Template

Neuroscientist — Competency Roadmap

Work towards being a neuroscientist: the biological, chemical and computational foundations of how nervous systems work, the experimental methods used to study them, and the research practice that turns questions into evidence.

This roadmap maps the complete territory from absolute fundamentals through advanced computational, cellular, systems, and cognitive neuroscience to rigorous experimental research design and application. Given your theory-first learning preference, each phase establishes rigorous mechanistic models before applying them to computational simulations, experimental protocol designs, or literature critiques. It explicitly balances independent tracks such as scientific programming alongside cellular biology. At 12 hours per week the tasks here come to roughly three to five months, which surveys the territory rather than replacing the years of laboratory training the field is normally entered through. You will develop the ability to critically dissect neuroscientific literature, simulate biophysical and neural network models, design rigorous experimental protocols across multiple recording modalities, and prepare competitive graduate research proposals.

By the end: You will be able to formulate novel neuroscientific hypotheses across cellular, circuit, and computational levels, implement computational simulations of neural dynamics in Python, and produce a publication-grade research proposal and analysis pipeline suitable for doctoral-level research applications.

Starting levelBeginnerStyleTheory first
12h / week8 phases32 tasks~189h total

This is the map — make this roadmap yours

It shows what this journey generally looks like. Tell Kaidoro your version of the goal and it builds the plan around where you are actually starting, what to do first, the hours you really have, and what you have already finished.

1

Foundations of Molecular Biology and Cellular Physiology

Establish the fundamental chemical and cell-biological concepts required to understand excitable cells. This phase provides the necessary foundation for neurophysiology and pharmacology.

  • Learn essential cellular biology and organic chemistry for neurobiology
    ~6hLearn1 resource

    Neuronal signalling is physically grounded in lipid membranes and embedded protein channels; without this, biophysics cannot be understood.

    You'll learn

    • Phospholipid bilayer — the amphipathic structure forming cellular membranes
    • Transmembrane proteins — proteins spanning the membrane acting as channels and pumps
    • Electrochemical gradient — the combined electrical and chemical potential driving ion movement

    Study cell structure, membrane mechanics, protein synthesis, and lipid bilayer properties. Understand how hydrophobic and hydrophilic interactions govern membrane permeability and protein channel embedding.

    Done when: You can draw a lipid bilayer with transmembrane proteins and accurately label phospholipid charges, hydrophobic cores, and aqueous boundary layers from memory.

    How to work through it

    1. Review fluid mosaic model and phospholipid bilayer properties
    2. Study amino acid side chains and tertiary protein structure folding
    3. Diagram active vs passive transport mechanisms across biological membranes
  • Diagram resting membrane potential and ionic equilibria
    ~5hLearn1 resource

    The resting potential is the baseline electrical state from which all neuronal computational signalling diverges.

    You'll learn

    • Nernst Equation — formula calculating electrical potential of a single ion at thermodynamic equilibrium
    • GHK Voltage Equation — formula determining resting potential across multi-permeable membranes
    • Reversal Potential — the membrane potential where net flow of a specific ion is zero

    Explore how unequal ion distributions (Na+, K+, Cl-, Ca2+) create cellular voltage. Master the Nernst equation for single ion equilibrium and the Goldman-Hodgkin-Katz (GHK) equation for whole-cell resting potential.

    Done when: You calculate resting membrane potential given non-standard intra- and extracellular ion concentrations within a written problem set with 100% mathematical accuracy.

    How to work through it

    1. Derive the Nernst equation from thermodynamic first principles
    2. Apply the Goldman-Hodgkin-Katz equation to calculate resting potential
    3. Calculate reversal potentials for Na+, K+, Cl-, and Ca2+ at physiological temperatures
  • Map the biochemistry of energy metabolism in the brain
    ~4hLearn

    Brain energy consumption directly constrains neural computation and forms the physiological basis of non-invasive neuroimaging.

    You'll learn

    • Na+/K+ ATPase — ATP-dependent pump maintaining high intracellular K+ and extracellular Na+
    • Astrocyte-Neuron Lactate Shuttle — metabolic coupling where astrocytes provide fuel to active neurons
    • BOLD signal coupling — the link between metabolic substrate delivery and fMRI vascular signals

    Investigate how the brain consumes glucose and oxygen to maintain ionic gradients via the Na+/K+ ATPase pump. Understand the astrocyte-neuron lactate shuttle hypothesis and the metabolic costs of action potentials.

    Done when: You write a 1-page summary contrasting neuronal vs astrocytic energy substrate utilisation and their connection to functional imaging signals (fMRI BOLD).

    How to work through it

    1. Trace glycolysis, the citric acid cycle, and oxidative phosphorylation in neurons
    2. Map astrocyte glycogen storage and the lactate shuttle mechanism
    3. Link ATP expenditure to the restoration of ionic gradients by Na+/K+ ATPase
  • Model steady-state ion flux in a computational notebook
    ~5hBuild

    Programming biophysical equations ensures conceptual equations translate directly into quantitative predictive models.

    You'll learn

    • Hyperkalemia — elevated extracellular potassium causing resting membrane depolarisation
    • NumPy array broadcasting — performing vectorised calculations over concentration arrays
    • Matplotlib subplots — plotting multi-variable biophysical state curves

    Translate GHK and Nernst equations into an interactive Python script using NumPy. Vary extracellular potassium and sodium concentrations to plot how resting membrane potential changes systematically.

