Neural structural representations are cerebral map- or model-like structures that structurally resemble what they represent. These representations are absolutely central to the “cognitive neuroscience revolution”, as they are the only type of representation compatible with the revolutionaries’ mechanistic commitments. Crucially, however, these very same commitments entail that structural representations can be observed in the swirl of neuronal activity. Here, I argue that no structural representations have been observed being present in our neuronal activity, no matter (...) the spatiotemporal scale of observation. My argument begins by introducing the “cognitive neuroscience revolution” (Sect. 1 ) and sketching a prominent, widely adopted account of structural representations (Sect. 2 ). Then, I will consult various reports that describe our neuronal activity at various spatiotemporal scales, arguing that none of them reports the presence of structural representations (Sect. 3 ). After having deflected certain intuitive objections to my analysis (Sect. 4 ), I will conclude that, in the absence of neural structural representations, representationalism and mechanism can’t go together, and so the “cognitive neuroscience revolution” is forced to abandon one of its commitments (Sect. 5 ). (shrink)

Multivariate pattern analysis, or MVPA, has become one of the most popular analytic methods in cognitive neuroscience. Since its inception, MVPA has been heralded as offering much more than regular univariate analyses, for—we are told—it not only can tell us which brain regions are engaged while processing particular stimuli, but also which patterns of neural activity represent the categories the stimuli are selected from. We disagree, and in the current paper we offer four conceptual challenges to the use of (...) MVPA to make claims about neural representation. Our view is that the use of MVPA to make claims about neural representation is problematic. (shrink)

Many human decisions and behaviors in daily life entail a cost-benefit analysis. From selecting what to eat for dinner to determining the career to pursue, we more or less assess the cost and benefit of each choice. Given the frequent occurrences of cost-benefit thinking in our minds, some intriguing questions arise: how do the cost-benefit thinking processes emerge? How does the brain function to generate such thoughts? Although these inquiries have yet to be thoroughly answered, scientists are adding new clues (...) to our knowledge, such as the neural representation of our subjective cost-benefit judgments. (shrink)

The recent resurgence of the language of thought (LOT) hypothesis has drawn much attention. The history of philosophy and cognitive science has provided us with various versions of LOT. From Sellars to Fodor, theorists have offered us considerations on the vehicles, content, and functions of such a representation format. However, it’s been more than 50 years since Fodor’s publication on LOT (1975), and the resurgence suggests that we need a modern iteration of LOT to fit with recent developments in the (...) study on cognition. Therefore, in this paper, I argue that we need a middle ground between Fodor and Sellars’s views on LOT – specifically, we should abandon Fodor’s view on the vehicle and content of LOT, and we should also acknowledge that perception might involve LOT (contrary to what Sellars thought). What we should focus on is LOT representations’ format’s function, that at least some neural processes are analogous to linguistic structures, and how such a function is able to facilitate a modern situational view of cognitive representation. Having this account of LOT, we can make it a more versatile tool fitting with the other cognitive frameworks, such as the free energy principle and Bayesian inference. (shrink)

We explore the contribution made by oscillatory, synchronous neural activity to representation in the brain. We closely examine six prominent examples of brain function in which neural oscillations play a central role, and identify two levels of involvement that these oscillations take in the emergence of representations: enabling (when oscillations help to establish a communication channel between sender and receiver, or are causally involved in triggering a representation) and properly representational (when oscillations are a constitutive part of (...) the representation). We show that even an idealized informational sender–receiver account of representation makes the representational status of oscillations a non-trivial matter, which depends on rather minute empirical details. (shrink)

In this paper I discuss one of the key issuesin the philosophy of neuroscience:neurosemantics. The project of neurosemanticsinvolves explaining what it means for states ofneurons and neural systems to haverepresentational contents. Neurosemantics thusinvolves issues of common concern between thephilosophy of neuroscience and philosophy ofmind. I discuss a problem that arises foraccounts of representational content that Icall ``the economy problem'': the problem ofshowing that a candidate theory of mentalrepresentation can bear the work requiredwithin in the causal economy of a mind (...) and anorganism. My approach in the current paper isto explore this and other key themes inneurosemantics through the use of computermodels of neural networks embodied and evolvedin virtual organisms. The models allow for thelaying bare of the causal economies of entireyet simple artificial organisms so that therelations between the neural bases of, forinstance, representation in perception andmemory can be regarded in the context of anentire organism. On the basis of thesesimulations, I argue for an account ofneurosemantics adequate for the solution of theeconomy problem. (shrink)

Despite its centrality in the philosophy of cognitive science, there has been little prior philosophical work engaging with the notion of representation in contemporary NLP practice. This paper attempts to fill that lacuna: drawing on ideas from cognitive science, I introduce a framework for evaluating the representational claims made about components of neural NLP models, proposing three criteria with which to evaluate whether a component of a model represents a property and operationalising these criteria using probing classifiers, a popular (...) analysis technique in NLP (and deep learning more broadly). The project of operationalising a philosophically-informed notion of representation should be of interest to both philosophers of science and NLP practitioners. It affords philosophers a novel testing-ground for claims about the nature of representation, and helps NLPers organise the large literature on probing experiments, suggesting novel avenues for empirical research. (shrink)

Many philosophers and psychologists have attempted to elucidate the nature of mental representation by appealing to notions like isomorphism or abstract structural resemblance. The ‘structural representations’ that these theorists champion are said to count as representations by virtue of functioning as internal models of distal systems. In his 2007 book, Representation Reconsidered, William Ramsey endorses the structural conception of mental representation, but uses it to develop a novel argument against representationalism, the widespread view that cognition essentially involves the (...) manipulation of mental representations. Ramsey argues that although theories within the ‘classical’ tradition of cognitive science once posited structural representations, these theories are being superseded by newer theories, within the tradition of connectionism and cognitive neuroscience, which rarely if ever appeal to structural representations. Instead, these theories seem to be explaining cognition by invoking so-called ‘receptor representations’, which, Ramsey claims, aren’t genuine representations at all—despite being called representations, these mechanisms function more as triggers or causal relays than as genuine stand-ins for distal systems. I argue that when the notions of structural and receptor representation are properly explicated, there turns out to be no distinction between them. There only appears to be a distinction between receptor and structural representations because the latter are tacitly conflated with the ‘mental models’ ostensibly involved in offline cognitive processes such as episodic memory and mental imagery. While structural representations might count as genuine representations, they aren’t distinctively mental representations, for they can be found in all sorts of non-intentional systems such as plants. Thus to explain the kinds of offline cognitive capacities that have motivated talk of mental models, we must develop richer conceptions of mental representation than those provided by the notions of structural and receptor representation. (shrink)

