Neuroscience + physiology¶

Weeks¶

Week 41¶

Spine: EPFL Neuronal Dynamics + OpenStax A&P

Reading: Neuronal Dynamics Ch. 1-4; A&P Ch. 12 nervous-system intro

Know: Model membrane potentials, spikes, integrate-and-fire behavior and neural coding fundamentals.

Reconstruct: Derive leaky integrate-and-fire equation from RC analogy and membrane time constant.

Do: Simulate spike response to noisy current; estimate firing-rate code and timing sensitivity.

Defend: What information is lost when spikes are reduced to firing rate?

Gate: Pass: relate circuit parameters to neural dynamics and data representation.

Source: source

Week 42¶

Spine: Neuronal Dynamics

Reading: Population coding, decoding, learning/plasticity selections

Know: Decode hidden variables from neural activity and understand population representations/adaptation.

Reconstruct: Derive simple linear population decoder/least-squares estimate and spike-train likelihood intuition.

Do: Build a synthetic BCI decoder and test drift, recalibration and distribution shift.

Defend: Why can a decoder improve benchmark accuracy while becoming worse for the user?

Gate: Pass: evaluate calibration, latency, robustness and adaptation-not accuracy alone.

Source: source

Week 43¶

Spine: OpenStax Anatomy & Physiology 2e

Reading: Ch. 17 endocrine + Ch. 19 heart + Ch. 20 vessels/circulation + Ch. 22 respiratory

Know: Understand multi-loop physiological homeostasis: endocrine signaling, circulation, gas exchange and autonomic regulation.

Reconstruct: Derive cardiac output and oxygen-delivery relations; sketch endocrine negative-feedback loop.

Do: Construct a lumped cardiovascular/respiratory control model under exercise or altitude.

Defend: Why is physiological control decentralized and multi-timescale?

Gate: Pass: model explains at least two interacting feedback loops and compensations.

Source: source

Week 44¶

Spine: OpenStax Anatomy & Physiology 2e

Reading: Ch. 25 urinary + acid-base/fluid regulation; integration with neural/endocrine systems

Know: Understand filtration, osmoregulation, electrolyte and acid-base homeostasis as controlled transport systems.

Reconstruct: Derive clearance concept and simple mass-balance model for body fluid compartment.

Do: Design an artificial-organ control toy model (dialysis/insulin/ventilation) with sensor delay and safety constraints.

Defend: What makes replacing an organ harder than matching its average throughput?

Gate: Module defense: physiological replacement must meet dynamic regulation, redundancy and failure constraints.

Source: source

Exit gate¶

Closed-book: 120 min: membrane dynamics, neural coding/decoding, cardiovascular/respiratory/endocrine/renal control.

Novel problem: Design a BCI or artificial-organ controller under sensor drift, biological adaptation and safety constraints.

Artifact: Neural decoder or physiological closed-loop simulation.

Defend: Defend what state is sensed, what remains hidden, compensation loops, latency and failure risk.

Pass criterion: Pass if device is judged by dynamic regulation, not average throughput/accuracy only.

Transfer problems¶

Try these before consulting solutions or asking for the complete answer.

1. Neuron: Simulate leaky integrate-and-fire response to step/noisy current.

2. Coding: Compare rate and temporal codes on a classification/estimation task.

3. Decoder: Fit neural population decoder and test drift.

4. Plasticity: Model how adaptation changes decoder calibration over time.

5. Cardiac: Compute cardiac output and oxygen delivery under exercise.

6. Respiratory: Model alveolar ventilation/gas-exchange response to altitude.

7. Endocrine: Draw and perturb a hormone feedback loop.

8. Renal: Construct a solute/water mass balance and clearance calculation.

9. Artificial organ: Design control targets/sensors/actuators for an insulin, dialysis or ventilation system.

10. Failure: Show how a compensatory physiological loop can mask device degradation until abrupt failure.

Textbooks¶

See the five-book resource page.