We begin by distinguishing computationalism from a number of other theses that are sometimes conflated with it. We also distinguish between several important kinds of computation: computation in a generic sense, digital computation, and analog computation. Then, we defend a weak version of computationalism—neural processes are computations in the generic sense. After that, we reject on empirical grounds the common assimilation of neural computation to either analog or digital computation, concluding that (...) class='Hi'>neural computation is sui generis. Analog computation requires continuous signals; digital computation requires strings of digits. But current neuroscientific evidence indicates that typical neural signals, such as spike trains, are graded like continuous signals but are constituted by discrete functional elements (spikes); thus, typical neural signals are neither continuous signals nor strings of digits. It follows that neural computation is sui generis. Finally, we highlight three important consequences of a proper understanding of neural computation for the theory of cognition. First, understanding neural computation requires a specially designed mathematical theory (or theories) rather than the mathematical theories of analog or digital computation. Second, several popular views about neural computation turn out to be incorrect. Third, computational theories of cognition that rely on non-neural notions of computation ought to be replaced or reinterpreted in terms of neural computation. (shrink)

The segregation of image parts into foreground and background is an important aspect of the neural computation of 3D scene perception. To achieve such segregation, the brain needs information about border ownership; that is, the belongingness of a contour to a specific surface represented in the image. This article presents psychophysical data derived from 3D percepts of figure and ground that were generated by presenting 2D images composed of spatially disjoint shapes that pointed inward or outward relative to (...) the continuous boundaries that they induced along their collinear edges. The shapes in some images had the same contrast (black or white) with respect to the background gray. Other images included opposite contrasts along each induced continuous boundary. Psychophysical results demonstrate conditions under which figure-ground judgment probabilities in response to these ambiguous displays are determined by the orientation of contrasts only, not by their relative contrasts, despite the fact that many border ownership cells in cortical area V2 respond to a preferred relative contrast. Studies are also reviewed in which both polarity-specific and polarity-invariant properties obtain perceptual figure-ground grouping results. The FACADE and 3D LAMINART models are used to explain these data. Keywords: figure-ground separation, border ownership, perceptual grouping, surface filling-in, V2, V4, FACADE Theory. (shrink)

Problems are raised with the global workspace hypothesis of consciousness, for example about exactly how global the workspace needs to be for consciousness to suddenly be present. Problems are also raised with Carruthers's version that excludes conceptual representations, and in which phenomenal consciousness can be reduced to physical processes, with instead a different levels of explanation approach to the relation between the brain and the mind advocated. A different theory of phenomenal consciousness is described, in which there is a particular (...) computational system involved in which Higher Order Syntactic Thoughts are used to perform credit assignment on first order thoughts of multiple step plans to correct them by manipulating symbols in a syntactic type of working memory. This provides a good evolutionary reason for the evolution of this kind of computational module, with which, it is proposed, phenomenal consciousness is associated. Some advantages of this HOST approach to phenomenal consciousness are then described with reference not only to the global workspace approach, but also to Higher Order Thought theories. It is hypothesized that the HOST system which requires the ability to manipulate first order symbols in working memory might utilize parts of the prefrontal cortex implicated in working memory, and especially the left inferior frontal gyrus, which is involved in language and probably syntactical processing. Overall, the approach advocated is to identify the computations that are linked to consciousness, and to analyze the neural bases of those computations. [copyright: Copyright © 2020 Rolls.]. (shrink)

Biological neural computation relies a great deal on architecture, which constrains the types of content that can be processed by distinct modules in the brain. Though artificial neural networks are useful tools and give insight, they cannot be relied upon yet to give definitive answers to problems in cognition. Knowledge re-use may be driven more by architectural inheritance than by epistemological drives.