We describe and try to motivate our project to build systems using both a knowledge based and a neural network approach. These two approaches are used at different stages in the solution of a problem, instead of using knowledge bases exclusively on some problems, and neural nets exclusively on others. The knowledge base (KB) is defined first in a declarative, symbolic language that is easy to use. It is then compiled into an efficient neural network (NN) representation, (...) run, and the results from run time and (eventually) from learning are decompiled to a symbolic description of the knowledge contained in the network. After inspecting this recovered knowledge, a designer would be able to modify the KB and go through the whole cycle of compiling, running, and decompiling again. The central question with which this project is concerned is, therefore, How do we go from a KB to an NN, and back again? We are investigating this question by building tools consisting of a repertoire of language/translation/network types, and trying them on problems in a variety of domains. (shrink)

Vehicle externalism maintains that the vehicles of our mental representations can be located outside of the head, that is, they need not be instantiated by neurons located inside the brain of the cogniser. But some disagree, insisting that ‘non-derived’, or ‘original’, content is the mark of the cognitive and that only biologically instantiated representational vehicles can have non-derived content, while the contents of all extra-neural representational vehicles are derived and thus lie outside the scope of the cognitive. In (...) this paper we develop one aspect of Menary’s vehicle externalist theory of cognitive integration—the process of enculturation—to respond to this longstanding objection. We offer examples of how expert mathematicians introduce new symbols to represent new mathematical possibilities that are not yet understood, and we argue that these new symbols have genuine non-derived content, that is, content that is not dependent on an act of interpretation by a cognitive agent and that does not derive from conventional associations, as many linguistic representations do. (shrink)

The higher-order thought (HOT) theory of consciousness is a reductive representational theory of consciousness which says that what makes a mental state conscious is that there is a suitable HOT directed at that mental state. Although it seems that any neural realization of the theory must be somewhat widely distributed in the brain, it remains unclear just how widely distributed it needs to be. In section I, I provide some background and define some key terms. In section II, I (...) argue against the view that HOT theory should treat first-order (i.e. world-directed) conscious states as requiring prefrontal cortical activity though it is reasonable to suppose that conscious states are realized in the brain. In section III, I then explore some of the key background metaphysical issues involved in understanding the nature of consciousness, such as the debate between realism and idealism as well as the prospects for solving the so-called “hard problem” of consciousness. Some of the differences in question often mirror the traditional differences between Western and Eastern perspectives on the nature of consciousness. Overall, I argue that some form of realism and physicalism is more plausible than the opposing views. I also argue that materialists (and especially HOT theorists) can offer plausible replies to the hard problem. (shrink)

A network of gene regulation organized in a hierarchical and combinatorial manner is crucially involved in the development of the neural network, and has to be considered one of the main substrates of genetic change in its evolution. Though qualitative features may emerge by way of the accumulation of rather unspecific quantitative changes, it is reasonable to assume that at least in some cases specific combinations of regulatory parts of the genome initiated new directions of evolution, leading to novel (...) capabilities of the brain. These notions are applied, in this paper, to the evolution of the capability of cognition-based human empa­thy. It is suggested that it has evolved as a secondary effect of the evolution of strategic thought. Development of strategies depends on abstract representations of one’s own pos­sible future states in one’s own brain to allow assessment of their emotional desirability, but also on the representation and emotional evaluation of possible states of others, allowing anticipation of their behaviour. This is best achieved if representations of others are con­nected to one’s own emotional centres in a manner similar to self-representations. For this reason, the evolution of the human brain is assumed to have established representations with such linkages. No group selection is involved, because the quality of strategic thought affects the fitness of the individual. A secondary effect of this linkage is that both the actual states and the future perspectives of others elicit vicarious emotions, which may contribute to the motivations of altruistic behaviour. (shrink)

The core objective of this project is to recognize and reconstruct distorted facial images, particularly in the context of accidents. This involves using deep learning techniques to analyze the features of a distorted face and regenerate it into a recognizable form. Deep learning models are wellsuited for this task due to their ability to learn complex patterns and representations from data the input data consists of distorted facial images, typically obtained from MRI scans of accident victims. These images may (...) contain various types of distortions such as swelling, bruising, or other injuries that affect facial appearance. (shrink)

The paper is dedicated to particular cases of interaction and mutual impact of philosophy and cognitive science. Thus, philosophical preconditions in the middle of the 20th century shaped the newly born cognitive science as mainly based on conceptual and propositional representations and syntactical inference. Further developments towards neural networks and statistical representations did not change the prejudice much: many still believe that network models must be complemented with some extra tools that would account for proper human cognitive (...) traits. I address some real implemented connectionist models that show how ‘new associationism ’ of the neural network approach may not only surpass Humean limitations, but, as well, realistically explain abstraction, inference and prediction. Then I stay on Predictive Processing theories in a little more detail to demonstrate that sophisticated statistical tools applied to a biologically realist ontology may not only provide solutions to scientific problems or integrate different cognitive paradigms but propose some philosophical insights either. To conclude, I touch on a certain parallelism of Predictive Processing and philosophical inferentialism as presented by Robert Brandom. (shrink)

