Promising Dogs and Music Studies

In my webinars on dogs and sound, I discuss the glaring inadequacies of the dogs and music studies. The problems are now being covered in detail by scholars in the field as well (Kriengwatana et al., 2022; Kriengwatana et al., 2025; Lindig et al., 2020; Snowdon, 2021). In my opinion, the most important issue is that the studies designed to assess whether dogs benefit from music have used entire pieces of music before testing dogs’ abilities to perceive the basics. Music is complex. No researcher started with simple melodies; no one separately tested whether dogs can distinguish between consonance and dissonance, which is key to the western classical and pop music they usually test. No one separated out rhythm to see what dogs could discern. No one tested to see if there were certain instruments or vocal sounds the dogs responded to. In other words, no one separated the variables. Instead, they compared dogs’ responses to whole playlists of, for example, “classical” music, pop, rock, or reggae.

The many other issues with the studies can’t be addressed until that one is.

The basics I listed above (and far more) have been tested with other species, but not dogs. I believe people were in a rush to find something that could help dogs in distressing situations, such as in shelters or the vet’s office. But the result of testing whole playlists before exploring dogs’ responses to the building blocks of music is that even if we were to find a piece of music that strongly and positively affected dogs, we wouldn’t know why. We wouldn’t know what aspect of the music was affecting them, because all variables were lumped together.

Although not a test of music, per se, I should mention the excellent work of Dr. Patricia McConnell (1990), who found that four short, rapidly repeating rising notes were more effective at prompting a dog to come when called and increased motor activity levels. This is the type of work we need when investigating dogs’ responses to music.

Two sets of music researchers have now performed some more basic studies. First, I’ll review what we know about dogs’ abilities to discriminate sound from older studies, then I’ll report on the good news.

What Information Do We Already Have?

Here are a few things we know about dogs’ hearing abilities as they apply to music. I cover this in a little more detail in my blog post, “How Does Dogs’ Hearing Compare To Humans’?”

Pitch

There is strong evidence that dogs can discriminate pitch down to tiny differences, much smaller than the Western diatonic music “half step” (Andreyev, 1934; Dworkin, 1935). The half step is generally the limit for testing human pitch discrimination or perfect pitch.

Speed/Rhythm

Andreyev (1934) also tested dogs’ ability to discriminate between different speeds of the ticks of a metronome, and they could detect a difference of 2 beats per minute.

Timbre

There is research evidence that dogs can discriminate differences in timbre. Timbre is defined as:

a sensory attribute of sound that enables one to judge differences between sounds having the same pitch, loudness, and duration (Gelfand 2010, p. 227).

Timbre is the quality that enables you to tell (or be able to learn) the difference between an oboe and a clarinet playing the same sequence of notes.

A study in 1929 showed that dogs could discriminate the difference between the same note played on a tuning fork or a keyboard instrument (Razran & Warden).

The ability to discriminate timbre is present when dogs can tell their family members’ voices apart.

These studies tell us nothing about dogs’ preferences in sound and music. But only by knowing that dogs can discriminate between these aspects of sound can we hypothesize that they might have preferences between them. So they are a start.

The “Modern” Studies on Music for Dogs

These studies started in 2002 with a study by Wells et al. I have a list of 23 of them here. In most of them, playlists or individual pieces in certain genres were played in the presence of dogs and their responses were recorded in various ways.

Most of these studies are the kind I disparaged above because they lumped together way too many variables.

But a study in 2020 and another in 2024 used better methods.

The Happy Exceptions

Here are the researchers who finally broke the trend.

The Amaya et al. study from 2020 played whole pieces of music but separated out variables to change one at a time. The Pinelli et al. study (2024) tested the ability of dogs to discriminate a building block of music: a short sequence of tones.

Amaya et al.

The study, “Effects of music pitch and tempo on the behaviour of kennelled dogs,” separated the variables of pitch and tempo when exposing dogs to playlists of music (Amaya et al., 2020). They used complex pieces of music, which is problematic for the reasons I stated above. But they showed good practice in many ways.

