From Basic Principles to Concept Maps

Sep 12, 2014 • Greg Wilson

Instructor training group 11 had their first meeting this week. We raced through a lot of foundational material, and had a lot of good questions. The first exercise is described below, followed by the notes participants took during the meeting; I look forward to seeing everyone again in two weeks.

Homework

  • Part 1
    • Pick a small topic from the Software Carpentry Version 5 lessons
      • Something you could teach in 5 minutes
    • Draw a concept map for it
      • Draw it freehand, then post a scan or photo — no computer drawing tools allowed
      • All links must be labelled: the relationships are as important as the concepts
      • Should not be strictly hierarchical — remember, densely connected is more memorable
    • Post it on the blog by Wednesday, Sept 17
  • See the “Concept Maps” category of this blog for examples from previous groups of learners
  • Part 2
    • Give feedback on at least three other concept maps
    • Substantive feedback: not just “I like it”, but:
      • Here’s where your map differs from my understanding of this problem
      • I think these links are misleading
      • You’ve missed this big idea
    • “Be the reviews you want to see in the world.”
    • All reviews posted by Tuesday, Sept 23
  • Part 3
    • Read chapters 1 and 2 of How Learning Works
    • Also read this article on the Software Carpentry blog, along with the comments on it and these replies
    • (Optional) add yourself to the “instructors” and “discuss” lists (which you can do here)
      • We use the first to advertise upcoming workshops: you’re welcome to start helping out when and where you can.
      • The second is for general discussion of Software Carpentry’s teaching, tools, procedures, and everything else.
  • We will meet again on Sept 24-25 (same times)

