a review of Mathematical Imagery

03 Aug 2018 / Leonardo Barichello

Mathematics Teaching has just published my review of the book Mathematical Imagery. For editorial reasons, the review was shortened but they authorized the publication of the full version online and here it is.

Review of the book Mathematical Imagery, written by Dave Hewitt and others

Initially, the book focuses on 5 processes that the authors identify as related to imagery, keeping in mind the almost self-evident but still very oblivious perception that one does not learn just by looking at an image but has to do some work on it. The processes are: Reconstruction/Construction (Deconstruction?), Moving from imagery to abstraction/symbolism, Awareness of awareness, Stressing and ignoring, and Images that provoke a need for…. The authors do not claim the list is exhaustive, but claim it illustrates, with examples of tasks and experiences, the types of process that may lay hidden behind imagery when it is used in mathematics classroom as more than just mere illustration.

The most important aspect of the book, in my opinion, is the fact that the authors do not evoke the well-known instructional approach of Concrete-Pictorial-Abstract (or any of its variations). Instead, the authors place imagery by itself as a central issue in teaching and learning mathematics and use it as a springboard to promote other skills, such as communication (what do you see in the image?), pattern seeking (what would come next?) and generalization (can you extend the image?).

This approach to imagery is coherent with the common view among mathematicians that there is something visual behind mathematical thinking and discovery; something that is often related to imagination, that has a more holistic character than symbolic representations and formal reasoning. But it is also consistent with new findings suggesting that our brains seem to rely on essentially visual models to represent basic mathematics concepts, such as a number line to represent quantities.

Beyond that, while reading the book something else got my attention. Something subliminal that resonated with the feeling I got after John Mason’s seminar “Teaching More by Teaching Less - Getting Learners to Make Use of Their Natural Powers” in November 2015. I am going to call it the de-numeration of mathematics. Despite the fact that several activities suggested in the second part of the book involve actual numbers, they are never central neither in the question nor in the solution methods I imagined while trying to solve the activities. It feels that the authors are saying, and I deeply agree with them, that there is a lot of mathematics that can be done without numbers or, more precisely, without arithmetic skills and knowledge.

This view seems to be especially relevant when we think about students struggling with prior mathematical knowledge. I would say that the approach suggested in the book offers a way out of the typical situation where a teacher avoids a topic because he has the impression that the students do not master the pre-requisites well enough. This seems to be the reason why students placed in low sets are often stuck revisiting the same topics over and over again. I do understand the dilemma; in a subject so hierarchically structured as (academic) mathematics, how can I move to a new topic if students are not fluent enough with the previous? I believe this book hints at an intriguing solution: why not do some de-numerated mathematics and, progressively, build new topics on the top of such knowledge? An interesting example is given by Louise Orr and her proposal of an algebraic model using Cuisenaire Rods.

This process of de-numerate mathematics could be a path to reduce the frustration caused by difficulties with times-tables and other arithmetic skills and to promote imagery as well as other visual skills, that are knowingly correlated to mathematical achievement and choice of STEM carriers.

The second part of the book consists of more than forty well illustrated images accompanied by prompts and questions to promote discussions that starts in what can be seen in the images and drifts to topics ranging from geometry, arithmetic, algebra, number theory and so on. All low threshold and high ceiling ideas. In conclusion, the authors say that:

We suspect that many successful mathematicians develop their own imagery for mathematical processes and, in the absence of such imagery, others are left mystified by where mathematical insight comes from. (p. 35)

If you agree with that proposal, this book offers a rich and intriguing set of images, tasks, experiences and thoughts to inspire you to explore a more de-numerated mathematics.

Peer Review

02 Aug 2018 / Leonardo Barichello

Comic strip straight from xkcd: https://xkcd.com/2025/

peer review

PS: if you do not know xkcd, there is a hidden joke when you leave your mouse over the image for a couple of seconds.

Where should I publish?