    Done when: The Jupyter notebook runs and generates interactive plots demonstrating membrane depolarisation in response to extracellular hyperkalemia.

    How to work through it

    1. Write Python functions for Nernst and GHK voltage calculations
    2. Simulate altered extracellular K+ concentrations and log changes in Vm
    3. Export figures comparing ideal vs simulated membrane voltage shifts
2

Scientific Computing and Mathematical Methods for Neuroscience

Build the foundational mathematics (calculus, linear algebra, probability) and scientific Python tooling needed for neural data analysis and biophysical modelling. Can be pursued concurrently with cellular biology.

  • Set up scientific Python environment with essential packages
    ~3hBuild

    Modern neuroscience depends heavily on scientific Python for data pipeline execution and theoretical modelling.

    You'll learn

    • Virtualenv / Conda — isolated environment managers preventing dependency conflicts
    • SciPy — Python library containing numerical integration and signal processing routines
    • Vectorised operations — high-speed array processing avoiding explicit Python loops

    Install Python, JupyterLab, NumPy, SciPy, Matplotlib, and Pandas. Configure a dedicated virtual environment for reproducible neuroscientific simulations.

    Done when: You run a test script importing all core libraries and plotting a basic sine wave without dependency errors.

    How to work through it

    1. Install Python 3 and configure Conda or venv virtual environment
    2. Install NumPy, SciPy, Matplotlib, Seaborn, and Pandas
    3. Verify environment by running a headless plotting and matrix multiplication test
  • Master differential equations and linear algebra for neural dynamics
    ~8hLearn1 resource

    Neural membrane voltage and network activity are dynamical systems governed by differential equations and matrix transformations.

    You'll learn

    • Forward Euler Method — first-order numerical procedure for solving ODEs with a given step size
    • Eigenvalues and Eigenvectors — matrix scalars and vectors describing stability and axes of transformation
    • Phase Portrait — visual representation of trajectories of a dynamical system on a state plane

    Study ordinary differential equations (ODEs), numerical integration (Euler, Runge-Kutta), eigenvalues/eigenvectors, and matrix transformations. These mathematical tools form the backbone of neural modelling.

    Done when: You solve a first-order leaky integrator ODE analytically and verify your solution against numerical integration in Python.

    How to work through it

    1. Review first-order differential equations and separation of variables
    2. Implement forward Euler numerical integration in Python
    3. Compute eigenvalues of 2D linear dynamical systems to classify stability fixed points
  • Study probability theory and signal processing fundamentals
    ~7hLearn

    Spike trains are stochastic point processes, and field potentials are time-series signals requiring rigorous spectral analysis.

    You'll learn

    • Poisson Point Process — stochastic model for generating inter-spike intervals (ISI)
    • Fourier Transform — mathematical transform decomposing a time series into constituent frequencies
    • Nyquist Frequency — the minimum sample rate required to avoid aliasing artifacts in recorded signals

    Master probability distributions, Bayes' rule, Poisson processes for spike trains, and Fourier/wavelet transforms for continuous neural signals like LFP and EEG.

    Done when: You implement a Discrete Fourier Transform (DFT) from scratch and filter simulated noisy neural oscillations into canonical frequency bands (theta, alpha, gamma).

    How to work through it

    1. Generate homogeneous and inhomogeneous Poisson spike trains in Python
    2. Compute power spectral density (PSD) using Welch's method
    3. Design and apply bandpass Butterworth filters to raw signal arrays
  • Build a neural signal filtering and spectral analysis pipeline
    ~6hBuild

    Signal processing is a core daily competency for electrophysiology and neuroimaging data pipelines.

    You'll learn

    • STFT (Short-Time Fourier Transform) — windowed Fourier transform representing frequency over time
    • Notch Filter — narrow band-stop filter used to eliminate 50/60 Hz electrical mains hum
    • Unit testing (pytest) — automated tests ensuring data transformation functions behave deterministically

    Write an end-to-end Python module that takes synthetic continuous multi-channel local field potential data, applies notch and bandpass filtering, extracts frequency power bands, and generates spectrograms.

    Done when: The pipeline takes raw time-series test arrays, outputs clean PSD plots and spectrogram figures, and passes automated unit tests.

    How to work through it

    1. Write classes for raw signal loading, baseline detrending, and notch filtering
    2. Implement Short-Time Fourier Transform (STFT) for spectrogram generation
    3. Package code into a modular script with input argument validation
3

Cellular Neurophysiology and Synaptic Transmission

Understand how neurons generate electrical impulses and communicate via chemical and electrical synapses.

  • Learn action potential biophysics and Hodgkin-Huxley dynamics
    ~6hLearn1 resource

    The action potential is the fundamental all-or-none unit of long-range neural communication.

    You'll learn

    • Hodgkin-Huxley model — mathematical model describing how action potentials are initiated and propagated
    • Inactivation Gate (h-gate) — time-dependent channel closure causing the absolute refractory period
    • Length constant (lambda) — distance over which passive voltage decays to 37% of original amplitude

    Examine voltage-gated ion channel kinetics (NaV and KV activation/inactivation gates), refractory periods, and passive cable theory (length and time constants).

    Done when: You explain the exact sequence of molecular conformational changes in voltage-gated sodium and potassium channels during each phase of an action potential.

