The fear response is a normal, adaptive behavior that helps protect the individual from harm; however, fear responses can be considered abnormal when they are repeatedly and consistently triggered by non-threatening stimuli, or when the level or duration of response is out of proportion compared to the actual threat level posed by the stimuli (1, 2). Sudden, loud noises are one of the most common triggers for fearful behavior in dogs (1, 3), and considerable research supports the concept that many companion dogs suffer from noise sensitivity and are affected by salient noises. Reported prevalence of noise sensitivities (marked and even extreme responses to noise) in domestic dogs varies by study, but often ranges as high as 50% (4–7).

Loud, sudden (and often unexpected) sounds have been used in studies of behavioral and physiological stress responses in dogs (8, 9). That research focuses on dramatic infrequent sounds—sounds that dogs do not encounter daily or on a regular basis, but rather occur as sporadic, unpredictable events [e.g., thunderstorms (8, 10, 11), fireworks (12, 13), and gunshots (9)]. Canine reactions associated with these stressors have been interpreted by professionals as expressions of stress, anxiety, and fear (11). For example, physiological responses to loud noises can include a dramatic (207%) increase in salivary cortisol, lasting for 40 mins or more (8); unexpected noises resulted in rapid responses including tachycardia, hypertension, and increased secretion of epinephrine and norepinephrine (14); and exposure to acute, irregular noises is associated with stimulation of the HPA axis and a rapid (within 15-min) increase in circulating cortisol concentrations (15, 16). Behavioral responses to loud noises include panting, hiding, pacing, cowering/lowered body posture, shaking/trembling, barking, escape attempts/retreating, and seeking out familiar people (2, 5, 8, 17).

Stress associated with fear and anxiety can have negative impacts on health, welfare, behavior and lifespan (18, 19), depending on both the nature of the stressor (intensity, duration, persistence, etc.) and the coping skills of the individual (20, 21). Every day, companion dogs are exposed to common household noises. Anecdotally [e.g., (22)], and based on reports of undesirable reactions to noises in the home [e.g., (23)], many of these noises may be causing fear and anxiety in the resident dogs. In this study, we define “common” household sounds as including both those that occur daily, and also other common, but non-daily, sounds that are heard in the home and confined to the space where a pet resides. The response of companion dogs to common household noises has not yet been studied in the dogs' home environments. Dogs' reactions to these household noises could have implications for their health and wellbeing in human homes. If these dogs are experiencing fear in response to regularly-occurring household stimuli, they may as a result be experiencing reduced welfare, and be at risk for the development of stress-related behavioral or physiological problems (24). Damage to the human-animal bond which can result from undesirable behaviors associated with fears and phobias can lead to decreased commitment to care of the dog, and/or an increased risk of relinquishment or euthanasia (19).

The purpose of this study was to investigate and describe, using survey data, whether household dogs display a fear/anxiety/stress response to common sounds occurring within homes. In addition, we recorded both canine and human behavior in publicly-shared videos and video clip compilations which featured companion dogs reacting strongly to household noises. Our goal was to describe both the prevalence and nature of the dogs' reactions to household noises seen in these videos, as well as the various interpretations that owners offered in response to their dogs' displays, possibly indicating fear, anxiety, or stress. Understanding the significant or insignificant effects of household sounds/noises on a companion dog's behavior and mental state will allow for better understanding of dogs' behaviors and how to improve their general welfare at home.

Our hypotheses were that: (1) Sounds that are common in households can elicit fear, anxiety, or stress responses in companion dogs; and (2) Owners can misinterpret or respond negatively to expressions of fear, anxiety, or stress in their companion dog if the stressor is considered “common.”