I defend a theory of mental representation that satisfies naturalistic constraints. Briefly, we begin by distinguishing (i) what makes something a representation from (ii) given that a thing is a representation, what determines what it represents. Representations are states of biological organisms, so we should expect a unified theoretical framework for explaining both what it is to be a representation as well as what it is to be a heart or a kidney. I follow Millikan in explaining (i) in (...) terms of teleofunction, explicated in terms of natural selection. -/- To explain (ii), we begin by recognizing that representational states do not have content, that is, they are neither true nor false except insofar as they both “point to” or “refer” to something, as well as “say” something regarding whatever it is they are about. To distinguish veridical from false representations, there must be a way for these separate aspects to come apart; hence, we explain (ii) by providing independent theories of what I call f-reference and f-predication (the ‘f’ simply connotes ‘fundamental’, to distinguish these things from their natural language counterparts). -/- Causal theories of representation typically founder on error, or on what Fodor has called the disjunction problem. Resemblance or isomorphism theories typically founder on what I’ve called the non-uniqueness problem, which is that isomorphisms and resemblance are practically unconstrained and so representational content cannot be uniquely determined. These traditional problems provide the motivation for my theory, the structural preservation theory, as follows. F-reference, like reference, is a specific, asymmetric relation, as is causation. F-predication, like predication, is a non-specific relation, as predicates typically apply to many things, just as many relational systems can be isomorphic to any given relational system. Putting these observations together, a promising strategy is to explain f-reference via causal history and f-predication via something like isomorphism between relational systems. -/- This dissertation should be conceptualized as having three parts. After motivating and characterizing the problem in chapter 1, the first part is the negative project, where I review and critique Dretske’s, Fodor’s, and Millikan’s theories in chapters 2-4. Second, I construct my theory about the nature of representation in chapter 5 and defend it from objections in chapter 6. In chapters 7-8, which constitute the third and final part, I address the question of how representation is implemented in biological systems. In chapter 7 I argue that single-cell intracortical recordings taken from awake Macaque monkeys performing a cognitive task provide empirical evidence for structural preservation theory, and in chapter 8 I use the empirical results to illustrate, clarify, and refine the theory. (shrink)

The cognitive basis of geometry is still poorly understood, even the ‘simpler’ issue of what kind of representation of geometric objects we have. In this work, we set forward a tentative model of the neural representation of geometric objects for the case of the pure geometry of Euclid. To arrive at a coherent model, we found it necessary to consider earlier forms of geometry. We start by developing models of the neural representation of the geometric figures of ancient (...) Greek practical geometry. Then, we propose a related model for the earliest form of pure geometry – that of Hippocrates of Chios. Finally, we develop the model of the neural representation of the geometric objects of Euclidean geometry. The models are based on the hub-and-spoke theory. In our view, the existence of specific models opens the possibility of addressing the relationship between geometric figures and geometric objects, in a novel way, in terms of their neural representation. (shrink)

A satisfactory solution to the problem of consciousness would take the form of a simple yet fully general model that specifies the precise conditions under which any given state of consciousness occurs. Science has uncovered numerous correlations between consciousness and neural activity, but it has not yet come anywhere close to this. We are still looking for the Newtonian laws of consciousness. -/- One of the main difficulties with consciousness is that we lack a language in which to formulate (...) illuminating generalizations about it. Philosophers and scientists talk about "what it’s like", sensations, feelings, and perceptual states such as seeing and hearing. This language does not allow a precise articulation of the internal structures of conscious states and their inter-relations. It is inadequate to capture relations of the kind we are looking for between conscious states and physical states. -/- In this thesis I refine and defend a theory of consciousness which promises to solve this regimentation problem: the representational theory of consciousness. I argue that the representational theory can solve the regimentation problem and smooth out other important obstacles to a fruitful study of consciousness. I also make a case for the theory independently of its payoffs, and I discuss the leading opposing theories at some length. (shrink)

Metacontrast is a visual illusion in which the visibility of a target stimulus is virtually lost when immediately followed by a nonoverlapping mask stimulus. For a colored target, metacontrast is color-selective, with target visibility markedly reduced when the mask and target are the same color, but only slightly reduced when the colors differ. This study investigated neural correlates of color-selective metacontrast for cone-opponent red and green stimuli in the human V1, V2, and V3 using functional magnetic resonance imaging. (...) class='Hi'>Neural activity was suppressed when the target was rendered less visible by the same-colored mask, and the suppression was localized in the cortical region retinotopically representing the target, correlating with the perceptual topography of visibility/invisibility rather than the physical topography of the stimulus. Retinotopy-based group analysis found that activity suppression was statistically significant for V2 and V3 and that its localization to the target region was statistically significant for V2. These results suggest that retinotopic color representations in early visual areas, especially in V2, are closely linked to the visibility of color. (shrink)

The main aim of this article is to present and defend a thesis according to which conceptual representations of some types of mental states are encoded in the same neural structures that underlie the first-personal experience of those states. To support this proposal here, I will put forth a novel account of the cognitive function played by ‘shared representations’ of emotions and bodily sensations, i.e. neural structures that are active when one experiences a mental state of (...) a certain type as well as when one observes someone else experiencing a state of the same type. I will argue that shared representations in fact constitute vehicles of certain mental state concepts (more precisely, concepts of specific types of emotions and somatosensory states). The main line of arguing for this will consist in showing that shared representations exhibit specific, ‘conceptual’ functional properties:(1) causal effect on forming metacognitive judgments, (2) cognitive penetrability, (3) diversity of input types. (shrink)

The neural vehicles of mental representation play an explanatory role in cognitive psychology that their realizers do not. In this paper, I argue that the individuation of realizers as vehicles of representation restricts the sorts of explanations in which they can participate. I illustrate this with reference to Rupert’s (2011) claim that representational vehicles can play an explanatory role in psychology in virtue of their quantity or proportion. I propose that such quantity-based explanatory claims can apply only to realizers (...) and not to vehicles, in virtue of the particular causal role that vehicles play in psychological explanations. (shrink)