26 Apr 2018 / Leonardo Barichello

A common recommendation for those starting to publish in academic journals is try to go for the journal you read the most. Of course this could mean a highly competitive journal, in which case you may need to consider other options, but the idea behind this advice is that by reading papers from a journal you get familiar with its style, purposes and so on and, therefore, all this would reflect in your paper.

When I heard this advice my computer programmer brain thought: I could write a script to do this for me!

Since I use Zotero and Zotero is a very friendly open source software, I decided to try to write a script in Python that would search its database. After checking the tables, I could identify where the information I needed was (see the query below).

SELECT * FROM itemData INNER JOIN itemDataValues ON itemData.valueID=itemDataValues.valueID
WHERE fieldID=12 ORDER BY value

For those interested: 12 is the id for the "publication" field, which contains the title of the journal for "journal article" items.

You can use the query above in any software that can read SQLite (I recommend SQLiteBrowser on Linux). But if you are not experienced with such things, you can just download the script available on my github or here (this link may be outdated).

After unzip it, you will get a .py file. Save it on main Zotero's folder (you will see a zotero.sqlite file in it) and run the script. It depends on some very basic Python library, which I believe are installed in most computers being regularly used. The screenshot below shows the result.


Now, you just need to scroll the list and find the most common journal in your library (pay attention to small variations in the title of the journal - the script is sensitive to them). For me, it was the Educational Studies in Mathematics.

Presentation using beamer via pandoc

09 Apr 2018 / Leonardo Barichello

Beamer is a tool to create slide presentations using Latex. Since I was "forced" to use Latex for a project I am involved with, I am trying to use it as much as possible to get acquainted with its quirks; which, by the way, are a lot.

Some time ago I heard of Pandoc. A tool that allows you to export text created with markdown (supposedly the most intuitive way to format text without using an interfaced software such as Libreoffice). The idea of exporting my notes, such as below, directly to a slide(image further down) is very appealing to me, so I decided to combine both to generate a pdf presentation directly from my notes.

## The lesson plans

- 12 lessons plans grouped into 3 packs (one per term);
- Worksheets + Comments for the teacher + cut-outs;
- Topics covered:
    - equivalent fractions,
    - comparison of fractions,
    - addition and subtraction,
    - mixed numbers and improper fractions,
    - word problems.

slide generated with pandoc and beamer

You can check the complete file with my notes in markdown (actually, just a part of it - just enough to show the most important codes in markdown) and the resulting pdf presentation.

Once you have pandoc and latex (with packages to run beamer) installed, use just have to use the following command to compile the presentation:

pandoc inputfile.md -t beamer -o output.pdf

Basically, pandoc generates a latex file and then uses latex+beamer to create a pdf formatted as a presentation.

Some comments

  • The first block of content is the metadata, which generates the front page and set some parameters regarding the layout. The slide_level parameter determines how many sections and subsections (marked by # in the md file) are before something is an actual slide. The header-includes are some parameter to change the layout of the theme I used. The others are sort of self explanatory;
  • I really do not like beamer's default themes, so I used metropolis. It is well documented and very elegant. It has some limitations and small bugs, but it does work and looks really nice;
  • The standout slide (last one) did not work perfectly for me. When I used it in the middle of the presentation it affected the title of the next slide. I guess it is a problem in the interaction between pandoc and the theme;
  • Beamer does not offer fields for supervisors and funders, so I had to improvise. Apparently it is possible to solve this issue by tweaking the theme and I will try to do so when I get some spare time.
  • I decided not to use any reference system because in presentations I use to include just a few references, but bibtex should work as usual;

The complete presentation can be seen here.

1 2 3 4 5 next ... end

Subscribe to my RSS:



english português música poesia coffee Ireland mathematics education café music movie opinion duas versões two versions research methodology recipe política matemática opinião nottingham visualization linux food educação programação receita comida viagem arte educação matemática beer libreoffice amsterdam pesquisa free software video vídeo mobile tourism coding restaurant comic