An attractor network model of serial recall

We present a neural network model of verbal working memory which attempts to illustrate how a few simple assumptions about neural computation can shed light on cognitive phenomena associated with the serial recall of verbal material. We assume that neural representations are distributed, that neural connectivity is massively recurrent, and that synaptic efficacy is modified based on the correlation between pre- and post-synaptic activity (Hebbian learning). Together these assumptions give rise to emergent computational properties that are relevant to working memory, including short-term maintenance of information, time-based decay, and similarity-based interference. We instantiate these principles in a specific model of serial recall and show how it can both simulate and explain a number of standard cognitive phenomena associated with the task, including the effects of serial position, word length, articulatory suppression (and its interaction with word length), and phonological similarity.     

A neural theory of visual attention: bridging cognition and neurophysiology

A neural theory of visual attention (NTVA) is presented. NTVA is a neural interpretation of C. Bundesen's (1990) theory of visual attention (TVA). In NTVA, visual processing capacity is distributed across stimuli by dynamic remapping of receptive fields of cortical cells such that more processing resources (cells) are devoted to behaviorally important objects than to less important ones. By use of the same basic equations used in TVA, NTVA accounts for a wide range of known attentional effects in human performance (reaction times and error rates) and a wide range of effects observed in firing rates of single cells in the primate visual system. NTVA provides a mathematical framework to unify the 2 fields of research--formulas bridging cognition and neurophysiology.     

Lesioning an attractor network: investigations of acquired dyslexia

A recurrent connectionist network was trained to output semantic feature vectors when presented with letter strings. When damaged, the network exhibited characteristics that resembled several of the phenomena found in deep dyslexia and semantic-access dyslexia. Damaged networks sometimes settled to the semantic vectors for semantically similar but visually dissimilar words. With severe damage, a forced-choice decision between categories was possible even when the choice of the particular semantic vector within the category was not possible. The damaged networks typically exhibited many mixed visual and semantic errors in which the output corresponded to a word that was both visually and semantically similar. Surprisingly, damage near the output sometimes caused pure visual errors. Indeed, the characteristic error pattern of deep dyslexia occurred with damage to virtually any part of the network.     

Seeing in three dimensions: the neurophysiology of stereopsis

From the pair of 2-D images formed on the retinas, the brain is capable of synthesizing a rich 3-D representation of our visual surroundings. The horizontal separation of the two eyes gives rise to small positional differences, called binocular disparities, between corresponding features in the two retinal images. These disparities provide a powerful source of information about 3-D scene structure, and alone are sufficient for depth perception. How do visual cortical areas of the brain extract and process these small retinal disparities, and how is this information transformed into non-retinal coordinates useful for guiding action? Although neurons selective for binocular disparity have been found in several visual areas, the brain circuits that give rise to stereoscopic vision are not very well understood. I review recent electrophysiological studies that address four issues: the encoding of disparity at the first stages of binocular processing, the organization of disparity-selective neurons into topographic maps, the contributions of specific visual areas to different stereoscopic tasks, and the integration of binocular disparity and viewing-distance information to yield egocentric distance. Some of these studies combine traditional electrophysiology with psychophysical and computational approaches, and this convergence promises substantial future gains in our understanding of stereoscopic vision.     

An acquisition and analysis system for on-line experiments on the neurophysiology of learning

An integrated hardware and software system is used in studying neural correlates of learning. The hardware system is designed to maximize experimental flexibility, while the software system is designed to maximize computer speed for data acquisition and programmer efficiency for data analysis.     

The hippocampus: hub of brain network communication for memory

A complex brain network, centered on the hippocampus, supports episodic memories throughout their lifetimes. Classically, upon memory encoding during active behavior, hippocampal activity is dominated by theta oscillations (6-10Hz). During inactivity, hippocampal neurons burst synchronously, constituting sharp waves, which can propagate to other structures, theoretically supporting memory consolidation. This 'two-stage' model has been updated by new data from high-density electrophysiological recordings in animals that shed light on how information is encoded and exchanged between hippocampus, neocortex and subcortical structures such as the striatum. Cell assemblies (tightly related groups of cells) discharge together and synchronize across brain structures orchestrated by theta, sharp waves and slow oscillations, to encode information. This evolving dynamical schema is key to extending our understanding of memory processes.     

Axiomatic world theory: An overview the general theory of evolution in brief

The world, its many subsystems and all their theories, starting with logic, can be reduced to two related functions: a combinatorial system generator and a hamiltonian system organizer. These can be derived, in turn, from an Axiom of Lawfulness, the expansion being guided by pseudo‐category and pseudo‐functor analysis to produce an axiomatic theory of the world or general theory of evolution. Specifically, world evolution is generated by a constrained combinatorial world generator, F:G_(X), deduced from two related axioms: I. The Axiom of World Lawfulness and II. The Axiom of World Constraint Constants, c = c₁, c₂, of primordial physical combinatee (substance), c₁, and physical combinator (motion), c₂.

