Neural Representation. A Survey-Based Analysis of the Notion

The motifs of radical embodied neuroscience

In this paper, I analyse how the emerging scientific framework of radical embodied neuroscience is different from contemporary mainstream cognitive neuroscience. To do so, I propose the notion of motif to enrich the philosophical toolkit of cognitive neuroscience. This notion can be used to characterize the guiding ideas of any given scientific framework in psychology and neuroscience. Motifs are highly unconstrained, open-ended concepts that support equally open-ended families of explanations. Different scientific frameworks-e.g., psychophysics or cognitive neuroscience-provide these motifs to answer the overarching themes of these disciplines, such as the relationship between stimuli and sensations or the proper methods of the sciences of the mind. Some motifs of mainstream cognitive neuroscience are the motif of encoding, the motif of input-output systems, and the motif of algorithms. The two first ones answer the question about the relationship between stimuli, sensations and experience (e.g., stimuli are input and are encoded by brain structures). The latter one answers the question regarding the mechanism of cognition and experience. The three of them are equally unconstrained and open-ended, and they serve as an umbrella for different kinds of explanation-i.e., different positions regarding what counts as a code or as an input. Along with the articulation of the notion of motif, the main aim of this article is to present three motifs for radical embodied neuroscience: the motif of complex stimulation, the motif of organic behaviour and the motif of resonance.

Representational formats of human memory traces

Neural representations are internal brain states that constitute the brain's model of the external world or some of its features. In the presence of sensory input, a representation may reflect various properties of this input. When perceptual information is no longer available, the brain can still activate representations of previously experienced episodes due to the formation of memory traces. In this review, we aim at characterizing the nature of neural memory representations and how they can be assessed with cognitive neuroscience methods, mainly focusing on neuroimaging. We discuss how multivariate analysis techniques such as representational similarity analysis (RSA) and deep neural networks (DNNs) can be leveraged to gain insights into the structure of neural representations and their different representational formats. We provide several examples of recent studies which demonstrate that we are able to not only measure memory representations using RSA but are also able to investigate their multiple formats using DNNs. We demonstrate that in addition to slow generalization during consolidation, memory representations are subject to semantization already during short-term memory, by revealing a shift from visual to semantic format. In addition to perceptual and conceptual formats, we describe the impact of affective evaluations as an additional dimension of episodic memories. Overall, these studies illustrate how the analysis of neural representations may help us gain a deeper understanding of the nature of human memory.

Neural representation in active inference: Using generative models to interact with-and understand-the lived world

This paper considers neural representation through the lens of active inference, a normative framework for understanding brain function. It delves into how living organisms employ generative models to minimize the discrepancy between predictions and observations (as scored with variational free energy). The ensuing analysis suggests that the brain learns generative models to navigate the world adaptively, not (or not solely) to understand it. Different living organisms may possess an array of generative models, spanning from those that support action-perception cycles to those that underwrite planning and imagination; namely, from explicit models that entail variables for predicting concurrent sensations, like objects, faces, or people-to action-oriented models that predict action outcomes. It then elucidates how generative models and belief dynamics might link to neural representation and the implications of different types of generative models for understanding an agent's cognitive capabilities in relation to its ecological niche. The paper concludes with open questions regarding the evolution of generative models and the development of advanced cognitive abilities-and the gradual transition from pragmatic to detached neural representations. The analysis on offer foregrounds the diverse roles that generative models play in cognitive processes and the evolution of neural representation.

Engram neurons: Encoding, consolidation, retrieval, and forgetting of memory

Tremendous strides have been made in our understanding of the neurobiological substrates of memory - the so-called memory "engram". Here, we integrate recent progress in the engram field to illustrate how engram neurons transform across the "lifespan" of a memory - from initial memory encoding, to consolidation and retrieval, and ultimately to forgetting. To do so, we first describe how cell-intrinsic properties shape the initial emergence of the engram at memory encoding. Second, we highlight how these encoding neurons preferentially participate in synaptic- and systems-level consolidation of memory. Third, we describe how these changes during encoding and consolidation guide neural reactivation during retrieval, and facilitate memory recall. Fourth, we describe neurobiological mechanisms of forgetting, and how these mechanisms can counteract engram properties established during memory encoding, consolidation, and retrieval. Motivated by recent experimental results across these four sections, we conclude by proposing some conceptual extensions to the traditional view of the engram, including broadening the view of cell-type participation within engrams and across memory stages. In collection, our review synthesizes general principles of the engram across memory stages, and describes future avenues to further understand the dynamic engram.

Investigating the concept of representation in the neural and psychological sciences

The concept of representation is commonly treated as indispensable to research on brains, behavior, and cognition. Nevertheless, systematic evidence about the ways the concept is applied remains scarce. We present the results of an experiment aimed at elucidating what researchers mean by "representation." Participants were an international group of psychologists, neuroscientists, and philosophers (N = 736). Applying elicitation methodology, participants responded to a survey with experimental scenarios aimed at invoking applications of "representation" and five other ways of describing how the brain responds to stimuli. While we find little disciplinary variation in the application of "representation" and other expressions (e.g., "about" and "carry information"), the results suggest that researchers exhibit uncertainty about what sorts of brain activity involve representations or not; they also prefer non-representational, causal characterizations of the brain's response to stimuli. Potential consequences of these findings are explored, such as reforming or eliminating the concept of representation from use.

