Computational models of episodic memory provide tools to better understand the latent neurocognitive processes underlying retention of information about specific events from one’s life. This chapter discusses the representations, associations, and dynamics of influential models of episodic memory, with particular emphasis on models of recognition and free recall tasks. In-depth discussion and model-fitting results of four models – the retrieving effectively from memory (REM) model, the bind cue decide model of episodic memory (BCDMEM), the search of associative memory (SAM) model, and the temporal context model (TCM) – are provided to facilitate understanding of these models, as well as similarities and differences between them. Alternative modeling frameworks, including neural network models, are discussed. Throughout, the importance of context in models of episodic memory is emphasized, particularly for free recall tasks.

Cognitive architectures are computational theories that attempt to cover as much of cognition as possible. Developers of cognitive architectures need to balance functionality of the theory against its explanatory and predictive power, which are often at odds with one another. This chapter discusses this balance with respect to working memory, cognitive performance, perceptual and motor systems, learning and neuroscience. It discusses which solutions six cognitive architectures offer to these areas, and illustrates this with a number of successful models.

This is a relatively comprehensive review of computational modeling work in social psychology and personality psychology, from the beginning of computer modeling in this area in the early sixties, shortly after the founding of artificial intelligence, to the current day.Among the major topics covered are social perception, group perception and stereotyping, attitudes and attitude change, social influence, group behavior, such as group formation and gossip, human mating strategies, culture, the self, and personality. The major modeling techniques used in this area are connectionist models and multi-agent systems.Occasionally researchers use mathematical models.Connectionist models are typically used to simulate intrapersonal processes, such as social perception and attitude change, whereas cellular automata and multi-agent models are typically used to simulate interpersonal processes, such as social influence, gossip, culture, and human mating strategies.

Recent decades have witnessed a rapid growth in computational emotion modeling. Models are being developed to enhance believability and autonomy of virtual agents and robots, and for basic research purposes, to help elucidate mechanisms mediating affective processes in biological agents.This chapter provides a comprehensive introduction and state-of-the-art overview of this emerging subdiscipline within the broader area of affective computing, focusing on models at the psychological (vs. neuroscience) level, and those that emphasize cognition emotion interactions.Following an overview of emotion research from psychology, the theoretical foundations for model design are discussed. An analytical framework is then introduced, to promote a more abstract perspective on model design and analysis, followed by a discussion of specific approaches to modeling emotion generation and emotion effects, along with examples of representative models.The chapter concludes with a discussion of model validation and evaluation, and highlights some of the open questions and key challenges.

In this chapter, we review computer models of cognition that have focused on the use of neural networks. These architectures were inspired by research into how computation works in the brain. The approach is called connectionism because it proposes that processing is characterized by patterns of activation across simple processing units connected together into complex networks, with knowledge stored in the strength of the connections between units. We place connectionism in its historical context, describing the “three ages” of artificial neural network research: from the genesis of the first formal theories of computation in the 1930s and 1940s, to the parallel distributed processing (PDP) models of cognition of the 1980s and 1990s, and the advances in “deep” neural networks emerging in the mid-2000s. Transition between the ages has been triggered by new insights into how to create and train more powerful artificial neural networks. We discuss important foundational cognitive models that illustrate some of the key properties of connectionist systems, and indicate how the novel theoretical contributions of these models arose from their key computational properties. We consider how connectionist modeling has influenced wider theories of cognition, and how in the future, connectionist modeling of cognition may progress by integrating further constraints from neuroscience and neuroanatomy.

This chapter first reviews advanced methods in reinforcement learning (RL), namely, hierarchical RL, distributional RL, meta-RL, RL as inference, inverse RL, and multi-agent RL. Computational and cognitive models based on reinforcement learning are then presented, including detailed models of the basal ganglia, variety of dopamine neuron responses, roles of serotonin and other neuromodulators, intrinsic reward and motivation, neuroeconomics, and computational psychiatry.

This chapter begins with an outline of logic and of the attempts to use it as a theory of human deduction. The fatal impediments to this approach led to the model theory in which models based on the meanings of premises yield deductive conclusions. And the chapter describes in detail the implementation of this theory’s account of deductions based on sentential connectives such as “if,” and how this simulation led to the discovery of systematic but compelling fallacies.The chapter outlines how algorithms based on models simulate deductions of the spatial relations among objects. And it points out the problems that need to be solved to extend accounts of elementary inferences from quantified assertions to deal with multiply-quantified relations. One alternative to the model theory is the idea that human deduction relies on probabilities. This approach concerns only which inferences people make, not the underlying mental processes by which they are made. The model theory fills the gap, because it applies to the deductions of probabilities, both those based on frequencies or proportions, and those based on evidence pertinent to unique events. The chapter ends with an account of why theories of human deduction need to be simulated in computer programs.

