You can walk to your kitchen in the dark. Somewhere in your head is a map that quietly tracks where you are. Deep in a brain region called the hippocampus, single neurons act like the "you-are-here" pin on a mall map: a place cell stays almost silent everywhere, then fires in a burst only when you stand in one particular spot. A neighbouring cell fires in a different spot. Between them, the spots they each "like" tile the whole room.
BUILD — one bump, one number
Call the spot a place cell loves $\mathbf{x}_0$. The further you wander from $\mathbf{x}_0$, the weaker its firing — a smooth hill, not a hard edge. In symbols the firing rate is $r=r_{max}\,e^{-d^{2}/2\sigma^{2}}$, where $d$ is your distance from the centre and $\sigma$ is the field width. At the centre ($d=0$) the cell fires flat-out, $r=r_{max}$. Step one width $\sigma$ away and the rate falls to $e^{-1/2}\approx 0.61\,r_{max}$ — about 61% — so a large $\sigma$ means a fuzzy, generous field and a small $\sigma$ a sharp, picky one.
DEEPEN — the hidden graph paper
One layer up, in the entorhinal cortex, grid cells fire not at one spot but at the vertices of a perfect hexagonal lattice covering the floor. You can build that pattern by adding just three cosine waves whose directions sit 60° apart: $r_{grid}(\mathbf{x})=\sum_{j=1}^{3}\cos(\mathbf{k}_j\!\cdot\!\mathbf{x}+\phi_j)$, where each wavelength sets the grid spacing $\lambda$ through $|\mathbf{k}_j|=4\pi/(\sqrt{3}\,\lambda)$. Reading position back out is just a weighted average of every active cell's favourite spot, $\hat{\mathbf{x}}=\sum_i r_i\mathbf{x}_i/\sum_i r_i$. And when the lights go out, the grid keeps updating by summing your own velocity — path integration — though small errors pile up, growing roughly as $\sqrt{\text{distance}}$, until a familiar landmark resets the map.
TRY THIS — in the sim above
Read Parameter 1 as the field width $\sigma$: widen it and picture each place-cell hill spreading until neighbours overlap and position turns blurry. Read Parameter 2 as the grid spacing $\lambda$: shrink it for a fine, dense lattice (sharp positions), grow it for a coarse one. Push Input I up to mimic a faster run and watch how much ground each cycle must cover — the basis of theta sequences.
CA1/CA3 Gaussian place fields (σ ≈ 0.5–5m). Population vector decodes position: x̂ = Σ_i r_i x_i / Σ_i r_i. Bayesian decoder: p(x|r) ∝ Π_i Poisson(r_i; f_i(x)). Precision ~cm for 100+ cells. Theta sequences: within each theta cycle, place cells fire in sequence encoding a ~3m future path.
STEP 2 — Grid Cell Mathematics
Hafting et al. (2005): MEC neurons with hexagonal periodic fields. Sum of 3 cosines at 60° intervals. Scale λ: 30 cm → 3 m (dorso-ventral MEC gradient). Hexagonal lattice = densest packing = maximally efficient positional code. Grid phases {φ_j} uniquely encode position via CRT-like combinatorics.
STEP 3 — Path Integration
Velocity-controlled oscillator or ring attractor model: activity bump on a torus moves with velocity. Grid phase encodes position without landmarks. Noise accumulates: error ∝ √(distance). ~10% position error after 8m straight-line path (rat data). Error corrected by place cell ↔ grid cell interaction.
🧠 ConceptualWhat is the core mathematical insight of Place Cells & Grid Cells?▼
The Place Cells & Grid Cells framework provides a rigorous bridge between microscopic neural parameters and macroscopic observables. The key insight is that complex emergent behaviours — oscillations, memory, coding precision — can be derived from simple mathematical rules governing individual neurons and their connections. The quantitative framework enables testable predictions about circuit function from experimentally measurable quantities.
Key: Mathematical rigour connects single-neuron parameters to network behaviour. Every emergent property (oscillation, attractor, code) is derivable from the microscopic rules.
