Parallel distributed processing and neuropsychology: A neural network model of Wisconsin card sorting and verbal fluency

Neural networks can be used as a tool in the explanation of neuropsychological data. Using the Hebbian Learning Rule and other such principles as competition and modifiable interlevel feedback, researchers have successfully modeled a widely used neuropsychological test, the Wisconsin Card Sorting Test. One of these models is reviewed here and extended to a qualitative analysis of how verbal fluency might be modeled, which demonstrates the importance of accounting for the attentional components of both tests. Difficulties remain in programming sequential cognitive processes within a parallel distributed processing (PDP) framework and integrating exceedingly complex neuropsychological tests such as Proverbs. PDP neural network methodology offers neuropsychologists co-validation procedures within narrowly defined areas of reliability and validity.     

Intermodal matching of vision and audition in infancy: A proposal for a new taxonomy

The aim of this review paper is first to devise a common framework for the various procedures used in experimental investigations of intermodal auditory – visual matching by human infants, and second to propose a new taxonomy of intermodal tasks in order to gain a better understanding of the perceptual-cognitive processing underlying these tasks and their relationships to other cognitive achievements such as the development of language. Based on an examination of the tasks used in the developmental literature, we suggest analysing them in terms of (a) their mode of presentation (simultaneous or sequential) for accessing the information in the two modalities (auditory and visual) and (b) the type of relation (amodal or arbitrary) between these two stimulus sources. A review of the literature in the light of this classification shows that most experimental studies employ parallel intermodal presentation (PIP) using both amodal and arbitrary relations between stimuli. Very few investigations have used sequential intermodal presentation (SIP) with amodal relations, and none have used SIP procedures with arbitrary relations. It is claimed here that this last approach might be the most appropriate one for furthering our understanding of (a) the categorical and semantic processing involved in intermodal matching and (b) how the developing infant learns to process words.     

States Are Not Basic Structures: Against State-Centric Political Theory

Abstract

Contemporary political philosophy often operates on a ‘two-tiered’ theoretical treatment of global politics, on which domestic political systems and the principles governing their internal dynamics constitute one tier, and on which the relationships between states and governing multinational institutions constitute a second. One way of grounding and justifying this approach, preferred by Rawls, is called constructivism. Constructivists describe the world as containing specific domains and domain-types of political and social interaction, and relativizes principles of justice to important versions of these—states, in the case of contemporary two-tiered political philosophy. In this paper I argue against the specifically Rawlsian account of uniting these three commitments (two-tiered political theory, constructivism, and statism) and gesture towards a general argument against the coherence of this bundle of views.     

Constraining the Ruler: On Escaping Han Fei's Criticism of Confucian Virtue Politics

One of Han Fei's most trenchant criticisms against the early Confucian political tradition is that, insofar as its decision-making process revolves around the ruler, rather than a codified set of laws, this process is the arbitrary rule of a single individual. Han Fei argues that there will be disastrous results due to ad hoc decision-making, relationship-based decision-making, and decision-making based on prior moral commitments. I lay out Han Fei's arguments while demonstrating how Xunzi can successfully counter them. In doing so, I argue that Xunzi lays out a political theory restricting the actions of the ruler through both the use of ritual and law, which allows him to develop a theory that legitimizes government while at the same time constraining itself. Xunzi's political theory makes important strides in its attempt to recognize the importance of the ruler as a moral exemplar while also restricting his control in the political process.     

Parallel versus serial processing and individual differences in high-speed search in human memory

Many mental tasks that involve operations on a number of items take place within a few hundred milliseconds. In such tasks, whether the items are processed simultaneously (in parallel) or sequentially (serially) has long been of interest to psychologists. Although certain types of parallel and serial models have been ruled out, it has proven extremely difficult to entirely separate reasonable serial and limited-capacity parallel models on the basis of typical data. Recent advances in theory-driven methodology now permit strong tests of serial versus parallel processing in such tasks, in ways that bypass the capacity issue and that are distribution and parameter free. We employ new methodologies to assess serial versus parallel processing and find strong evidence for pure serial or pure parallel processing, with some striking apparent differences across individuals and interstimulus conditions.     

A parallel distributed processing approach to automaticity.