For decades, symbolic models of cognition were the dominant computational approaches of cognition. Today they coexist with subsymbolic, statistical, and hybrid models, but they are still the de facto standard for modeling human reasoning processes. This chapter summarizes important aspects of symbolic and hybrid models of cognition, approaching the topic from different perspectives. After some discussion on historical aspects and the theoretical basis of symbolic models of cognition, cognitive architectures as models for intelligent agents are examined. Subsequently, the role of symbolic computational approaches towards processing natural language, representation of human knowledge, and commonsense reasoning are considered. Then the focus is put on the crucial question of learning new representations and theories, before finally looking at hybrid and neural-symbolic systems combining reasoning and learning and bridging between symbolic and subsymbolic elements.

Computational psycholinguistics seeks to develop computational theories and implemented models of the cognitive systems that map an unfolding linguistic signal onto a mental representation of its meaning. Focusing primarily on language comprehension, this chapter begins with early theories of sentence processing, before reviewing several prominent implemented computational models. These accounts are largely informed by reading-time studies that seek to establish the strategies and constraints that determine how people resolve ambiguity. This review concludes with a more in-depth discussion of rational probabilistic accounts, for which there has been considerable consensus in recent years, and surprisal theory, which formally links these models with measures of human comprehension effort, such as reading times and brain potentials. Finally, an implemented neurobehavioral model of language comprehension is presented in greater detail, illustrating the benefit of linking computational models with several behavioral and neurophysiological indices of human comprehension, as well as the importance of looking beyond syntactic processing alone to the modeling of semantic comprehension and the role of world knowledge.

This chapter introduces computational models of decision making as worthy successors to the traditional, algebraic utility framework that has dominated the field. It provides an overview of several different computational modeling approaches before providing a detailed example of perhaps the most well-established of these, based on sequential sampling of information and evidence accumulation. It is shown how this approach can account for common paradoxes in decision behavior, and how it can be extended to a variety of tasks and response modes while retaining the same basic cognitive principles. The chapter concludes with an illustration of how process-tracing methods that capture the information acquisition and response processes can help to evaluate computational models of decision making and discriminate among them.

Social Justice Education (SJE) has become the defining orientation of many educators and educational researchers, but is not without its detractors. Because of its overt political investments, SJE has been accused of brainwashing students and violating the terms of democratic legitimacy. In this chapter, I offer a philosophical defense of some SJE. Using Canada as an example and comprehensive liberalism as a framework, I argue that many practices that we wish to protect under the banner of SJE can be defended by appeal to the foundational values that are common to liberal democracies and find expression in contemporary legislation. I suggest five criteria for distinguishing between defensible and indefensible forms of political education, allowing that not all self-proclaimed SJE will be defensible, and some less progressive education will be. I conclude by anticipating two objections to this strategy.

This chapter presents a systematic theory of generalized (or universal) Fechnerian scaling, based on the intuition underlying Fechner’s original theory. The intuition is that subjective distances among stimuli are computed by cumulating small discriminability values between “neighboring” stimuli. A stimulus space is supposed to be endowed by a dissimilarity function, computed from a discrimination probability function for any pair of stimuli chosen in two distinct observation areas. On the most abstract level, one considers all possible chains of stimuli leading from stimulus a to stimulus b and back to a, and takes the infimum of the sums of the dissimilarities along these chains as the subjective distance between a and b. In arc-connected spaces, the cumulation of dissimilarity values along all possible chains reduces to their cumulation along continuous paths, leading to a fully fledged metric geometry. In topologically Euclidean spaces, the cumulation along paths further reduces to integration along smooth paths, and the geometry in question acquires the form of a generalized Finsler geometry. The chapter also discusses Fechner’s original derivation of his logarithmic law, observation sorites paradox, a generalized Floyd--Warshall algorithm for computing metric distances from dissimilarities, and an ultra-metric version and data-analytic application of Fechnerian scaling.

This chapter offers an overview of some of the more important approaches to these questions in contemporary, mostly anglophone, conceptions of educational justice in primary and secondary education. Section 16.2 starts with some provisions of some important goals of education. Section 16.3 turns to educational justice in general. Section 16.4 asks about the spheres of educational justice: is it education and socialization in general, or the school system in particular? Section 16.5 distinguishes three different levels of education: basic education for all; the cultivation of individual talents and capacities; and selection for higher education and the job market. Section 16.6 outlines the differences between five principles of justice and equality in the field of education: strict equality; a conception of fair equality of opportunity, iii) a conception of luck-egalitarian equality of opportunity; iv) a prioritarian conception of educational justice; and democratic adequacy as a conception of educational justice.