Book review of the adaptable mind: What neuroplasticity and neural reuse tells us about language and cognition

Xueyao Pan, Xi LiFrontiers in Neuroscience 2022 · DOI

TL;DR

This paper reviews John Zerilli’s book, The Adaptable Mind, which challenges the traditional "Swiss Army knife" model of the brain. Instead of isolated, hardwired modules for functions like language, the author argues for a model of "neural reuse" and neuroplasticity, where brain regions are flexible, recycled, and interconnected to support complex cognition.

What Problem Does This Paper Address?

For decades, cognitive science has been dominated by the "modularity of mind" hypothesis, championed by figures like Noam Chomsky and Jerry Fodor. This view suggests that the brain consists of functionally and anatomically isolated subsystems—essentially "organs" dedicated to specific tasks like language or vision.

However, modern neuroscience increasingly contradicts this "hard" modularity. We now see evidence of neural reuse, where the same brain regions are "recycled" by different neural communities to perform various tasks. The problem the book (and this review) addresses is how to reconcile the apparent specialized nature of human language with the reality of a highly plastic, interconnected brain. If there is no "language organ," how does language actually work?

Methods

As this is a book review, the "methods" consist of a critical synthesis and summary of John Zerilli’s interdisciplinary arguments. Zerilli draws from:

  • Neuroscience: Evidence of synaptic and cortical map plasticity.
  • Evolutionary Biology: The "Massive Redeployment Hypothesis," which suggests evolution favors reusing existing neural circuits over building new ones from scratch.
  • Graph Theory: Using network science to define "soft" modules as communities of interconnected nodes rather than isolated physical regions.
  • Philosophy of Mind: Re-evaluating the "Multiple Realization" hypothesis and the relationship between psychological states and neural substrates.

Datasets

Not discussed in the paper. As a theoretical book review, it synthesizes existing research and literature rather than introducing new empirical datasets.

Key Insights and Ideas

  • From "Hard" to "Soft" Modularity: Zerilli proposes a "soft" version of modularity. Instead of anatomically fixed boxes, modules should be viewed as functionally interconnected "communities" of nodes. The specificity of a module comes from the global structure of the network, not the individual neurons themselves.
  • The Metamodal Brain: The brain may not be domain-specific (e.g., a "language" area) but rather modality-specific or process-specific. For example, a region might be specialized for processing sequences, which is a requirement for both motor control and syntax.
  • Neural Reuse vs. Redundancy: The review distinguishes between these two. Neural reuse means the same neural "tokens" (specific circuits) are recruited for different tasks. Neural redundancy means the same neural "types" (similar structures) are replicated across the cortex, allowing one area to take over if another is damaged.
  • Language as a "Latecomer": From an evolutionary perspective, language is a relatively recent development. It is more likely that language "hijacked" and adapted existing neural resources (like those used for motor planning or tool use) rather than evolving a brand-new, dedicated hardware module.

Key Results

The review highlights several qualitative conclusions from Zerilli’s work:

  • Refutation of the "Language Organ": The traditional view of a hardwired language module is likely incorrect. Even "classic" areas like Broca’s area are now known to be involved in non-linguistic action-related tasks.
  • Support for Grounded Cognition: The book provides theoretical weight to "embodied" or "grounded" cognition—the idea that high-level language processing is not self-sufficient but depends on lower-level sensory and motor systems.
  • The "Global Workspace" View: Rather than a single module, language is better understood as a large-scale network involving the orbital frontal-temporal occipital network and the middle temporal lobe.

Limitations and Future Work

The reviewers note a few areas where the book leaves questions unanswered:

  • Lack of a "Smoking Gun": While the book effectively refutes traditional modularity, it does not provide a definitive, "clear-cut answer" for the exact neural architecture of language.
  • Complexity of Substrates: The specific neural substrates that make language unique remain a subject of intense debate.
  • Practical Applications: The reviewers suggest that future work should focus on how these "neural reuse" insights can be applied to practical fields like language rehabilitation (post-stroke), language learning, and the development of more flexible Artificial Intelligence.

Summary updated

Abstract

From the works of Noam Chomsky to Jerry Fodor, the modularity of mind has been deeply rooted in cognitive science. From the “modular” perspective, cognition consists of functionally and anatomically isolable subsystems. Therefore, it is reasonable to assume a language module or a hardwired language organ that is functionally and anatomically different from the other cognitive modules (e.g., vision, olfaction, motor). However, evidence from neuroscience casts doubt on this modular hypothesis. It has been shown that many brain regions are likely to be reused and recycled by various neural communities in order to serve various cognitive functions. If this is the case, language and other cognitive faculties should not be considered modules, since they cannot be realized in special-purpose, special-structure regions. Therefore, it is time to change our understanding of our brains.
BibTeX
@inproceedings{pan2022book,
  author = {Xueyao Pan and Xi Li},
  title = {{Book review of the adaptable mind: What neuroplasticity and neural reuse tells us about language and cognition}},
  booktitle = {Frontiers in Neuroscience},
  year = {2022},
  doi = {10.3389/fnins.2022.978196},
}
Figure