Voice and hearing are the most common things people attribute to their poor singing ability. They may deem themselves “tone deaf,” or have “no voice.” However, a current study at the University of Buffalo suggests that singing ability may not be a reflection on the voice or ears, but on the brain.

UB’s Auditory Perception and Action Lab directed by Dr. Peter Pfordresher is looking at the how people match pitches and melodies with their voice.

“[Pitch matching,] seems like the aspect of singing that is most problematic for a lot people. Also we think that it’s a very interesting thing to understand because the manipulation of pitches with your voice is something that is really hard to get a handle on. You’re using muscles that you cannot see, and muscles that don’t give a lot of feed back to your brain and yet it’s something that we all learn early on in life, learning how to talk, learning how to sing, and some people end up getting much better at it than other people do,” said Pfordresher.

The research evidence suggests that bad pitch is not a problem with hearing or muscle movement (vocal chord), but in the way that people process the tones that they both hear and sing. The study has come to this theory by examining someone’s “pitch discrimination ability.” “This is whether two pitches are the same or different. It’s an easy task but some people really struggle with it. Often labeled as “tone deaf,” it would make sense that someone with this problem wouldn’t be able to sing the pitches well,” said Pfordresher. “As it turns out there are two problems with that. First off, there are a whole lot of poor singers that do fine with the pitch discrimination task, so we find there is not a significant relationship with your ability to discriminate pitches and your ability to match pitches. On the flip side there’s a small segment of the population that can sing accurately, but are not able to discriminate pitches. So they can match pitches with their voice but when they’re told to label the changes are unable to accurately do so.”

Other findings by the Auditory Perception and Action Lab lead researches to believe it’s more than just having a poor voice as well. “The other side of it is the motor control issue. This is not surprising, but poor singers not only sing the wrong pitches, but the range of pitches they produce is much smaller,” said Pfordresher. “When you look at data, an intuitive conclusion would be the bad singer simply cannot move their vocal chords as well. In our experiments we give people tasks where they have to move their voice up and down in pitch. We have them go really low, really high, go up and down as fast as they can, as it turns out that task is also unrelated to the ability to match pitches. After looking at perception alone and the vocal action tasks alone, what we’re left with is this notation that perhaps one’s ability to accurately match pitches is something with their ability to link perception with action. One fruitful avenue we think has to do with the role of mental imagery.”

Mental imagery is a common process for our brains to go through before doing many tasks. When playing tennis one might self-generate the image of serving the ball before doing so, the same thing happens during a dance routine or playing an instrument, your mind pictures the motions before you physically do them. These are all motor skill tasks, but the same happens within our auditory skills when imagining a song in our head. “Using ‘Jingle Bells’ as an example, we ask someone is the word “snow” associated with a higher pitch or a lower pitch than “sleigh.” Most people will sing the lyrics in their head to find out,” said Pfordresher. “If you look at brain activity while they’re doing that, there is activity not just in areas related to hearing, but in areas associated with action planning. That suggests that auditory imagery has a motor planning component to it. For the less accurate singers there might be less of an activation in the motor areas of the brain when they imagine melodies.

While Pfordresher and his colleagues are currently looking for funding to gain more evidence, the Auditory Perception and Action Lab has been performing tests on auditory imagery that back up that theory on linking perception to action. “We give a self report measure designed by my collaborator, Andrea Halpern at Bucknell University. This asks people very basic questions like: ‘Think about a trumpet playing the national anthem, how vivid is the sound? Was it like you were hearing it right next to you or was it hard to hear it at all?’ As it turns out people who are more accurate at singing, rate themselves having more vivid auditory imaginations. That is a correlational finding that we think provides a clue into what’s going on in poor singers. They may have less of an ability to generate a vivid mental image of what they’re about to do. This is associated with difficulty linking hearing something to how they would reproduce it,” said Pfordresher.

The majority of the study’s experimentation is done at the Auditory Perception and Action Lab. Though some subjects are experienced musicians, most are musically untrained singers enrolled in UB’s psychology program. Participants are brought into a recording studio environment; melodies are played for subjects who are then asked to sing those melodies back into a high quality microphone. Lab attendants digitally record the singing. The acoustic signal of the recording is then analyzed to determine if they are in fact a strong or poor singer.

While the majority of test subjects are college students, with the help from a collaborator at Northwestern University’s music education department, the study has been able to expand to other age groups as well. “In a recent study with a collaborator of mine from music education named Steven Demorest, we’ve done some similar tasks with kindergarteners and sixth graders. Those have been done in schools, recording students in quiet environments,” said Pfordresher.

The findings from the preliminary testing do not exactly relate to the theory on pitch matching but add another wrinkle in how we perceive singing ability. “What the early findings suggest is that pitch matching and singing is not something that comes with maturity alone, it’s more of a use it or lose it situation,” said Pfordresher. “When you look at the three groups that we’ve looked at, you go from kindergartners who are generally poor at singing, to sixth graders where the singing provenience improves greatly. The unfortunate part is when you go from sixth graders to college students; the college student’s singing ability goes down again, dropping to roughly midway between the sixth grader and the kindergartener. It makes sort of a U-shape to this trend rather than a consistent improvement with age.”

The thought behind this decrease after middle school relates to the American school system. In kindergarten and younger elementary school grades, kids are singing in class often, but sixth grade is typically the last year that music is part of general education. “At that point kids are starting to self identify as ‘what kind of group am I going to belong to?’ Some kids stick with music and some kids go on in other directions, and at the same time that happens, voices change,” said Pfordresher. “You have this complicated system of voice and pitch matching that you’ve been using and then suddenly it changes on you, and at the same time it changes you are no longing getting experience using it in the same way as part of general education.”

While the early findings strongly support the theories that singing ability is a “use it or lose it tool” and that problems with matching pitches comes from problems linking perception to action, both studies are in early phases. Pfordresher and his collaborators are in the process of creating a online test taking tool that could potentially greatly expand their subject pool.

“We are in the final stages of an online screening message called the ‘The Seattle Singing Accuracy Protocol.’ It was designed at a conference in Seattle—thus the name. This will be something people can use on their computers as it leads you through a series of tasks similar to the tests we do on campus. There will be some simple vocalization tasks, some pitch discrimination tasks, singing familiar tunes from memory, and matching the pitches in melodies that you hear,” said Pfordresher. “Our hope is that we will be able to look at a very board range of ages and backgrounds, so we can really understand how vocal imitation and singing ability changes across a lifespan, and as a function of experience too.” The team is currently working on the final pieces of the website and hopes it will be live soon. Those interested in participating in the online study can send an email to singingaccuracy@gmail.com and will receive an update when the website is up and running.