First, rather than choosing a genre of music, which can cover a broad range of tonality, rhythm, instrumentation, and more, they stipulated characteristics of the music that they deemed desirable. They filtered Spotify to find such pieces. They selected pieces with a tempo of 70 or fewer beats per minute, a “positive/euphoric valence” as determined by Spotify’s rating system, and “low energy,” again determined on a numerical scale by Spotify. In addition, they selected pieces by (usually) one instrument, the piano, and selected pieces that all started at approximately the same pitch to be uniform when the pitch was changed.

But the standout characteristic of this study is that they took the same pieces of music and changed only one variable at a time. They played the music/sound for dogs under six conditions: a control (unaltered playlist), white noise with no music, the playlist at both a faster tempo and a slower one, and the playlist with raised pitch and lowered pitch.

This study had surprising results. Only one condition appeared to affect the dogs’ behavior. The lowered pitch condition was accompanied by behavior indicating the dogs were not as relaxed. It was speculated that dogs may associate lower pitches with growling.

Several of the “music for dogs” companies advertise that their dog music is lowered in pitch to promote relaxation. Perhaps now they are scrambling to change their claims.

Another great thing about this study is that the alterations to the music were described well and were replicable. I replicated them myself.

I found one of the pieces from their playlist: Lavender Hills by Brian Crain.

I lowered the pitch using Audacity as they described in the paper and offer this excerpt under fair use for educational purposes. You can compare it to the full original at the link in the previous paragraph. Note the very low frequencies when you listen to the altered version. This is the transformation that may have aroused or bothered the dogs.

Their descriptions of the classical pieces were not as precise as they should have been. They used titles like “etudes” and “piano sonata” without further specifying the composer and other identifying information. That makes replication impossible because there are hundreds of pieces with these names. Perhaps they offered more information on the pieces in supplemental materials, but I didn’t find them.

But this study is superior to the other playlist studies because the researchers used the same pieces of music in different conditions, separating one variable at a time. They described their methods well. And the possibility of lower-pitched music causing stress rather than relaxation to dogs is an important finding.

Pinelli et al.

“In ‘Tone’ with Dogs: Exploring Canine Musicality” is the first research I know of that used a small building block of music in its study of dogs’ perceptions (Pinelli et al., 2024). The goal of the study was to determine whether dogs might possess an ability called relative pitch.

Even though this study is groundbreaking in several ways, there are some frustrating things. The researchers never defined relative pitch (see below for some definitions); they just claimed that if a dog could make certain discriminations, they had the ability. Also, because of certain characteristics of the tone sequences they used, there is another possible explanation for the dogs’ ability to discriminate besides relative pitch. I’ll get to that. Even so, anything we learn about dogs’ basic abilities with sound and music is useful.

You can read a decent definition of relative pitch here, but I’ll elaborate. Most musicians have relative pitch. One way it is tested for musicians is that a note is played, and the person is told the identity of the note, e.g., “the D above middle C, or “D4.” Then, another note is played. A person with relative pitch can tell you the identity of that note because of the audible relationship between the two. If they know the beginning note, they can write in musical notation a sequence of notes you play for them. These ways of testing require the person to understand some music theory.

But people without this music theory knowledge can be tested, as well. One way would be to introduce a sequence of a few notes, a melody. Ask the person to sing or whistle it a few times. Then give them a different starting note, and ask them to sing or whistle the melody starting on that note instead. This is called transposition. A person who can transpose can hear the relationships (intervals) between different tones. People with relative pitch transpose without thinking about it if they start to hum or whistle a tune in a range that is higher or lower than they can execute. “Oops, I can’t reach those high notes, I’d better start lower.” Here is a page where humans can train and test their relative pitch abilities.

The researchers tested the dogs using transposition as well.

First, they started by teaching the dogs to discriminate between two sequences of four notes.

Here are the original sequences. I generated them as sinusoidal waves as described in the study and with the duration of the notes and silence stipulated there.