    How to work through it

    1. Study Hodgkin-Huxley gating variables (m, h, n) and their voltage dependence
    2. Calculate membrane time constant (tau) and length constant (lambda)
    3. Analyse the role of myelin and nodes of Ranvier in saltatory conduction
  • Investigate neurotransmitter release and receptor kinetics
    ~5hLearn

    Synaptic transmission determines the computational rules of integration and plasticity across circuits.

    You'll learn

    • SNARE Complex — protein machinery (synaptobrevin, syntaxin, SNAP-25) driving vesicle fusion
    • NMDA Receptor — voltage-dependent glutamate receptor acting as a molecular coincidence detector
    • Spatial and Temporal Summation — algebraic integration of concurrent and sequential PSPs

    Examine SNARE-complex mediated vesicular exocytosis, calcium triggering, ionotropic vs metabotropic (GPCR) receptors, and postsynaptic potentials (EPSPs/IPSPs).

    Done when: You produce a comparative diagram detailing ionotropic vs metabotropic downstream pathways for glutamate (AMPA/NMDA) and GABA (GABA-A/GABA-B).

    How to work through it

    1. Trace SNARE-mediated vesicle docking, priming, and Ca2+-triggered fusion via synaptotagmin
    2. Compare fast ligand-gated channels with second-messenger cascades (cAMP, IP3/DAG)
    3. Detail the magnesium block and coincidence detection properties of the NMDA receptor
  • Study mechanisms of synaptic plasticity: LTP, LTD, and homeostatic scaling
    ~5hLearn

    Plasticity is the biological basis of learning, memory, and adaptive circuit refinement.

    You'll learn

    • LTP (Long-Term Potentiation) — persistent strengthening of synapses based on recent patterns of activity
    • CaMKII — enzyme acting as a molecular memory device via autophosphorylation
    • Homeostatic plasticity — cellular mechanisms stabilising overall firing rates across fluctuating input

    Explore Long-Term Potentiation (LTP) and Long-Term Depression (LTD) mechanisms in hippocampal CA1/CA3 pyramidal neurons, including CaMKII activation, AMPA receptor trafficking, and retrograde retrograde messengers.

    Done when: You write a mechanistic explanation of how calcium influx dynamics through NMDARs determine whether LTP or LTD is induced.

    How to work through it

    1. Analyse high-frequency stimulation (tetanus) vs low-frequency stimulation protocols
    2. Trace CaMKII autophosphorylation and AMPA receptor insertion into the postsynaptic density
    3. Examine homeostatic synaptic scaling and metaplasticity rules
  • Simulate a Hodgkin-Huxley neuron model in Python
    ~6hBuild

    Building the Hodgkin-Huxley model solidifies the transition from biophysical theory to predictive computational formulation.

    You'll learn

    • F-I Curve — firing rate as a function of injected input current
    • Gating kinetics — temperature- and voltage-dependent opening and closing rates of ion channels
    • Numerical stiffness — stability considerations when choosing ODE solver step sizes

    Implement the full Hodgkin-Huxley differential equations using numerical integration in Python. Inject varying step currents to simulate firing thresholds, action potential waveforms, and refractory periods.

    Done when: Your simulation reproduces authentic action potentials, shows spike frequency adaptation/refractoriness, and generates a current-frequency (F-I) curve.

    How to work through it

    1. Define differential equations for V, m, h, and n variables with voltage-dependent rate functions
    2. Run numerical ODE integration across a simulated 100 ms timeframe
    3. Plot voltage traces for subthreshold, threshold, and continuous firing current inputs
4

Neuroanatomy, Sensory Systems, and Circuit Architectures

Survey the macro- and micro-structural organisation of the nervous system, tracing sensory pathways from periphery to neocortex.

  • Master gross neuroanatomy and directional axes
    ~5hLearn1 resource

    Anatomical localisation is fundamental to all circuit-level, electrophysiological, and functional imaging work.

    You'll learn

    • Telencephalon — embryonic division giving rise to the cerebral cortex and basal ganglia
    • Coronal plane — vertical cross-section dividing structure into anterior and posterior parts
    • Ventricular system — connected cavities in the brain containing cerebrospinal fluid

    Learn the anatomical axes (rostral/caudal, dorsal/ventral, medial/lateral), major brain divisions (telencephalon, diencephalon, mesencephalon, metencephalon, myelencephalon), ventricular system, and cranial nerves.

    Done when: You pass a self-assessment identifying 30 major anatomical structures and directional orientations on unlabelled human and rodent MRI slice diagrams with 95% accuracy.

    How to work through it

    1. Memorise anatomical planes (coronal, sagittal, axial/horizontal)
    2. Trace cerebrospinal fluid (CSF) flow through the ventricular system
    3. Map the functional topography of the cerebral cortex lobes and basal ganglia
  • Trace canonical sensory pathways and cortical column organisation
    ~6hLearn

    Sensory circuits provide the classical paradigms for understanding receptive field properties, hierarchical processing, and cortical computation.

    You'll learn

    • Receptive field — specific region of sensory space in which a stimulus modifies a neuron's firing
    • Laminar organisation — 6-layered histological structure of mammalian neocortex
    • Topographic mapping — orderly spatial mapping of sensory surfaces onto cortical areas (e.g., retinotopy)

    Examine visual, auditory, and somatosensory transduction and ascending circuitry. Study the 6-layered neocortex architecture, feedforward/feedback thalamocortical loops, and receptive fields.

    Done when: You produce a detailed anatomical wiring diagram tracing visual inputs from photoreceptors through the LGN to V1 layers 4C, 2/3, 5, and 6, annotating receptive field transformations.

