Skip to content
Tutorials Center
My learning

Topics

Lesson 20 of 43 ยท Neuroscience Foundations

Chapter 4 review: Neuroscience foundations

Review neurons, action potentials, brain lobes, famous patients and imaging methods, work through reverse-inference arithmetic and finish with a ten-question quiz.

This review ties together the neuroscience of chapter 4: the cell, the circuit, the map and the methods. It also gives you a calculation that explains why a glowing region on a brain scan tells you less than you might think.

Four recap cards: neuron signaling, brain map, famous cases, and measurement methods, each with the key facts and numbers.
Chapter 4 on one page: cell, map, cases and methods.

What you will learn

  • The key facts from each lesson of the chapter.
  • How timescales and distances shape what neurons can do.
  • How to calculate how much an activation tells you about a mental process.

Recap table

Lesson Key facts
Overview About 86 billion neurons; brain is about 2% of body mass and uses about 20% of resting energy; levels from synapse to system
Neurons and signaling Rest about -70 mV; threshold about -55 mV; spikes are all or none; synapse gap about 20 nm; glutamate excites, GABA inhibits; Hodgkin and Huxley 1952
Brain organization Four lobes; Broca 1861, Wernicke 1874, H.M. 1953; double dissociation; networks, not isolated centers
Methods EEG (ms, cm), fMRI (s, mm), single neurons (ms, micrometers), TMS tests causes; beware reverse inference and multiple comparisons

Key terms

Action potential: all-or-none electrical spike.

Double dissociation: opposite impairments in two patients on two tasks.

BOLD: blood-oxygen-based fMRI signal.

Reverse inference: inferring a mental process from a brain activation.

Timescales and distances

Neural signaling has a speed limit. A spike lasts about a millisecond, a fast myelinated axon carries it at around 100 meters per second, and a thin unmyelinated fiber may manage only about 1 meter per second. Over a distance of 0.1 meter (across the brain), a fast axon takes about 1 millisecond, a slow one 100 milliseconds. These delays are part of why reaction times of hundreds of milliseconds (chapter 2) involve only a modest number of sequential neural steps, a point made forcefully by Jerome Feldman and Dana Ballard in their "100-step rule" for the brain: tasks done in about a tenth of a second can involve only a limited chain of sequential neural steps, so the brain must rely on massive parallelism.

Reverse inference by the numbers

Suppose a region R is active while a participant performs a task, and you want to infer whether the task engaged a particular process Y (say, fear processing). Bayes' rule gives the probability of Y given activation:

Text
P(Y | active) = P(active | Y) P(Y) / [P(active | Y) P(Y) + P(active | not Y) P(not Y)]

If R is active nearly every time Y occurs (0.9) but also half the time when Y does not occur (0.5), with a prior of 50% that Y was engaged, the posterior is only about 64%. If R is selective (active only 10% of the time without Y), the posterior is 90%. Selectivity matters more than activation.

Common misconceptions

"A larger, brighter activation means a stronger effect." Display thresholds and statistics determine blob size. Effect size and sample size matter.

"Neurons are binary, so the brain is a digital computer." Spikes are all or none, but timing, rates, synaptic strengths and graded potentials carry information in analog ways.

Try it

A conduction-time calculator. Enter a distance and several fiber speeds to see how long a spike needs. The speeds are typical ranges, not fixed values.

Python How long does a spike take to arrive?

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

A reverse-inference calculator. Change the numbers to see how selectivity changes the conclusion. The probabilities are made-up examples chosen to illustrate the arithmetic.

Python Reverse inference calculator

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Check your understanding

Quick check ยท 10 questions

of 10 answered You scored / 10 ยท perfect!

  1. 1. What is the approximate resting membrane potential of a neuron?

    Inside is about 70 millivolts negative relative to outside.
  2. 2. Which pair of terms describes the strength of evidence from a double dissociation?

    Reversed patterns across patients argue against a single common mechanism.
  3. 3. Which technique best combines millisecond timing with non-invasive use in humans?

    EEG and MEG track activity in real time but with coarse localization.
  4. 4. A region is active in many different tasks. What does its activation tell you about any one process?

    Low selectivity makes reverse inference weak.
  5. 5. Which neurotransmitter is the main inhibitory one in the adult brain?

    GABA mediates most fast inhibition.
  6. 6. Hodgkin and Huxley's 1952 model explained

    They modeled the squid giant axon's ionic currents.
  7. 7. Which pairing of patient and finding is correct?

    H.M.'s amnesia was selective, sparing procedural learning.
  8. 8. Why does fMRI have poor temporal resolution?

    The hemodynamic response peaks several seconds after the neural event.
  9. 9. The dead salmon study warns against

    Thousands of voxel tests yield false positives without correction.
  10. 10. What is the best way to test whether a region is necessary for a function?

    Causal tests require manipulation, not just observation.

Summary

  • Neurons integrate inputs and fire all-or-none spikes; synapses are chemical and plastic.
  • The brain has four lobes plus deeper structures, and famous patients taught us about specialization.
  • Methods trade space for time, and stimulation or lesions test causation.
  • Reverse inference depends on selectivity; always check sample size, statistics and corrections.

Next: Chapter 5 begins the next part of the course, on models of cognition. See the full plan on the Cognitive Science course page.