Place Field

Reify a place cell's spatial tuning as a measurable object — the bounded region where its firing rate is reliably elevated — turning 'the hippocampus represents space' into a battery of scalars (size, peak, stability, remapping) that travel across preparations.

Core Idea

A place field is the bounded region of an environment in which a given place cell's firing rate is reliably elevated above its baseline; it is the spatial-tuning property that defines the cell and gives the place-cell framework its empirical grip. In standard rodent arenas, fields are roughly 2D bell-shaped regions with diameters on the order of tens of centimetres, but field size varies systematically along the hippocampal long axis: dorsal CA1 cells carry small, precise fields; ventral CA1 cells carry large, coarsely-grained ones — a gradient that maps onto the precision-generalisation trade-off in spatial cognition. A single cell typically has one field in a given environment but may have distinct fields across environments, and the fields of simultaneously recorded cells tile the arena with overlapping coverage, enabling accurate population-vector decoding of location. Within a field, firing is not rate-uniform: as the animal traverses the field in a single theta cycle, the cell fires progressively earlier in the theta oscillation with each successive position step — a phenomenon called theta-phase precession — which encodes within-field position at finer temporal resolution than the firing rate envelope alone. Fields are stable across visits to the same environment in healthy animals but remap when the environment changes substantially; instability of fields under fixed environmental conditions is a sensitive marker of hippocampal dysfunction, observed in Alzheimer's disease mouse models before overt behavioural deficits appear. Field formation in a novel environment proceeds over minutes to hours as an initially noisy population converges on a stable spatial code, a process dependent on LTP-like plasticity at entorhinal-hippocampal synapses and on off-line replay during subsequent sleep.

Structural Signature

Sig role-phrases:

• the underlying place cell — the hippocampal neuron whose firing is being characterized as a property distinct from the cell itself
• the bounded region — the spatial extent within which firing rate is reliably elevated above baseline, allocentrically referenced
• the tuning scalars — the measurable battery the field reifies: size, peak rate, shape, location, stability across visits, remapping behaviour
• the size-position gradient — field size scaling along the hippocampal long axis (small/sharp dorsal, large/coarse ventral), reading a precision-generalisation trade-off off anatomy
• the within-field theta-phase precession — firing progressively earlier in each successive theta cycle, multiplexing fine within-field position onto a second channel above the rate envelope
• the stability baseline — the calibrated expectation that fields hold steady within an environment, so instability under unchanged conditions reads as circuit dysfunction
• the two-level discipline — single-field tuning properties held distinct from population properties (coverage, overlap, decoding accuracy)
• the formation trajectory — convergence from an initially noisy population onto stable fields over minutes to hours via LTP-like plasticity and off-line replay

What It Is Not

• Not the same thing as the place cell. The cell is a fixed piece of tissue; the field is a property it expresses in a particular environment — one it can lose, relocate, or duplicate when the environment changes. Reifying the field as a separable, sized object is exactly what makes spatial tuning measurable; collapsing it back into the neuron forfeits that grip.
• Not a region the cell or animal "draws" on space. The bounded field is a consequence of receptive-field-like tuning — where firing rate happens to exceed baseline — not an act of agentive boundary-drawing or segmentation. The edge is a tuning curve falling off, not a partition imposed on the environment.
• Not incidental in its size. Field size is an explanatory variable, not noise: it scales systematically along the hippocampal long axis (small and sharp dorsally, large and coarse ventrally), reading a built-in precision–generalisation trade-off straight off anatomy. Dismissing size as measurement scatter discards a structural signal about what the field is for.
• Not fully captured by the firing-rate envelope. Within a field, firing is not rate-uniform: theta-phase precession multiplexes finer within-field position onto a second, temporal channel, so two moments the rate map cannot separate are distinguishable by where in the theta cycle the spikes fall. The rate map is one channel, not the whole code.
• Not "remapping means dysfunction." Remapping when the environment changes substantially is normal, expected behaviour. The pathological signature is the opposite — field instability under unchanged conditions (enlarged, drifting fields) — which marks circuit dysfunction and can precede overt behavioural deficits. Confusing the two reads healthy environmental coding as failure.

