Skip to content
Tutorials Center
My learning

Topics

Lesson 43 of 43 ยท Projects and Practice

Further Study: Where to Go After Cognitive Science

Plan your next steps after this course: neuroscience, linguistics, cognitive psychology, machine learning and psychology, with tracks on this site and how to read research.

Cognitive science is a meeting place. No one person masters all of it, and most researchers pick one angle: brains, language, behavior, computation or philosophy. This last lesson helps you choose a path and gives you habits for reading the research on your own.

Five study paths around a central node.
Cognitive science connects to many fields.

What you will learn

  • Which neighboring fields match which interests.
  • Where to find related tracks on this site.
  • How to read an empirical paper and judge its evidence.

Pick a path

If you enjoyed... Try next Why
Neurons, vision pathways, dopamine, consciousness research Neuroscience The implementational level in depth
Parsing, priming, language acquisition Linguistics Systematic analysis of sound, structure and meaning
Memory, attention, reasoning experiments Cognitive psychology The experimental core of the field
The broader behavior and development of people Introduction to psychology The wider context of the mind
Perceptrons, backpropagation, networks Machine learning and deep learning Build and train the models you read about
Language models and text Natural language processing How machines handle language
Edge detection and vision models Computer vision Engineering view of what V1 and the ventral stream do

How to read a research paper

You do not need to understand every equation to judge a study. Ask these questions in order:

  1. What is the question and the hypothesis? Is it stated before the data?
  2. Who or what was studied? How many participants, and are they typical of the population the claim covers?
  3. What was measured? Does the measure capture the concept, or just something nearby?
  4. What was compared? A good design has a control condition.
  5. How large is the effect? Look beyond p-values to effect sizes and confidence intervals.
  6. Has it been replicated? Prefer results seen by independent teams, and be alert to preregistered or multi-lab replications.
  7. What do the authors say are the limits? Honest limits sections are a sign of quality.

Keep a project going

The best way to learn is to build. Ideas that extend what you did in this chapter:

  • Extend the Stroop demo with a neutral condition and more trials, and analyze your own data using the paired test from the project.
  • Modify the Hopfield network to store larger patterns, or add noise during updates, and study spurious states.
  • Re-implement the dopamine TD model with a delay that varies from trial to trial, and see how the error behaves.
  • Write a one-page critique of one claim from the consciousness chapter, stating which evidence would change your mind.

Try it

Which path suits you?
JS A paper-reading checklist

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

No output.

Check your understanding

Quick check ยท 3 questions

of 3 answered You scored / 3 ยท perfect!

  1. 1. Which field gives the deepest look at the implementational level of explanation?

    Neuroscience studies how neurons and circuits implement cognitive functions.
  2. 2. Why prefer results that have been independently replicated?

    Independent replication is a strong check on chance findings and hidden flaws.
  3. 3. What should you look at besides a p-value?

    They tell you how large and how precisely estimated the effect is.

Summary

  • Cognitive science has many neighbors, and you can choose a path by interest.
  • This site has tracks on neuroscience, linguistics, cognitive psychology, psychology, machine learning, deep learning, NLP and computer vision.
  • Reading papers critically is a skill with a simple checklist.
  • Keep building: small projects teach more than passive reading.

Next: Back to the Cognitive Science course