We MUST do Speech Perception Testing

Jane Madell
June 23, 2020

I just read an article about using speech perception testing to validate hearing aid outcomes for adults. The article has a large box on the front page which says that it is not customary to do speech perception testing unless a patient is struggling enough to be considered for a cochlear implant. The author, Dr. Anna L. Mc Craney, is suggesting that we rethink this viewpoint.

I am a pediatric audiologist so my viewpoint may be different than that of audiologists who primarily work with adults. I have always had to be certain that we really tested function completely, because how a child heard was going to result in how they learn to listen and talk, and eventually, read. So in the clinics in which I was the director, we did extensive functional testing of every child, and also, of every adult. It was just the clinic protocol.

Does it take time? Of course!! Is it worth the time? Absolutely!!

I am not going to talk about what happens with adults. That is someone else’s bailiwick, but I want to discuss what MUST happen with children.


Why do we test speech perception?


If we do not know how a child is hearing with their technology (and without technology for a child with an auditory disorder other than hearing loss, such as auditory processing disorder), we are being derelict in our responsibilities. Period. You cannot assume. Speech perception testing tells us what auditory information is available to the child, what they are hearing, and how much auditory information is available for them to use for language learning, academic learning, and literacy.

Most speech perception tests require that the child (or adult) repeat the test word or sentence. These tests require that a child have the vocabulary to understand the words. Some children, especially those who are in auditory based, listening and spoken language, therapy programs will repeat what they hear, even if it is not a word.

While most speech perception testing requires that a child hear and understand words, some do not. Some use nonsense syllables, and for very little kids, there we can test detection of various phonemes allowing us to know if a child hears the sound.


Testing speech perception


The first task is to select the appropriate test. We should not be using a pre-school test on a child who is in 3rd grade. It will not tell us how this child will manage in the 3rd grade classroom. (This blog is not the place to discuss speech perception tests in detail but for more comprehensive information see Madell and Schafer, Speech Perception Testing in Children in Madell, Flexer, Wolfe and Schafer; 2019, Pediatric Audiology published by Thieme Medical.)

Once the appropriate test is selected, we need to test in all the necessary conditions. We need to test at normal conversational levels (50 dBHL), and soft conversational levels (35 dBHL). Most of what kids learn they learn by overhearing.

If a child cannot hear soft speech they will not be able to overhear resulting in no incidental learning. We also need to test in competing noise since, unfortunately, the world is noisy and we have to listen and learn in noise. We need to know what the child is hearing in each ear and with both ears together.

The table below is what testing we need to do (note grayed out boxes represent tests to conduct if time allows):

We expect testing in both ears to be better than each ear alone. If not, we need to know why. If one ear is significantly poorer than the other ear, we need to know why. The first thing we need to know is whether technology is doing what it needs to do. If not, we need to adjust or change the technology. If technology is where it should be, then we need to work on listening with the poorer ear alone.


Using speech perception information


When we score testing we need to look at the errors a child is making, not just the total score.
The specific phoneme errors will give us a clue about what should need to be adjusted in the technology. If a child is confusing voiced and non-voiced phonemes, they have either too much or too little information between 200-300 Hz.

If they are not using plurals, and missing /s/, they likely do not have sufficient auditory access in the area of 4000-6000 Hz. Use this information to modify technology settings and improve auditory access.

  • If a child has excellent speech perception (90-100%) on an appropriate language level test, we assume that the child will hear will in the classroom and be able to follow classroom activity (with the use of a remote microphone.)
  • If a child has good speech perception (80-89%) we expect that the child will hear most of what is happening in the classroom and with a RM and other assistance will do well.
  • If a child has fair speech perception (70-79%), she is going to be struggling to follow classroom activity.
  • A child with speech perception that is poor (below 70%) will struggle even more and will need more assistance.


Just do it!!!


I know it takes time, but the information gained is amazing important in improving performance. For a child it is essential. It will determine how much language they learn and what opportunities they will have in life.

If all necessary testing cannot be accomplished in one appointment, just schedule two. It can be done. (I would argue that speech perception testing is just as important in adults. They are not learning language but if we tested them, and they were not hearing optimally, settings could be changed. Maybe we could have more satisfied hearing aid users. But this is not a topic for discussion in the Hearing and kids blog.)

When it comes to kids, there are no good reasons not to do test speech perception. Just do it!!!


Leave a Reply