Publicly-shared videos do not represent systematic or random sampling, and so cannot be used to estimate the true prevalence of fear in response to common household noises in companion dogs. Owners of dogs who exhibit more extreme reactions to common household items that the owners consider “normal” are presumably far more likely to film and share their dog's behavior online. Nonetheless, numerous canine behaviors frequently associated with fear and anxiety (both overt and subtle) were observed in these videos, in response to both regular (e.g., microwaves, vacuum cleaners) and irregular but “normal” household noises (e.g., smoke detector low-battery warning beeps or chirps, smoke alarms). Dog behaviors in response to these noises included proximity-seeking directed toward their owners, an attachment-related behavior seen in dogs (and humans) as a means of coping with present stress (26–28), as well as other recognized signs of canine fear and anxiety, such as panting, howling, and lip licking. Behavioral signs of fear and anxiety in response to household noises were also reported by dog owners in the survey data, particularly when the sound was loud/infrequent (Tables 3, 4), although our brief survey did not enable detailed analyses of differences between the types of behavioral reactions associated with specific sound sources. While some of the behaviors recorded in the videos (e.g., barking, lunging, jumping on owner) can reflect other canine emotions such as excitement, many of the behaviors seen (e.g., trembling, panting, retreating from the sound source) are commonly associated with noise phobias (29) and are likely to reflect distress in these videos, given the context in which the behaviors were seen. Our Hypothesis 1 was supported by our data: certain types of household noises do cause marked fear in some dogs, and future research into the true prevalence of this issue, and the associated welfare issues, is warranted.

Some categories of household noises appear to be more problematic for the dogs. More intense signs of fear, such as trembling, were significantly more likely to be seen in the presence of sounds characterized as high frequency intermittent (HFI), such as smoke detector beeps, in the videos; and, in the owner surveys, extreme reactions were significantly more likely to be reported by dog owners in response to these types of sounds. Behaviors associated with arousal, agitation, and excitement (barking, lunging) were more commonly seen in the presence of low frequency constant (LFC) sound sources, such as vacuum cleaners, although behavioral signs of fear (such as lip licking, ears tucked back) were also seen with LFC sounds. This pattern is also seen in the differences in dogs' behavior in the compilations (which all featured LFC sound sources, primarily vacuum cleaners) vs. the individual videos (the majority of which featured HFI sound sources). For example, lunging was the most commonly observed behavior in the video compilation clips (Table 8), but was very rare in the individual videos (Table 6). Dogs' reactions to vacuum cleaners in these videos, exhibited in behaviors such as lunging and barking, may stem from emotional reactions other than fear, as has been noted elsewhere (7). Videos consistently presented the more extremely stressed dogs responding to the higher frequency beeps occurring sporadically, whereas the compilations depicted dogs' less-extreme responses to lower frequency appliance sounds that occur on a more regular basis.

The more extreme reactions to HFI sounds in this study are in agreement with the literature on noise sensitivity in dogs: loud noises, especially when unpredictable and when the dog concerned cannot control their exposure to the sound (as is the case here, for dogs confined inside a human dwelling), can cause intense fear reactions in dogs, detectable in both physiological and behavioral measures (15, 24, 30). The frequency, and constancy, of the sounds is an important factor in the dog's reaction. High frequency sounds are believed to be more salient for many species (i.e., attracting greater attention and prompting greater alertness) than lower frequency sounds (31); for example, the cries of a human baby will cause distress to a dog, despite the fact that human infant vocalizations are not of direct evolutionary importance to dogs (32, 33). Dogs' sensitivity to high frequency sounds is greater than humans', whereas at lower frequencies, the sensitivities of dogs vs. humans may not differ much (31, 34). In addition, the amplification processes of the ear may be stronger in dogs than humans (31), thus dogs could be more adversely affected by high frequency sounds in the home than their human companions.

Smoke detectors and smoke alarms, while an important safety measure in the home, are a particular concern for companion dogs. These devices are designed to effectively wake up people from deep sleep, and are thus commonly set at a frequency of around 3,000 Hz, designed to achieve a sound level of at least 75 dBA at a person's pillow (necessarily being much higher at the source) (35). Dogs, regardless of their body size, have particularly high sensitivity for hearing in the range 1,000–8,000 Hz (36), and dogs generally have better hearing in these high ranges than humans (31). Further, a significant proportion of adult humans over 48 years of age experience some degree of hearing loss (37): most likely involving sensitivity to higher frequencies. Thus, smoke alarms that are somewhat loud to most humans are likely to be painfully loud to most dogs.

This influence of the nature (frequency, constancy, predictability) of the sound source may also explain differences between our study and Gin et al. (16), who found that companion dogs did not have a significant stress response to vacuum cleaner sounds in a veterinary hospital setting, based on serum cortisol levels collected after exposure to the sounds. Gin et al. (16) noted a number of possible explanations for the lack of anticipated increases in cortisol levels, including lack of pre-exposure cortisol measurements, familiarity of these particular dogs with the clinic setting, and presence of owners as a mitigating influence on the dogs' stress levels. In addition, however, it may be that the characteristics of the vacuum used in that study may not have been the characteristics which cause the strongest fear reactions in dogs. Similarly, Stellato et al. (38) found that only respiratory rate increased when dogs were exposed to background noises (people talking, dogs barking, metal doors banging) during veterinary exams in a clinic setting. Behavior and other physiological variables (e.g., temperature, heart rate) were unaffected by noise levels, leading the authors to conclude that certain handling aspects during exams were a more powerful stressor than noise levels. Again, the differences between Stellato et al.'s (38) findings and the present study may be related to setting (veterinary clinic vs. home), but also to the nature of the sound stimuli to which dogs were exposed.