Notes

  • http://en.wikipedia.org/wiki/Teaching_reading:_whole_language_and_phonics (Phonics popular today, teaches by elements — sounds -> words. Whole language popular in 60s, learn to recognize whole words first)
    • Dr. Seuss (example of whole-language approach)
    • However: teacher enthusiasm is more important than the instructional model for success in learning to read!
  • Computing education & learning far less studied than mathematics education! (so we’ll draw on research from other disciplines)Premises:
  • Participants are supposed to be well fed, etc.. We will not be focusing on the social aspects underlying education.
  • Terms:
    • cognitivist approach: thinking of the problem in terms of brain science.
    • educational psychology: study of how brains learn. Tells us a lot about what could work but under-constrains the problem. It is the Science. Covers neurobiology to sociology.
    • instructional design: Application of educational psychology. Engineering. The role is to feel, test and compare different educational tools. Take something that sounds plausible, take it out and test it.
    • reflective practice
    • concept map
    • conceptual “chunking” and “chunks”: Sequences of things that commonly occur together eventually get remembered as a single object rather than constituent parts
    • others?
  • 7+/-2 concepts fit in working memory
    • Working memory has limited storage but fast (>1ms) access. Long term (persistent) memory is, as far as we’re concerned, unbounded but has slow (~100ms) access. Research predicts that we can fit about 7 +/- 2 concepts in working memory at a time.
    • best: new concepts for 5 minutes, then absorb for 5 minutes, repeat (consolidation and retrieval is needed; consolidation is a bit like packaging it for storage, and retrieval is getting it back, as in using it or a test.)
    • Students need to be able to recognize patterns in order to consolidate it into long term memory.
    • Teachers try to steer students into learning chunks, so they have an easier time consolidating information into long term memory. Chunking is inevitable- it is important to guide students to make useful chunks.
    • modern, longer classes based upon human convenience, but often introduce too many concepts
  • 3 Stages as we learn (simplified from 5-stage model): Patricia Brenner, 1980s
  • “Never hesitate to sacrifice truth for clarity” — Words of Wisdom from Greg’s dad
    1. Novice — doesn’t know what he/she knows (apparent bc they don’t ask meaningful questions) Lacks a conceptual model of the topic. They are not a blank slate. They are taking the information you give them and putting them into boxes (their mental models) they currently have. So they are mis-categorizing and mis-labeling information. Every person places the information in different boxes.
      • Don’t give them lots of information too soon. If they place too much information in the wrong boxes, it reinforces misconceptions.
      • Goal: Help them construct a correct conceptual model.
      • need a conceptual model to make learning efficient (e.g. discussing the meaning of commands as they are introduced)
      • If they have a model, they can acquire information rapidly on their own time. Without a model, we’ll have given them parts of a watch without telling them what a watch is.
      • Many tutorials go wrong because they think they exist to give people commands.
      • give a map of terrain
      • Voodoo programming.
    1. Competent Practitioner — has a workable model of problem to do most things most of the time, without thinking much
      • Not stunt (James Bond ;-)) drivers but can navigate in typical city traffic and not crash.
      • mostly ignored in SWC training
      • Good teachers for novices (still remember what it’s like not to know the subject)
    1. Expert — has many more links between concepts/facts
      • Can handle the oddball cases.
      • seem to have ‘leaps of intuition’
      • Say you store information as a graph. You have a bunch of nodes connected by links. Barack Obama --- United States. Key difference between the expert and the competent practitioner is that the graph is more densely connected. There might be more nodes, but the key thing about expertise is that the expert has to go fewer hops to go from what she knows to what she needs to know. Experts can go from A ---> Z in one leap. In their minds those concepts are adjacent. For competent practitioners, they have to do a search and manually go from A ---> B ---> C, etc.
      • Have difficulty explaining their reasoning.
      • Most experts backfill a plausible path, not explain their actual path. They try to guess at what their path is — related to chunking.
      • Experts are fantastic debuggers, much better at working backwards.
      • no longer remember what it’s like to not have a correct working model
      • not inherently good teachers
  • SC is not about creating experts.
    • What does it take to become an expert? You can drive every day and not become James Bond.
    • What it takes is reflective practice (deliberate practice). You need to reflect on your practice in order for it to improve.
    • The problem is that as soon as you become competent in something, you often stop thinking about it.
    • It means you will never become better at chopping onions.
  • Steps students go through:
    • Novice —> Expert: “Deliberate Practice”
    • you have to think about and reflect on it
      1. You are getting feedback from somebody else. You are the recipient of feedback.
      1. You start to give people feedback. I.e. In athletics, the teenagers start coaching the preteens.
      1. Epiphany: People start taking the skill of giving feedback and** apply it to themselves**.
  • The problem is that this reflective loop is absent in the English speaking school system with respect to teachers.
    • This is why in SC, you are not only doing the teaching exercises but evaluating others exercises.
    • And that’s why it’s equally important to practice evaluating others (and then yourself) as doing the actual exercise.
    • Learning how to say something negative in a constructive way is a skill.
    • Learning how to hear criticism and evaluate and apply it is also a skill.
  • Teaching is a performing art.
    • Everyone who teaches SWC feels as a fraud. That’s normal, but try getting better and helping others.
    • Student evaluations have been shown many times to be uncorrelated with how much students have learned.
    • Their purpose is to make students feel like their feedback is being listened to
  • The goal is to make it as densely connected as possible.
    • “What are the ideas that need to be in this lesson and how can they be connected?”
    • With point form notes, not only are you committing to the things you’re going to teach but also the order that you’re going to teach.
    • With concept maps, you separate the lessons from the order. You find the ideal path.
    • The connections are at least as important as the facts.
  • can be used to (1) develop curriculum for single learning session
    • to plan lectures, assess complexity, divide and rearrange ideas
  • useful for (2) scoping, making sure there’s not too much in each lesson: “Don’t put more nodes than you have fingers on one hand on a concept map”
  • if the concept map is too complex, it’s too much for learners to absorb
  • Ideally, you want to present memory through two channels so that it’s reinforced. A concept map drawn live or integrated in your presentation is useful for this.
  • You want to go at the speed of learners learning not teachers speaking.
  • can be used (3) as a teaching tool in class:
    • hand out at beginning of lecture
    • can draw the concept map as you instruct
  • can be used (4) as a feedback tool: ask learners to draw a concept map of what they learned (can be used by instructor to correct misconceptions)
  • For short course format use “one good thing one bad thing” feedback notes after each session.
  • Jeff Johnson, UI designer.
    • “The more polished something is the less honest the feedback will be.”
    • This is why it’s more useful to give someone a rough draft to edit the text rather than a beautifully designed Hollywood quality essay.
    • This applies to teaching tools as well.
    • Designing with the Mind in Mind