Psycholinguistics seeks to use psychological techniques to test, measure and investigate linguistic phenomena. How is language represented and processed in the brain? How is language learnt? How far does this depend on structures defined by the human genome – what is known as the nature / nurture issue? How different is production from comprehension? Or are these just different reflections of how the faculty of language is used?

Neurolinguistics, or the investigations of these things at a neurological level, is a significant sub-field.

In the 1930s and 40s, the theory of behaviorism was broadly accepted in psychology in the English-speaking world. The notion of linguistic elements as intrinsically meaningful was roundly dismissed as ‘mentalism’ . At around the same time, various dissatisfactions were emerging in linguistics in relation to the notion of distinct levels, one defining the phoneme as an abstract entity, one defining it exclusively in relation to spoken pronunciation. And the term psycholinguistics was coming into use.

But the new approach to linguistics being developed by Noam Chomsky and a handful of close colleagues implicitly questioned many of the emerging canons of psycholinguistics.

One hugely significant discovery in psycholinguistics was made in 1958 by Jean Berko-Gleason with her ‘Wugs test’. Berko-Gleason discovered that rather than learning the forms of words, comes, goes, came, went, and so on one by one, children learn the overall patterns by which words are formed. Berko-Gleason drew imaginary creatures, one of which she called a ‘wug’, and she devised imaginary activities for it. Taking care to ensure that children all understand what she was expecting of them, she drew them into conversations like “This is a wug. Now there is another one. There are two of them. There are two ___”. Or of a man with a steaming pitcher on his head “This is a man who knows how to spow. He is spowing. He did the same thing yesterday. Yesterday he ___?”

By a supreme intellectual feat, Berko-Gleason devised this clever test before the field which she was implicitly testing, now known as ‘generative phonology’, had emerged.

Not rushing to conclusions

The technicalities of linguistics can seem overwhelming. And the differences between alternative approaches and theories can be confusing. Perhaps reacting to such thoughts, it is sometimes concluded, quite mistakenly in my view, that psycholinguistics gets closer than linguistics to the ultimate empirical reality of language.

Just one of the dangers here arises from the number of factors that are involved in any processing of language. Many tests and assessments of language attempt to remove all such factors. But they do so at the expense of naturalness. And given an unnatural seeming task subjects can reconstrue it in what seem to be natural ways – with unpredictable results. One such factor is plausibility.

One commonly used test of whether children understand passives with familiar, regular verbs involves pictures of a boy or girl being pushed or pulled in a toy truck by another child of the opposite sex with a prompt like “Show me the boy being pushed by the girl.” Such a test is designed to be as neutral as possible. But the child may have his or her own ideas about who should do the pushing in such a situation. And the response may be influenced by prior notions of appropriate gender roles rather than the grammar of the prompt.

In a similarly subtle way, if a tester or experimenter shows a child a bunch of interesting and desirable toy animals, shows the child a bag, saying: “This is my special hiding place”, puts one in the bag,  and asks, “Which one has been hidden?” this is more plausible than asking “Which animals have not been hidden?” in an effort to see whether the child can cross multiply negation and passivisation. A wrong answer could be by misconstruing the plausibility of the question rather than by a failure to understand a negated passive.

In psycholinguistic practice it is very hard to avoid such biases. Seemingly robust conclusions can be rebutted on such grounds.