Option for an algorithm on unreviewed cards past its due date

I think this is a very helpful tool! I will give an example to picture out my thoughts.

Let’s say you have a card. The last time you reviewed the card, you thought that you had a good grasp of it, and clicked the option to review it again in two weeks time. But, instead of reviewing it within 2 weeks, it was only after 1 month that you reviewed the card. So when you are faced with the card, there are two possible scenarios, it is either:

  1. You forgot every content of that card; or
  2. You remembered it - whether fully, or partially

There is no problem when you forgot the card, since you would just opt to recall it immediately and the algorithm does it job.

The issue comes in the second result when you remember it fully or partially. Because once you remembered it, the algorithm will still treat it as if you had revised it within the due date. Meaning, the algorithm would set its next due date by a month, or further. However, this could be a problem because going back to the foundations, the forgetting curve would have already set in by failure to review it within its due date. Yet, you would still be required to recall it after one month. Just imagine the forgetting curve. Given the situation above you should have been in the lower spectrum already of the graph, hence, it is not optimal to set its next revision after one month.

So, to resolve it, I was hoping that there will be an algorithm that will adjust the due dates automatically if it fails to be reviewed within its due date. Going back to the example above, instead that the card would then be reviewed after one month, it would prompt you that it will still be reviewed after 2 weeks or whatever is optimal. What the algorithm would do would be to decrease the review date until to whatever it thinks is optimal.

This was my thought when I wrote the following for this feature request Rem Cluster - a small group of rems that can be reviewed together in sequential order. Why are these tools not already smart enough to do all these!!

  1. What if I forget one card from a cluster?
    Ideally there should be a setting so that the user can decide what to do when this happens. One option is to consider the whole cluster as forgotten, but that is not ideal. Another smarter option would be to take into account the next repetition time of the cluster and schedule reviews for the forgotten card in such a way that by the time the cluster gets reviewed next time, you would remember the forgotten card.

You do realise this is what the “hard” option is for? Much like “easy” is to push cards further if they are, and this is nuts, too easy, “hard” schedules the next repetition sooner. You may play around with the ease multipliers in the options if the default ones aren’t exactly to your liking.

I don’t think the hard option is the solution for what I was thinking. As you can see, if a card had been left unreviewed, the algorithm of its SRS remains the same from when you left it. So when you left it, the same was due for 2 weeks, the next time you open the card, the following options will have the respective due dates “Again”=Immediate, “Hard”=21 Days, “Good”=1 month, “Easy”=1.4 months. But if you forgot to review the card on its due date, and only reviewed it after a month from its due date, you would still have same options for Again, Hard, Good, and Easy since you left it.

However, as I have mentioned, the forgetting curve should have set in already. Meaning, the optimal review dates should not have been 21 days, 1 month, or 1.4 months, respectively, but something lower than those numbers.

The numbers are mere suggestions. Since you make the cards and you recall them, only you can truly know whether it was hard or good, for whatever reason. In your case, a previously “good” card turned “hard”, and you may correct it as such. Once it turns “good” or “easy”, you may correct it again. The algorithm is there to work for you, not the other way around.

found this from here; but not sure how much of an effect it has on the intervals