    How to work through it

    1. Trace the visual pathway: retina, lateral geniculate nucleus (LGN), and primary visual cortex (V1)
    2. Compare Hubel & Wiesel simple vs complex cell receptive fields
    3. Analyse the canonical microcircuit model of cortical laminar connectivity
  • Examine motor control systems and subcortical modulation
    ~5hLearn

    Motor circuits demonstrate how subcortical nuclei modulate cortical representations to execute adaptive behaviour.

    You'll learn

    • Direct Pathway — striatal circuit facilitating movement via disinhibition of thalamus
    • Purkinje Cell — principal inhibitory projection neuron of the cerebellar cortex
    • Dopaminergic modulation — regulation of circuit excitability by substantia nigra pars compacta projections

    Study upper and lower motor neurons, corticospinal tracts, basal ganglia direct and indirect pathways, and cerebellar computation for error correction and motor coordination.

    Done when: You create a circuit diagram of the basal ganglia motor loops, explaining how dopamine modulation via D1 and D2 receptors shifts balance between action initiation and suppression.

    How to work through it

    1. Map primary motor cortex (M1) organisation and descending corticospinal pathways
    2. Detail striatal pathways: striatonigral (direct) vs striatopallidal (indirect/hyperdirect)
    3. Examine cerebellar Purkinje cell, climbing fibre, and mossy fibre architecture
  • Construct an annotated connectomic circuit diagram for a sensory-motor reflex
    ~6hBuild

    Translating anatomical literature into structured circuit models is essential for formulating testable circuit-level research questions.

    You'll learn

    • Tract tracing — histological technique using viral or chemical tracers to reveal neural connectivity
    • Barrel cortex — somatosensory cortex region specialised for processing rodent whisker inputs
    • Cell-type taxonomy — classification of neurons based on morphology, transcriptomics, and electrophysiology

    Select a well-characterised sensory-motor loop (e.g., rodent whisker-barrel circuit or optokinetic reflex). Synthesise published histological and tract-tracing data into a structured connectivity map specifying cell types, neurotransmitters, and laminar projections.

    Done when: You produce a verified vector schematic and legend citing primary literature for every synaptic connection and neurotransmitter identity in the circuit.

    How to work through it

    1. Select a classic model circuit from primary neuroanatomy literature
    2. Identify each node, projection neuron type, synaptic sign (excitatory/inhibitory), and target layer
    3. Format the schema into publication-standard figures using vector design or schematic tools
5

Modern Experimental Methods and Neurotechnologies

Survey the experimental toolkit used to observe and manipulate neural activity across spatial and temporal scales.

  • Learn electrophysiology techniques: patch clamp to high-density silicon probes
    ~6hLearn1 resource

    Electrophysiology remains the gold standard for measuring action potentials and synaptic potentials at millisecond temporal resolution.

    You'll learn

    • Voltage Clamp — technique clamping membrane potential to measure ionic currents across the membrane
    • Spike Sorting — algorithmic classification of extracellular waveforms to individual single units
    • Neuropixels — high-density silicon probes recording hundreds of single neurons simultaneously

    Study whole-cell patch-clamp recording (voltage-clamp, current-clamp), extracellular local field potentials, multi-electrode arrays, and modern high-density probes (Neuropixels).

    Done when: You compare whole-cell vs extracellular recording in a written technical brief detailing spatial resolution, temporal resolution, signal-to-noise ratio, and invasive limits.

    How to work through it

    1. Examine patch clamp configurations: cell-attached, whole-cell, inside-out, outside-out
    2. Understand spike sorting principles: filtering, waveform detection, feature extraction (PCA), and clustering
    3. Review Neuropixels probe architecture and multi-site recording capabilities
  • Study optical imaging and optogenetic perturbation
    ~6hLearn

    Optical imaging and optogenetics allow cell-type specific recording and causal manipulation of intact neural circuits.

    You'll learn

    • Two-photon microscopy — optical technique using near-infrared light for deep in vivo brain imaging
    • GCaMP — engineered fluorescent indicator binding calcium to report action potential bursts
    • Optogenetics — using light-sensitive ion channels to excite or inhibit genetically targeted neurons

    Explore fluorescence microscopy, two-photon calcium imaging using GCaMP indicators, voltage imaging, and optogenetic/chemogenetic perturbation tools (Channelrhodopsin, Halorhodopsin, DREADDs).

    Done when: You design an optogenetic stimulation protocol specifying light wavelength, opsin expression targeting via Cre-Lox, and power density calculations to avoid thermal tissue damage.

    How to work through it

    1. Study two-photon excitation physics and depth penetration in scattering brain tissue
    2. Trace genetically encoded calcium indicators (GCaMP) kinetics and dF/F signal processing
    3. Examine viral delivery systems (AAVs, Lentivirus) and transgenic Cre-driver mouse lines
  • Understand non-invasive human neuroimaging: fMRI, EEG, and MEG
    ~6hLearn

    Human cognitive and clinical neuroscience relies on non-invasive imaging modalities to map network-level representations.

    You'll learn

    • Hemodynamic Response Function (HRF) — temporal profile of blood oxygenation changes following neural activity
    • General Linear Model (GLM) — statistical framework modelling fMRI time-series as linear combinations of predictors
    • Forward / Inverse Problem — mathematical challenge of reconstructing source dipole locations from surface EEG/MEG signals

    Study functional Magnetic Resonance Imaging (BOLD physics, echo-planar imaging, GLM analysis), Electroencephalography (event-related potentials, source localisation), and Magnetoencephalography.