The human eyes and brain, which have finite boundaries, create a ‘‘virtual’’ space within our central nervous system that interprets and perceives a space that appears boundless and infinite. Using insights from studies on the visual system, we propose a novel fast processing mechanism involving the eyes, visual pathways, and cortex where external vision is imperceptibly processed in our brain in real time creating an internal representation of external space that appears as an external view. We introduce the existence of (...) a three-dimension default space consisting of intrapersonal body space that serves as the framework where visual and non-visual sensory information is sensed and experienced. We propose that the thalamus integrates processed information from corticothalamic feedback loops and fills-in the neural component of 3D default space with an internal visual representation of external space, leading to the experience of visual consciousness. This visual space inherently evades perception so we have introduced three easy clinical tests that can assist in experiencing this visual space. We also review visual neuroanatomical pathways, binocular vision, neurological disorders, and visual phenomenon to elucidate how the representation of external visible space is recreated within the mind. (shrink)

In artificial intelligence, recent research has demonstrated the remarkable potential of Deep Convolutional Neural Networks (DCNNs), which seem to exceed state-of-the-art performance in new domains weekly, especially on the sorts of very difficult perceptual discrimination tasks that skeptics thought would remain beyond the reach of artificial intelligence. However, it has proven difficult to explain why DCNNs perform so well. In philosophy of mind, empiricists have long suggested that complex cognition is based on information derived from sensory experience, often appealing (...) to a faculty of abstraction. Rationalists have frequently complained, however, that empiricists never adequately explained how this faculty of abstraction actually works. In this paper, I tie these two questions together, to the mutual benefit of both disciplines. I argue that the architectural features that distinguish DCNNs from earlier neural networks allow them to implement a form of hierarchical processing that I call “transformational abstraction”. Transformational abstraction iteratively converts sensory-based representations of category exemplars into new formats that are increasingly tolerant to “nuisance variation” in input. Reflecting upon the way that DCNNs leverage a combination of linear and non-linear processing to efficiently accomplish this feat allows us to understand how the brain is capable of bi-directional travel between exemplars and abstractions, addressing longstanding problems in empiricist philosophy of mind. I end by considering the prospects for future research on DCNNs, arguing that rather than simply implementing 80s connectionism with more brute-force computation, transformational abstraction counts as a qualitatively distinct form of processing ripe with philosophical and psychological significance, because it is significantly better suited to depict the generic mechanism responsible for this important kind of psychological processing in the brain. (shrink)

Two universal functional principles of Grossberg’s Adaptive Resonance Theory [19] decipher the brain code of all biological learning and adaptive intelligence. Low-level representations of multisensory stimuli in their immediate environmental context are formed on the basis of bottom-up activation and under the control of top-down matching rules that integrate high-level long-term traces of contextual configuration. These universal coding principles lead to the establishment of lasting brain signatures of perceptual experience in all living species, from aplysiae to primates. They are (...) re-visited in this concept paper here on the basis of examples drawn from the original code and from some of the most recent related empirical findings on contextual modulation in the brain, highlighting the potential of Grossberg’s pioneering insights and groundbreaking theoretical work for intelligent solutions in the domain of developmental and cognitive robotics. (shrink)

Recent breakthroughs in artificial intelligence have primarily resulted from training deep neural networks (DNNs) with vast numbers of adjustable parameters on enormous datasets. Due to their complex internal structure, DNNs are frequently characterized as inscrutable ``black boxes,'' making it challenging to interpret the mechanisms underlying their impressive performance. This opacity creates difficulties for explanation, safety assurance, trustworthiness, and comparisons to human cognition, leading to divergent perspectives on these systems. This chapter examines recent developments in interpretability methods for DNNs, with (...) a focus on interventionist approaches inspired by causal explanation in philosophy of science. We argue that these methods offer a promising avenue for understanding how DNNs process information compared to merely behavioral benchmarking and correlational probing. We review key interventionist methods and illustrate their application through practical case studies. These methods allow researchers to identify and manipulate specific computational components within DNNs, providing insights into their causal structure and internal representations. We situate these approaches within the broader framework of causal abstraction, which aims to align low-level neural computations with high-level interpretable models. While acknowledging current limitations, we contend that interventionist methods offer a path towards more rigorous and theoretically grounded interpretability research, potentially informing both AI development and computational cognitive neuroscience. (shrink)

Symmetry in biological and physical systems is a product of self-organization driven by evolutionary processes, or mechanical systems under constraints. Symmetry-based feature extraction or representation by neural networks may unravel the most informative contents in large image databases. Despite significant achievements of artificial intelligence in recognition and classification of regular patterns, the problem of uncertainty remains a major challenge in ambiguous data. In this study, we present an artificial neural network that detects symmetry uncertainty states in human observers. (...) To this end, we exploit a neural network metric in the output of a biologically inspired Self- Organizing Map Quantization Error (SOM-QE). Shape pairs with perfect geometry mirror symmetry but a non-homogenous appearance, caused by local variations in hue, saturation, or lightness within and/or across the shapes in a given pair produce, as shown here, a longer choice response time (RT) for “yes” responses relative to symmetry. These data are consistently mirrored by the variations in the SOM-QE from unsupervised neural network analysis of the same stimulus images. The neural network metric is thus capable of detecting and scaling human symmetry uncertainty in response to patterns. Such capacity is tightly linked to the metric’s proven selectivity to local contrast and color variations in large and highly complex image data. (shrink)

