Does the human Brain learn language and math in the same way?

Jamal (Jimmy) Esmaily

Ludwig Maximilian Universität, München, Germany

A brief commentary on Sara Stillesjö, Linnea Karlsson Wirebring, Micael Andersson, Carina Granberg, Johan Lithner, Bert Jonsson, Lars Nyberg, and Carola Wiklund-Hörnqvist (2021). Active math and grammar learning engages overlapping brain networks. PNAS, 118 (46). https://doi.org/10.1073/pnas.2106520118

Math and Language: how similar can they be?

Math and languages! Which one could you learn better? Probably most of you would say ‘I am awesome at both OR I am awful at both!‘. Probably you remember yourself and a former classmate and think: Yes! It seems to be the case! Actually, new research suggests that indeed it probably is the case: most people are either good at both or not so good at both. But why? Could it be that our brain does the same kind of things when doing math as when learning a new language?

Studies of IQ and general intelligence have shown some behavioural correlations between learning math and language. An important question that remains unclear is why this is the case. A very recent study by Stillesjö and colleagues tried to answer this question. To understand that study, we first need to clarify a few things about what do we mean by “learning” languages and math.

Languages are learned in two ways: active and passive learning. Let’s say you want to learn a new word in a foreign language, for example, “Ketab” which is the Persian word for book. Learning passively happens if the word pair [Ketab – book] is presented to you repeatedly, for example if you are listening to Farsi-English podcast while cooking your dinner. Active learning happens when you sit in a Farsi language class and retrieve the meaning of “Ketab” repeatedly, look it up in a dictionary, make sentences with it, and so on. In active learning, you are not just being exposed to the Ketab-Book pairing but in fact use the two words actively. Of course, active and passive learning of a language can happen at the same time, are not confined to vocabulary and can happen for other areas of language such as grammar, spelling etc.

Active and passive learning happen in domains other than language too and math is one such area. You could learn how to solve a math problem via demonstrated, step-by-step formulas of the solution (passive) or by actively trying to come up with the solution yourself (active).

The study I discuss here asked if learning math and grammar would involve similar or different brain activity. To answer this, they conducted two experiment in which they tested high school students:

Experiment 1: vocabulary learning
Experiment 2: math learning

In Experiment 1, before the brain scanning, Swedish participants first learned 60 vocabulary items of a foreign language (Swahili). Half of these items were learned passively: the Swahili-Swedish pair was presented repeatedly to the participants. For the other, active half, experimenters asked participants to complete the Swedish counterpart of a given Swahili word cue test with feedback. One week later, participants went inside an MRI scanner to measure their brain activity. During the scan, students were presented with the same 60 word Swahili words and for each one, they chose the correct Swedish pair from multiple choice options.

Experiment 2 was very similar to the first, but here people solved math problems in active and passive learning styles.

The behavioural results showed that participants performed better for the active learning trials rather than passive trials. Active learning was more effective than passive. This is not new. Many studies have shown a similar outcome before. Doing more effort gives you better results.

But what they really wanted to know was the brain question. They wanted to know:

Which brain areas are more active during active (vs passive) learning?
Do similar brain areas show activity during vocabulary and math learning?

Figure 1: six overlapping areas that would be activated during active learning in math and language. LA: Language Active, LP: Language Passive, MA: Math Active, MP: Math Passive.

The results show that there are at least 6 areas in the brain whose responses increase during active learning BOTH for math AND language. These six areas are illustrated in figure 1A. This shows that there are similarities between what our brain does when we learn a language and when we learn math. Not only are these brain areas similar, the pattern of each area’s activity (Figure 1B) is also similar when learning math and language. The experiment succeeded!

So, back to our first question: Why are people good/bad at math and language at the same time? We now have a very good answer supported by biological evidence: because both engage similar brain areas. In conclusion, if you are good at one of these things (math or language) you could safely assume you will do great in the other one as well!