    Done when: You outline the General Linear Model (GLM) formulation for an event-related fMRI experiment, defining design matrices, hemodynamic response function (HRF) convolution, and statistical contrast vectors.

    How to work through it

    1. Learn BOLD contrast biophysics and hemodynamic response characteristics
    2. Understand spatial/temporal resolution trade-offs across fMRI, EEG, MEG, and intracranial stereo-EEG
    3. Review preprocessing steps: motion correction, slice-timing correction, co-registration, and spatial smoothing
  • Analyse an open-access calcium imaging dataset
    ~7hBuild1 resource

    Working directly with real open-source neurophysiology data builds the core data processing skills expected in contemporary labs.

    You'll learn

    • NWB (Neurodata Without Borders) — standardized data format for cellular neurophysiology and behavior
    • dF/F — normalised change in fluorescence relative to baseline, reflecting intracellular calcium changes
    • Tuning curve — graphical representation of a neuron's response magnitude across a range of stimulus values

    Download an open-access two-photon calcium imaging dataset (e.g., from the Allen Brain Observatory). Write a Python script using standard tools (like CaImAn or Suite2p concepts) to extract spatial footprints, compute dF/F traces, and correlate fluorescence with presented stimuli.

    Done when: You generate a raster plot and average tuning curves of extracted neural regions of interest (ROIs) responding to sensory stimuli.

    How to work through it

    1. Load Allen Brain Observatory NWB (Neurodata Without Borders) calcium imaging files
    2. Extract raw ROI time series and calculate baseline fluorescence (F0) to obtain dF/F
    3. Plot tuning curves showing response amplitude across different stimulus orientations
6

Cognitive and Systems Neuroscience

Investigate how neural populations encode information, make decisions, navigate space, and execute cognitive functions.

  • Learn neural population coding and dimensionality reduction
    ~6hLearn

    High-dimensional neural recordings must be conceptualised as low-dimensional collective dynamics within state space.

    You'll learn

    • Neural Trajectory — the path traversed by a population state vector through time in neural state space
    • PSTH (Peri-Stimulus Time Histogram) — trial-averaged spike rate aligned to specific sensory or motor events
    • Latent variable — unobserved underlying variable driving coordinated activity across multiple recorded neurons

    Study rate codes vs temporal codes, population vectors, state-space representations, and dimensionality reduction techniques (PCA, t-SNE, UMAP, dPCA) applied to multi-neuron recordings.

    Done when: You implement Principal Component Analysis on synthetic multi-neuronal firing rates to project population trajectories during a simulated decision task into low-dimensional state space.

    How to work through it

    1. Compare single-neuron tuning vs population subspace representations
    2. Apply PCA to multi-unit peri-stimulus time histograms (PSTHs)
    3. Visualise neural trajectories in low-dimensional latent spaces across distinct task conditions
  • Study the neural mechanisms of spatial navigation and episodic memory
    ~6hLearn

    The hippocampal-entorhinal system is the premier model system for understanding cognitive maps and episodic memory mechanics.

    You'll learn

    • Place Cell — hippocampal pyramidal neuron firing when an animal is in a specific spatial location
    • Grid Cell — medial entorhinal cortex neuron firing at periodic triangular vertices across an environment
    • Sharp-Wave Ripple (SWR) — high-frequency synchronous oscillation in the hippocampus linked to memory replay

    Examine place cells (hippocampus), grid cells (entorhinal cortex), head-direction cells, and sharp-wave ripples (SWRs) supporting memory consolidation and replay.

    Done when: You explain the continuous attractor network model of grid cell generation and describe how optogenetic disruption of sharp-wave ripples impairs memory retention in behavioural tasks.

    How to work through it

    1. Study O'Keefe place fields and Moser 2D hexagonal grid field geometries
    2. Analyse continuous attractor network (CAN) theories for path integration
    3. Examine the role of hippocampal theta-gamma phase-amplitude coupling and SWR replay
  • Explore perceptual decision making and executive control circuits
    ~7hLearn

    Decision making links sensory representations, internal value states, and motor command execution through quantifiable computational models.

    You'll learn

    • Drift-Diffusion Model (DDM) — continuous stochastic model of two-choice decision-making processes
    • Drift Rate — parameter representing the speed and quality of sensory evidence accumulation
    • Working memory persistent firing — sustained cellular and recurrent circuit activity maintaining representations without external input

    Study evidence accumulation models (drift-diffusion model), frontal eye fields, lateral intraparietal area (LIP), prefrontal cortex working memory circuits, and attractor dynamics.

    Done when: You fit a Drift-Diffusion Model (DDM) to synthetic reaction-time and choice data in Python, recovering drift rate and decision boundary threshold parameters.

    How to work through it

    1. Study classic Shadlen random-dot motion paradigms in non-human primates
    2. Derive drift-diffusion equations for bounded accumulation of noisy evidence
    3. Fit DDM parameters to behavioural accuracy and response time distributions using HDDM or custom Python optimization
  • Perform a critical paper dissection of a major systems neuroscience study
    ~5hPractice

    Critical literature appraisal is how practicing neuroscientists evaluate evidence and discover viable research gaps.

