Effect of Ultrasonic Devices on the Distraction Behaviour of Guide Dogs

There are numerous electronic devices that emit ultrasound that are designed to repel household and garden pests and to discourage gatherings of young people. Some of these operate at frequencies within the auditory range of the dog, reported to be between 67.0 Hz and 44.0 kHz (Heffner, 1998). Concerns have been expressed that these devices could influence the behaviour of guide dogs resulting in safety implications for the guide dog owner or at least a reluctance of dogs to work within certain areas.

In order to investigate this, two ultrasonic devices were tested, namely a garden pest-repeller “Pest-Stop Outdoor” (Procter Brothers Ltd., n.d.) and a young person deterrent “Mosquito™ Teenage Deterrent” (Compound Security Systems, n.d.). The Pest-Stop Outdoor has a motion sensor that detects movement up to 21.3 m from the unit and is designed to scare animals from a garden (Procter Brothers Ltd., n.d.). For the large animal (dog) setting the device emits a continuous burst of ultrasound, measured within our laboratory environment to be 17.5 kHz and 63 dB at 1.0 m distance from the device. The Mosquito™ has an effective range of 15 to 20 m (Compound Security Systems, n.d.) and can be set to emit a continuous sound, measured within our laboratory environment to be 16.8 kHz and either 74 dB or 88 dB at 1.0 m from the device.

The devices were tested under controlled conditions in an outdoor, purpose-built environment used when assessing canine behaviour. The test areas contained naturally occurring flora, such man-made structures as a fenced area, and were next to a road. Therefore, dogs were exposed to some road noise. Excluding road noise and sounds made by the devices, there were no other unnatural sound distractions. The test areas were unfamiliar to the dogs and there were no other dogs or humans in the area aside from the handler, the observer, and the study dog.

The Labrador, Golden Retriever, and German Shepherd dogs belonging to the Guide Dogs for the Blind Association used in the study were all accommodated in kennels at the Guide Dogs Breeding Centre and included bitches in season, entire dogs, and spayed bitches. The dogs were split into groups in random order. Dogs were collected from their kennels and walked to the test area on a half check collar and leather lead. The test area was a two minute walk away from the kennel block and the route taken was the same for each dog. Twenty-nine dogs (nine male, 20 female, age range 2.5 to 12.0 years, mean 4.3 years) were used in a cross-over study to investigate the pest-repeller. The dogs were allocated to five groups and dogs in each group were individually exposed to the device twice over a 2.5 day period, once when the device was emitting sound (treatment) and once when the device was off (control) (Table 1).

Table 1.

Dogs were walked at a steady pace by an experienced handler from a consistent starting point, through an 8.4 m x 2.0 m hard surface walkway bordered on both sides by raised banks covered in natural flora. The device was situated on the right hand side at the end of the walkway and mounted on a pole that was placed in the ground as per the manufacturer’s instructions. The device was set at the large animal setting, which was that most audible to the human ear of the handler and observer. For treatment days, the passive infra-red sensor on the device detected the dog’s approach once the dog had walked approximately one metre along the walkway toward the device, which caused the device to emit a continuous sound. The dog’s distraction behaviour was scored by an independent trained observer using Guide Dogs criteria for assessing canine behavioural response to stimuli (Table 2).

Table 2.

Twenty-four dogs (10 male, 14 female, age range 1.0 to 8.0 years, mean 4.1 years) were used in a cross-over study to investigate the person deterrent. The 24 dogs were allocated to three groups and dogs in each group were individually exposed to the device three times over a 1.5 day period; when the device was off (control), emitting at 74 dB (setting 1) and emitting at 88 dB (setting 2) (Table 3).

Table 3.

The device was fixed at 1.68 m height and angled five degrees downwards from the horizontal within a semi-closed area constructed of fence panels at right angles to each other. The handler and dog walked to the area and stood 2.0 m perpendicular from the device and remained at that point for two minutes. During the two minutes, the dog’s distraction behaviour was scored by an independent trained observer using Guide Dogs criteria for assessing canine behavioural response to stimuli (Table 4). The handler stopped and stood with their back to the device at the same point in the test area with each dog. On the occasions when the device was on, it was turned on prior to the dog walking toward the area and turned off after the dog had left the area.

Table 4.

Data were analysed using Chi-Square or Fishers Exact Analysis to determine whether or not there was a significant difference in the behavioural scores between the control and treatment for the two devices. Values were considered significant when P<0.05.

There was a significant effect of the pest-repeller on the behavioural score of dogs (P<0.001) with 93.1% of dogs scoring a ‘2’ when the device was on compared with 100.0% scoring a ‘1’ when the device was off. Despite the significant difference in scores, the magnitude of the effect was marginal and indicated that the dogs could hear but were not adversely affected by the device. No association was found between age, breed, or sex of the dogs and the scores they received (P>0.05).

There was no significant difference between the behaviour of dogs exposed to the person deterrent at either intensity setting or the control. More dogs scored a ‘1’ when the device was on (22/24 and 24/24 for 74 dB and 88 dB respectively) compared with the control (21/24). The device was observed to produce a low frequency beep every minute to signify that it was on. The dogs were observed to respond to this sound with an ear twitch or small head movement. There was no association between the breed, age, or sex of the dogs and the scores they received (P>0.05).

It is possible that due to the person deterrent being on prior to the dogs entering the area, that dogs may have become habituated to the sound before they reached the area and before scoring commenced. Therefore, a greater difference in behaviour may have been observed had the device been turned on once the dogs reached the area. The device in the study was closer to the dog (at a height of 1.68 m) than the device would be if placed on a building, where a working guide dog would encounter the sound. Therefore, the sound pressure level to which a working guide dog would be exposed would be less than the sound pressure level to which dogs were exposed in this study. Combined with the results of this study, it is suggested that guide dogs working in an environment with an operational person deterrent would not likely be affected by the sound produced.

This study did not test the effect of the devices on dogs in a laboratory setting because this would not have reflected the environment in which guide dogs are likely to encounter the device. Further observations of guide dogs working past devices on buildings or in gardens in urban environments could provide results that more accurately reflect guide dogs’ reactions when encountering the sounds. However, overall the results indicate that both devices are noticed by dogs, but do not produce a distracting effect of a significant magnitude. Neither device produced an adverse effect on dog behaviour, and therefore seems unlikely that there would be an impact on the working ability of a guide dog.