Deriving additional measures of semantic processing in animal fluency

This project will produce new information pertaining to performance on an animal fluency task, in which people must name all the animals they can think of in one minute. This task is used to identify language deficits in people who have developed cognitive problems, for example due to Alzheimer’s disease. We will examine the types of animals that people name and the order in which they are produced. Changes in the way this task is performed may provide important information about a person’s cognitive status.

This project examined performance on an animal fluency task, in which people have one minute to name as many animals as they can think of. These tasks are usually scored in terms of the total number of words produced; however, examining the actual items produced has the potential to provide important information about how people are performing this task.

Using machine learning approaches, we examined the types of animals that people name and the order in which they are produced, how this changes with age, and whether speaking more than one language has an effect on performance. We found that as people get older, they tend to produce fewer words, and those words tend to be more frequently used in everyday language, and closer in meaning to many other items. We found the same pattern in monolinguals compared to bilinguals. These findings indicate subtle differences in fluency performance with age and bilingualism, and may provide important information about a person’s cognitive status, taking into account their age and the languages they speak.