Language Interference and Cognition
At the end of the last post, I stated that linguistic interference was often used as an argument against the interaction of language and thought, but that Lupyan turns this around and uses it as support for this very theory. Let us take a look at how this works.
This is the second post in a three-part series on Gary Lupyan’s label-feedback hypothesis. Before getting into the interesting psychology of the label-feedback hypothesis, I would like to make sure that you have read my previous post, Is linguistic information part of every cognitive process?, which lays out the basics of this interesting theory.
First, a quick overview of linguistic interference. When experiment subjects are taking part in a task, they are generally free to use any sort of cognitive strategy they want. If they are doing a memory task, they can repeat the name of one of the items over and over to help them remember it, for example. If they are classifying images, they can give them labels — they can think of one as “the ladder one” and another as “the shoe one.”
Linguistic interference seeks to disrupt these strategies by requiring the participant to engage in a linguistic task throughout the experiment, essentially “using up” their available linguistic resources so they cannot be used during the task. An example of this is repeating nonsense syllables over and over. The idea is that if you have to say “la lo li do ba da na lu na” repeatedly throughout a task, you will not be able to apply linguistic labels to any parts of the task.
Previous research has repeatedly used this fact to look into the relationship between language and thought. For example, the words that English speakers and Greek speakers use to describe motion are different (the exact grammatical difference is a bit complicated; if you arre interested, leave a comment and I can explain it). One study applied both linguistic and non-linguistic interference in an effort to see how it would affect participants. They concluded, based on their results, that linguistic strategies were only used in high-cognitive-load situations, and that it was a transient phenomenon. They argued that this provided evidence for the separation of language and thought.
Lupyan, however, sees it differently. In his hypothesis, language is always involved in categorizing unless it is disrupted, which means that, in everyday situations, language is affecting our “non-linguistic” cognition. It is only in cases of aphasia or interference that it goes away. Essentially, he is coming at the problem of language and thought from the other direction. Researchers who take the interpretation above tend to think of adding language to the non-linguistic process of categorization as the exception to the rule, while Lupyan considers the subtraction of language as the exception.
The label-feedback hypothesis essentially does away with the idea of a separation between linguistic and non-linguistic thought, which will likely make a number of cognitive scientists and psycholinguists quite uncomfortable. It looks to me like he might be onto something, though: The reconceptualization of linguistic interference and the reinterpretation of its effects are certainly appealing, and the highly interactive cognitive system that he proposes does seem to agree with some newer models of both neural computation and distributed cognitive processing.
What do you think? Is there a difference between linguistic and non-linguistic thought? Do they interact?
Check back soon for the final post in this series, where I will discuss some of the cognitive implications of Lupyan’s theory.
Lupyan G (2012). Linguistically modulated perception and cognition: the label-feedback hypothesis. Frontiers in psychology, 3 PMID: 22408629
Trueswell, J. C., & Papafragou, A. (2010). Perceiving and remembering events cross-linguistically: Evidence from dual-task paradigms. Journal of Memory and Language, 63(1), 64–82. doi: 10.1016/j.jml.2010.02.006