We consider how a particular set of information processing principles, developed within the parallel distributed processing (PDP) framework, can address issues concerning automaticity. These principles include graded, activation-based processing that is subject to attentional modulation; incremental, connection-based learning; and interactivity and competition in processing. We show how simulation models, based on these principles, can account for the major phenomena associated with automaticity, as well as many of those that have been troublesome for more traditional theories. In particular, we show how the PDP framework provides an alternative to the usual dichotomy between automatic and controlled processing and can explain the relative nature of automaticity as well as the fact that seemingly automatic processes can be influenced by attention. We also discuss how this framework can provide insight into the role that bidirectional influences play in processing: that is, how attention can influence processing at the same time that processing influences attention. Simulation models of the Stroop color-word task and the Eriksen response-competition task are described that help illustrate the application of the principles to performance in specific behavioral tasks.     

The effects of feedback format on dynamic decision making

The tendency for people to perform poorly on dynamic decision-making tasks has been attributed to their inability to form adequate mental models of dynamic systems (Sterman, 1994). An alternative explanation is that the tabular feedback formats used in previous studies do not facilitate mental representations of the dynamic task and lead to processing overload. Drawing on research in the human factors area, this study examined the impact of graphical versus tabular feedback formats on performance, response times, and learning on a dynamic inventory control task under differing levels of complexity. The graphical feedback group performed better but the tabular group showed stronger evidence of learning. The two groups did not differ in response times. Records of decision rationales showed that participants were aware of complexity factors but were unable to cope with them effectively under either feedback format. The proximity compatibility principle ( Wickens & Carswell, 1995) from human factors, which argues that feedback must be matched to the processing requirements of the task, provides a useful theoretical framework for the design and testing of feedback formats for organizational tasks and criteria.     

An experimental comparison of viewpoint-specific and viewpoint-independent models of object representation

Four experiments that investigate the cognitive representation of objects in human observers are reported. Two broad classes of theory were examined: viewpoint-specific and viewpoint-independent models. The former postulate that the data structures underpinning object recognition correspond to discrete views and require additional processing to access them from unfamiliar viewpoints. The latter postulate data structures that are independent of any particular viewpoint and can be directly accessed from a wide range of viewpoints. Two experimental tasks were used: a sequential matching paradigm and a cognitive learning paradigm. Findings favour viewpoint-specific models over viewpoint-independent models.     

Social significance of tasks,routines, and pragmatic schemas in distribution activities

Past research on social marking and pragmatic reasoning schemas suggests that cognitive processing modes are first elaborated by children when they are carrying out regulating social routines in the course of which they learn to produce the responses that satisfy the demand of their environment. The data of the two experimental studies reported here with 4–5 year old children, for object distribution tasks, show that the social routines evoked by the objects to be processed have a dual effect, influencing both the representation of the partition to be made, as well as the procedures used to make them. It can be hypothesised that the utilization of routine-evoking tasks, which children are capable at a very young age, of accomplishing by activating efficient social-goal oriented procedures, promote the attribution of an operative meaning to the linguistic expressions used to characterize the states to be attained, and the transformations to be applied in order to attain them.     

Modeling parallelization and flexibility improvements in skill acquisition: from dual tasks to complex dynamic skills

Emerging parallel processing and increased flexibility during the acquisition of cognitive skills form a combination that is hard to reconcile with rule-based models that often produce brittle behavior. Rule-based models can exhibit these properties by adhering to 2 principles: that the model gradually learns task-specific rules from instructions and experience, and that bottom-up processing is used whenever possible. In a model of learning perfect time-sharing in dual tasks (Schumacher et al., 2001), speedup learning and bottom-up activation of instructions can explain parallel behavior. In a model of a complex dynamic task (Carnegie Mellon University Aegis Simulation Program [CMU-ASP], Anderson et al., 2004), parallel behavior is explained by the transition from serially organized instructions to rules that are activated by both top-down (goal-driven) and bottom-up (perceptually driven) factors. Parallelism lets the model opportunistically reorder instructions, leading to the gradual emergence of new task strategies.     

Learning and production of movement sequences: behavioral, neurophysiological, and modeling perspectives

A wave of recent behavioral studies has generated a new wealth of parametric observations about serial order behavior. What was a trickle of neurophysiological studies has grown to a steady stream of probes of neural sites and mechanisms underlying sequential behavior. Moreover, simulation models of serial behavior generation have begun to open a channel to link cellular dynamics with cognitive and behavioral dynamics. Here we review major results from prominent sequence learning and performance tasks, namely immediate serial recall, typing, 2 x N, discrete sequence production, and serial reaction time. These tasks populate a continuum from higher to lower degrees of internal control of sequential organization and probe important contemporary issues such as the nature of working-memory representations for sequential behavior, and the development and role of chunks in hierarchical control. The main movement classes reviewed are speech and keypressing, both involving small amplitude movements amenable to parametric study. A synopsis of serial order models, vis-a-vis major empirical findings leads to a focus on competitive queuing (CQ) models. Recently, the many behavioral predictive successes of CQ models have been complemented by successful prediction of distinctively patterned electrophysiological recordings. In lateral prefrontal cortex, parallel activation dynamics of multiple neural ensembles strikingly matches the parallel dynamics predicted by CQ theory. An extended CQ simulation model--the N-STREAMS neural network model--exemplifies ongoing attempts to accommodate a broad range of both behavioral and neurobiological data within a CQ-consistent theory.     