In a recent paper, Kaplan (Synthese 183:339–373, 2011) takes up the task of extending Craver’s (Explaining the brain, 2007) mechanistic account of explanation in neuroscience to the new territory of computational neuroscience. He presents the model to mechanism mapping (3M) criterion as a condition for a model’s explanatory adequacy. This mechanistic approach is intended to replace earlier accounts which posited a level of computational analysis conceived as distinct and autonomous from underlying mechanistic details. In this paper I discuss work in (...) computational neuroscience that creates difficulties for the mechanist project. Carandini and Heeger (Nat Rev Neurosci 13:51–62, 2012) propose that many neural response properties can be understood in terms of canonical neural computations. These are “standard computational modules that apply the same fundamental operations in a variety of contexts.” Importantly, these computations can have numerous biophysical realisations, and so straightforward examination of the mechanisms underlying these computations carries little explanatory weight. Through a comparison between this modelling approach and minimal models in other branches of science, I argue that computational neuroscience frequently employs a distinct explanatory style, namely, efficient coding explanation. Such explanations cannot be assimilated into the mechanistic framework but do bear interesting similarities with evolutionary and optimality explanations elsewhere in biology. (shrink)

Computationalism about the brain is the view that the brain literally performs computations. For the view to be interesting, we need an account of computation. The most well-developed account of computation is Turing Machine computation, the account provided by theoretical computer science which provides the basis for contemporary digital computers. Some have thought that, given the seemingly-close analogy between the all-or-nothing nature of neural spikes in brains and the binary nature of digital logic, neural (...) class='Hi'>computation could be a species of digital computation. A few recent authors have offered arguments against this idea; here, I review recent findings in neuroscience that further cement the implausibility of this view. However, I argue that we can retain the view that the brain is a computer if we expand what we mean by “computation” to include analog computation. I articulate an account of analog computation as the manipulation of analog representations based on previous work on the difference between analog and non-analog representations, extending a view originally articulated in Shagrir :271–279, 2010). Given that analog computation constitutes a significant chapter in the history of computation, this revision of computationalism to include analog computation is not an ad hoc addition. Brains may well be computers, but of the analog kind, rather than the digital kind. (shrink)

The relationship between the neurobiology of speech and music has been investigated for more than a century. There remains no widespread agreement regarding how (or to what extent) music perception utilizes the neural circuitry that is engaged in speech processing, particularly at the cortical level. Prominent models such as Patel’s Shared Syntactic Integration Resource Hypothesis (SSIRH) and Koelsch’s neurocognitive model of music perception suggest a high degree of overlap, particularly in the frontal lobe, but also perhaps more distinct representations (...) in the temporal lobe with hemispheric asymmetries. The present meta-analysis study used activation likelihood estimate analyses to identify the brain regions consistently activated for music as compared to speech across the functional neuroimaging (fMRI and PET) literature. Eighty music and 91 speech neuroimaging studies of healthy adult control subjects were analyzed. Peak activations reported in the music and speech studies were divided into four paradigm categories: passive listening, discrimination tasks, error/anomaly detection tasks and memory-related tasks. We then compared activation likelihood estimates within each category for music versus speech, and each music condition with passive listening. We found that listening to music and to speech preferentially activate distinct temporo-parietal bilateral cortical networks. We also found music and speech to have shared resources in the left pars opercularis but speech-specific resources in the left pars triangularis. The extent to which music recruited speech-activated frontal resources was modulated by task. While there are certainly limitations to meta-analysis techniques particularly regarding sensitivity, this work suggests that the extent of shared resources between speech and music may be task-dependent and highlights the need to consider how task effects may be affecting conclusions regarding the neurobiology of speech and music. (shrink)

Aicardi et al. look to neuroscience to mitigate the limitations of current robotics technology. They propose that robotics technology guided by neuroscience has the capacity to create intelligent robots that function with awareness and capacity for abstraction and reasoning. As neurorobotics extends the capability of robotics technology, it introduces new social and ethical concerns, in particular co-opting civilian applications for military use, conflicts between industry and the academy, and data security. However, here we argue that empirical evidence has shown that (...) human cognition is faulty; therefore there is not a clear motivation to build intelligent robots on a human model; representation of meaning in the brain is not well-understood; therefore neuro-robotics is limited; and to the extent that intelligent robots become a reality, the ethics of robot rights will be of central concern. (shrink)

We present techniques for integrating low‐level neurobiological constraints into high‐level, functional cognitive models. In particular, we use these techniques to construct a model of eyeblink conditioning in the cerebellum based on temporal representations in the recurrent Granule‐Golgi microcircuit.