Scope of Application

The place field lives across the systems, computational, behavioural, and translational subfields of neuroscience; its reach is within that domain, bounded by the hippocampal, allocentric, theta-organised substrate — the receptive-field skeleton it instantiates is what travels to vision tuning, CNN filters, and kernel methods, while "place field," as named, stays in the cognitive substrate.

• Systems neuroscience — characterising a place cell is characterising its field, reading size, peak rate, shape, location, stability, and remapping as the cell's spatial-tuning property.
• Computational neuroscience — models of how grid-cell plus boundary input produces a field, and of how field size and overlap set population-decoding accuracy.
• Behavioural neuroscience — field stability and remapping used as readouts of memory and environmental representation in spatial-learning studies.
• Translational research — enlarged, drifting, unstable fields in Alzheimer's disease mouse models predicting spatial-memory deficits before overt behaviour fails.
• Within-circuit coding analogues — the same field machinery reappearing as "time fields" in time cells and "concept fields" in concept cells, with the size-to-role gradient, stability baseline, and theta-phase-precession diagnostics carrying intact.

Clarity

Treating the place field as an object distinct from the place cell is what turns spatial coding into something measurable. The cell is a fixed piece of tissue; the field is a property it expresses in a particular environment — one the cell can lose, relocate, or duplicate when the environment changes. Reifying that property gives the experimenter a handful of quantities to put numbers on — field size, peak rate, shape, location, stability, remapping behaviour — and so converts "the hippocampus represents space" into a battery of comparable measurements that travel across experiments, species, and disease models. It also licenses a clean two-level analysis: single-field properties (how sharply this cell is tuned, how it precesses in theta phase within its field) sit at one level, population properties (coverage, overlap, decoding accuracy) at another, and confusions about hippocampal coding often turn out to be a failure to say which level is meant.

Naming the field also makes its own size a variable to be explained rather than a fixed fact. Because fields scale systematically along the hippocampal long axis — small and precise dorsally, large and coarse ventrally — the framework exposes a precision-generalisation trade-off built into the anatomy, and lets a researcher ask what field size is for rather than treating it as noise. And by making field stability a measurable baseline, it sharpens a diagnostic question: instability under unchanged conditions is not normal remapping but a signature of circuit dysfunction, detectable before behaviour visibly fails.

Manages Complexity

"The hippocampus represents space" is, on its own, an unmanageable claim: it names no quantity, supports no measurement, and travels nowhere across the welter of arenas, species, recording rigs, and disease models in which spatial coding is studied. Reifying the field — treating the bounded region of elevated firing as an object the cell expresses, separable from the cell itself — is what tames that sprawl, by replacing the amorphous claim with a fixed, finite battery of scalars every recording can be reduced to: field size, peak rate, shape, location, stability across visits, remapping behaviour, and within-field theta-phase precession. The unbounded question "how does this circuit encode the world?" collapses into reading off a handful of numbers per cell, and because the same numbers are defined identically everywhere, they travel — a dorsal-CA1 field in a rat, a 3D field in a bat, a field in a human intracranial recording, and a field in an Alzheimer's mouse model become directly comparable measurements rather than incommensurable observations.

What the analyst then tracks, and what they read off, separates cleanly into two levels the construct keeps from blurring. At the single-field level sit tuning quantities — size, peak, shape, precession — and field size in particular becomes an explanatory variable rather than noise: because it grows systematically along the hippocampal long axis, small and sharp dorsally to large and coarse ventrally, the analyst reads a built-in precision-generalisation trade-off straight off anatomical position, predicting fine spatial resolution dorsally and broad generalisation ventrally without re-deriving it per cell. At the population level sit coverage, overlap, and decoding accuracy, read from the ensemble of fields jointly. Crucially, stability becomes a calibrated baseline against which a single diagnostic branch is read: fields are expected to hold steady across repeated visits to one environment and to remap when the environment changes substantially — so instability under unchanged conditions is not normal remapping but a signature of circuit dysfunction, and reads as an early marker of pathology detectable before behaviour visibly fails. The move is from an unmeasurable global assertion about spatial representation to a small parameter set — a per-field tuning vector plus population coverage plus a stability baseline — whose values the experimenter reads to compare circuits across substrates, infer what a field's size is for, and flag a failing hippocampus, instead of re-characterising "spatial coding" anew in every preparation.