Thinking Differently Principles of Process in Living Systems and the Specificity of Being Known

As a way of integrating emerging knowledge of biological systems, developmental process, and therapeutic process, we identify principles in the process of exchange between organism and its context of life support that are present at all levels of complexity in living systems, from the cellular to the organization of consciousness. These principles range from specificity, rhythmicity, recurrence, and pattern to coherence, wholeness, and a relative unity in the organization of component parts. By proposing that these principles are also governing the exchange between mother and infant as they negotiate a sequence of essential tasks of adaptation, or “fitting-together” between them over the first years of life, the author suggests that the biological level becomes integrated with the developmental. A sequence of adaptive tasks extends from specificity of recognition in the newborn state, to recognition of inner awareness, purpose, and intention—shaping conscious organization. The bridge to the therapeutic level is constructed as therapist and patient build increasingly inclusive and coherent moments of recognition between themselves at the level of conscious organization, which act as corrective experiences, bringing the patient's own senses of “true self” and of “agency-to-initiate” to new levels of validity and competence.     

RATIONALITY IN POLITICS

This essay examines current decision-making procedures in politics, especially those employed in parliamentary procedure, with a view to determining the extent to which they contribute to the making of rational political decisions. It concludes that political decision-making procedures are, on the whole, inferior to court-trial procedures, and proceeds to exploit this conclusion by describing a new method of political decision-making based on the concept of a political jury. This method, it is claimed, is more likely than present methods to produce sound legislation.     

Democracy and scientific expertise: illusions of political and epistemic inclusion

Realizing the ideal of democracy requires political inclusion for citizens. A legitimate democracy must give citizens the opportunity to express their attitudes about the relative attractions of different policies, and access to political mechanisms through which they can be counted and heard. Actual governance often aims not at accurate belief, but at nonepistemic factors like achieving and maintaining institutional stability, creating the feeling of government legitimacy among citizens, or managing access to influence on policy decision-making. I examine the traditional relationship between inclusiveness and accuracy, and illustrate this connection by discussing empirical work on how group decision-making can improve accuracy. I also advance a Generic Epistemic Principle that any evidence-based decision-making procedures must embrace. Focusing on policy-making, I then measure the distance between these standards and the ones actually implemented in U.S. political settings. Psychological research on individual and group decision-making is a source of normative assessment for existing policy judgment, but it neither rationalizes nor legitimates the actual and typical processes used in U.S. institutions of political decision making. To establish this point, I focus on one characteristic government institution—the U.S. House of Representatives Committee on Science, Space, and Technology—that displays deliberative processes at odds with the sciences they advocate, and with the Generic Epistemic Principle. I explain this discouraging condition in terms of several inveterate factors in U.S. politics: a limitlessly money-driven and endless campaigning process that effectively forces elected representatives to align themselves with money and vote strategically, the use of procedural arrangements known to make people feel politically included when they are not, and the unresponsiveness of a majoritarian (vs. consensus) democracy.     

TOKEN REINFORCEMENT: A REVIEW AND ANALYSIS

Token reinforcement procedures and concepts are reviewed and discussed in relation to general principles of behavior. The paper is divided into four main parts. Part I reviews and discusses previous research on token systems in relation to common behavioral functions—reinforcement, temporal organization, antecedent stimulus functions, and aversive control—emphasizing both the continuities with other contingencies and the distinctive features of token systems. Part II describes the role of token procedures in the symmetrical law of effect, the view that reinforcers (gains) and punishers (losses) can be measured in conceptually analogous terms. Part III considers the utility of token reinforcement procedures in cross‐species analysis of behavior more generally, showing how token procedures can be used to bridge the methodological gulf separating research with humans from that with other animals. Part IV discusses the relevance of token systems to the field of behavioral economics. Token systems have the potential to significantly advance research and theory in behavioral economics, permitting both a more refined analysis of the costs and benefits underlying standard economic models, and a common currency more akin to human monetary systems. Some implications for applied research and for broader theoretical integration across disciplines will also be considered.     

