreamlike-report content indicate that it may be useful to distinguish different kinds of nonreporters. (17 ref.) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

A search for correlates of intra-individual response variability.

"… evidence does not support… finding[s] that variability on free response tests is positively related to… Tolerance of Ambiguity… and is negatively related to… Gregariousness… . The problem is that if intra-individual variability is true variance it seems odd that we cannot find any correlates of it. It is… possible that variability scores, along with response set scores, are measuring aspects of personality that do not appear in self-report scores. The search for personality-measure correlates of variability should, perhaps, be directed toward other measures of style rather than content." 16 refs. (PsycINFO Database Record (c) 2010 APA, all rights reserved)