Abstract Reasoning

Reifying the field as a measurable object licenses the field's most distinctive move: inferring a cell's functional role from its anatomical position by way of field size. Because fields scale systematically along the hippocampal long axis — small and sharp in dorsal CA1, large and coarse in ventral CA1 — the analyst reasons that field size is not noise but an explanatory variable, and reads a built-in precision-generalisation trade-off straight off where the cell sits: a dorsal cell is inferred to provide fine spatial resolution, a ventral cell broad generalisation. The inference runs from a structural fact (position on the long axis) to a computational one (the granularity of the spatial code the cell contributes), and it lets the researcher ask what a given field size is for rather than treating size as an incidental property. This is reasoning the place cell framing does not supply; it depends on treating the field as a graded, sized object distinct from the neuron expressing it.

The field's signature diagnostic is reading stability against a calibrated baseline. The framework establishes the normal expectation — fields hold steady across repeated visits to one environment and remap only when the environment changes substantially — so the analyst can interpret a deviation from it as a hidden cause. Field instability under unchanged conditions is inferred not as ordinary remapping but as a signature of circuit dysfunction, and because it shows up before overt behavioural deficits (enlarged, drifting fields in Alzheimer's disease mouse models), it functions as an early marker: the analyst reasons from a degraded field property to a failing hippocampus in advance of the behaviour that would otherwise reveal it. The move requires the baseline to be quantified, which is exactly what reifying the field provides.

A third move reads within-field position at finer resolution than the firing-rate envelope by exploiting theta-phase precession. Because the cell fires progressively earlier in each successive theta cycle as the animal advances through the field, the analyst infers that the phase of firing carries positional information the rate alone does not — so two moments that the rate map cannot distinguish (both near the field's peak) are separable by where in the theta cycle the spikes fall. The reasoning is that the field encodes position on two multiplexed channels, a coarse rate envelope and a fine temporal phase, and the analyst extracts the finer estimate from the phase relationship rather than from rate.

These inferences come with the construct's two-level discipline and a substrate boundary, both themselves reasoning moves. The discipline is to keep single-field properties (size, peak rate, shape, precession) distinct from population properties (coverage, overlap, decoding accuracy), and the analyst must say which level a claim is pitched at, because confusions about hippocampal coding often reduce to conflating the tuning of one cell with the representation carried by the ensemble. The construct also supports an order-of-events inference about field formation: a novel environment yields an initially noisy population that converges over minutes to hours onto stable fields, a process the analyst attributes to LTP-like plasticity at entorhinal-hippocampal synapses and to off-line replay during subsequent sleep — so a freshly-formed, still-noisy field is read as early in that consolidation trajectory rather than as malfunction. The boundary is that all of this is anchored to the hippocampal, allocentric, theta-organised substrate: the size-to-role inference, the stability diagnostic, and the phase-precession read have force because the field is a receptive field in allocentric space within the hippocampal circuit, and it is those substrate-specific commitments — not the generic notion of a bounded tuning region — that make the predictions about size, stability, and within-field coding load-bearing.

Knowledge Transfer

Within neuroscience the place-field construct transfers as mechanism, and because the field is a reified, measurable property it travels as a fixed battery of scalars rather than a slogan. The same quantities — field size, peak rate, shape, location, stability across visits, remapping behaviour, theta-phase precession — are defined identically across systems neuroscience (characterising a place cell is characterising its field), computational neuroscience (modelling how grid-cell plus boundary input produces a field, how field size and overlap set decoding accuracy), behavioural neuroscience (field stability and remapping as memory readouts), and translational research (enlarged, drifting fields predicting spatial-memory deficits before behaviour fails). That common definition is exactly what lets a dorsal-CA1 field in a rat, a 3D field in a bat, a field in a human intracranial recording, and a field in an Alzheimer's mouse model be directly compared. The construct has also transferred within the home circuit across coding content: the same field machinery reappears as "time fields" in time cells and "concept fields" in concept cells, and the diagnostics (read the size-to-role gradient along the long axis, flag instability against the stability baseline, extract within-field position from theta phase) carry with it — but this is reach within the hippocampal, allocentric, theta-organised substrate.

