We are pleased to present the results of a new study recently published in the Annals of the New York Academy of Sciences journal. This scientific paper is the result of the meticulous work of Pascale Tremblay, director of the laboratory; Lydia Gagnon, former research assistant in our laboratory; Edith Durand, speech-language pathologist and professor at the Université du Québec à Trois-Rivières (UQTR); as well as Joël Macoir, professor at the School of Rehabilitation Sciences at Université Laval.

Introduction

Verbal fluency is a task used in speech-language pathology and neuropsychology that involves generating words quickly according to specific rules. These rules can be, among others, phonemic (based on speech sounds) or semantic (based on a category of objects or items). Here are two examples of instructions for performing this type of task:

1. Name as many words as possible that begin with the sound /p/ in 60 seconds (phonological constraint)
2. Name as many animals as possible in 60 seconds (semantic constraint)

If you try this task yourself, you will notice that it can be more difficult than it seems. Indeed, verbal fluency measures several language and cognitive abilities:

a. Lexical access, that is, the ability to retrieve known words stored in our brain;

b. Working memory, which allows us to temporarily hold and manipulate several items in memory;

c. Inhibition, which helps, among other things, avoiding repeating items already produced or producing words that do not follow the instructions;

d. Cognitive flexibility, which enables us to adapt to different rules and constraints within a task.

(To learn more about cognitive functions, see this article: Cognitive functions)

Some people spontaneously use strategies to facilitate word retrieval. For example, in a semantic verbal fluency task, they organize their responses into subcategories (e.g., farm animals, savanna animals). Others use self-directed speech (talking to oneself) to try to generate ideas.

Errors made during this task can be classified into different categories. In our study, three types of errors were measured:

1. Intrusion errors, when a person names a word that does not belong to the requested category;
2. Repetition errors, when a person produces a word already produced;
3. Non-words, that is, the production of nonexistent words.

This complex task is particularly interesting to study because it can measure changes in cognitive and language abilities that occur during normal and pathological aging [1–2]. For instance, a study recently published by Dr. Joël Macoir and Dr. Carol Hudon, professors at Université Laval, showed that people living with mild cognitive impairment produce fewer correct responses in verbal fluency compared to people with good cognitive health [3].