# The Big Picture Problem with Anki

**URL:** <https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516>\
**Category:** Learning Effectively\
**Created:** [June 27, 2024, 3:29pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516 "2024-06-27T15:29:56Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![DerIshmaelite](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/derishmaelite/32/17983_2.png) [@DerIshmaelite](https://forums.ankiweb.net/u/DerIshmaelite)\
**Post date:** [June 27, 2024, 3:29pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/1 "2024-06-27T15:29:56Z")

</div>

It is generally recognized as a rule of thumb that cards in Anki should be done according to the minimum information principle, being as atomized as possible, with not so many pieces of information being asked all at the same instance at once and with just the right amount of cues.

I have made my cards to try to abide by this principle as much as I could get with cloze cards. However, I do often find myself suffering from the problem of losing the Big Picture which links everything with each other.

I just learn the cards as their own and not pertaining to a greater, bigger topic. And generally, the bigger topic I am learning is, the worse this problem becomes.

* * *

I find a text, copy it, read it, and try to understand it and learn it verbatim by making cloze cards out of it. I do not want to read the entire topic every single time a single card in the text is being asked. That defeats a major benefit of spaced repetition.

So how do you find the balance between the atomization principle and maintaining the Big Picture all at the same time. ❓

---

<div class="post-metadata">

**Author:** ![guillempalausalva](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/guillempalausalva/32/34258_2.png) [@guillempalausalva](https://forums.ankiweb.net/u/guillempalausalva)\
**Post date:** [June 27, 2024, 3:42pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/2 "2024-06-27T15:42:11Z")

</div>

> [@DerIshmaelite](#):
>
> learn it verbatim

Do you mean you do not adjust the text when formulating your cards?

---

<div class="post-metadata">

**Author:** ![DerIshmaelite](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/derishmaelite/32/17983_2.png) [@DerIshmaelite](https://forums.ankiweb.net/u/DerIshmaelite)\
**Post date:** [June 27, 2024, 3:48pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/3 "2024-06-27T15:48:27Z")

</div>

Sometimes. But most of the time not. If you mean making my own notes then absolutely not.

Sometimes I highlight linking words to try to keep track of what some words are referring to but nothing much…so yeah.

---

<div class="post-metadata">

**Author:** ![ferophila](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/ferophila/32/10566_2.png) [@ferophila](https://forums.ankiweb.net/u/ferophila)\
**Post date:** [June 27, 2024, 3:57pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/4 "2024-06-27T15:57:12Z")

</div>

For me, learning with cloze texts is often not effective, as the learning content is only recalled within the context of the card. In this respect, with the exception of some of my language learning cards, I don’t use cards with cloze texts.

In order not to lose sight of the big picture, I create several cards from different perspectives for many topics. On the answer page of my cards, a field is often displayed that contains extra information on the topic (text, images, audio) providing a larger context.

---

<div class="post-metadata">

**Author:** ![DerIshmaelite](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/derishmaelite/32/17983_2.png) [@DerIshmaelite](https://forums.ankiweb.net/u/DerIshmaelite)\
**Post date:** [June 27, 2024, 3:59pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/5 "2024-06-27T15:59:15Z")

</div>

Well, I am doing also that, in a way at least. Take a look at my cloze note.

Here is a cloze card inside my cloze note, which essentially contains the text of the entire topic. Anki is essentially my textbook at this point.

 ![image](https://us1.discourse-cdn.com/flex002/uploads/anki2/original/3X/3/e/3ef5426dab04599de220832ec4240e7fbcb0baf5.png)

---

<div class="post-metadata">

**Author:** ![ferophila](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/ferophila/32/10566_2.png) [@ferophila](https://forums.ankiweb.net/u/ferophila)\
**Post date:** [June 27, 2024, 4:00pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/6 "2024-06-27T16:00:11Z")

</div>

What is the question and what the answer side?

---

<div class="post-metadata">

**Author:** ![DerIshmaelite](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/derishmaelite/32/17983_2.png) [@DerIshmaelite](https://forums.ankiweb.net/u/DerIshmaelite)\
**Post date:** [June 27, 2024, 4:01pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/7 "2024-06-27T16:01:01Z")

</div>

The beige cloze card is my question side.  
Here is the answer side.

 ![image](https://us1.discourse-cdn.com/flex002/uploads/anki2/original/3X/6/5/65df0e650fa356ef92e03c09be09484c1252f304.png)

---

<div class="post-metadata">

**Author:** ![guillempalausalva](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/guillempalausalva/32/34258_2.png) [@guillempalausalva](https://forums.ankiweb.net/u/guillempalausalva)\
**Post date:** [June 27, 2024, 4:04pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/8 "2024-06-27T16:04:25Z")

</div>

That would be the opposite side of the spectrum regarding the minimum information principle. It would be best if you had simple Q&A or cloze deletions.

