Infants struggle with unfamiliar male voices

Researchers examine whether 4.5-month-old infants can identify unfamiliar male speakers.


A new study by Madeleine Yu from the University of Toronto Mississauga shows that babies have more trouble telling apart the voices of unfamiliar men than unfamiliar women.

Madeleine Yu, a University of Toronto Mississauga student, did not expect these results. She works in a Child Language and Speech Studies (CLASS) lab. The study was published in a journal called JASA Express Letters.

Yu said, “It seems like there is something going on with this difference in how infants respond to unfamiliar male voices.

Yu worked with Elizabeth Johnson, who leads the CLASS lab. Johnson is a professor at the University of Toronto Mississauga. They also collaborated with Natalie Fecher, who previously studied how well babies can recognize unfamiliar female voices in 2019. In this new study, Yu repeated Fecher’s procedures but used male speakers instead to see if the babies’ responses were different depending on the speaker’s gender.

The study included 48 babies aged four to five months from the Greater Toronto Area who only spoke English. Most of the babies (46 out of 48) were primarily cared for by their mothers.

Each babysat on their caregiver’s lap in a room designed to block out outside noise. They listened to recordings of 40 simple sentences spoken by pairs of adult men. These men were all from Canada, spoke English as their first language, had similar-sounding voices, and spoke in a regular tone.

When a recording played, a colorful flashing pattern appeared on a screen with a blinking red star in the middle. If the baby paid attention to the sound and looked at the screen, the researchers measured how long they looked at it.

First, the babies listened repeatedly to recordings from one of the men to get familiar with his voice. Then, the researchers tested the babies in four rounds: two rounds with the same voice they had heard before and two rounds with a new voice they had not heard yet.

The researchers discovered that the babies looked at the screen simultaneously in both trials. This differs from Johnson and Fecher’s study in 2019, where the babies looked much longer during the same-voice trial than in the different-voice trial.

This led the researchers to think that the babies could have been better at noticing when the speaker changed with the male voices compared to the female voices.

Yu said, “It is possible that the infants’ greater exposure to females [as their primary caregivers] shaped their perception to tell apart better female versus male voices.”

Johnson suggests another possible explanation, which could be related to evolution. She said, “We know babies’ hearing develops before their vision, and they must recognize their caregiver for survival. Throughout history, women have often been the primary caregivers.”

Yu and Johnson faced challenges when studying such young babies. They had to plan their experiments around the infants’ unpredictable sleep, feeding, and moods. However, Yu believes their hard work is worthwhile. This study addresses a significant gap in research on how babies recognize voices, which has primarily focused on female voices. It offers valuable insights into how infants perceive the world.

“I find it interesting that by observing how long babies look at things, Yu said, we can understand how they perceive things at just four months old.”

This study adds to Yu’s previous research on speech processing and recognizing different speakers. She studied cognitive and behavioral science at the University of California, San Diego, for her bachelor’s degree and psychology at the University of Toronto for her master’s degree.

Currently, she is investigating how we recognize voices with familiar and unfamiliar accents, which is the main focus of her doctoral dissertation.

Journal reference:

  1. Madeleine E. Yu, Natalie Fecher et al., Learning to identify talkers: Do 4.5-month-old infants distinguish between unfamiliar males? JASA Express Letters. DOI: 10.1121/10.0024271.