How Well Do Dogs and Other Animals Hear?

Reporting the frequency range for hearing in dogs and other species is not a straightforward task - the "how" of determining hearing frequency ranges must first be explained. Testing in animals differs from the method commonly used with humans of voluntarily reporting if a sound is heard. When determining the frequency range in animals, an investigator usually must first train the animal to respond to a presented sound stimulus by selecting between two actions using rewards. Often this response is to try to drink or eat from one of two dispensers when a sound is heard. The sounds are randomly presented from one side or the other, and the subject must select the right dispenser (on the same side as the stimulus) to get the reward; otherwise no food or drink is dispensed. This is done with the animal hungry or thirsty to motivate responding. Stimuli are different pure tones at varied frequencies (units of Hertz [Hz] - or kilohertz [kHz]) and at different loudness intensities (units of decibels [dB] - a logarithmic measure). The investigator then plots the responses on an audiogram, a graph of the softest intensity at which the subject was able to detect a specific. The plot of responses is a bowl-shaped curve, steeper on the high frequency end. A series of five typical audiograms for different dogs (Canis canis) is shown in the figure below.
(right click image to see it more clearly)
These audiograms are from a book compiling thousands of published references into a single difficult to find source (Fay, 1988). This particular audiogram compiles data on the dog from two published sources: one reporting an average from 11 dogs of unspecified breeds (Lipman & Grassi, 1942) and one reporting results from single dogs of four breeds (Heffner, 1983). Frequency is displayed on a logarithmic scale from 10 Hz to 100,000 Hz (100 kHz), while stimulus intensity is displayed (in dB sound pressure level) from -30 to 80 dB. Curve 1 was from the Lipman study, while curve 2 (Poodle), curve 3 (Dachshund), curve 4 (Saint Bernard) and curve 5 (Chihuahua) were from the Heffner study. In general, dogs had slightly greater sound sensitivity (detected lower intensity sounds) than humans, and cats had greater sensitivity than dogs, indicated by how low on the y-axis points were located.

It can be seen that the lowest intensity detected differs between the two studies; I place greater reliance on the Heffner study because it is more current and because he is a widely published and respected audiology researcher. It can also be seen that the greatest sensitivity (i.e. the frequencies that can be detected at the lowest intensities) is in the frequency range of 4-10 kHz. One dog (the Poodle) heard a tone at the low frequency of 40 Hz, but an intensity of 59 dB was required for it to be detected; most of the other dogs didn't respond until the stimulus frequency reached 62.5 Hz. Three dogs (the Poodle, Saint Bernard, and Chihuahua) heard a tone at the highest frequency of 46 kHz, requiring intensities of 64-73 dB. On the other hand, the Poodle heard a 4 kHz tone when it was -4 dB (since dB are logarithmic units based on a ratio of the stimulus intensity compared to a standard intensity, any stimulus smaller than the standard results in a ratio less than one, and the logarithm of a number smaller than one is a negative number; therefore a -4 dB stimulus intensity is a VERY soft one!) and an 8 kHz tone when it was -3.5 dB. There was no systemic relation seen among the four breeds between high frequency hearing sensitivity and head size, body weight, or tympanic membrane area.

From the figure it can be seen that choosing the frequencies for reporting the frequency range for dogs is hard - presumably lower frequencies could have been detected if a loud enough stimulus was used, and likewise for high frequencies. Nevertheless, the following table reports the approximate hearing range for different species with an attempt to apply the same cut-off criteria to all, using data from Fay (1988) and Warfield (1973). Since different experimental methods were used in these different studies, too much value should not be placed on comparing species.

Species Approximate Range (Hz) 
human 64-23,000
dog 67-45,000 
cat 45-64,000 
cow 23-35,000 
horse 55-33,500 
sheep 100-30,000 
rabbit 360-42,000 
rat 200-76,000 
mouse 1,000-91,000 
gerbil 100-60,000 
guinea pig 54-50,000 
hedgehog 250-45,000 
raccoon 100-40,000
ferret 16-44,000 
opossum 500-64,000 
chinchilla 90-22,800 
bat 2,000-110,000 
beluga whale 1,000-123,000
elephant 16-12,000 
porpoise 75-150,000
goldfish 20-3,000 
catfish  50-4,000 
tuna  50-1,100 
bullfrog  100-3,000 
tree frog 50-4,000 
canary  250-8,000 
parakeet  200-8,500 
cockatiel 250-8,000 
owl  200-12,000 
chicken  125-2,000 


  • RR Fay. 1988. Hearing in Vertebrates: a Psychophysics Databook. Hill-Fay Associates, Winnetka IL.
  • D Warfield. 1973. The study of hearing in animals. In: W Gay, ed., Methods of Animal Experimentation, IV. Academic Press, London, pp 43-143.
  • RR Fay & AN Popper, eds. 1994. Comparative Hearing: Mammals. Springer Handbook of Auditory Research Series. Springer-Verlag, NY.
  • CD West. 1985. The relationship of the spiral turns of the cochela and the length of the basilar membrane to the range of audible frequencies in ground dwelling mammals. Journal of the Acoustic Society of America 77:1091-1101.
  • EA Lipman & JR Grassi. 1942. Comparative auditory sensitivity of man and dog. Amer J Psychol 55:84-89.
  • HE Heffner. 1983. Hearing in large and small dogs: Absolute thresholds and size of the tympanic membrane. Behav Neurosci 97:310-318.
  • Return to Deafness in Dogs and Cats

    June 3, 2003