    You'll learn

    • Causal vs Correlational inference — distinguishing observation of activity from necessity/sufficiency tests
    • Statistical power & multiple comparisons correction — methods like FDR and Bonferroni preventing false discovery
    • Construct validity — the degree to which an experimental test truly measures the psychological construct intended

    Select a landmark paper integrating behaviour, large-scale recording, and circuit perturbation (e.g., from Nature or Neuron). Write a structured 3-page critique dissecting hypotheses, controls, methodological limitations, and statistical validity.

    Done when: Your written dissection accurately details the core causal claims, evaluates whether controls ruled out alternative explanations, and proposes one follow-up experiment.

    How to work through it

    1. Read the paper across three passes: abstract/figures, experimental logic, and supplementary methods
    2. Analyse whether recording and perturbation techniques causally support the conclusions
    3. Draft a critique structured into: Core Claim, Methodological Strengths, Uncontrolled Confounders, and Next Experiments
7

Computational Neuroscience and Theoretical Models

Formulate mathematical and theoretical models of single neurons, recurrent neural networks, and reinforcement learning.

  • Study simplified spiking neuron models and synaptic plasticity rules
    ~6hLearn1 resource

    Simplified phenomenological models enable large-scale network simulations that are computationally prohibitive with full Hodgkin-Huxley equations.

    You'll learn

    • Leaky Integrate-and-Fire (LIF) — computationally efficient model treating a neuron as a leaky capacitor with reset
    • STDP (Spike-Timing-Dependent Plasticity) — biological plasticity rule where synaptic modification depends on millisecond spike ordering
    • Izhikevich model — two-dimensional ODE model combining biological plausibility with computational efficiency

    Examine Leaky Integrate-and-Fire (LIF), Izhikevich, and Wilson-Cowan population models, alongside Spike-Timing-Dependent Plasticity (STDP) and Hebbian learning rules.

    Done when: You simulate an LIF neuron in Python and demonstrate how relative spike timing in STDP systematically strengthens or weakens synaptic weights.

    How to work through it

    1. Derive the subthreshold voltage dynamics and reset conditions of the LIF model
    2. Implement STDP exponential learning window functions
    3. Simulate pre- and post-synaptic paired spikes to generate an STDP plasticity curve
  • Build and train a Recurrent Neural Network (RNN) on a cognitive task
    ~8hBuild

    Task-trained RNNs serve as synthetic model organisms to generate testable hypotheses about biological circuit computations.

    You'll learn

    • CTRNN (Continuous-Time RNN) — dynamical network model obeying biophysically-inspired differential equations
    • Attractor Dynamics — stable states or trajectories towards which a dynamical system converges over time
    • Task-driven representation — latent representations emerging naturally under task performance constraints

    Construct a continuous-time recurrent neural network (CTRNN) using PyTorch. Train it to perform a perceptual decision or working memory task (e.g., delayed match-to-sample), then analyse the emergent network dynamics and fixed points.

    Done when: Your trained RNN solves the working memory task with >90% accuracy, and you plot the low-dimensional state space fixed points demonstrating attractor dynamics.

    How to work through it

    1. Define a CTRNN architecture with recurrent excitatory and inhibitory weight matrices in PyTorch
    2. Train the network using gradient descent to hold stimulus cues across a variable delay period
    3. Extract hidden unit activation states and perform PCA to visualise persistent activity attractors
  • Study Reinforcement Learning and the dopamine reward prediction error
    ~6hLearn1 resource

    Dopaminergic TD learning is the most successful bridge connecting formal computational theory, neurophysiology, and animal behaviour.

    You'll learn

    • Temporal Difference (TD) Error — difference between expected and received value, computing updating signals
    • Value Function V(s) — expected cumulative discounted future reward from a given state
    • Phasic Dopamine — transient burst or pause in midbrain firing signalling unexpected reward presence or omission

    Examine Markov Decision Processes, Temporal Difference (TD) learning, value iteration, and the neurobiological mapping of TD error to midbrain dopamine neuron firing (Schultz, Dayan, Montague).

    Done when: You code a tabular TD(0) reinforcement learning agent in Python and demonstrate that simulated prediction errors match the empirical firing patterns of dopamine neurons across unpredicted, predicted, and omitted rewards.

    How to work through it

    1. Derive Bellman equations and the TD prediction error formula: delta = r + gamma*V(s') - V(s)
    2. Simulate classical Pavlovian conditioning in a temporal-difference agent
    3. Plot simulated agent delta values across acquisition, extinction, and reward omission trials
  • Simulate balanced excitation-inhibition (E-I) network dynamics
    ~7hBuild

    E-I balance is a fundamental organising principle of mammalian cortical dynamics, preventing runaway excitation or quiescence.

    You'll learn

    • Asynchronous Irregular (AI) State — physiological regime characterised by high spike irregularity and low correlation
    • E-I Balance — dynamic tracking where inhibitory currents closely cancel excitatory fluctuations
    • Fano Factor — ratio of spike count variance to mean, measuring firing variability

    Construct a network of 1,000 LIF neurons (80% excitatory, 20% inhibitory) with random sparse connectivity in Python. Tune synaptic conductances to achieve asynchronous irregular (AI) firing states resembling in vivo cortical activity.

    Done when: The simulation outputs a spike raster and population voltage trace demonstrating asynchronous irregular firing, stable firing rates, and balanced excitatory and inhibitory currents.