Beyond that substrate the honest reading is shared abstract mechanism (case B), and the place field is an unusually clean instance of it because its generalisation already has a name. What travels is the receptive-field skeleton — a unit tuned to a bounded region of some input space, with population coverage of that space yielding accurate decoding — and that skeleton genuinely recurs as co-instances across distinct substrates: sensory-neuron tuning curves, convolutional-network receptive fields, radial-basis-function centres, Gaussian-process kernels, and attentional regions are all the same pattern, not metaphors for it. The place field is one instance of that pattern, specialised to allocentric space in the hippocampus. So the cross-domain lesson should carry the receptive-field / population_coding parent, which transfers literally wherever a population of bounded tunings covers an input space; the home-bound cargo — what makes a place field a place field — is the allocentric world-centred reference frame, environment-specific remapping, theta-phase precession, and the hippocampal circuit, none of which leaves the cognitive substrate. ML architectures that borrow "location-specific bounded responses" (RBF networks, location-aware embeddings) are therefore instantiating the general receptive-field pattern, not reusing the place-field-specific commitments. The candidate's broadest hint — that a substrate-general receptive_field pattern subsumes vision tuning, CNN filters, kernel methods, and attention — is correct, but that is the parent that should travel; "place field," as named, is the hippocampal instance and should not.

Examples

Canonical

The seminal demonstration is O'Keefe and Dostrovsky's 1971 recording from the hippocampus of freely-moving rats. Sampling single units in the dorsal hippocampus while the animal explored an enclosure, they found neurons that fired vigorously only when the rat occupied a particular part of the environment and were near-silent elsewhere — the first place cells, each expressing a bounded region of elevated firing. What made the finding tractable was treating that region as a measurable object: one could plot where in the arena the cell fired, mark the boundary where rate fell back to baseline, and read off the field's size, location, and peak rate. From that single observation grew the whole battery of scalars — a discovery that earned O'Keefe a share of the 2014 Nobel Prize and turned "the hippocampus represents space" into something one measures cell by cell.

Mapped back: The CA1 neuron that fires only in one part of the enclosure is the underlying place cell; the patch of arena where its rate is elevated above baseline is the bounded region, allocentrically referenced to the environment. Plotting that region and marking where firing returns to baseline is exactly the act of reifying the tuning scalars — size, location, peak rate — that the rest of the framework then measures and compares across preparations.

Applied / In Practice

Translational neuroscience turns field stability into a diagnostic read against the healthy baseline. Cacucci and colleagues (2008) recorded hippocampal place cells in aged Tg2576 mice — a transgenic model that accumulates amyloid — and found their place fields markedly less stable and less spatially precise than those of wild-type controls across repeated visits to the same, unchanged environment. Because a healthy field is expected to hold steady when nothing about the environment changes, this instability could not be read as ordinary remapping; it was a signature of a failing circuit. Crucially, the degree of place-field disruption tracked both the animals' spatial-memory impairment and their amyloid plaque burden, positioning unstable fields as a physiological readout of pathology. Subsequent AD models likewise reported enlarged, drifting fields emerging before overt behavioural deficits — the field property degrading ahead of the behaviour it supports.

Mapped back: The expectation that fields hold steady across repeated visits to one unchanged arena is the stability baseline; the transgenic animals' drifting, imprecise fields are a deviation from it read not as normal remapping but as circuit dysfunction. The disrupted properties are the tuning scalars (stability, precision) degrading, and their appearance before behaviour visibly fails is the early-marker inference the reified, quantified baseline makes possible.

Structural Tensions

T1: Reification versus over-objectification (treating the field as a thing wins measurement but courts a category slip). Separating the field from the cell — the field a property the cell expresses, losable, relocatable, duplicable — is exactly what turns "the hippocampus represents space" into a battery of scalars that travel across preparations. That reification is the construct's whole grip. But objectifying the field invites the slip the entry warns against: reading the bounded region as something the cell or animal "draws" on space, an act of segmentation, rather than as a tuning curve falling off where firing happens to drop below baseline. The more concrete and thing-like the field is made in order to measure it, the easier it becomes to attribute agency or intent to what is only a receptive-field consequence. The edge is a threshold crossing, not a partition imposed. Diagnostic: Is the field's boundary being treated as a tuning curve returning to baseline, or as a line the system actively drew on the environment?