Introduction There are some exceptions, which we shall see below, but virtually all theories in psychology and cognitive science make use of the notion of representation. Arguably, folk psychology also traffics in representations, or is at least strongly suggestive of their existence. There are many different types of things discussed in the psychological and philosophical literature that are candidates for representation-hood. First, there are the propositional attitudes – beliefs, judgments, desires, hopes etc. (see Chapters 9 and 17 of this (...) volume). If the propositional attitudes are representations, they are person-level representations – the judgment that the sun is bright pertains to John, not a subpersonal part of John. By contrast, the representations of edges in V1 of the cerebral cortex that neuroscientists talk about and David Marr’s symbolic representations of “zero-crossings” in early vision (Marr 1982) are at the “sub-personal” level – they apply to parts or states of a person (e.g. neural parts or computational states of the visual system). Another important distinction is often made among perceptual, cognitive, and action-oriented representations (e.g. motor commands). Another contrast lies between “stored representations” (e.g. memories) and “active representations” (e.g. a current perceptual state). Related to this is the distinction between “dispositional representations” and “occurrent representations.” Beliefs that are not currently being entertained are dispositional, e.g. your belief that the United States is in North America - no doubt you had this belief two minutes ago, but you were not consciously accessing it until you read this sentence. Occurrent representations, by contrast, are active, conscious thoughts or perceptions. Which leads us to another important distinction: 1 between conscious and non-conscious mental representations, once a bizarre-sounding distinction that has become familiar since Freud (see Chapter 4 of this volume). I mention these distinctions at the outset to give you some idea of the range of phenomena we will be considering, and to set the stage for our central “problem of representation”: what is a mental representation, exactly, and how do we go about deciding whether there are any? We know there are public representations of various kinds: words, maps, and pictures, among others.. (shrink)

A model of consciousness and conscious experience is introduced. Starting with a non-Lipschitz Chaotic dynamics of neural activity, we propose that the synaptic transmission between adjacent as well as distant neurons should be regulated in brain dynamics through quantum tunneling. Further, based on various studies of different previous authors, we consider the emergence of very large quantum mechanical system representable by an abstract quantum net entirely based on quantum-like entities having in particular the important feature of expressing self-reference similar (...) to what occurs in consciousness. The properties of such quantum-like mind entities are discussed in detail. A quantum-like model of conscious experience is also discussed. It is shown that such quantum mechanical entities are able to arrange themselves alternatively on the basis of the subject story, memory, and pain-pleasure in response to an external stimulus, thus giving the subject the possibility to response to the stimulus on the basis of his emotion as well as cognitive state. Finally, we discuss the possible connections between the quantum-like model introduced in this paper and the chaotic behaviors often identified experimentally in studies on brain dynamics. Part I of this article contains: Introduction; 1. Non Lipschitz Terminal Dynamics of Single Neuron Activity; and References; 2. Quantum Mechanical Properties of Neuron Dynamics; and 3. A Quantum Model of Consciousness I. (shrink)

We discuss cases where subjects seem to enjoy conscious experience when the relevant first-order perceptual representations are either missing or too weak to account for the experience. Though these cases are originally considered to be theoretical possibilities that may be problematical for the higher-order view of consciousness, careful considerations of actual empirical examples suggest that this strategy may backfire; these cases may cause more trouble for first-order theories instead. Specifically, these cases suggest that (I) recurrent feedback loops to V1 (...) are most likely not the neural correlate of first-order representations for conscious experience, (II) first-order views seem to have a problem accounting for the phenomenology in these cases, and either (III) a version of the ambitious higher-order approach is superior in that it is the simplest theory that can account for all results at face value, or (IV) a view where phenomenology is jointly determined by both first-order and higher-order states. In our view (III) and (IV) are both live options and the decision between them may ultimately be an empirical question that cannot yet be decided. (shrink)

The success of deep learning in natural language processing raises intriguing questions about the nature of linguistic meaning and ways in which it can be processed by natural and artificial systems. One such question has to do with subword segmentation algorithms widely employed in language modeling, machine translation, and other tasks since 2016. These algorithms often cut words into semantically opaque pieces, such as ‘period’, ‘on’, ‘t’, and ‘ist’ in ‘period|on|t|ist’. The system then represents the resulting segments in a dense (...) vector space, which is expected to model grammatical relations among them. This representation may in turn be used to map ‘period|on|t|ist’ (English) to ‘par|od|ont|iste’ (French). Thus, instead of being modeled at the lexical level, translation is reformulated more generally as the task of learning the best bilingual mapping between the sequences of subword segments of two languages; and sometimes even between pure character sequences: ‘p|e|r|i|o|d|o|n|t|i|s|t’ → ‘p|a|r|o|d|o|n|t|i|s|t|e’. Such subword segmentations and alignments are at work in highly efficient end-to-end machine translation systems, despite their allegedly opaque nature. The computational value of such processes is unquestionable. But do they have any linguistic or philosophical plausibility? I attempt to cast light on this question by reviewing the relevant details of the subword segmentation algorithms and by relating them to important philosophical and linguistic debates, in the spirit of making artificial intelligence more transparent and explainable. (shrink)

This article addresses computational procedures that are no longer constrained by human modes of representation and considers how these procedures could be philosophically understood in terms of ‘algorithmic thought’. Research in deep learning is its case study. This artificial intelligence (AI) technique operates in computational ways that are often opaque. Such a black-box character demands rethinking the abstractive operations of deep learning. The article does so by entering debates about explainability in AI and assessing how technoscience and technoculture tackle the (...) possibility to ‘re-present’ the algorithmic procedures of feature extraction and feature learning to the human mind. The article thus mobilises the notion of incommensurability (originally developed in the philosophy of science) to address explainability as a communicational and representational issue, which challenges phenomenological and existential modes of comparison between human and algorithmic ‘thinking’ operations. (shrink)

The aim of this paper is to develop and defend an Attentional View of bodily awareness, on which attention is necessary for bodily awareness. The original formulation of the Attentional View is due to Marcel Kinsbourne. First, I will show that the Attentional View of bodily awareness as formulated by Kinsbourne is superior to other accounts in the literature for characterizing the relationship between attention and bodily awareness. Kinsbourne’s account is the only account in the literature so far which can (...) accommodate key neurological diseases such as personal neglect. Second, when I consider Kinsbourne’s view in more detail, I will argue that Kinsbourne’s Attentional View faces problems because it is too reductive. Kinsbourne deviates from the standard taxonomy on which there is a body schema and a body image. Instead he reduces the body image to the neural representation of the body in the somatosensory cortex, the body schema and attentional shifts. I will present two challenges to Kinsbourne’s view which demonstrate that Kinsbourne’s reduction of the body image is unsuccessful. Finally, I will present a revised version of the Attentional View that is both empirically adequate and philosophically satisfactory. (shrink)