> [@DerIshmaelite](#):
>
> If you mean making my own notes then absolutely not.
> 
> Sometimes I highlight linking words to try to keep track of what some words are referring to but nothing much…so yeah.

If by notes you mean anki notes (that generate anki card via card types and templates). then that is the main reason you are not building the big picture.

---

<div class="post-metadata">

**Author:** ![DerIshmaelite](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/derishmaelite/32/17983_2.png) [@DerIshmaelite](https://forums.ankiweb.net/u/DerIshmaelite)\
**Post date:** [June 27, 2024, 4:11pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/9 "2024-06-27T16:11:39Z")

</div>

Here is what my card looks like:

> [@The Big Picture Problem with Anki](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/5):
>
> Well, I am doing also that, in a way at least. Take a look at my cloze note. Here is a cloze card inside my cloze note, which essentially contains the text of the entire topic. Anki is essentially my textbook at this point.

> [@The Big Picture Problem with Anki](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/7):
>
> The beige cloze card is my question side. Here is the answer side.

---

<div class="post-metadata">

**Author:** ![ferophila](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/ferophila/32/10566_2.png) [@ferophila](https://forums.ankiweb.net/u/ferophila)\
**Post date:** [June 27, 2024, 4:34pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/10 "2024-06-27T16:34:33Z")

</div>

It has been a while since I dealt with Schlundbögen and -taschen. Try to make ten to fifteen cards. Ask short questions that can be answered in a short amount of time:

- Was sind Schlundbögen?
- Wieviele Schlundbögen gibt es?
- was entsteht aus dem ersten Schlundbogen?
- Was entsteht aus dem zweiten Schlundbogen?
- Was bildet sich aus der ersten Schlundtasche?
- Woraus entwickeln sich die C-Zellen der Schilddrüse?

Give a short answer and add extra information for context to the back of your card, e.g. some fine Prometheus images or some mnemonic. Study your cards. If you struggle with some aspect, create another card addressing this aspect.

---

<div class="post-metadata">

**Author:** ![DerIshmaelite](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/derishmaelite/32/17983_2.png) [@DerIshmaelite](https://forums.ankiweb.net/u/DerIshmaelite)\
**Post date:** [June 27, 2024, 10:26pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/11 "2024-06-27T22:26:37Z")

</div>

Well, these sorts of questions are already covered by my cards requiring answers that could be formulated within \<10secs.  
The issue is not about the cards themselves: that is not my point.

My point is that my brain keeps learning these facts separately from each other due to this very reason, when in fact, in a real-life setting it would not require you to explain things individually, but in a larger context, spontaneously.

This is particularly problematic when it comes to subjects that require a lot of piecing facts together like parts of a puzzle, like Physiology, not as in Anatomy where facts could be simply learned on their own (to some extent, after which this applies no more as well).

---

<div class="post-metadata">

**Author:** ![ferophila](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/ferophila/32/10566_2.png) [@ferophila](https://forums.ankiweb.net/u/ferophila)\
**Post date:** [June 27, 2024, 10:49pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/12 "2024-06-27T22:49:42Z")

</div>

You need other modalities of learning then, like reading in context, explaining to fellow students, simulating oral examinations or creating individual notes (in the classic sense of the word).

---

<div class="post-metadata">

**Author:** ![medley](https://avatars.discourse-cdn.com/v4/letter/m/5fc32e/32.png) [@medley](https://forums.ankiweb.net/u/medley)\
**Post date:** [June 27, 2024, 11:11pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/13 "2024-06-27T23:11:28Z")

</div>

I would look into some sort of software where you can implement a PKM (personal knowledge management) system. I think that you need to organize all your thoughts and create connections between them in the form of notes AND use Anki to create atomic cards. I think if you organize (via note-taking), memorize (via Anki), and apply (via practice problems, projects, real-word tasks) you will have a very solid understanding of whatever you are trying to learn.

---

<div class="post-metadata">

**Author:** ![Shigeyuki](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/shigeyuki/32/23878_2.png) [@Shigeyuki](https://forums.ankiweb.net/u/Shigeyuki)\
**Post date:** [June 29, 2024, 1:48pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/14 "2024-06-29T13:48:58Z")

</div>

These add-ons may be helpful to memorize the big picture.