    How to work through it

    1. Set up population arrays of excitatory and inhibitory LIF neurons with conductance-based synapses
    2. Wire random recurrent connections using sparse probability matrices (e.g., p = 0.05)
    3. Compute Fano factors and inter-spike interval coefficient of variation (CV) to confirm AI state
8

Research Rigour, Experimental Design, and Translational Pathways

Master experimental design, open science standards, scientific ethics, and translational career routes across academia, neurotechnology, and clinical science.

  • Master statistical experimental design, power calculations, and preregistration
    ~5hPractice1 resource

    Rigorous experimental design and statistical power prevent irreproducible findings and waste of scientific resources.

    You'll learn

    • Statistical Power (1 - beta) — probability of correctly rejecting a false null hypothesis
    • Preregistration — time-stamped publication of experimental hypotheses and analysis plans before data collection
    • Effect size (Cohen's d) — standardized quantitative measure of the magnitude of an experimental effect

    Learn statistical power analysis (G*Power / Python pingouin), sample size estimation, avoiding p-hacking, factorial designs, and preregistration frameworks (OSF).

    Done when: You write a formal preregistration plan for a hypothetical mouse behavioural experiment, calculating statistical sample size for a targeted effect size (Cohen's d) and alpha level.

    How to work through it

    1. Perform statistical power analyses to determine animal/subject cohort sizes
    2. Define explicit inclusion/exclusion criteria and blinded randomised allocation procedures
    3. Draft a full preregistration protocol using Open Science Framework (OSF) templates
  • Explore neuroethics, animal welfare regulations, and human IRB governance
    ~4hLearn

    Neuroscience research operates under strict ethical and legal regulatory frameworks that every researcher must navigate.

    You'll learn

    • The 3Rs — ethical principles guiding animal research: Replacement, Reduction, and Refinement
    • IRB (Institutional Review Board) — committee applying ethical standards to research involving human subjects
    • Humane endpoints — predetermined criteria for ending an experimental animal procedure to avoid unnecessary distress

    Study institutional animal care and use regulations (3Rs: Replacement, Reduction, Refinement), human ethics committees (IRB/Ethics Boards), and emerging neuroethical issues in neural decoding and BCIs.

    Done when: You complete a written mock animal protocol justification application detailing the 3Rs principles and humane endpoints for an in vivo electrophysiology experiment.

    How to work through it

    1. Review institutional animal welfare mandates (IACUC / Home Office guidelines)
    2. Apply the 3Rs principles to an invasive rodent recording paradigm
    3. Analyse ethical concerns regarding neural privacy, cognitive enhancement, and brain organoids
  • Evaluate neurotech, industry, and clinical neuroscience career landscapes
    ~4hLearn

    Understanding the ecosystem of industry, neurotech, and clinical research allows strategic alignment of personal projects and career targets.

    You'll learn

    • Neural Decoding / BCI — translating neural signals into control signals for external actuators or prostheses
    • Translational Research — moving fundamental biological discoveries towards clinical diagnostic and therapeutic tools
    • IND (Investigational New Drug) — FDA regulatory approval to begin clinical trials with an experimental drug

    Analyse the alternative career paths beyond pure academia: brain-computer interface (BCI) industry (Neuralink, Blackrock Neurotech), neuropharmacology/biotech, clinical trial design for neurological disorders (Alzheimer's, Parkinson's), and data science.

    Done when: You write a 2-page comparative analysis mapping how core neuroscience competencies translate into specific roles across academic PI tracks, BCI systems engineering, and biotech drug discovery pipelines.

    How to work through it

    1. Map key industry sectors: closed-loop neurostimulation, neurodiagnostics, and neurological therapeutics
    2. Identify technical and regulatory differences between academic basic research and FDA-cleared medical device pipelines
    3. Contrast funding cycles, publishing pressures, and project timelines across academia vs industry
  • Write a complete doctoral-grade research proposal and analysis plan
    ~12hApply1 resource

    A comprehensive research proposal synthesises every domain of your training into verifiable evidence of independent scientific thinking.

    You'll learn

    • Specific Aims — concise statement of project goals, hypotheses, and experimental strategies
    • Falsifiability — capacity of a scientific hypothesis to be proven wrong through empirical testing
    • Contingency Planning — alternative experimental pathways planned in advance in case primary methods fail

    Formulate a novel, hypothesis-driven neuroscience research proposal. Integrate a specific biological question across cellular, circuit, and computational levels. Specify experimental design, recording modalities, computational analysis pipelines, controls, power calculations, and potential pitfalls.

    Done when: You complete a 5-to-8 page formal research grant proposal structured with Specific Aims, Background/Significance, Experimental Design, Analytical Methods, and Alternative Approaches, formatted for graduate school or funding applications.

    How to work through it

    1. Identify an unresolved question at the boundary of circuit physiology and computational modelling
    2. Draft two or three testable Specific Aims with explicit falsifiable hypotheses
    3. Detail experimental methods (recording, optogenetics, behavioural tasks) and statistical analysis pipelines
    4. Outline potential pitfalls and contingency plans for each aim

How the plan fits together

8 phases in 6 stages. Anything on the same row can be worked on at the same time, and 2 of them can start straight away.

An arrow points from a phase to the work it unlocks: before starting any phase, every phase with an arrow into it has to be finished first.