Contextual signals might supervise the discovery of coherently varying information between cortical modules computing different functions of their receptive field input. This hypothesis is explored in two sets of computational experiments, one studying the effects on learning of long-range unidirectional contextual signals mediated by intervening processors, and the other showing contextually supervised discovery of a high-order variable in a multilayer network.

Although it has been argued that mechanistic explanation is compatible with abstraction, there are still doubts about whether mechanism can account for the explanatory power of significant abstract models in computational neuroscience. Chirimuuta has recently claimed that models describing canonical neural computations must be evaluated using a non-mechanistic framework. I defend two claims regarding these models. First, I argue that their prevailing neurocognitive interpretation is mechanistic. Additionally, a criterion recently proposed by Levy and Bechtel to legitimize mechanistic abstract models, (...) and also a criterion proposed by Chirimuuta herself aimed to distinguish between causal and non-causal explanation, can be employed to show why these models are explanatory only under this interpretation. Second, I argue that mechanism is able to account for the special epistemic achievement implied by CNC models. Canonical neural components contribute to an integrated understanding of different cognitive functions. They make it possible for us to explain these functions by describing different mechanisms constituted by common basic components arranged in different ways. (shrink)

A wide range of systems appear to perform computation: what common features do they share? I consider three examples, a digital computer, a neural network and an analogue route finding system based on soap-bubbles. The common feature of these systems is that they have autonomous dynamics — their states will change over time without additional external influence. We can take advantage of these dynamics if we understand them well enough to map a problem we want to solve onto (...) them. Programming consists of arranging the starting state of a system so that the effects of the system''s dynamics on some of its variables corresponds to the effects of the equations which describe the problem to be solved on their variables. The measured dynamics of a system, and hence the computation it may be performing, depend on the variables of the system we choose to attend to. Although we cannot determine which are the appropriate variables to measure in a system whose computation basis is unknown to us I go on to discuss how grammatical classifications of computational tasks and symbolic machine reconstruction techniques may allow us to rule out some measurements of a system from contributing to computation of particular tasks. Finally I suggest that these arguments and techniques imply that symbolic descriptions of the computation underlying cognition should be stochastic and that symbols in these descriptions may not be atomic but may have contents in alternative descriptions. (shrink)

``Neural computing'' is a research field based on perceiving the human brain as an information system. This system reads its input continuously via the different senses, encodes data into various biophysical variables such as membrane potentials or neural firing rates, stores information using different kinds of memories (e.g., short-term memory, long-term memory, associative memory), performs some operations called ``computation'', and outputs onto various channels, including motor control commands, decisions, thoughts, and feelings. We show a natural model of (...) neural computing that gives rise to hyper-computation. Rigorous mathematical analysis is applied, explicating our model's exact computational power and how it changes with the change of parameters. Our analog neural network allows for supra-Turing power while keeping track of computational constraints, and thus embeds a possible answer to the superiority of the biological intelligence within the framework of classical computer science. We further propose it as standard in the field of analog computation, functioning in a role similar to that of the universal Turing machine in digital computation. In particular an analog of the Church-Turing thesis of digital computation is stated where the neural network takes place of the Turing machine. (shrink)

I address whether neural networks perform computations in the sense of computability theory and computer science. I explicate and defend
the following theses. (1) Many neural networks compute—they perform computations. (2) Some neural networks compute in a classical way.
Ordinary digital computers, which are very large networks of logic gates, belong in this class of neural networks. (3) Other neural networks
compute in a non-classical way. (4) Yet other neural networks do not perform computations. Brains may well (...) fall into this last class. (shrink)

