Jun 07 2011

Language and Thinking

It has been known for many years that language has a profound effect on the way we think. Words and syntax not only give form to our thoughts – they constrain and influence them. Words are anchors for ideas, and as we expand our vocabulary, we expand our intellectual repertoire.

The extent of this effect, however, is still a matter of debate and research. Does the syntax of language, for example, influence our thinking in other domains, such as math, music, or visuo-spatial reasoning? A new study suggests that various types of abstraction are linked in the brain, perhaps they are even variations on a theme.

The study in question looked at a process called priming. In psychological studies, if subjects are made to think in a certain way, that will prime them to think in a similar way on a later task. With respect to language, if subjects use a certain syntax, they are more likely to use it again when asked to construct a new sentence. This is not that surprising – memories of our recent behavior influence (or prime) our current behavior.

Researchers used this priming effect to ask a question about the relationship between math and language – can doing a certain type of math problem prime a sentence construction? If so that implies a certain abstract connection between math and language. Perhaps the “language of math” really is akin to what we normally think of as language.

The specific effect they looked at was “attachment,” – the relationship of a word or term to previous parts of a sentence or equation. In math, low attachment refers to an operation that affects only the most recent number, while high attachment affects a larger portion of the previous equation. The example given is as follows: 80 – (5+15) / 5 is a high attachment equation, because the “/5” refers to a large portion of the prior equation – “(5+15).” While 80-5+15/5 is a low attachment equation because the “/5” only refers to the 15.

With language a high attachment phrase will refer to a large portion of the sentence preceding it. For example “The bells of the church rang loudly,” is high attachment, because “rang loudly” refers to “the bells of the church.” Meanwhile, “The bells of the church on the hill” is low attachment because “on the hill” refers only to “church.”

The question is – are high and low attachment in mathematical equations analogous in the brain to high and low attachment in language? If so, that implies a connection. The researchers gave subjects equations to solve with either high or low attachment. They then gave them a partial sentence (The bells of the church) and asked them to complete it. They found that low attachment equations primed subjects to prefer low attachment sentence completion, while high attachment equations primed subjects to prefer high attachment sentence completion.

As always, it is difficult to draw firm and wide ranging conclusions from any single study, especially in the complex area of human psychology. But there does appear to be an effect here, and it suggests a mental connection between the syntax of math and the syntax of language.  This supports the notion of a far ranging effect of language on how we think. This is definitely the direction in which the research is heading.

An interesting next step would be to find the neurological correlates of this psychological effect – perhaps researchers can use fMRI to see what’s happening inside the brain during priming tasks for math and language. Do the same parts of the brain light up, or different parts? How much of the language cortex is engaged when solving math equations, and does high and low attachment in math have similar brain correlates to high and low attachment in sentence structure? Psychological effects, like priming, are very interesting, but far more interesting (to me) is what they tell us about how the brain is organized and how it functions.

36 responses so far