STARTSTAGE 2STAGE 3STAGE 4STAGE 5STAGE 61Foundations of MolecularBiology and CellularPhysiology4 tasks · ~20h2Scientific Computing andMathematical Methods forNeuroscience4 tasks · ~24h3Cellular Neurophysiologyand Synaptic Transmission4 tasks · ~22h4Neuroanatomy, SensorySystems, and CircuitArchitectures4 tasks · ~22h5Modern ExperimentalMethods andNeurotechnologies4 tasks · ~25h6Cognitive and SystemsNeuroscience4 tasks · ~24h7Computational Neuroscienceand Theoretical Models4 tasks · ~27h8Research Rigour,Experimental Design, andTranslational Pathways4 tasks · ~25h

Resources

16 in this plan's library, beyond the links on individual tasks.

Books, Courses & Fundamentals

Primary neuroscience textbooks, computational lecture series, and documentation.

  • From Neuron to Brain

    Emphasizes the foundational classic experiments behind membrane excitability, synaptic vesicle release, ionotropic/metabotropic signaling, and visual system sensory architectures.

    global.oup.com · Sinauer Associates / Oxford University Press · Book · ~$130–$200

  • Fundamentals of Neuroscience (3-Part Course Series)

    Takes learners from resting membrane potentials, Nernst/Goldman equations, and action potential kinetics to synaptic transmission and microcircuits up to gross neuroanatomy and sensory perception.

    edx.org · HarvardX · Course · Free to audit

  • Guide to Research Techniques in Neuroscience

    Provides an overview of two-photon calcium imaging, optogenetics, chemogenetics, multi-electrode arrays, single-cell RNA-seq, and behavioral tracking pipelines.

    elsevier.com · Academic Press (Elsevier) · Book · ~$45–$75

  • Mathematics for Neuroscientists

    Introduces ordinary differential equations, cable theory mathematics, probability theory, Fourier analysis, and linear algebra directly within the context of electrophysiology and neural modeling.

    elsevier.com · Academic Press (Elsevier) · Book · ~$80–$120

  • Molecular Biology of the Cell

    Before diving into excitable membranes, you need a rigorous grounding in lipid bilayer biophysics, ion channel transport proteins, second-messenger signaling cascades, and gene expression.

    wwnorton.com · W. W. Norton & Company · Book · ~$100–$180

  • Neuromatch Academy: Computational Neuroscience Course

    Open-source curriculum spanning model fitting, Generalized Linear Models, dimensionality reduction, dynamical systems, continuous attractor models, and reinforcement learning in neural populations.

    compneuro.neuromatch.io · Neuromatch · Course · Completely free and open-source courseware / optional paid interactive summer school

  • Neuronal Dynamics: From Single Neurons to Networks and Models of Cognition

    Covers leaky integrate-and-fire dynamics, Hodgkin-Huxley systems, population rate models, synaptic plasticity, attractor networks, and decision-making models with simulation exercises.

    neuronaldynamics.epfl.ch · Cambridge University Press / EPFL · Book · Free online HTML version / paid print edition

  • Principles of Neural Science

    The standard comprehensive reference text across cellular electrophysiology, neuroanatomy, and systems neuroscience.

    McGraw-Hill · Book · ~$130 for textbook print/digital editions · Intermediate

  • Theoretical Neuroscience: Computational and Mathematical Modeling of Neural Systems

    The definitive graduate-level reference on neural encoding/decoding, information theory, receptive field estimation, network dynamics, and classical conditioning models.

    mitpress.mit.edu · The MIT Press · Book · ~$60–$90

Open Datasets & Repositories

Open neurophysiology, calcium imaging, and connectomics datasets.

  • Allen Brain Map & AllenSDK

    Provides standardized open access to petabytes of high-density extracellular electrophysiology, 2-photon in vivo calcium imaging, patch-seq cell morphology, and single-cell transcriptomics.

    portal.brain-map.org · Allen Institute for Brain Science · Open Dataset & Repository · Free

  • DANDI Archive (Distributed Archives for Neurophysiology Data Integration)

    The central repository for raw and processed neurophysiology datasets structured using the Neurodata Without Borders (NWB) standard.

    dandiarchive.org · NIH BRAIN Initiative / DANDI Team · Open Dataset & Repository · Free

  • OpenNeuro

    Open-platform archive for validated neuroimaging and neural recording datasets conforming to the BIDS standard.

    openneuro.org · Center for Reproducible Neuroscience, Stanford University · Data Repository · Free · Intermediate

Scientific Software & Simulators

Python libraries, neural simulators, and analysis packages.

  • Brian 2 Neural Simulator

    A Python-based simulation environment that lets you define custom biological neuron and synapse models using standard mathematical equations with C++ code generation.

    briansimulator.org · Brian Simulator Development Team / eLife · Scientific Software & Simulator · Free

  • MNE-Python Documentation and Tutorials

    Analysis and visualization package for human neurophysiological data including EEG, MEG, and intracranial electrophysiology.

    mne.tools · MNE Community · Software & Documentation · Free and open-source · Intermediate

Societies & Scientific Conferences

Professional associations, conferences, and open science communities.

  • COSYNE (Computational and Systems Neuroscience)

    The premier international scientific conference bringing together experimentalists and theoretical neuroscientists to present research on systems-level neural coding, circuit computation, and decision-making.

    cosyne.org · Cosyne Organization · Scientific Conference · Paid registration / free talk archives

  • Society for Neuroscience (SfN)

    The world's largest association of scientists and physicians devoted to understanding the brain and nervous system, offering key professional development and networking.

    sfn.org · Society for Neuroscience · Society & Scientific Conference · Paid membership and registration / free educational materials