T2: The scalar battery versus what the numbers omit (comparability bought by discarding structure). Reducing each recording to size, peak, shape, location, stability, and remapping is what makes a rat field, a bat field, a human intracranial field, and an Alzheimer's-model field directly comparable. But a fixed finite battery is a decision about what counts, and firing within a field is not rate-uniform: theta-phase precession multiplexes finer within-field position onto a temporal channel the rate scalars do not capture. So the very reduction that lets fields travel as comparable measurements also flattens a second coding channel into invisibility unless precession is separately tracked. The battery's portability and its incompleteness are the same move — every scalar added for comparability is a commitment that the code lives in these quantities, and the temporal channel shows that commitment can under-count the real code. Diagnostic: Does the claim at hand rest only on the rate-envelope scalars, or does it require the within-field phase information the standard battery leaves out?

T3: Stability as diagnostic baseline versus remapping as normal function (the same instability reads two opposite ways). Making stability a calibrated baseline is what lets field drift under unchanged conditions flag circuit dysfunction before behaviour fails — a genuine early-marker payoff. But fields are supposed to remap when the environment changes substantially; remapping is healthy environmental coding, not failure. The diagnostic therefore rides entirely on a judgment the construct cannot itself certify: whether the environment truly held constant. Misjudge that — an unnoticed contextual change, a shift in the animal's internal state — and normal remapping is misread as pathology, or genuine dysfunction is excused as expected remapping. The pathological signature and the adaptive one are the same observable (a field that moved); only the assumed constancy of conditions separates them. Diagnostic: Was the environment genuinely unchanged between visits, or has an uncontrolled contextual shift made healthy remapping look like a failing circuit?

T4: Anatomy-reads-role versus overgeneralized gradient (the long-axis inference is powerful and coarse). The size-to-role move — small sharp dorsal fields for fine resolution, large coarse ventral ones for generalization — lets the analyst read a precision-generalization trade-off straight off anatomical position without re-deriving it per cell. That is reasoning the place-cell framing cannot supply. But it is a statistical gradient, not a law: it licenses inferring a cell's computational granularity from where it sits, and a systematic gradient invites treating position as destiny, ignoring the within-region scatter and the possibility that field size is set by inputs (grid scale, boundary geometry, experience) that anatomy only correlates with. The inference's strength — reading function off structure cheaply — is exactly what makes it prone to over-attributing role to location. Diagnostic: Is this cell's field size being predicted from its long-axis position as a strong default, or being explained by it as though anatomy fixed the code?

T5: Two-level discipline versus the pull to conflate them (single-field tuning against population representation). Keeping single-field properties (size, peak, precession) distinct from population properties (coverage, overlap, decoding accuracy) is load-bearing: the entry notes that confusions about hippocampal coding often reduce to conflating one cell's tuning with the ensemble's representation. But the construct's own success pressures the conflation — because the reified field is so measurable at the single-cell level, it is tempting to narrate "the representation of space" in the vocabulary of one field's scalars, or conversely to attribute population-level decoding accuracy to individual field sharpness. The discipline that makes the two-level analysis clean is precisely what the vividness of the single-field object works to erode. Diagnostic: Is this claim pitched at the tuning of one cell or at the representation carried by the ensemble — and does the argument silently slide between the two?

T6: Autonomy versus reduction (a hippocampal place field or the receptive-field pattern it instantiates). "Place field" is a named, Nobel-anchored systems-neuroscience object with proprietary commitments — allocentric world-centred reference, environment-specific remapping, theta-phase precession, the hippocampal circuit — and within neuroscience it travels as mechanism, the same scalar battery comparing rat, bat, human, and disease-model fields, and reappearing as "time fields" and "concept fields" within the circuit. But beyond that substrate the honest traveler is the receptive-field / population_coding skeleton it instantiates: a unit tuned to a bounded region of some input space, with population coverage yielding accurate decoding. That skeleton recurs as genuine co-instances — sensory tuning curves, CNN receptive fields, RBF centres, Gaussian-process kernels, attentional regions — none metaphors, all the same pattern. An RBF network and a CA1 cell tile an input space the same way, but there is no place field in the network, only the receptive-field pattern made concrete. Diagnostic: Resolve toward the parent (receptive_field / population_coding) when asking what carries outside the hippocampus; toward the named place field when the allocentric frame, remapping, and theta organization are doing the work in situ.

