Akaike's Information Criterion and Recent Developments in Information Complexity

In this paper we briefly study the basic idea of Akaike's (1973) information criterion (AIC). Then, we present some recent developments on a new entropic or information complexity (ICOMP) criterion of Bozdogan (1988a, 1988b, 1990, 1994d, 1996, 1998a, 1998b) for model selection. A rationale for ICOMP as a model selection criterion is that it combines a badness-of-fit term (such as minus twice the maximum log likelihood) with a measure of complexity of a model differently than AIC, or its variants, by taking into account the interdependencies of the parameter estimates as well as the dependencies of the model residuals. We operationalize the general form of ICOMP based on the quantification of the concept of overall model complexity in terms of the estimated inverse-Fisher information matrix. This approach results in an approximation to the sum of two Kullback-Leibler distances. Using the correlational form of the complexity, we further provide yet another form of ICOMP to take into account the interdependencies (i.e., correlations) among the parameter estimates of the model. Later, we illustrate the practical utility and the importance of this new model selection criterion by providing several real as well as Monte Carlo simulation examples and compare its performance against AIC, or its variants. Copyright 2000 Academic Press.     

旧哲学和新哲学的比較

在本文中作者把他所谓的新哲学(又称之为科学的哲学)和旧哲学(或传统哲学、思辨哲学)加以比较。他认为新哲学的贡献乃在于对传统哲学所提出的问题作了科学的回答,他所举的得到回答的问题有:几何知识的起源问题,因果关系的有效性问题,物质的本质的问题,支配进化的原理的问题;逻辑的本质的问题,关于预言的知识的问题,外在世界和人的心灵的存在的问题,伦理学的本质的问题。作者用了较多的篇幅阐述他对伦理学的本质的看法。他认为要用类似于解决几何学问题的方法来解决伦理学的问题。他主张区分开伦理学的公理(或前提)和伦理学的蕴涵,认为只有蕴涵是能用逻辑证明的。他认为几何学与伦理学有一个基本的差别:几何公理可能成为真命题,含有经验的真理;而伦理学的公理完全不能成为认识的命题,不能根据任何一种解释而被称为真的,它们是意志的决定。     

邏輯經驗主义的科学的哲学中的几个主要爭論点及其發展

费格尔(Herbert Feigl)是维也纳学派的创始人之一,后迁居美国,1940年起任明尼苏答大学哲学教授。此文是1954年他在科学的哲学国际大会上的一篇演说的修正和扩大稿。载《明尼苏答科学的哲学研究》第一卷(明尼苏答大学出版社,1956)。他首先简单地介绍了一下逻辑经验主义的历史,以及它在当前的三大趋向。然后着重讨论了科学的哲学运动中三个主要争论:第一个争论是关于分析命题和综合命题的区别是否必要的问题,作者认为是必要的,因为这是区别形式科学和事实科学的基础;第二个争论是关于经验主义的意义标准问题;第三个争论是关于科学是否需要形而上学前提的问题。讨论是极其烦琐的,但从讨论中也反映出逻辑实证主义的一些重要观点,如经验的证实,否认客观世界的规律性等等。最后作者指出了当前逻辑经验主义的一些特点:更加逻辑的、更加肯定的、更加经验的。     

给一位苏联哲学家的信

本文作者于一九六一年访问苏联,与苏联哲学研究所穆施文尼拉茲教授作了会说,双方约定以公开信的形式陈述各自的观点,达里译出的是作者写给穆施文尼拉茲的公开信。作者否认存在和意识问题是哲学的根本问题,甚至认为“不该把它看作一个有意义的问题”。普通语义学的根本问题牵涉到精神健全的性质,它指示出在创立一门关于人的科学以及在促进健全思想方面必须遵循的方向。作者佯称承认客观世界的存在乃是普通语义学的基础,但又否认这个承认构成了“哲学根本问题的一个解决”。他把这种承认简单地看作是一个公理,一种约定,一个方便的起点,或者是为了使哲学讨论能够平静地进行而大家同意保持的一个契约。     

无产阶级专政与全民国家

近几年来,我们党作了巨大的理论工作。苏共第二十二次代表大会给马克思列宁主义的宝库带来了特别重大的贡献。达次代表大会提出了一系列对共产主义建设实践有极其重要意义的新原理和结论。这些结论之一是:无产阶级专政的国家由于社会主义的完全和最终的胜利而转变为全民国家。达一结论完全建筑在马克思列宁主义的根本原理之     

Time estimation: does the reference memory mediate the effect of knowledge of results?

This study presents evidence for the role of knowledge of results (KR) in the estimation of medium time intervals (4-12 s durations). Experiment 1 tested the hypothesis that KR operates at the same stage of temporal processing as attention. The absence of an interaction of attention and KR is taken as evidence against this hypothesis. Experiments 2 and 3 further studied the effects of KR in a reproduction and a production task, respectively. The effects of KR were small in the reproduction and substantial in the production task. The results are interpreted in terms of a clock-based timing model and the findings taken together converge on the interpretation that KR affects the reference memory (RM), rather than the other components of the model.     

Spatial selection in peripheral letter recognition: in search of boundary conditions

Recognition performance for a peripherally presented target letter embedded in a letter string is worse than for targets presented in isolation. This lateral masking effect is commonly attributed to impairments when identifying flanked letters. The hypothesis that also failures during spatial selection of the target underlie lateral masking effects was examined by varying the brightness of the flankers (Experiment 1), by cueing the target position at various times before, during, or after presentation of the string (Experiment 2) and by avoiding the abrupt on- and offset of the flankers (Experiments 3 and 4). Whereas the flankers' brightness had no effect on and cueing only slightly reduced the lateral masking effects, it was substantially reduced by avoiding the abrupt on- and offset of the flankers. The results show that spatial selection is impaired in peripheral vision, and that this is one important factor underlying lateral masking effects.     

The priority of primary process categorizing: experimental evidence supporting a psychoanalytic developmental hypothesis

Earlier work has provided experimental evidence for the existence of the primary and secondary process mental organization posited by Freud and has demonstrated that primary process effects are the more active unconsciously (Brakel et al. 2000). Primary and secondary processes were assessed by a categorization test in which qualitatively different principles could be used. In new experiments using the same stimuli, another significant implication of Freud's model was tested: that primary process mental organization has developmental priority. In these experiments, which studied 559 participants ranging in age from 3 to 80, it was found (1) that primary process mentation predominates in preschoolers; (2) that it is not until around age 7 that primary process organization is supplanted by secondary process organization; and (3) that after age 7 the predominance of secondary process organization remains remarkably stable throughout the life span.     

Mourning analysis: the post-termination phase

One hundred twenty-one analytic candidates who had completed training analysis responded to a survey about their post-termination experience. Seventy-six percent of respondents experienced a mourning process that lasted on average between six months and a year, while 24 per cent experienced no discernible sense of painful loss. Twenty candidates were interviewed to obtain a deeper understanding of the mourning process that follows analysis. During the post-termination phase, the analysand's self-analytic capacity is tested in the struggle to contain and understand feelings about the loss of the analyst, as well as transference reactions triggered by that loss. After a "good-enough analysis," the analysand internalizes not only the analyst's functions and attitudes toward him or her, but also a sustaining, positive internal image of the analyst. Four cases illustrate unexpected difficulties that may emerge during the post-termination phase when the loss of the analyst is experienced as a repetition of earlier, traumatic losses or as a rupture of an unanalyzed, selfobject transference.