Are Flashcards, Memory Palaces, Or Flashcards + Memory Palaces Better For Language Learning?

It's inevitable once you get into the world of language learning you will come across the problem: how should I learn vocabulary?

By far the most common answer you will find is god ol' flashcards. But if you did a little deeper, you might discover the world of memory palaces and even deeper memory palaces + flashcards.

Here's the golden question: out of flashcards, memory palaces, and memory palaces + flashcards, which is the most effective?

I set on a self-experiment over 12 days to answer this by using each of the three methods to memorize Spanish vocabulary. Why am I learning Spanish? I'm doing it for fun so I can travel to South America like a boss, talk with my Spanish-speaking friends, and learn more about meta-learning. I know the sample size is N = 1, but regardless, I hope my self-experiment motivates further study and introspection in the language learning community.

Here's how it went.

Methods

The experiment took place over 12 days. Every day, I memorized vocabulary for fifteen minutes using one of the types of recall, making for a total of 4 days using each technique. On the first day, I used flashcards, second memory palaces, third memory palaces with flashcards, and so on and so on. This fifteen minutes included the time it took me to go back through the words I memorized in the last sessions using that technique and memorizing new words.

I memorized words from my personalized vocabulary list--a list of Spanish words I write down throughout my days whenever I'm reading, playing video games, writing, talking, or listening to someone and realize I would really like to know a word. This is much more effective than memorizing words from some list online because it's in the context of what I will use them for.

At the end of the 12 days, I spent day 13 going through and testing every word I memorized, from every technique, both from Spanish to English and English to Spanish. I gave myself a maximum of seven seconds to remember a word; if I couldn't remember it by then, I counted it as 0 "points". Any word remembered with incorrect spelling or wrong accents counts as 0.5 "points."

Results

After testing myself on all the words I had memorized on day 13 here were the results I got.

• Flashcards:
  • Total Proportion Correctly Recalled: 118/144
  • Total words memorized: 59
• Memory Palaces:
  • Total Proportion Correctly Recalled: 49.5/52
  • Total words memorized: ~25
• Flashcards + Memory Palaces:
  • Total Proportion Correctly Recalled: 57.5/60
  • Total words memorized: ~29

We can use a statistical test that compares proportions to analyze the effectiveness of the three memory methods (flashcards, memory palaces, and the combination of flashcards and memory palaces) in your vocabulary learning experiment. I'll admit I used GPT-4 for these calculations, but the statistics were pretty basic, so I think it got them correct. The data provided includes the total number of words successfully recalled for each method out of the total attempted, allowing for a direct recall rate comparison.

Calculation of Recall Rates

• Flashcards Recall Rate: 81.9%
• Memory Palaces Recall Rate: 95.2%
• Flashcards + Memory Palaces Recall Rate: 95.8%

Statistical Test: Chi-Square Test for Independence

Given that we have categorical data (recalled vs. not recalled) across three independent groups, a Chi-Square test is appropriate to determine if there are statistically significant differences in the recall rates among the three methods.

Steps in Chi-Square Test

1. Observed Frequencies: Your data as observed counts.
2. Expected Frequencies: Assuming no difference in effectiveness, expected frequencies are calculated based on the overall recall rate.
3. Chi-Square Statistic: Calculated to see if the observed frequencies significantly deviate from the expected frequencies.

Hypotheses

• Null Hypothesis (H0): There is no difference in the effectiveness of the three memory methods.
• Alternative Hypothesis (H1): At least one method differs significantly in effectiveness from the others.

The Chi-Square test yielded a statistic of approximately 10.95 with a p-value of 0.0042. The p-value of 0.0042 is less than the commonly used significance level of 0.05, indicating that there is a statistically significant difference in the effectiveness of the three memory methods used in the study. The observed recall rates were significantly different from what would be expected if there were no differences among the methods (as shown by the expected frequencies).

Discussion

The results suggest that both the Memory Palaces method and the combined Flashcards + Memory Palaces method are more effective than using Flashcards if your goal is recall rate.

Flashcards recall rate was only 81.9% where as memory palaces were 95.2% and memory palaces + flashcards were 95.8%.

But if your goal is to learn the most words—at the sacrifice of recall rate—it's abundantly clear Flashcards are the winner.

Using Flashcards, I could memorize 59 words compared to ~25 for memory palaces and ~29 words for flashcards + memory palaces.

However, both results disregard the fun factor of using the methods. Personally, I found memory palaces and memory palaces + flashcards to be much more engaging and joy-filled, albeit in a cognitively difficult manner, for me to memorize vocabulary. There's something about simply memorizing vocab with Anki straight that sucks your soul. Creating personalized, memorable images is much more fun in comparison.

So, to answer the question, are flashcards, memory palaces, or flashcards + memory palaces better? ** it's up to you to decide what you find more important**. If you value recall rate, fun, and engagement, perhaps you should choose memory palaces or flashcards + memory palaces. But if you just want to memorize the most words possible—maybe you're cramming for your Spanish exam in a few weeks, you naughty person—perhaps flashcards are best.

Limitations

It's important to note there are some major limitations to this study.

Firstly, I did it on one person—ME! To get more generalizable results, we need to have the study done on more people. I've been using memory palaces for a while. Someone who has never used them might find them more difficult than flashcards. Some people might react differently to different techniques. Who knows?!

Secondly, the study should be done for longer. 12 days isn't a very long amount of time, especially when you consider language learning is a never-ending endeavor. I would like a study that's at least 30 days.

Finally, future studies might explore how different languages and vocabulary affect results. I could see memory palaces being more useful for something like Chinese if you're an American speaker. The words are so different that flashcards might not be enough to get you to memorize them effectively. What if vocabulary was thematic around a certain topic instead of personalized and pretty all over the place like it was for me? Perhaps memory palaces would do better than them as they are particularly valuable for connected information.

The future is rich for the field of language learning. I'm excited to see what technique reigns supreme for others. Or maybe a fourth hidden technique will come to the fore and blow the others out of the water. Adventure!