Axiom I postulates a lawful analysis by an analyzer adhering to appropriate coordinate systems, CS, of a lawful analysand obeying a conservation law, X = X. The analysand consists of a base combinatee (the set and elements), X = {x₁, x₂,… x_n}, and a base combinator, namely, the universal Boolean operator, NOR = NOT + OR. Base combinatee and combinator both have attributes of quantity combinatorially generated by NOR operating on the universal number, 1, and of quality generated by NOR operating on the universal dimensions, MLT (mass, length, time), including the null sets.

Axiom II fixes the base constants, c, = c₁, c₂, thereby converting X to material substance using c₁ and NOR to material motion using c₂. This comprehensive, quality and quantity‐competent foundational science is called Universal Combinatorics. Its elements comprise the logical alphabet or metavector, A = {c, 1, MLT; X, NOR}, where c is obtained from the remaining terms. These give: (1) the attributive pseudo‐functor, F = P_(c,1,MLT), where P is the power set of the indicated attributes, and (2) the logic generator, G_(X), where G = NOR_(NOR). F then maps G_(X) into world evolution, F:G_(X) → world evolution, as follows:

Expanding the abstract generator, F₁:G,_(x), with world constants eliminated, i.e., c = 0, generates Universal Grammar consisting of (1) the substantive content of the abstract science chain running from linguistic grammar to mathematics and logic and (2) a comprehensive epistemology equivalent to an explicit theory of the strategic aspects of the scientific method, including a universal hamiltonian theory structure informally related to a mathematical category. The four epistemological theorems are:

1. I. The Combinatorial System Generator, F:G_(X), (read as “The attributive functor, F, maps the logic generator, G_(X), into world theory” or “The world is an attributive combinatorial function of logic").

2. II. The Hamiltonian System Operating Theorem, h (an abstract theory‐category structure).

3. III. The System Stability Theorem, PI?, where PI is the extremal Performance Index or controlling law.

4. IV. The Intersystem Abstraction Ranking Theorem given by the Attributive Functor/ Function, F.

F₂ admits the world constants, c > 0, to materialize the grammar generator, G_(X), to an homologous concrete Euler combinatorial physical wave generator, namely, the superstring equation of quantum theory, E_(NI) = A(σ,τ), where E is the permutational function, NI, is the set of nonintegers and the solution is the dual amplitude, σ,τ. Expanding generates the elementary particles of nonadaptive physics and, by inference, the substantive content of Universal Physics consisting of three additional primary systems comprising the world, where a primary system is defined as one having a distinct but derivative extremal controlling law:

1. I. Nonadaptive physics and chemistry (harmonic hamiltonian wave systems) : Minimize Action, subject to conservation constraints.

2. II. Adaptive physics or biology (membrane bound duplicating polymer‐copolymer hamiltonian systems) : Maximize Survival, subject to energy constraints.

3. III. Sentient physics or sociopsychology (neuromatrix hamiltonian systems) : Maximize subjective Happiness, subject to survival constraints and

4. IV. Representational physics or language (a symbolic combinatorial routine): Maximizes the Information Gain, subject to happiness constraints.

The world can then be viewed as a perpetual superfluid computer implicitly using the epistemology of Axiom I as a world program to process the physical data base, c > 0, of Axiom II into world evolution. After evolving through Systems I and II, mankind, i.e., System III, evolves as an internal metacomputer which makes the combinatorial program explicit and uses it to put all four primary systems in standard hamiltonian theory (pseudo‐category) format and terminology. This can be viewed as a generalization of the Darwinian variation‐and‐selection theme in which combinatorial‐variation is recursively hamiltonian‐selected thereby incrementing world logic and logic constraints on successive primary systems. Because Universal Physics and Universal Grammar are functor‐related homologous concrete and abstract combinatorial pseudocategories, related by a pseudo‐functor, thus, differing only in the presence and absence, respectively, of the World Constants, c ≥ 0, they constitute, ipso facto, Universal Science (Formal Philosophy, World Evolution, World Unification, Explicit Theory of Everything, ETOE, or Axiomatic World Theory).

QED: Because intricate verified predictions, ranging from particles to personality types, mental disorders, political parties and the abstract sciences, result from a system which is merely expanding to fill its possibility set, it is concluded that the world is lawful and that this means it is an object deterministic but not fully analytically determinable combinatorial system. In the object domain, the world is system‐number complete at four. Dually, in the analytical codomain, understanding of it is approximately complete, as measured by a world information gain function. Hence, the dualistic, analysand‐analyzer world program is finite and has dualistic completion criteria, as required of an involuted program.     

Semantic information and the network theory of account

The article addresses the problem of how semantic information can be upgraded to knowledge. The introductory section explains the technical terminology and the relevant background. Section 2 argues that, for semantic information to be upgraded to knowledge, it is necessary and sufficient to be embedded in a network of questions and answers that correctly accounts for it. Section 3 shows that an information flow network of type A fulfils such a requirement, by warranting that the erotetic deficit, characterising the target semantic information t by default, is correctly satisfied by the information flow of correct answers provided by an informational source s. Section 4 illustrates some of the major advantages of such a Network Theory of Account (NTA) and clears the ground of a few potential difficulties. Section 5 clarifies why NTA and an informational analysis of knowledge, according to which knowledge is accounted semantic information, is not subject to Gettier-type counterexamples. A concluding section briefly summarises the results obtained.     