Muscles or Movements? Representation in the Nascent Brain Sciences

The idea that the brain is a representational organ has roots in the nineteenth century, when neurologists began drawing conclusions about what the brain represents from clinical and experimental studies. One of the earliest controversies surrounding representation in the brain was the "muscles versus movements" debate, which concerned whether the motor cortex represents complex movements or rather fractional components of movement. Prominent thinkers weighed in on each side: neurologists John Hughlings Jackson and F.M.R. Walshe in favor of complex movements, neurophysiologist Charles Sherrington and neurosurgeon Wilder Penfield in favor of movement components. This essay examines these and other brain scientists' evolving notions of representation during the first eighty years of the muscles versus movements debate (c. 1873-1954). Although participants agreed about many of the superficial features of representation, their inferences reveal deep-seated disagreements about its inferential role. Divergent epistemological commitments stoked conflicting conceptions of what representational attributions imply and what evidence supports them.

Getting language right: Relating individual differences in right hemisphere contributions to language learning and relearning

Language, or the diverse set of dynamic processes through which symbolic, perceptual codes are linked to meaning representations in memory, has long been assumed to be lateralized to the left hemisphere (LH). However, after over 150 years of investigation, we still lack a unifying account of when, and for whom, a particular linguistic process relies upon LH or right hemisphere (RH) computations, or both. With a focus on individual differences, this article integrates existing theories of hemispheric contributions to language and cognition into a novel proposed framework for understanding how, when, and for whom the RH contributes to linguistic processes. We use evidence from first and second language learning and language relearning following focal brain damage to highlight the critical contributions of the RH.

Three aspects of representation in neuroscience

Neuroscientists often describe neural activity as a representation of something, or claim to have found evidence for a neural representation, but there is considerable ambiguity about what such claims entail. Here we develop a thorough account of what 'representation' does and should do for neuroscientists in terms of three key aspects of representation. (i) Correlation: a neural representation correlates to its represented content; (ii) causal role: the representation has a characteristic effect on behavior; and (iii) teleology: a goal or purpose served by the behavior and thus the representation. We draw broadly on literature in both neuroscience and philosophy to show how these three aspects are rooted in common approaches to understanding the brain and mind. We first describe different contexts that 'representation' has been closely linked to in neuroscience, then discuss each of the three aspects in detail.

New insights on single-neuron selectivity in the era of population-level approaches

In the past, neuroscience was focused on individual neurons seen as the functional units of the nervous system, but this approach fell short over time to account for new experimental evidence, especially for what concerns associative and motor cortices. For this reason and thanks to great technological advances, a part of modern research has shifted the focus from the responses of single neurons to the activity of neural ensembles, now considered the real functional units of the system. However, on a microscale, individual neurons remain the computational components of these networks, thus the study of population dynamics cannot prescind from studying also individual neurons which represent their natural substrate. In this new framework, ideas such as the capability of single cells to encode a specific stimulus (neural selectivity) may become obsolete and need to be profoundly revised. One step in this direction was made by introducing the concept of “mixed selectivity,” the capacity of single cells to integrate multiple variables in a flexible way, allowing individual neurons to participate in different networks. In this review, we outline the most important features of mixed selectivity and we also present recent works demonstrating its presence in the associative areas of the posterior parietal cortex. Finally, in discussing these findings, we present some open questions that could be addressed by future studies.

Predictive brain: Addressing the level of representation by reviewing perceptual hysteresis

In recent years, the idea that the prediction of sensory input is one of the major computational goals of the nervous system led to the development of several large-scale theories of brain functioning, such as different versions of the Bayesian approach to brain functions, predictive coding theories of cognition and the Free-energy principle. During the years, various empirical phenomena have been re-interpreted within such frames, and have been considered as consequences of predictive processing. Here we focus on perceptual hysteresis, or serial dependence, as an exemplary case. We unravel a potential gap in the predictive frameworks and raise the idea that alternative explanations of this effect can solve this issue, as they address the type of cognitive and neural representations involved.

Multiscale modeling of cortical gradients: The role of mesoscale circuits for linking macro- and microscale gradients of cortical organization and hierarchical information processing

The gradient concept in neuroscience describes systematic and continuous progressions of features of cortical organization across the entire cortex. Recent multimodal studies revealed a macroscale gradient from primary sensory to transmodal association areas which is linked to increasing representational abstraction along the cortical hierarchy, and which is paralleled by microscale gradients of cytoarchitecture and gene expression profiles. Convergent or divergent evidence from these multimodal studies is then used to support inferences about the existence of one common or multiple scale-specific gradients of hierarchical information processing. This paper evaluates the validity of such inferences within the framework of multiscale modeling. In branches of physics and biology where multiscale modeling techniques are used, the simple averaging of microscale details can introduce errors in macroscale modeling if it ignores structures at the intermediate mesoscales of organization which affect system behavior. Conversely, information about mesoscale structures can be used to determine which microscale details are actually relevant to macroscale behavior. In this paper, I similarly argue that multiscale modeling of cortical gradients needs to take organization of mesoscale circuits into account if it affects the structure-function relation that the models describe. Information about these circuits provides crucial evidence for evaluating inferences from micro- and macroscale data to the role of cortical gradients in hierarchical information processing. My application of the multiscale modeling framework reveals that the gradient concept tracks multiple overlapping progressions of cortical properties, rather than one overall gradient of hierarchical information processing. I support this argument by proposing a mesoscale gradient of connectivity which describes architectural differences between granular and agranular circuits, and which helps us better understand the relation between neural connectivity and hierarchical information processing.