CAUTION: These could scare some dogs who are triggered by high-pitched digital sounds.

The dogs were trained to go to a bowl on one side of the owner for a treat when hearing one sequence and to a bowl on the other side when hearing the other. Training discriminations is challenging, and only 2 out of the 16 dogs that started the study mastered that exercise and went on to the transposition phase.

Here is a sample transposition of both sequences.

The notes have the same relationship to each other (that’s the “relative” part), but they are different notes. Both dogs could respond to the transposed sequences and go to the correct dish at a percentage better than chance. If the transposed sequences sound awfully high to you—that’s another good move by the researchers. Dogs’ prime hearing range is higher than ours (and even higher than those notes).

The researchers go into some detail regarding the challenges in the study, and it’s an interesting read. Among other things, they detected a Clever Hans effect, so they included a condition where the owners were hidden behind a barrier.

The scientists appear to have missed the implications of one characteristic of the sequences they chose. That is that the four-note sequences either go exclusively up in pitch (each note higher than the last) or exclusively down, the reverse. So, I don’t believe they ruled out the possibility that the dogs were simply detecting the difference between a sequence that rose in pitch or fell. A dog might hear that aspect without detecting the relationships between the notes. That would not be a demonstration of relative pitch. The researchers may have chosen the sequences because they would probably be easier to discriminate than a four-note sequence that didn’t go in the same “direction.” But even that finding would be valuable. And perhaps a future study could explore different sequencs.

Conclusion

I’ve been studying the dogs and music papers since 2012. I’ve branched out into reading the studies about all animals and music and touched on the human music therapy literature as well, since it is highly relevant and often neglected. As I mentioned above, I present in detail in my webinars the flaws in the dogs and music studies, and I’m pretty intense about reaming out some companies. The baseless claims the companies make are simply astonishing.

I’ll be writing more about this, but I decided to start by highlighting these better studies before diving into the problems. So stay “tuned”!

Copyright 2025 Eileen Anderson

References

Amaya, V., Descovich, K., Paterson, M. B., & Phillips, C. J. (2020). Effects of music pitch and tempo on the behaviour of kennelled dogs. Animals, 11(1), 10.

Andreyev, L. A. (1934). Extreme limits of pitch discrimination with higher tones. Journal of Comparative Psychology, 18(3), 315-332. METRONOME & PITCH DISC

Dworkin, S. (1935). Alimentary motor conditioning and pitch discrimination in dogs. American Journal of Physiology-Legacy Content, 112(2), 323-328. PITCH DISC.

Gelfand, S. (2010). Hearing: An introduction to psychological and physiological acoustics. Informa Healthcare.

Lindig, A. M., McGreevy, P. D., & Crean, A. J. (2020). Musical dogs: A review of the influence of auditory enrichment on canine health and behavior. Animals, 10(1), 127.

Kriengwatana, B. P., Mott, R., & ten Cate, C. (2022). Music for animal welfare: A critical review & conceptual framework. Applied Animal Behaviour Science, 105641.

Kriengwatana, B. P., Nager, R. G., South, A., Ullrich, M., & Doolittle, E. L. (2025). Playing music to animals: An interdisciplinary approach to improving our understanding of animals’ responses to music. Animal Behaviour, 221, 123074.

McConnell, P. B. (1990). Acoustic structure and receiver response in domestic dogs, Canis familiaris. Animal Behaviour, 39(5), 897-904.

Pinelli, C., Scandurra, A., Giacoma, C., Di Lucrezia, A., & D’Aniello, B. (2024). In “Tone” with dogs: exploring canine musicality. Animal Cognition, 27(1), 38.

Razran, H. S., & Warden, C. J. (1929). The sensory capacities of the dog as studied by the conditioned reflex method (Russian schools). Psychological Bulletin, 26(4), 202.

Snowdon, C. T. (2021). Animal signals, music and emotional well-being. Animals, 11(9), 2670.