It is often assumed that neural representation, with content that is in principle detachable from the flow of natural-factive information, is necessary to perceptually experience an object. In this paper we present and discuss two cases challenging this assumption. We take them to show that it is possible to experience an object with which you are interacting through your sensory systems without those systems constructing a representation of the object. The first example is viewing nearby medium-sized groups of objects. (...) The second is hearing objects through misbound sounds. These cases bring out two different ways object representation can fail while object experience persists, suggesting that object experience requires only that the object be revealed through sensory information, not full-blown representation. Constructing object representations is one way sensory systems reveal objects, but it is not the only way. Object representation in neural sensory systems is relatively demanding, rare, and fragile; object experience is relatively easy, pervasive, and robust. We conclude that even minimal forms of neural phenomenal internalism are false. (shrink)

In this article, I link the empirical hypothesis that neural representations of sensory stimulation near the body involve a unique motor component to the idea that the perceptual field is structured by skillful bodily activity. The neurophenomenological view that emerges is illuminating in its own right, though it may also have practical consequences. I argue that recent experiments attempting to alter the scope of these near space sensorimotor representations are actually equivocal in what they show. I propose (...) resolving this ambiguity by treating these representations as responsive to the development or degeneration of know-how—which can be isolated as an appropriate object for scientific investigation. (shrink)

Mainstream teleosemantics is the view that mental representation should be understood in terms of biological functions, which, in turn, should be understood in terms of selection processes. One of the traditional criticisms of teleosemantics is the problem of novel contents: how can teleosemantics explain our ability to represent properties that are evolutionarily novel? In response, some have argued that by generalizing the notion of a selection process to include phenomena such as operant conditioning, and the neural selection that underlies (...) it, we can resolve this problem. Here, we do four things: we develop this suggestion in a rigorous way through a simple example, we draw on recent neurobiological research to support its empirical plausibility, we defend the move from a host of objections in the literature, and we sketch how the picture can be extended to help us think about more complex “conceptual” representations and not just perceptual ones. (shrink)

This commentary (1) raises the question about the possible conflation of core affect with the neural representation of interoceptive changes in regard to whether biological value is subpersonal or must be experienced, and (2) proposes that Wundt’s third dimension of core affect – strain-relaxation – can be accounted for in the target model under a generalised predictive model of attention.

Mental representations remain the central posits of psychology after many decades of scrutiny. However, there is no consensus about the representational format(s) of biological cognition. This paper provides a survey of evidence from computational cognitive psychology, perceptual psychology, developmental psychology, comparative psychology, and social psychology, and concludes that one type of format that routinely crops up is the language-of-thought (LoT). We outline six core properties of LoTs: (i) discrete constituents; (ii) role-filler independence; (iii) predicate–argument structure; (iv) logical operators; (v) (...) inferential promiscuity; and (vi) abstract content. These properties cluster together throughout cognitive science. Bayesian computational modeling, compositional features of object perception, complex infant and animal reasoning, and automatic, intuitive cognition in adults all implicate LoT-like structures. Instead of regarding LoT as a relic of the previous century, researchers in cognitive science and philosophy-of-mind must take seriously the explanatory breadth of LoT-based architectures. We grant that the mind may harbor many formats and architectures, including iconic and associative structures as well as deep-neural-network-like architectures. However, as computational/representational approaches to the mind continue to advance, classical compositional symbolic structures – that is, LoTs – only prove more flexible and well-supported over time. (shrink)

The idea that only complex brains can possess genuine representations is an important element in mainstream philosophical thinking. An alternative view, which I label ‘liberal representationalism’, holds that we should accept the existence of many more full-blown representations, from activity in retinal ganglion cells to the neural states produced by innate releasing mechanisms in cognitively unsophisticated organisms. A promising way of supporting liberal representationalism is to show it to be a consequence of our best naturalistic theories of (...) representation. However, several philosophers and scientists have recently argued against this strategy. In the paper I counter these objections in defense of liberal representationalism. (shrink)

The purpose of this paper is to present a general mechanistic framework for analyzing causal representational claims, and offer a way to distinguish genuinely representational explanations from those that invoke representations for honorific purposes. It is usually agreed that rats are capable of navigation because they maintain a cognitive map of their environment. Exactly how and why their neural states give rise to mental representations is a matter of an ongoing debate. I will show that anticipatory mechanisms (...) involved in rats’ evaluation of possible routes give rise to satisfaction conditions of contents, and this is why they are representationally relevant for explaining and predicting rats’ behavior. I argue that a naturalistic account of satisfaction conditions of contents answers the most important objections of antirepresentationalists. (shrink)

The representations of deep convolutional neural networks (CNNs) are formed from generalizing similarities and abstracting from differences in the manner of the empiricist theory of abstraction (Buckner, Synthese 195:5339–5372, 2018). The empiricist theory of abstraction is well understood to entail infinite regress and circularity in content constitution (Husserl, Logical Investigations. Routledge, 2001). This paper argues these entailments hold a fortiori for deep CNNs. Two theses result: deep CNNs require supplementation by Quine’s “apparatus of identity and quantification” in order (...) to (1) achieve concepts, and (2) represent objects, as opposed to “half-entities” corresponding to similarity amalgams (Quine, Quintessence, Cambridge, 2004, p. 107). Similarity amalgams are also called “approximate meaning[s]” (Marcus & Davis, Rebooting AI, Pantheon, 2019, p. 132). Although Husserl inferred the “complete abandonment of the empiricist theory of abstraction” (a fortiori deep CNNs) due to the infinite regress and circularity arguments examined in this paper, I argue that the statistical learning of deep CNNs may be incorporated into a Fodorian hybrid account that supports Quine’s “sortal predicates, negation, plurals, identity, pronouns, and quantifiers” which are representationally necessary to overcome the regress/circularity in content constitution and achieve objective (as opposed to similarity-subjective) representation (Burge, Origins of Objectivity. Oxford, 2010, p. 238). I base myself initially on Yoshimi’s (New Frontiers in Psychology, 2011) attempt to explain Husserlian phenomenology with neural networks but depart from him due to the arguments and consequently propose a two-system view which converges with Weiskopf’s proposal (“Observational Concepts.” The Conceptual Mind. MIT, 2015. 223–248). (shrink)

