Computational Mind vs. Navigating Mind: The Metaphor Fight Reshaping How Neuroscience Thinks About Thought

For more than half a century, neuroscience has framed the brain as something like a computer: an organ that processes information, manipulates symbols, and computes outputs from inputs. That metaphor has organized research across "highly diverse panels loosely stitched together into a patchwork quilt," as Churchland and Abbott put it in words quoted by the historian Matthew Cobb. But a countervailing idea is gaining a hearing: that the mind is better understood not as a calculator but as a navigator, an embodied system continually orienting itself in space and time. The question is no longer whether the computational metaphor is useful—it plainly has been—but whether it should remain the field's central organizing picture.

Introduction: A Shifting Paradigm in Neuroscience

The debate is, at its core, about metaphors. Not loose ones, but the generative images that decide what questions researchers ask, what models they build, and what counts as an explanation. The computational picture has held that position for decades. The alternative, articulated by Andrea Hiott in a conversation with Paul Middlebrooks on The Transmitter's Brain Inspired podcast, proposes a "navigational approach to the mind," one that treats cognition as a matter of orienting and moving rather than computing.

What makes the moment notable is that the challenge comes from inside the field's own successes. Hiott's argument draws on research related to the hippocampus, a brain region long studied for its role in spatial navigation, and asks whether navigation might be the more fundamental function from which other cognitive abilities grow. As she frames it, now is the time to embrace what she calls "seemingly paradoxical findings in neuroscience" rather than force them into a computational template. This article lays out both paradigms, the criticisms each faces, and what is genuinely at stake for how research gets done.

The Dominant View: The Computational Mind

The computational theory of mind, in its classic philosophical form, holds that mental activity is computation: that thinking consists in the manipulation of symbolic representations according to formal rules. The Stanford Encyclopedia of Philosophy's entry on the computational mind traces this tradition through several variants, including connectionism, which models cognition using networks of simple units rather than explicit symbol manipulation.

A more recent descendant is computational neuroscience, which builds mathematical models of actual neural systems. As the Stanford Encyclopedia explains, this approach differs from earlier versions of the theory in a crucial respect: it abandons multiple realizability, the idea that the same computation could run on silicon as well as on neurons. Because computational neuroscientists cite specific neurophysiological properties and processes, their models do not apply equally well to, say, a sufficiently different silicon-based creature. Philosophers once found that substrate-independence attractive; computational neuroscientists accept the sacrifice in exchange for insight into neurophysiological underpinnings.

There is also a levels-of-explanation version of the view. As one framework published through PMC describes it, mental events and mechanistic brain events can be treated as different levels of explanation of the operation of a single computational system—linked, not by one causing the other, but by the computational process being performed. On this account, the computational framing offers a way between dualism and reductive physicalism.

Critiques of the Computational Metaphor

The strongest criticisms do not claim the computational picture is false; they claim it is partial in a way that misdirects research. The Stanford Encyclopedia records the complaint in its sharpest form: the theory "treats mental activity as static symbol manipulation detached from the embedding environment," neglecting the myriad complex ways the environment causally or constitutively shapes mental activity. The critics' proposed replacement "emphasizes continuous links between mind, body, and environment," with agent-environment dynamics, not internal mental computation, holding the key to understanding cognition.

Computationalists have a reply. The same entry notes that they argue computational models can take into account how mind, body, and environment continuously interact—so the dispute is over emphasis and research priorities, not a simple refutation.

Matthew Cobb, writing in the Guardian, presses a different worry: the metaphor's cultural overreach. Some researchers, he observes, treat the mind as an operating system implemented on neural hardware, implying that a mind, "seen as a particular computational state, could be uploaded on to some device or into another brain." He calls that wrong, or at best hopelessly naive, on the materialist hypothesis that brains and minds are identical. Uploading would require understanding neuronal function far beyond anything currently envisaged, plus a simulation precisely mimicking the structure of the brain in question. His verdict: mind-uploading can be dismissed as fantasy, at least until the far future.

Cobb also notes that skepticism about machine metaphors is old. In 1951, the pioneer neuroscientist Karl Lashley argued against any machine-based metaphor at all; in 2015 the roboticist Rodney Brooks chose the brain-as-computer idea as his "pet hate"; and decades earlier the historian S. Ryan Johansson argued that endlessly debating the truth or falsity of the metaphor was "a waste of time."

The Emerging Alternative: The Navigating Mind

Hiott's navigational approach starts from a different image. Where the computational mind processes information, the navigating mind orients: it regulates the body, interacts with the world, and makes predictions from a position embedded in space and time. The hippocampus—long known for its role in spatial navigation—anchors the empirical side of the proposal, suggesting that navigation may be a foundational function from which other capacities can be understood.

Two cautions belong here. First, this framework is presented as an emerging alternative, and it does not yet carry the depth of historical literature that the computational theory has accumulated across philosophy, connectionism, and computational neuroscience. Second, the debate itself is a disagreement about emphasis and fruitfulness, not a settled verdict; Cobb notes that the brain-as-computer metaphor "retains its dominance, although there is disagreement about how strong a metaphor it is."

Paradigm Shift: Computational vs. Navigating Mind

The table below juxtaposes the two paradigms across four dimensions: their core metaphor, their fundamental unit of cognition, their key brain focus, and their philosophical implications. Reading down each column shows the substance of the disagreement—two different answers to the same questions about what a mind is for.

Core tenets of the computational mind versus the navigating mind
Dimension	Computational Mind	Navigating Mind
Core metaphor	The brain as a computational system, akin to a computer, processing information — the dominant metaphor for decades.	The brain as a navigating system, constantly orienting itself in space and time — presented as an alternative metaphor for understanding cognition.
Fundamental unit of cognition	Information processing, modeled mathematically at multiple scales from neurons to behavior, using theoretical analysis and simulation.	Regulation of the body and interaction with the world, with the mind continuously making predictions — emphasizing continuous links between mind, body, and environment.
Key brain focus	Neurophysiological properties and processes, modeled computationally; mental and mechanistic levels linked by the computational process being performed.	The hippocampus and spatial-navigation research, treated as a possible foundation for broader cognitive functions.
Philosophical implication	Critics argue it treats mental activity as static symbol manipulation detached from the environment, and can invite naive ideas such as mind-uploading.	Shifts focus toward the brain as a biological organ interacting with its environment — agent-environment dynamics rather than isolated symbol manipulation.

Read together, the rows make the fault line visible. The computational tradition explains cognition by what happens inside the system—representations manipulated, computations performed. The navigational tradition explains it by the system's ongoing relationship to its body and surroundings. Neither column is empty; the question is which framing generates the better research program going forward.

Navigating the Future of Neuroscience

Neuroscience researchers and students should consider both the established strengths of the computational paradigm and the emerging insights offered by the navigating mind framework when approaching studies of cognition and behavior. The computational view has delivered decades of formal models and, in its neuroscientific form, genuine insight into neurophysiological underpinnings; the navigational view promises to reframe paradoxical findings—especially around the hippocampus and spatial cognition—as something other than anomalies to be forced into an inherited template.

The debate will not be settled by argument alone. The observable signal to watch is the increasing publication of research papers and review articles that explicitly compare or integrate aspects of both computational and navigational approaches to brain function, or that propose novel experimental designs informed by the navigating mind metaphor. If the navigational framing is more than a podcast-level provocation, it will show up there first—as testable designs, not just a new metaphor for old questions.