In this paper I criticize a view of functional localization in neuroscience, which I call “computational absolutism”. “Absolutism” in general is the view that each part of the brain should be given a single, univocal function ascription. Traditional varieties of absolutism posit that each part of the brain processes a particular type of information and/or performs a specific task. These function attributions are currently beset by physiological evidence which seems to suggest that brain areas are multifunctional—that they process distinct information (...) and perform different tasks depending on context. Many theorists take this contextual variation as inimical to successful localization, and claim that we can avoid it by changing our functional descriptions to computational descriptions. The idea is that we can have highly generalizable and predictive functional theories if we can discover a single computation performed by each area regardless of the specific context in which it operates. I argue, drawing on computational models of perceptual area MT, that this computational version of absolutism fails to come through on its promises. In MT, the modeling field has not produced a univocal computational description, but instead a plurality of models analyzing different aspects of MT function. Moreover, CA cannot appeal to theoretical unification to solve this problem, since highly general models, on their own, neither explain nor predict what MT does in any particular context. I close by offering a perspective on neural modeling inspired by Nancy Cartwright’s and Margaret Morrison’s views of modeling in the physical sciences. (shrink)

Cognitive control has long been one of the most active areas of computational modeling work in cognitive science. The focus on computational models as a medium for specifying and developing theory predates the PDP books, and cognitive control was not one of the areas on which they focused. However, the framework they provided has injected work on cognitive control with new energy and new ideas. On the occasion of the books' anniversary, we review computational modeling in the study of cognitive (...) control, with a focus on the influence that the PDP approach has brought to bear in this area. Rather than providing a comprehensive review, we offer a framework for thinking about past and future modeling efforts in this domain. We define control in terms of the optimal parameterization of task processing. From this vantage point, the development of control systems in the brain can be seen as responding to the structure of naturalistic tasks, through the filter of the brain systems with which control directly interfaces. This perspective lays open a set of fascinating but difficult research questions, which together define an important frontier for future computational research. (shrink)

It is generally accepted that, in the cognitive and neural sciences, there are both computational and mechanistic explanations. We ask how computational explanations can integrate into the mechanistic hierarchy. The problem stems from the fact that implementation and mechanistic relations have different forms. The implementation relation, from the states of an abstract computational system to the physical, implementing states is a homomorphism mapping relation. The mechanistic relation, however, is that of part/whole; the explaining features in a mechanistic explanation are (...) the components of the explanandum phenomenon and their causal organization. Moreover, each component in one level of mechanism is constituted and explained by components of an underlying level of mechanism. Hence, it seems, computational variables and functions cannot be mechanistically explained by the medium-dependent states and properties that implement them. How then, do the computational and the implementational integrate to create the mechanistic hierarchy? After explicating the general problem, we further demonstrate it through a concrete example, of reinforcement learning, in the cognitive and neural sciences. We then examine two possible solutions. On one solution, the mechanistic hierarchy embeds at the same levels computational and implementational properties. This picture fits with the view that computational explanations are mechanistic sketches. On the other solution, there are two separate hierarchies, one computational and another implementational, which are related by the implementation relation. This picture fits with the view that computational explanations are functional and autonomous explanations. It is less clear how these solutions fit with the view that computational explanations are full-fledged mechanistic explanations. Finally, we argue that both pictures are consistent with the reinforcement learning example, but that scientific practice does not align with the view that computational models are merely mechanistic sketches. (shrink)

Scientists, engineers, and healthcare professionals are currently developing a variety of new devices under the category of brain-computer interfaces (BCIs). Current and future applications are both medical/assistive (e.g., for communication) and non-medical (e.g., for gaming). This array of possibilities comes with ethical challenges for all stakeholders. As a result, BCIs have been an object of both hope and concern in various media. We argue that these conflicting sentiments can be productively understood in terms of personhood, specifically the impact of BCIs (...) on what it means to be a person and to be recognized as such by others. To understand the dynamics of personhood in the context of BCI use and investigate whether ethical guidance is required, a meeting entitled "BCIs and Personhood: A Deliberative Workshop" was held in May 2018. In this article, we describe how BCIs raise important questions about personhood and propose recommendations for BCI development and governance. (shrink)

Fluid neural networks can be used as a theoretical framework for a wide range of complex systems as social insects. In this article we show that collective logical gates can be built in such a way that complex computation can be possible by means of the interplay between local interactions and the collective creation of a global field. This is exemplified by a NOR gate. Some general implications for ant societies are outlined. ©.