In 2014, John O’Keefe was awarded the Nobel Prize in Physiology or Medicine for his work on the hippocampus and its role in encoding map-like representations. His contributions were significantly influenced by Morris water maze studies. O’Keefe himself acknowledged the pivotal role of the Morris water maze, stating that it remains the preeminent behavioral assay for assessing hippocampal function. Indeed, thousands of researchers have turned to the Morris water maze for evidence about navigation abilities and the effects that stress, (...) lesions, pharmaceutical interventions, and more can have on them. This body of studies constitutes an important scientific enterprise. -/- This dissertation is about some of the philosophical dimensions of Morris water maze studies. Chapter 1 is about the different types of information states (including representational states) found across hypotheses about Morris water maze performance. Chapter 2 is about systematic task failures reported in Morris water maze studies. I argue these impose a constraint on what can count as an explanation (a good hypothesis) of task success. While this has the air of a chopping block for scientific hypotheses, I see things a little differently. In Chapter 3, I argue that satisfying the constraint is a formidable challenge for any hypothesis. One that should make researchers second-guess the concepts and strategies employed to explain the roles information states play in the maze task. Chapter 4 focuses on two spatial concepts. I argue that researchers working definitions for the spatial concepts of distal (far) and proximal (near) in maze studies are problematic. They are not ecologically valid, and so claims about them do not generalize to real-life navigation behaviors like migration or scavenging behaviors. Following this, I present alternative definitions in terms of neural information about visual cues. The neural information relevant to this account is non-conceptual, and so it provides a sketch of the ways in which information states can fruitfully contribute to explanations of rat success and failure while maintaining ecological validity. -/- To sum, this dissertation navigates important philosophical dimensions of Morris water maze studies, illustrating the challenges and opportunities involved in using information states to explain and understand animal navigation. (shrink)

Endowing artificial systems with explanatory capacities about the reasons guiding their decisions, represents a crucial challenge and research objective in the current fields of Artificial Intelligence (AI) and Computational Cognitive Science [Langley et al., 2017]. Current mainstream AI systems, in fact, despite the enormous progresses reached in specific tasks, mostly fail to provide a transparent account of the reasons determining their behavior (both in cases of a successful or unsuccessful output). This is due to the fact that the classical problem (...) of opacity in artificial neural networks (ANNs) explodes with the adoption of current Deep Learning techniques [LeCun, Bengio, Hinton, 2015]. In this paper we argue that the explanatory deficit of such techniques represents an important problem, that limits their adoption in the cognitive modelling and computational cognitive science arena. In particular we will show how the current attempts of providing explanations of the deep nets behaviour (see e.g. [Ritter et al. 2017] are not satisfactory. As a possibile way out to this problem, we present two different research strategies. The first strategy aims at dealing with the opacity problem by providing a more abstract interpretation of neural mechanisms and representations. This approach is adopted, for example, by the biologically inspired SPAUN architecture [Eliasmith et al., 2012] and by other proposals suggesting, for example, the interpretation of neural networks in terms of the Conceptual Spaces framework [Gärdenfors 2000, Lieto, Chella and Frixione, 2017]. All such proposals presuppose that the neural level of representation can be considered somehow irrelevant for attacking the problem of explanation [Lieto, Lebiere and Oltramari, 2017]. In our opinion, pursuing this research direction can still preserve the use of deep learning techniques in artificial cognitive models provided that novel and additional results in terms of “transparency” are obtained. The second strategy is somehow at odds with respect to the previous one and tries to address the explanatory issue by avoiding to directly solve the “opacity” problem. In this case, the idea is that one of resorting to pre-compiled plausible explanatory models of the word used in combination with deep-nets (see e.g. [Augello et al. 2017]). We argue that this research agenda, even if does not directly fits the explanatory needs of Computational Cognitive Science, can still be useful to provide results in the area of applied AI aiming at shedding light on the models of interaction between low level and high level tasks (e.g. between perceptual categorization and explanantion) in artificial systems. (shrink)

This article re-examines Ned Block‘s ( 1997 , 2007 ) conceptual distinction between phenomenal consciousness and access consciousness. His argument that we can have phenomenally conscious representations without being able to cognitively access them is criticized as not being supported by evidence. Instead, an alternative interpretation of the relevant empirical data is offered which leaves the link between phenomenology and accessibility intact. Moreover, it is shown that Block’s claim that phenomenology and accessibility have different neural substrates is highly (...) problematic in light of empirical evidence. Finally, his claim that there can be phenomenology without cognitive accessibility is at odds with his endorsement of the 'same-order-theory' of consciousness. (shrink)

The computational metaphor for mind is still the central guiding idea in cognitive science despite many insightful and well-founded rejections of it. There is good reason for its staying power: when we are at our cognitive best, we reason about our world with our concepts. But the challengers are right, I argue, in insisting that no reductive account of that capacity is forthcoming. Here I describe an externalist account that grounds representations in organism-level engagement with its environment, not in (...) its neural activity. (shrink)