- [🔗Anki Note Linker](https://ankiweb.net/shared/info/1077002392)
- [Mindmap from tag hierarchy](https://ankiweb.net/shared/info/965278890)
- [📊Graph View - Link notes and visualize connections](https://ankiweb.net/shared/info/1068714931)

---

<div class="post-metadata">

**Author:** ![DerIshmaelite](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/derishmaelite/32/17983_2.png) [@DerIshmaelite](https://forums.ankiweb.net/u/DerIshmaelite)\
**Post date:** [June 29, 2024, 2:29pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/15 "2024-06-29T14:29:08Z")

</div>

Thanks a lot for these suggestions. However, tags don’t work great with cloze cards. Is there a different solution for that, like some sort of AI that can analyze text and create a mindmap based on a topic automatically outside of Anki and then use image occlusion on that mindmap to make learning much more visually appealing?

---

<div class="post-metadata">

**Author:** ![ferophila](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/ferophila/32/10566_2.png) [@ferophila](https://forums.ankiweb.net/u/ferophila)\
**Post date:** [June 29, 2024, 2:57pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/16 "2024-06-29T14:57:52Z")

</div>

Don’t fall prey to the popular misconception among students that all learning must be digital and don’t waste your time to make advanced technological systems work for your study progress.

Draw yourself a mindmap or some other visual overview of „the big picture“. You still can take a photo of it and attach it to your cards.

---

<div class="post-metadata">

**Author:** ![Shigeyuki](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/shigeyuki/32/23878_2.png) [@Shigeyuki](https://forums.ankiweb.net/u/Shigeyuki)\
**Post date:** [June 29, 2024, 3:36pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/17 "2024-06-29T15:36:59Z")

</div>

Currently I do not think there are any tools to create mind maps with AI.  
(Though I think there are tools that create cards from PDFs with AI or automatically create Image Occlusion from images).  
Recently many AI tools have been developed, but there are many criticisms, and it is still unclear which ones are useful.

- [Reddit : Best AI for creating flashcards as of now?](https://www.reddit.com/r/Anki/comments/16oev43/comment/k1k7hcb/?utm_source=reddit&utm_medium=web2x&context=3)

The OCR add-on may make it a little easier to extract text from images.  
But these are initially in English (maybe it doesn’t work for German), and don’t work on Mac.

- [📸Anki OCR - Generate OCR text from images](https://ankiweb.net/shared/info/546383173)
- [📸PasteOCR - Paste image as text](https://ankiweb.net/shared/info/1808435406)

In my case I use the incremental reading add-on for such long texts.(Guillempalausalva in this thread is an IR expert)  
In short, I use OCR to extract text from PDFs etc., put it in IR, and create cards little by little.

- [Incremental Reading v4.11.9 (unofficial clone)](https://ankiweb.net/shared/info/999215520)

But this IR add-on is designed to extract cards manually, so I do not think it will be very effective in getting the big picture or saving time.

---

<div class="post-metadata">

**Author:** ![Keks](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/keks/32/20537_2.png) [@Keks](https://forums.ankiweb.net/u/Keks)\
**Post date:** [June 29, 2024, 7:19pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/18 "2024-06-29T19:19:39Z")

</div>

Export a set of cards as txt and ask different neural networks to draw it based on this document.

---

<div class="post-metadata">

**Author:** ![DerIshmaelite](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/derishmaelite/32/17983_2.png) [@DerIshmaelite](https://forums.ankiweb.net/u/DerIshmaelite)\
**Post date:** [June 29, 2024, 7:52pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/19 "2024-06-29T19:52:49Z")

</div>

Can you give an example please. That sounds really interesting.

---

<div class="post-metadata">

**Author:** ![Keks](https://sea2.discourse-cdn.com/flex002/user_avatar/forums.ankiweb.net/keks/32/20537_2.png) [@Keks](https://forums.ankiweb.net/u/Keks)\
**Post date:** [June 29, 2024, 8:49pm UTC](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516/20 "2024-06-29T20:49:58Z")

</div>

> **[ChatGPT - Mindmap Master I Diagrams, Charts, Planner 🧠](https://chatgpt.com/g/g-Gv8NBj862-mindmap-master-i-diagrams-charts-planner)**
>
> Generate an ultra-cool mind map from your concept, a link or an upload file. Support Mindmap, Flowchart, Sequence Diagram, Gantt Chart, Class Diagram, State Diagram, Pie Chart, Bar Chart and more.

I don’t have any examples. I don’t do that. Here is one of the ready-made solutions for you.

[Next page](https://forums.ankiweb.net/t/the-big-picture-problem-with-anki/46516.md?page=2)