🌍 AppliedHow is Place Cells & Grid Cells used in neurotechnology?▼
The Place Cells & Grid Cells framework is directly applied in brain-computer interfaces and computational medicine. Mathematical models derived from these principles optimise stimulation parameters for DBS, predict drug effects on network dynamics, decode neural signals for BCI applications, and guide the design of neuromorphic hardware. The quantitative framework enables model-based optimisation rather than empirical trial-and-error approaches in clinical settings.
Key: Direct applications in BCI design, DBS parameter optimisation, pharmacological modelling, and neuromorphic chip architecture.
🔬 SimulationHow do I use the simulation for Place Cells & Grid Cells?▼
Configure the parameter sliders on the right panel, select a preset to set up a canonical example, then press Play. The Main View shows the primary mathematical relationship; the Phase Plane shows geometric structure; the Time Series shows temporal evolution; the Parameter Sweep reveals sensitivity to inputs. All displayed quantities correspond directly to terms in the equations in Section 2. Export the data as CSV for further analysis.
Key: Each tab maps to a specific aspect of the mathematics. Use Parameter Sweep to find bifurcation points and Phase Plane to understand stability.
💡 Non-ObviousWhat is counterintuitive about Place Cells & Grid Cells?▼
The most surprising result in Place Cells & Grid Cells research is that small parameter changes can produce qualitatively different dynamics (bifurcations). A system that is stable for one parameter value can suddenly oscillate, burst, or become chaotic with a tiny perturbation. This sensitivity to parameters is the hallmark of nonlinear systems and has profound implications: the same neural circuit can perform completely different computations depending on its neuromodulatory state, which shifts parameters near bifurcation boundaries.
Key: Bifurcations: tiny parameter changes → qualitative behavioural change. Neuromodulators exploit this by operating near bifurcation points to switch neural circuit function.
📐 ComputationalWhat numerical methods and pitfalls apply to Place Cells & Grid Cells?▼
The appropriate numerical method depends on stiffness: for near-spike dynamics (HH, AdEx) use RK4 with dt ≤ 0.025 ms or implicit methods; for mean-field equations (WC, Amari) use Euler or RK4 with dt ≤ 0.5–1 ms. Always verify accuracy by halving dt and checking solution convergence. The most common errors: wrong units (mixing mV, ms, nS), missing initial conditions (not setting w₀=b×v₀ in Izhikevich), and incorrect boundary conditions in Fokker-Planck or cable equation simulations.
Key: Verify stability by halving dt. Most errors: wrong units, incorrect initial conditions, violated model assumptions. Check the model validity range before trusting any result.
§ 04 Resources
Gerstner et al. — Neuronal Dynamics. Free: neuronaldynamics.epfl.ch
Dayan & Abbott — Theoretical Neuroscience, MIT Press, 2001
Izhikevich — Dynamical Systems in Neuroscience. Free: dynamicalsystems.org
Neuromatch Academy — neuromatch.io (free comp neuro tutorials)
§ 05
Misconceptions & Common Errors
Sub-block A — Conceptual
❌"The Place Cells & Grid Cells model is only theoretical with no experimental support."
✅The Place Cells & Grid Cells framework has been extensively validated. Key predictions have been confirmed in electrophysiology, imaging, and pharmacological experiments. Modern neuroscience uses these models as quantitative tools integrated with the experimental cycle — for prediction, interpretation, and guidance of experiments.
❌Applying Place Cells & Grid Cells equations outside their range of validity (wrong N, dt, or approximation regime).
✅Every model has a validity regime. Always check: (1) Is N large enough for mean-field? (2) Is dt small enough (halve dt and verify solution converges)? (3) Are units consistent (mV, ms, nS throughout)? (4) Are initial conditions correct? When in doubt, compare to direct simulation of the full system.
🔍 Why: Violating assumptions gives quantitatively or qualitatively wrong results — and the error is often invisible without explicit validation.