Do You Really Hear What You Think You Hear?

Hearing and Cognitive Penetration

Posted Feb 12, 2015

Wikipedia Commons
Source: Wikipedia Commons

It is no secret that most research on perception is on vision. This phenomenon applies across the fields from neuroscience to philosophy.

One of the issues that has been a controversial topic in philosophy and psychology is that of whether our understanding, beliefs and knowledge can alter visual perception—particularly visual perception of low-level features, such as color and shape.

This phenomenon, which is sometimes thought of as theory-laden perception, is also known as cognitive penetration. One of my collaborators and I have recently argued that there is no good evidence showing that color experience is cognitively penetrated.

The question remains, however, whether auditory perception, or hearing, is cognitively penetrated. Do the things we know about languages affect what we hear?

There is no doubt that it feels differently to listen to a language you know and a language you don't know. But can that kind of knowledge influence the sounds we hear?

When we learn a language, we become better at detecting the phonemes of the language. This suggests that simply learning a language that contains different phonemes might alter what we hear at the sensory level.

But what if we set those changes aside? Is there still a case to be made for hearing being cognitively penetrated? Two cases seem to provide the strongest evidence for this hypothesis. One is that of sine wave speech. Watch the following demonstration.

Basically by being told what we hear, we are suddenly able to identify spoken words embedded in what earlier sounded like gibberish.

Another case in point is the little known language Food Tongue. It's a language that was originally made by people in a math camp. I have written a couple of blog posts about it here and here.

Basically, the constituents of Food Tongue are food words in English, such as "cherrypie" and "kiwi." But the food words don't retain their ordinary meaning, and there are grammatical rules for how you can combine them.

I once made an attempt at learning Food tongue, and I must admit that the spoken version of the language started to sound different. The very sounds, for example the sounds of "apple" or "kiwi" suddenly began to have a different "ring" to them.

But in these cases the phonemes are exactly the same as they are in English. So, it cannot be the detection of new sounds that contributes to the change. That would seem to suggest that what we hear is affected by what we believe, understand and know, even when our ability to detect sounds doesn't change.