A basic deep neural network (DNN) is trained to exhibit a large set of input–output dispositions. While being a good model of the way humans perform some tasks automatically, without deliberative reasoning, more is needed to approach human‐like artificial intelligence. Analysing recent additions brings to light a distinction between two fundamentally different styles of computation: content‐specific and non‐content‐specific computation (as first defined here). For example, deep episodic RL networks draw on both. So does human conceptual reasoning. Combining (...) the two takes advantage of the complementary costs and benefits of each. It also offers a better model of human cognitive competence. (shrink)

Nijhawan argues that neural compensation is necessary to account for couplings of perception and action. Although perhaps true in some cases, computational tolerance for asynchronously arriving continuous information is of more importance. Moreover, some of the everyday venues Nijhawan uses to argue for the relevance of prediction and compensation can be better ascribed to skill.

Neurulation, the curling of the neuroepithelium to form the neural tube, is an essential component of the development of animal embryos. Defects of neural tube formation, which occur with an overall frequency of one in 500 human births, are the cause of severe and distressing congenital abnormalities. However, despite the fact that there is increasing information from animal experiments about the mechanisms which effect neural tube formation, much less is known about the fundamental causes of neural (...) tube defects (NTD). The use of computer models provides one way of gaining clues about the ways in which neurulation may be compromised. Here we employ one computer model to examine the robustness of different cellular mechanisms which are thought to contribute to neurulation. The model, modified from that of Odell et al (Odell, G.M., Oster, G., Alberch, P. and Burnside, B., (1981)) mimics neurulation by laterally propagating a wave of apical contraction along an active zone within a ring of cells. We link the results to experimental evidence gained from studies of embryos in which neurulation has been perturbed. The results indicate that alteration of one of the properties of non-neural tissue can delay or inhibit neurulation, supporting the idea, gained from observation of embryos bearing genes which predispose to NTD, that the tissue underlying the neuroepithelium may contribute to the elevation of the neural folds. The results also show that reduction of the contractile properties of a small proportion of the neuroepithelial cell population may have a profound effect on overall tissue profiling. The results suggest that the elevation of the neural folds, and hence successful neurulation, may be vulnerable to relatively minor deficiencies in cell properties. (shrink)

An ideal vision model accounts for behavior and neurophysiology in both naturalistic conditions and designed lab experiments. Unlike psychological theories, artificial neural networks (ANNs) actually perform visual tasks and generate testable predictions for arbitrary inputs. These advantages enable ANNs to engage the entire spectrum of the evidence. Failures of particular models drive progress in a vibrant ANN research program of human vision.

The computer distance teaching system teaches through the network, and there is no entrance threshold. Any student who is willing to study can log in to the network computer distance teaching system for study at any free time. Neural network has a strong self-learning ability and is an important part of artificial intelligence research. Based on this study, a neural network-embedded architecture based on shared memory and bus structure is proposed. By looking for an alternative method of exp (...) function to improve the speed of radial basis function algorithm, and then by analyzing the judgment conditions in the main loop during the algorithm process, these judgment conditions are modified conditionally to reduce the calculation scale, which can double the speed of the algorithm. Finally, this article verifies the function, performance, and interface of the computer distance education system. (shrink)

Adaptive dynamic programming, which belongs to the field of computational intelligence, is a powerful tool to address optimal control problems. To overcome the bottleneck of solving Hamilton–Jacobi–Bellman equations, several state-of-the-art ADP approaches are reviewed in this paper. First, two model-based offline iterative ADP methods including policy iteration and value iteration are given, and their respective advantages and shortcomings are discussed in detail. Second, the multistep heuristic dynamic programming method is introduced, which avoids the requirement of initial admissible control and achieves (...) fast convergence. This method successfully utilizes the advantages of PI and VI and overcomes their drawbacks at the same time. Finally, the discrete-time optimal control strategy is tested on a power system. (shrink)