Daily Class Preparation:   cs.wellesley.edu/~cs125/reading.html

This page provides links to readings, videos, and questions to think about for each class, with the most recent classes at the end. The date indicates the class for which the assigned activities should be completed. Each class entry is linked directly from the course schedule page.

Wednesday, January 25
  • Read the first part of this online Introduction to Neurons and Neuronal Networks by John Byrne, to the end
    of the section on The Neuron, i.e. up to the start of the section on Neuronal Networks (don't worry about
    following all the details - we'll review the key properties of neurons in Lab)    . As you complete the reading,
    think about the following questions:
    1. What were neurons invented for? In other words, what do animals use neurons for?
    2. What do you use your brain for? Can you answer this in a way that applies both to you and to the
      simplest creatures that use neurons, like jellyfish?
  • For fun, view the video by Prof. Nancy Kanwisher (assisted by Wellesley alum, Rosa Lafer-Sousa!),
    The neuroanatomy lesson (Director’s cut) (about 5 mins)
Friday, January 27
  • View the TED talk by Prof. Nancy Kanwisher (MIT), A neural portrait of the human mind (about 17 mins)
    Think about the following questions as you view the video:
    1. What are some examples of cognitive processes for which the brain appears to have specialized regions?
    2. Why do you think the brain would develop specialized neural circuitry for tasks like this?
    3. What did you learn from the example of a potential "food" region?
  • Read the article, Fei-Fei Li: If we want machines to think, we need to teach them to see from Wired magazine,
    about an interview with Prof. Fei-Fei Li, Director of the Stanford Artificial Intelligence Lab
Tuesday, January 31
  • View the video by Prof. Nancy Kanwisher (MIT), What you can learn from studying behavior (about 25 mins)
    Think about the following question as you view the video:
    • What do you think it means to process a face "holistically"?
Wednesday, February 1
  • Measure your ability to recognize famous faces and newly learned faces by running the following two online
    face recognition experiments. Record your results for each experiment, and bring these results to Lab 2: Remember that face recognition ability has no correlation with IQ and other measures of intelligence!
Friday, February 3
  • View the video about the work of Dr. David Hubel and Dr. Misha Pavel (about 6 minutes)
    Think about the following questions as you view the video:
    1. Dr. Hubel says that "we're only at a very elementary stage when it comes to understanding something
      like how you recognize a face."       At what "level" does Dr. Hubel aspire to understand such visual processes?
    2. The narrator says that Dr. Pavel has "demonstrated how the visual system breaks down visual stimulation
      into millions of bits of information and recombines them into a coherent image that we recognize."
      Why do you think the brain breaks the image into bits if it is going to recombine them?
Tuesday, February 7
  • See the reading, videos, and questions listed for Friday, February 10. You may want to begin this preparation early.
Wednesday, February 8
  • Read the article "100% Accuracy in Automatic Face Recognition" by Jenkins and Burton that appeared in Science
    (vol. 319, 2008, page 435). Think about the following question as you complete the reading:
    • The authors have proposed that for the purpose of recognizing faces, we store an "average" appearance of each
      individual person — why might this be beneficial?
Friday, February 10
  • Read the article "Face-Blind" by Oliver Sacks that appeared in The New Yorker (August 10, 2010, page 36)
  • View the following videos by Prof. Nancy Kanwisher (MIT):
    1. What is fMRI? (about 6 minutes)
    2. Explaining a very simple fMRI experiment (about 10 minutes)
    3. Discovering a face specific region with fMRI (about 9 minutes)
    4. What happens when you stimulate the face area? (about 3 minutes)

    Think about the following questions as you complete the reading and videos:

    1. When designing an fMRI experiment, why is it essential to compare the fMRI signal obtained for two different
      conditions?
    2. fMRI studies reveal neurons in the FFA that respond more strongly to faces than other objects — why was the
      patient study also important to establish the role of the FFA in face recognition?
    3. What evidence does Dr. Sacks cite suggesting that the fusiform face area can be used for non-face objects,
      contrary to Dr. Kanwisher's claim that this area is primarily for faces?
Friday, February 17
  • Read the article "How the Brain Recognizes Faces" that appeared in MIT News (December 1, 2016) about some
    recent work from Prof. Tomaso Poggio's research group at MIT. The article also contains an embedded video (about
    5 minutes) that you should view. Think about the following question as you complete this reading and video:
    • Using machine learning to implement a new model of face recognition in the brain, these researchers
      discovered that their model developed an intermediate processing step that paralleled a behavior that had
      puzzled neuroscientists studying face selective neurons in the brain — what was this unexpected stage of
      processing?
Tuesday, February 28 — Wednesday, March 1
Friday, March 3
  • View the TEDx talk by Prof. Matt Wilson (MIT), Reading the Minds of Rats, (about 21 mins). Think about the
    following question as you view the video:
    • What possible connection(s) between hippocampus, navigation, memory, and sleep, are suggested by the
      work summarized in this talk?
Friday, March 10
Friday, March 17
  • Listen to the middle segment of the WNYC radio podcast, Words that Change the World, which is part of the
    RADIOLAB series. The middle segment (about 11 minutes, from 10:38-21:45) begins with, Consider what
    happens when you put words together...     It describes a simple task that children are only able to do when
    they begin to utter phrases like, the biscuit is left of the blue wall. As you listen to this segment, think about
    what are the two different types of knowledge being combined here, that enable children to perform this task
    (that rats cannot do!)
  • There are two other segments of the podcast that you may be interested in, but these are optional. The first
    describes a man who was born deaf and learns for the first time that words are symbolic of things in the world,
    and the last segment is about the many new words that Shakespeare introduced to the English language by
    creating "mash-ups" of existing words, like "eyeball."
Tuesday, March 21
  • The paper by Ernst Davis and Gary Marcus, Commonsense Reasoning and Commonsense Knowledge in
    Artificial Intelligence    , reviews the state-of-the-art of research in this area, its challenges and successes.
    Just skim through this paper and think about what kind of commonsense knowledge and reasoning is
    needed to build a language understanding system that converses naturally about everyday things. Also
    think about how this relates to your experience with the Mitsuku chatbot that you explored in Lab 7.
Friday, April 7
  • On Friday, we will have a class discussion of the somewhat provocative article, Can Machines be Conscious?
    by Christof Koch and Giulio Tononi that appeared in a 2008 issue of IEEE Spectrum.

    In an early part of the article, Koch and Tononi argue that there are six functions or processes that are not
    necessary     for consciousness: (1) sensory input and motor output, (2) emotion, (3) attention, (4) explicit
    or working memory, (5) self-reflection, and (6) language. You will each be assigned to one of these functions
    and should be prepared to briefly summarize the author's argument about this function in class. (You may
    also want to comment on whether you think the evidence for their claim is strong or weak.)

    What is necessary for consciousness in humans or animals, according to the authors? What do the authors
    think is necessary for consciousness more generally (e.g. in a machine)?

    In addition, find at least one claim in the article that the authors do not directly support by citing evidence
    or an argument.

Friday, April 28