Skip to main content Link Menu Expand (external link) Document Search Copy Copied
I2 Intro Neuro/AI

Unit 12: Human Characteristics of the Brain

Several traits of humans appear to be unique to our species and may be essential in designing an intelligence similar to our own. This chapter is dedicated to the study of these phenomena of our brains. We begin by trying to peel back the layers of humans’ strong social tendencies:

Task 1: Complete the following lecture and synthesis questions.

20. Theory of Mind & Mentalizing

Synthesis Questions:

  • Describe the false belief paradigm and what function it serves.
  • What part of the brain is/are specifically involved in thinking about others' thoughts? Where is it located? What are several close by modules and their functions?
  • Google the terms TMS, EEG, and DBS. What are several differences and similarities between these cognitive science methods?

Task 2: Depending on your level of comfort with languge modeling, take a look at the following resources below, and then watch the Attention and Awareness lecture. If you are familiar with Attention in Transformers, try comparing and contrasting to what the lecture says about the brain. If you are not familiar with Attention in Transformers, just learn what you can from the video purely from a neuroscience standpoint.

If we have not completed the LM unit yet, Attention in ML may not be the most familiar. If you would like, here are some resources to prime your understanding:

Optional: A very comprehensive and visually helpful intuition for what the attention mechanism actually does: Attention Mechanism In a Nutshell

Optional: If you want to play with the math behind LLMs, this is pretty cool: 3Blue1Brown: Attention in Transformers

Optional: Self Attention Colab Notebook

For a deeper dive, feel free to take a look at the Language Modeling chapter of this document. Now we will dive into the capabilities of the brain and its own beautiful, endogenous attention mechanisms!

24. Attention and Awareness

Synthesis Questions:

  • Describe the brain's ability to multitask? What are some scenarios where parallel processing is feasible and some where it is not?
  • Describe covert and overt attention.
  • Describe "priming the visual cortex" why it might be useful.
  • "You have 10x as many connections going down from cortex, down to the LGN ([lateral geniculate nucleus](https://en.wikipedia.org/wiki/Lateral_geniculate_nucleus))… than going forward. One of the things you're doing is setting up selective filters so that only the stuff you want to process makes it to higher stages." Compare this fact to deep learning algorithms.
  • Describe the role of the Fronto-Parietal Attention Network
  • If familiar with Attention in ML: Research some machine learning attention mechanisms. Compare and contrast the mechanism and the capabilities of one with what you learned about our biological attention mechanism.

Project Spec

Consider, for a moment, the ways in which there are or are not parallels between our current mechanisms of machine learning and the various modules and functions of the brain that we have learned about. In your opinion, are some crucial for intelligence? Which, if any, have we managed to emulate with algorithms?

Now consider attention. The brain seems to block out unwanted information from ever being processed (about minute 8:00, Attention and Awareness). Are there ML algorithms that do this? Should they? What should the function of attention be? Does it depend on the context or is it a fixed algorithm for all contexts?

Write some thoughts (200+ words) on these questions. Then, finally, find one article on Grey Matters that interests you and consider how it relates to the above questions. Provide the link and some thoughts :)