Blog/2025-01-15/Spatial-Textual Memory

Right where I expected it

A recent bit of news has Sweden switch off computers for studying and homework to paper. The usual criticism of computer homework and studying is:

• Eye-strain from backlit displays
• Trouble with focusing on a general-purpose device with recreation tools built-in

These are legitimate concerns, for sure, but one thing that I've found interesting in my current use of e-readers over paper books is the location of the text in the book and related cues.

Place on the Page

I recall that in the past, when reading books I could search for a section very easily by skipping back to that part of the text. This search was insanely effective. Often I'd open the book to the page that the text was on the first time, and other times I'd be a few pages off and able to search down pretty quickly after that. The memory worked the other way too: for sections that I recall with importance, I could mentally visualize the book's shape (how much read, how much unread), the page and the text's location, and the rest of the words would come to me almost magically.

Part of the change from then to now is that I am getting older and have suffered many concussions, but certainly one thing I find disorienting when using e-readers (which are otherwise incredibly convenient) is that they reflow text in some situations. I might set aside the reader and come back and the text won't be at the position I recall. Either I'm hitting some text-reflow shortcut accidentally or the text is reflowed automatically. In either case, the experience is strangely disorienting.

The other thing is that I have lost this spatial intuition in the book. I can no longer switch to the section or perform the memory tricks from before. I was curious about this so I've got a couple of paper books and one thing that I find quite interesting is that even on the first reading, I have a memory of random things. Precious Little Sleep has a distinction between SIDS and SUID in the bottom left of a page in the beginning of the book. I can see it in my mind's eye without picking up the book.

In the research, this is called 'place-on-the-page' memory and apparently, while a weak addition to normal recall of concepts and text it does actually work as an index into the haystack. Looking through the work on memory and reading in that text it seems that this is quite common among people. Certainly it is then not strange that some of us do it more than others, since few human abilities land within a tight distribution.

Place in the Class

One strange related anecdote is that I've only taken notes during lectures once in class: in Algebraic Geometry, my worst class ever. For most lectures I can usually recall all sorts of other related information that cues me into a memory. Often this is how much of the blackboard is full, where the lecturer was standing next to the board or in which orientation, and sometimes the position in the textbook where I next encountered a concept first discussed. Within the first couple of classes, we encountered Hilbert's Nullstellensatz which I distinctly recall as on the top of the page in Hartshorne's Algebraic Geometry. On the third class or so, taken by the theory that note-taking improves memory and so on, I applied myself to rigorously taking notes.

Unused to writing down things while paying attention, I found myself falling behind in class. But the research showed that note-taking was good, so I persisted. The result was disastrous: my worst grade in any class since middle-school and with the last surviving memory of the class being the position of the Nullstellensatz in the textbook. This was also a lesson in abandoning techniques once they start failing you, since I committed to it far too long, and then persisted in it because if I abandoned the technique then it just meant I was bad at Algebraic Geometry: which might well be the case.

I have since persisted in my original method of class attendance: 100% attention in class with zero note-taking except to write down ideas that follow that I want to explore. This worked well for the classes that I took after and in the Computer Science degree at North Carolina State University that I had to acquire to move to the US, which was admittedly far less rigorous than the Mathematics post-graduate degree at IIT Madras.

All in all, it strikes me that if most people use place-in-the-page as an index like I do, then they probably also use place-in-the-class like I do. The notes, in the end, were never as useful as self-study post-hoc and notes I created from the textbook rather than the lecture. I haven't yet found anything in pedagogy about this zero notes approach, though some people online do mention it as an aside in other contexts.