Belief-desire reasoning as a process of selection

Human learning may depend upon domain specialized mechanisms. A plausible example is rapid, early learning about the thoughts and feelings of other people. A major achievement in this domain, at about age four in the typically developing child, is the ability to solve problems in which the child attributes false beliefs to other people and predicts their actions. The main focus of theorizing has been why 3-year-olds fail, and only recently have there been any models of how success is achieved in false-belief tasks. Leslie and Polizzi (Inhibitory processing in the false-belief task: Two conjectures. Developmental Science, 1, 247-254, 1998) proposed two competing models of success, which are the focus of the current paper. The models assume that belief-desire reasoning is a process which selects a content for an agent's belief and an action for the agent's desire. In false belief tasks, the theory of mind mechanism (ToMM) provides plausible candidate belief contents, among which will be a 'true-belief.' A second process reviews these candidates and by default will select the true-belief content for attribution. To succeed in a false-belief task, the default content must be inhibited so that attention shifts to another candidate belief. In traditional false-belief tasks, the protagonist's desire is to approach an object. Here we make use of tasks in which the protagonist has a desire to avoid an object, about which she has a false-belief. Children find such tasks much more difficult than traditional tasks. Our models explain the additional difficulty by assuming that predicting action from an avoidance desire also requires an inhibition. The two processing models differ in the way that belief and desire inhibitory processes combine to achieve successful action prediction. In six experiments we obtain evidence favoring one model, in which parallel inhibitory processes cancel out, over the other model, in which serial inhibitions force attention to a previously inhibited location. These results are discussed in terms of a set of simple proposals for the modus operandi of a domain specific learning mechanism. The learning mechanism is in part modular--the ToMM--and in part penetrable--the Selection Processor (SP). We show how ToMM-SP can account both for competence and for successful and unsuccessful performance on a wide range of belief-desire tasks across the preschool period. Together, ToMM and SP attend to and learn about mental states.     

Towards ethical decision-making in counselling research

This paper explores the process of ethical decision-making in counselling research and asks the question: after following ethical guidelines, reviewing ethical principles and consulting codes of practice, to what extent is our decision-making based on intuitive thinking? The paper begins by examining Kitchener's model of ethical justification and the ethical principles upon which it rests. Bond's approach to ethical problem-solving is also considered. The second part of the paper explores the sequential decision-making process suggested by Gellatt, upon which both the Kitchener and Bond models seem to rest, and compares this with Gellatt's later revision of his model. The process by which we arrive at an ethical decision is then explored in relation to these. The final part of the paper argues that we reach ethical decisions through intuition, informed by ethical principles, codes of practice and reference to the laws of society. In order to safeguard the well-being of our clients and/or research participants, we need to be aware of this and offer up our decision-making process to the scrutiny of peers and the public at large. Although much of the paper explores this process at a general level, research undertaken into children's perception of stress will be used as an example.     

Dissociating automatic associations: Comparing two implicit measurements of race bias

Weak correspondence across different implicit bias tasks may arise from the contribution of unique forms of automatic and controlled processes to response behavior. Here, we examined the correspondence between estimates of automatic and controlled processing derived from two sequential priming tasks with identical structure and timing designed to separately measure stereotypic (Weapons Identification Task; WIT) and evaluative (Affective Priming Task; APT) associations. Across two studies using predominantly White samples, three consistent patterns emerged in the data: (a) stereotypic bias was stronger for Black targets, whereas evaluative bias was stronger for White targets; (b) overall response accuracy bias correlated modestly across the two tasks; and (c) multinomial processing tree estimates of controlled processing corresponded much more strongly than estimates of automatic processing. These findings support models positing distinct learning and memory systems for different forms of race bias, and suggest that these differing forms contribute to estimates of automatic associations.     

Model-Based Wisdom of the Crowd for Sequential Decision-Making Tasks

We study the wisdom of the crowd in three sequential decision-making tasks: the Balloon Analogue Risk Task (BART), optimal stopping problems, and bandit problems. We consider a behavior-based approach, using majority decisions to determine crowd behavior and show that this approach performs poorly in the BART and bandit tasks. The key problem is that the crowd becomes progressively more extreme as the decision sequence progresses, because the diversity of opinion that underlies the wisdom of the crowd is lost. We also consider model-based approaches to each task. This involves inferring cognitive models for each individual based on their observed behavior, and using these models to predict what each individual would do in any possible task situation. We show that this approach performs robustly well for all three tasks and has the additional advantage of being able to generalize to new problems for which there are no behavioral data. We discuss potential applications of the model-based approach to real-world sequential decision problems and discuss how our approach contributes to the understanding of collective intelligence.