Running memory span: A comparison of behavioral capacity limits with those of an attractor neural network

We studied a computational model of short term memory capacity that performs a simulated running memory span task using Hebbian learning and rapid decay of connection strengths to keep recent items active for later recall. This model demonstrates recall performance similar to humans performing the same task, with a capacity limit of approximately three items and a prominent recency effect. The model also shows that this capacity depends on decay to release the model from accumulating interference. Model findings are compared with data from two behavioral experiments that used varying task demands to tax memory capacity limits. Following additional theoretical predictions from the computational model, behavioral data support that when task demands require attention to be spread too thin to keep items available for later recall, capacity limits suffer. These findings are important both for understanding the mechanisms underlying short term memory capacity, and also to memory researchers interested in the role of attention in capacity limitations.     

身与身外:儒家的一个未决问题

Fei Xiaotong's thoughts on the Confucian system of interpersonal relationships actually indicated that the Confucian theory of social cooperation leads itself to an unsettled paradox, that is, there is a lack of universal theoretical construction in the Confucian moral system. Confucian theory does not extend beyond practical circumstances. Instead, its universal principles always disappear in specific circumstances. Because of its long established position in mainstream dialogue, Confucianism failed to reflect on its flaws, but this paradox has been revealed in the face of modern challenges. Translated by Yan Xin from Zhongguo Renmin Daxue Xuebao 中国人民大学学报 (Journal of Renmin University of China), 2007, (1): 15–21     

心理病理学网络理论、方法与挑战

对于精神障碍这一概念的理解, 传统DSM-ICD分类诊断系统和研究领域标准RDoC均基于潜变量视角, 认为精神障碍的症状由其潜在共同原因所致。这2种观点都忽略了症状间的相互作用。不同于分类和维度视角, Borsboom在2008年对精神障碍的概念化提出了的全新视角——心理病理学网络理论。此理论的核心观点是症状之间的动态因果关系构成了精神障碍。基于心理病理学网络理论的网络分析方法, 主要以结合EBIC的glasso算法估计症状间的偏相关网络, 并通过网络中节点中心性与网络连接性等指标, 来考查精神障碍症状的不同特性。近几年来, 研究者发现心理病理学网络分析方法在对症状间因果关系的推断、核心症状的识别和网络结构的可靠性与可重复性方面仍面临一些挑战。这些挑战为心理病理学网络理论与方法指明了未来可能的发展方向。     

Self-disclosure and liking: An integration theory approach

Integration theory models predict that sequences increasing or decreasing in intimacy should have less impact on attraction than constant intimate or nonintimate disclosures. Male and female subjects (N= 159) conversed with a videotaped confederate. Each confederate relayed six disclosures in one of five intimacy sequences: increasing, constant high, decreasing, constant low, and random. The confederate either had a choice of topics and sequence or they were imposed. Substantial support was found for integration theory models. The constant high disclosers by choice were liked more than their no-choice controls and the constant low disclosers by choice were disliked more than their no-choice controls. The mixed sequences (increasing, decreasing, and random) produced no significant differences.     

Interactional value theory: An interpretation

Summary The imperatives of organization, technology, and planning operate similarly in both capitalist and Marxist systems. Differences in behavioural outputs (such as organizational productivity, industrial relations behaviour, or the outcomes of different health services systems) can be explained by adopting a framework of an interactional value theory which accepts convergence of different value systems and points out and analyzes differences in outputs in light of interactions between available climatic techniques and preferred values. The interactional approach links together ethical theory and various issues in social philosophy, especially in the context of the resolution of conflict among values. The dynamics of economic and social change stem from differing values held by key actors in society - government, business and the public. Values are not categorical but integrative, and values held by various parties converge. Further research should be focussed to clarify value-beliefs of government, business and other key players in identifying the interactional characteristics of public policy issues.     

Social network therapies and society: An overview

In this overview of the preceding papers, the author points to common themes: the complex nature of social network intervention; positive and negative effects; and different network constellations that vary with different clinical problems. The author raises the variables of ethnicity and social change as important variables of network effects which remain to be explained.     

At the root of embodied cognition: cognitive science meets neurophysiology

Recent experimental research in the field of neurophysiology has led to the discovery of two classes of visuomotor neurons: canonical neurons and mirror neurons. In light of these studies, we propose here an overview of two classical themes in the cognitive science panorama: James Gibson's theory of affordances and Eleanor Rosch's principles of categorization. We discuss how theoretical perspectives and neuroscientific evidence are converging towards the current paradigm of embodied cognition. From this perspective, we discuss the role of action and simulation in cognitive processes, which lead to the perceptual recognition of objects, and actions and to their conceptual categorization.