Computational Neuroscience & Vision
An independent one-hour primer on retinal computation, neural coding, adaptation, experimental neuroscience, interpretable models, and neural perturbation, designed around research themes relevant to the Stephen A. Baccus Lab.
3 modules · 6 lessons · 1h · mastery threshold 80
Watch videos free — no sign-inBackground for Research in the Baccus Lab This independent educational primer introduces scientific concepts relevant to research themes in the Stephen A. Baccus Lab at Stanford University. It is not an official Stanford University or Baccus Lab course and does not imply endorsement or affiliation. For ambitious high-school students and early undergraduates. Prerequisites: high-school biology; no neuroscience or programming background required. About 45 minutes of video, two brief checks per lesson, and a 10-question final assessment. Total learner time: about one hour. No lab, capstone, or Research Defense is required; certificate eligibility is mastery-only. Course arc: function → computation → mechanism → model → perturbation → behavior.
Module 1
What it means for a circuit to compute, and why the retina is a neural computer.
Module 2
How retinal circuits adapt and predict, and how scientists measure neural computation.
Module 3
From predictive and interpretable models to cortex, behavior, and ultrasound perturbation.
Electrophysiology and neural population recording
Interpret spike trains, raster plots, and firing rates from electrophysiology and imaging.
Function, computation, and mechanism
Distinguish the function, computation, and mechanism of a circuit.
Interpretable models and mechanistic hypotheses
Explain why predictive accuracy does not by itself establish biological mechanism.
Linking neural representation to behavior
Explain how researchers connect stimulus, neural activity, and behavior.
Neural adaptation and sensitization
Explain adaptation and sensitization as responses to changing stimulus statistics.
Neural computation as input-output transformation
Describe a neural computation as a transformation from input to output.
Neural perturbation and ultrasound neurostimulation
Describe focused ultrasound as an experimental tool for neural perturbation and its current limits.
Observation, perturbation, and causal evidence
Explain why perturbation provides stronger mechanistic evidence than recording alone.
Prediction and object-motion computation
Describe how retinal circuits use regularities to anticipate motion and separate object from global motion.
Predictive models of neural responses
Describe a linear-nonlinear model r(t) = f(k * s(t)) and how it is evaluated on held-out data.
Receptive fields and visual feature extraction
Predict ganglion-cell responses from center-surround receptive fields.
Retinal circuitry and ganglion-cell output
Trace visual signals from photoreceptors to ganglion cells and the optic nerve.