Structural–Framed Character

Place field sits toward the structural end of the spectrum but stops short of the pole — best read as mixed-structural, like the place cell whose tuning it reifies. Its structural credentials hold on four of the five criteria. Its evaluative weight is nil: a bounded region of elevated firing is neither good nor bad, and the scalar battery (size, peak, shape, stability) measures rather than judges — even the instability that flags a failing circuit is read diagnostically, not as a verdict the concept renders. It is not human-practice-bound: fields form, hold steady, and remap in the animal's hippocampus regardless of who is watching; theta-phase precession multiplexes within-field position on the circuit's own clock, not on an observer's. Its institutional origin is none: the field is a consequence of receptive-field-like tuning — where firing rate exceeds baseline — a fact of physiology rather than an artifact of a survey or convention (the framework reifies a property nature already expresses, it does not invent it). And within its range cross-domain reuse is recognition rather than import: the same field machinery is recognized intact across species, disease models, and within-circuit analogues (time fields, concept fields), carrying its diagnostics with it.

What keeps it off the structural pole is vocab_travels, which it fails. The distinctive vocabulary — allocentric world-centred reference, environment-specific remapping, theta-phase precession, the hippocampal long-axis size gradient, CA1 — is pinned to the neural substrate and does not float free; beyond the hippocampus the import_vs_recognize mark flips to analogy. The portable structural skeleton is the receptive_field pattern (with population_coding as its ensemble face): a unit tuned to a bounded region of some input space, many such tunings covering the space to yield accurate decoding. That skeleton genuinely recurs as co-instances — sensory tuning curves, CNN filters, RBF centres, Gaussian-process kernels, attentional regions — but it is precisely what the place field instantiates from its parent, not what makes "place field" travel: the cross-domain reach belongs to the receptive-field / population-coding umbrella, while the allocentric, remapping, theta-organised specifics are the domain accent that stays home. Its character: a real, evaluatively-neutral, recognized-in-nature spatial-tuning object whose portable spine is the receptive-field pattern, expressed in hippocampal vocabulary that keeps "place field" itself in the cognitive substrate — mixed-structural, not a prime.

Structural Core vs. Domain Accent

This section decides why place field is a domain-specific abstraction and not a prime — separating the portable tuning skeleton from the hippocampal object that reifies it.

What is skeletal (could lift toward a cross-domain prime). Strip the neuroscience and a thin relational structure survives: a unit responds selectively over a bounded region of some input space, its response falling back to baseline outside that region, and a population of such bounded tunings covering the space yields accurate read-out of the current input. That is the receptive-field pattern — selectivity as a localized bump on an input dimension, coverage by many overlapping bumps, decoding from the ensemble — with the reifying move that turns a tuning into a measurable object carrying scalars (extent, peak, location) that let one tuning be compared to another. This skeleton is genuinely substrate-portable, which is exactly why the entry names receptive_field as the parent that travels, with population_coding as its ensemble face. That portable core is what place field shares, not what makes it a place field.

What is domain-bound. Almost everything that makes the construct this object is hippocampal furniture and none of it survives extraction: the allocentric world-centred reference frame that keys the bump to external space rather than the body; the environment-specific remapping by which a cell can lose, relocate, or duplicate its field; the theta-phase precession that multiplexes fine within-field position onto a temporal channel above the rate envelope; the long-axis size gradient (small/sharp dorsal, large/coarse ventral) that reads a precision–generalisation trade-off off anatomy; the stability baseline against which drift under unchanged conditions flags circuit dysfunction; and the LTP-plus-replay formation trajectory. These are the worked vocabulary, the instruments (arena recording, population decoding, disease-model assays), and the empirical cases the discipline studies. The decisive test: remove the allocentric frame, the remapping, and the theta organisation and what remains is no longer a place field but a bare bounded tuning — a receptive field in some space, the looser and more general thing.

