I have been studying geography with the Image Occlusion+ note type. I have a map with names of places on it. I cover the place names with rectangles, which then creates cards where one rectangle is highlighted and I have to guess what is under it.
My feature request
I would like the reverse of the Image Occlusion+ note type. By which I mean I want to be asked to point out the location of something on the map.
To be a bit more specific, the front of the card could say something like “Point out Rwanda” and show the map with rectangles covering all the names. The back of the card can then highlight the right rectangle in red or something.
It would of course require a way to associate a text (“Rwanda”) with a specific rectangle, which might make it a bit harder to implement.
Other ways to achieve the goal
I considered just editing the image and putting numbers on the map instead of names of places. The front card could then say “Where is Rwanda” and the back just says “13” or whatever the right number is on the map (the same map image can be used for all cards to save storage space).
The drawback with this is that it is slower, since you need the extra step of comparing the number on the map with the number on the answer card, rather than immediately seeing a highlighted red rectangle. It costs even more effort if you got it wrong and have to look around for the right number.
Another way would be to just manually make a bunch of images myself with various regions highlighted, but this would require too much storage space.
Other suggestions
There might be some smart workflow I’m not thinking of that would allow me to do this kind of “Point out X in this image” type of study. I’m equally happy for such suggestions.
Whether it is important depends on the demand I guess. Let’s see if a lot of people leave hearts on the post.
I agree that my described case is similar to the reverse. If you know the place name/location in one way, you should be able to figure out the reverse.
On the other hand, I tried sorting a deck I’ve been studying that has both types of cards by number of lapses. (I created the cards manually in that case using Wikimedia images, but this is only possible sometimes). The most lapsed 50% of cards were mostly the ‘say what is under this rectangle’ type. And the least lapsed 50% were mostly of the ‘point out this place on the map’ type.
I have been desiring something like this for quite some time. My map-based notes are configured in the same way as gvorst’s. Unfortunately, I have found the resulting cards to be ineffective. I have tons of lapses. My most common means of answering correctly is something like: “I remember getting ‘Rwanda’ incorrect a few weeks ago, so that rectangle must be Rwanda.”
I wish this were true! Alas, I have not found it to be the case. In general, knowing A→B does not automatically mean knowing B→A. For example, knowing that “hola” means “hello” does not guarantee that I know “hello” means “hola”.
To my way of thinking, the prompts are opposites and hence quite different:
Card 1: Given a rectangle, choose a name
Card 2: Given a name, choose a rectangle
Although I have wondered—perhaps Card 2 should not even display the rectangles on the front side of the card; rather, the user must visualize where the rectangle would go. If the front shows all the possible rectangles, the card becomes a species of multiple choice (which generally seems ineffective).