The thresholds of human observers detecting line targets improve significantly when the targets are presented in a spatial context of collinear inducing stimuli. This phenomenon is referred to as spatial facilitation, and may reflect the output of long-range interactions between cortical feature detectors. Spatial facilitation has thus far been observed with luminance-defined, achromatic stimuli on achromatic backgrounds. This study compares spatial facilitation with line targets and collinear, edge-like inducers defined by luminance contrast to spatial facilitation with targets and inducers defined (...) by color contrast. The results of a first experiment show that achromatic inducers facilitate the detection of achromatic targets on gray and colored backgrounds, but tend to suppress the detection of chromatic targets. Chromatic inducers facilitate the detection of chromatic targets on gray and colored backgrounds, but tend to suppress the detection of achromatic targets. Chromatic spatial facilitation appears to be strongest when inducers and background are isoluminant. The results of a second experiment show that spatial facilitation with chromatic targets and inducers requires a longer exposure duration of the inducers than spatial facilitation with achromatic targets and inducers, which is already fully effective at an inducer exposure of 30 ms only. The findings point towards two separate mechanisms for spatial facilitation with collinear form stimuli: one that operates in the domain of luminance, and one that operates in the domain of color contrast. These results are consistent with neural models of boundary and surface formation which suggest that achromatic and chromatic visual cues are represented on different cortical surface representations that are capable of selectively attracting attention. Multiple copies of these achromatic and chromatic surface representations exist corresponding to different ranges of perceived depth from an observer, and each can attract attention to itself. Color and contrast differences between inducing and test stimuli, and transient responses to inducing stimuli, can cause attention to shift across these surface representations in ways that sometimes enhance and sometimes interfere with target detection. (shrink)

I will review the most widely held account of phantom sensations; the “re-mapping hypothesis.” According to the re-mapping hypothesis, amputation is followed by significant neural reorganization that, over time, restores the alignment between the brain’s representation of and the actual condition of the body. Implicit in the re-mapping hypothesis is the view that the brain’s primary function is to accurately represent the body. In response, I propose an alternative theory, the “preservation hypothesis.” The preservation hypothesis argues that the primary (...) function of the brain is to preserve the entirety of the brain’s structures and functional capacities. While these issues concern empirical matters, assessing our views on the subject discloses deeply held assumptions regarding the brain’s primary function; does the brain operate to provide an accurate representation of the body? Or, does the brain operate solely to preserve its structures and functional capacities in their entirety? (shrink)

I argue that Leibniz’s doctrine of sensory representation is intended in part to close an explanatory gap in his philosophical system. Unlike the twentieth century explanatory gap, which stretches between neural states on one side and phenomenal character on the other, Leibniz’s gap lies between experiences of secondary qualities like color and taste and the objects that cause them. The problem is that the precise arrangement and distribution of such experiences can never be given a full explanation. In response, (...) Leibniz appeals to representation. I argue that the sense in which Leibnizian sensations are representations is too weak to close his explanatory gap. In the end, he must appeal to the doctrine of the best possible world. (shrink)

How can we explain the intentional nature of an expert’s actions, performed without immediate and conscious control, relying instead on automatic cognitive processes? How can we account for the differences and similarities with a novice’s performance of the same actions? Can a naturalist explanation of intentional expert action be in line with a philosophical concept of intentional action? Answering these and related questions in a positive sense, this dissertation develops a three-step argument. Part I considers different methods of explanations in (...) cognitive neuroscience (Bennett & Hacker’s philosophical, conceptual analysis; Marr’s three levels of explanation; Neural Correlates of Consciousness research; mechanistic explanation), defending ‘mechanistic explanation’ as a method that provides the necessary tools for integrating interdisciplinary insights into human action. Furthermore, a dynamic, explanatory mechanism allows the assessment of the impact of learning and development on expert action in a valuable way that other methods don’t. Part II continues by scrutinizing several cognitive neuroscientific theories of learning and development (neuroconstructivism; dual-processing theories; simulation theory; extended mind/cognition hypothesis), arguing for the complex interactions between different types of processing and different action representations involved in expert action performances. Moreover, according to our discussion of a particular ‘simulation theory’ these interactions can be influenced in several ways with the use of language, allowing an agent to configure a specific action representation for performance at a later stage. The results of Parts I and II are then applied in Part III to a parallel discussion of philosophical analyses of intentional action (discussing i.a. Frankfurt, Bratman, Pacherie and Ricoeur) and cognitive neuroscientific insights in it. Both approaches are found to converge in emphasizing the importance for an expert to develop stable patterns of actions that comply maximally not only with his intentions, but also with his motor expertise and with situational conditions. Consequently, his actions – automatic, or not – rely on this ‘sculpted space of actions’. (shrink)

I reply to comments on my book Perception (Routledge 2021) by Umrao Sethi, Jeff Speaks and Brian Cutter. Sethi objects to my representational view of perception on the ground that that having an experience of a color or shape can enable you to know what that color or shape is like only if it is actually present in the experience. Speaks has a very interesting discussion of my puzzle of the laws of appearance for the representational view. And Cutter asks (...) what I have against 'neural sense datum theory'. In my responses, I try to take the discussion forward. At the end, I develop a new argument against reductive physicalist theories of sensible qualities based on a principle I call "qualitative closure". (shrink)

The notion of “hierarchy” is one of the most commonly posited organizational principles in systems neuroscience. To this date, however, it has received little philosophical analysis. This is unfortunate, because the general concept of hierarchy ranges over two approaches with distinct empirical commitments, and whose conceptual relations remain unclear. We call the first approach the “representational hierarchy” view, which posits that an anatomical hierarchy of feed-forward, feed-back, and lateral connections underlies a signal processing hierarchy of input-output relations. Because the representational (...) hierarchy view holds that unimodal sensory representations are subsequently elaborated into more categorical and rule-based ones, it is committed to an increasing degree of abstraction along the hierarchy. The second view, which we call “topological hierarchy", is not committed to different representational functions or degrees of abstraction at different levels. Topological approaches instead posit that the hierarchical level of a part of the brain depends on how central it is to the pattern of connections in the system. Based on the current evidence, we argue that three conceptual relations between the two approaches are possible: topological hierarchies could substantiate the traditional representational hierarchy, conflict with it, or contribute to a plurality of approaches needed to understand the organization of the brain. By articulating each of these possibilities, our analysis attempts to open a conceptual space in which further neuroscientific and philosophical reasoning about neural hierarchy can proceed. (shrink)