Why this does not clear the prime bar. A prime is a relational structure whose vocabulary travels and whose cross-domain transfer is recognition of the same mechanism, not analogy. Place field's transfer is bimodal. Within the hippocampal, allocentric, theta-organised substrate it travels intact — the same scalar battery compares a dorsal-CA1 field in a rat, a three-dimensional field in a bat, a human intracranial field, and an Alzheimer's-model field, and the machinery reappears as "time fields" and "concept fields," genuine recognition of one mechanism. Beyond that substrate it travels only by analogy: architectures that borrow "location-specific bounded responses" — RBF centres, location-aware embeddings — rename every component, and what actually recurs there is the receptive-field pattern, not the place field's proprietary commitments. And when the bare structural lesson is needed cross-domain — a bounded tuning whose population covers an input space — it is already carried, in more general form, by the receptive_field / population_coding umbrella, which subsumes sensory tuning curves, CNN filters, kernel centres, and attentional regions as true co-instances, not metaphors. The cross-domain reach belongs to that parent; "place field," as named, carries allocentric, remapping, theta-precession, hippocampal baggage that should stay home.

Relationships to Other Abstractions

Current abstraction Place Field Domain-specific

Parents (1) — more general patterns this builds on

• Place Field is a kind of Receptive Field Prime

  A place field is a receptive field specialized to the bounded spatial region in which a hippocampal place cell responds above baseline.

Children (1) — more specific cases that build on this

• Place Cell Domain-specific is part of Place Field

  A place field is the defining spatial-response region contained in the place-cell abstraction and is the measurable property by which the cell is identified.

Hierarchy path (1) — routes to 1 parentless root

• Place Field → Receptive Field → Boundary

Not to Be Confused With

• Place cell. The neuron itself — a fixed piece of hippocampal tissue. The place field is the property that cell expresses in a particular environment: the bounded region of elevated firing, a sized object separable from the neuron and losable, relocatable, or duplicable when the environment changes. Reifying the field as distinct from the cell is exactly what makes spatial tuning measurable; collapsing the two forfeits that grip. Tell: is the referent the unit of tissue (place cell) or the measurable region it produces, with size/peak/stability scalars (place field)?

• Receptive field. The substrate-neutral parent pattern — a unit selectively responsive over a bounded region of some input space, with a population of such tunings covering the space to yield decoding. The place field is the hippocampal instance of that pattern, specialized to allocentric physical space and carrying remapping and theta organization. The receptive field is the umbrella that genuinely travels (to sensory tuning curves, CNN filters, RBF centres, GP kernels); the place field is one co-instance. Treated more fully in the Knowledge Transfer and Structural Core vs. Domain Accent sections. Tell: strip the allocentric frame, remapping, and theta precession and what remains is a bare receptive field in some space — the parent, not the place field.

• Grid field. The firing region of a grid cell — but a grid cell has many such fields arranged in a repeating hexagonal lattice tiling the whole arena, forming a periodic metric. A place field is (typically) a single bounded region per environment with no lattice periodicity. Tell: is the region one of a hexagonally-repeating set spanning the environment (grid field), or a lone bounded patch (place field)?

• Remapping. Not an object but a behaviour of fields — the wholesale reshuffling (global) or rate change (rate) of fields when the environment changes. Remapping is something place fields do; it is measured as one of the field's tuning scalars, not a separate structure. Crucially, remapping under a changed environment is normal, whereas field instability under unchanged conditions is the pathological signature. Tell: are you naming the reified region (place field) or the environment-driven reconfiguration of that region (remapping)?

• Theta-phase precession. A within-field temporal phenomenon — the cell fires progressively earlier in each successive theta cycle as the animal advances through the field — that multiplexes fine positional information onto a channel above the rate envelope. It is a coding property inside the field, not the field itself; the rate-defined bounded region can be characterized without it, and the standard scalar battery leaves it out. Tell: is the referent the spatial extent of elevated rate (place field) or the temporal phase relationship of spikes within that extent (precession)?

• Time field / concept field. Within-circuit analogues in which the same reified-tuning machinery is applied to elapsed time (time cells) or an abstract concept space rather than allocentric location. These share the field's diagnostics (size-to-role gradient, stability baseline, precession) but tile a non-spatial dimension. The place field is the spatial member of that family. Tell: does the bounded tuning cover a region of physical space (place field) or of elapsed time / concept space (time field, concept field)?

Neighborhood in Abstraction Space

Place Field sits in a crowded region of the domain-specific corpus (39^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Neural Topographic Maps (7 abstractions)

Nearest neighbors

• Place Cell — 0.92
• Somatotopy — 0.86
• Grid Cell — 0.86
• Retinotopy — 0.85
• Spike-Timing-Dependent Plasticity — 0.84

Computed from structural-signature embeddings · 2026-07-12