Monkeys and Nambees

A Study on Two English Dialects of Child- and Adult-Directed Speech

One of the most interesting aspects of studying child language acquisition is examining the differences between how adults speak to other adults versus how they speak to children. With babies and young children, many of us may use a high-pitched, softer, more sing-song-like voice, with special words like “bye-bye” and “doggie”, without even thinking about it. This manner of speaking is called child-directed speech, or CDS (also known as infant-directed speech (IDS), motherese, or parentese). Although it is not used in all cultures, CDS is helpful for children learning language thanks to its distinct intonational properties, and affects their development positively: for example, the use of repetition in CDS helps predict their rate of vocabulary growth in the future. 

Two of our requirements for our participants are that the mother must have grown up speaking Canadian English, and that her infant must be between 18- and 24-months-old. Before coming into the lab with her infant, we offer the mother the opportunity to complete a Communicative Development Inventory (CDI). The CDI is a checklist composed of words that most infants may have been exposed to. In our study, mothers were asked to choose an animal (e.g., monkey) that the infant would be familiar with. Once the mother chose 3 out of 6 animals (butterfly, tiger, monkey, chicken, elephant, turtle) then she sends the responses back to the lab. Note that data sent to the lab are then stored and secured. Based on the infant’s knowledge, we choose stories and ‘fact’ sheets created for the study which feature real animals and imaginary animals which correspond to each real animal: for example, a “nambee” is very similar to a monkey in terms of appearance and behavior. 

Before beginning, the mother’s are informed about the scope of the study and what they will be doing as participants. They provide their consent to allow their data to be used for the purposes of this study. They can also consent to their data being added to a database for other future studies, which is optional. 

The mother is given the stories about the animals, based on which animal words their infant is familiar with, along with the corresponding imaginary animal stories. The mother is then guided to a sound-proof booth and connected with a recording device so that we can collect their readings. The study then begins one of two ways. The mother will read a set of the stories by herself in the booth, with the directions to read the stories as if she is reading them to another adult. During this time their child will be supervised by one of the researchers in the waiting room, with many toys and games to entertain them while their mother is in the booth. 

Afterwards, the infant will be brought into the booth for the second reading. This time, with the infant in the booth, the mother will be directed to read the stories as if she is reading them to her child. These two trials are counterbalanced across sessions, meaning that sometimes the mother reading to her child will be done first and the reading to an adult done second, or vice versa. 

Once this study is completed, the data collected from the mothers’ recordings will be anonymized and run through software called Praat that allows us to analyze the acoustic properties of the Mothers’ speech segments. When analyzing this data, we are focusing on the formants of stressed vowels in the target words, which are the animal names in the stories. These differences in stress are compared between the infant-directed storytelling and the adult-directed storytelling. Vowels are spoken across a continuum or gradient, meaning that some vowels are exaggerated or hyper-articulated more than others, especially in IDS, as a means of conveying important linguistic information to infants. By analyzing the vowels in the mothers’ speech, we can learn information about how pitch, vowel variability, and other prosodic characteristics are used in IDS to promote infant learning of speech.

The goal of this study is to investigate the use of specific acoustic cues and prosodic characteristics in the mothers’ voices when they are reading to adults versus when reading to children to observe the differences between IDS and ADS and how this differs across dialects of the same language; in this case, English. A previous study was conducted in the UK examining this same hypothesis. The data collected in this Canadian study will be compared to the UK study’s data to examine differences in ID and AD speech across different dialects of English.

Brooklyn Mackown & Catriona Carmichael-Hauer

July 2023 

Linguistics Students at the University of Alberta