The survey results revealed a mismatch between the owners' perception of fearfulness in their dogs, and the amount of fearful behavior exhibited by their dogs, suggesting that many owners are underestimating fearfulness and anxiety in their dogs. While (as noted above) publicly-shared videos and compilations represent a biased sample, these videos demonstrate that some proportion of dog owners react to their dogs' fearful and defensive reactions with amusement, rather than concern. Concern was only expressed by owners/narrators in 17.5% of the individual videos, and was not expressed in any of the video clip compilations used in this study. Our Hypothesis 2 was thus also supported by these data: owners can misinterpret or respond negatively to expressions of fear, anxiety, or stress in their companion dog, particularly if the stressor is considered “common.” These findings underscore calls for better education for dog owners with regard to accurately interpreting canine body language, particularly the more subtle signs of stress and anxiety, to both safeguard welfare and minimize development of anxiety-related behavior problems [e.g., (39)].

Other studies have demonstrated that non-experts (i.e., the average companion dog owner, as opposed to a canine behaviorist or knowledge-assessed trainer) may be less able to identify signs of fear and anxiety in their dogs, particularly when the behaviors are subtle (39, 40). Because of this inability to accurately read canine body language, many owners appear to perceive their dogs as less fearful than they actually are, and may therefore not intervene in response to early or subtle stress signals to mitigate or remove sources of stress for their dogs (39). A particular extreme example of a dog's reaction to household noise bears mentioning here: in Bender and Strong (22), canine behavioral consultant Emily Strong describes a case working with a dog displaying signs of intense, chronic anxiety (pacing, whining, restlessness, lack of interest in interacting with his owners). The owners of this dog were mystified by the dog's behavior, and had spent thousands of dollars ruling out medical issues with this dog; no medical problems had been found. The behavior had started suddenly a few months prior to the first consult, and continued to get worse; the owners had lived with the dog since puppyhood, and prior to this change, the dog had never exhibited any anxiety-related or behavioral problems. Tellingly, the dog was relaxed and happy when away from the home, but resistant to returning home after walks. After much questioning, it was discovered that the owners had installed a sonic (very high frequency) pest repellant device in the home a few months before; upon turning off the device, the dog immediately began to relax, and returned to his normal behavior over the next few days. These owners were, by all accounts, very dedicated to the welfare of their dog, but they could not hear the device; the dog, on the other hand, was highly bothered by it [(22); p. 6]. Ultrasonic pest deterrents are generally inaudible to humans, although van Wieringen and Glorieux (41) noted that some younger individuals were able to hear them and found them disturbing, even during relatively short exposures (20 mins). Many everyday household devices produce ultrasound (42), and ultrasound, particularly when very loud, has been used successfully to repel dogs from tested areas (4). In humans, exposure to ultrasound has been associated with annoyance, disorientation, headache, fatigue, nausea and high arousal (31).

Individual dogs vary in their ability to cope with stressors (43), and not all dogs display marked sensitivity or phobias toward noise. Dogs suffering from physical pain may be at increased risk of noise sensitivity; Fagundes et al. (44) reported fear responses to loud noises in 10/10 noise-sensitive dogs concurrently experiencing pain, vs. in 6/10 noise-sensitive control dogs not experiencing pain. Up to 50% of dogs may experience noise sensitivity and/or extreme reactions to noise in their lifetimes (5, 7). Many companion dogs are primarily contained within human households (alone, or with human caretakers unable to detect these sounds, or to accurately interpret the dog's fearfulness), and are unable in many cases to control their own level of exposure to these sounds. This issue, therefore, may represent a significant welfare concern for affected dogs. In conclusion, and as noted earlier, additional targeted research into companion dogs' reactions to household sounds, owner perceptions of (and reactions to) their dogs' behavior, and the associated welfare implications, is warranted.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by UC Davis Institutional Research Board. The survey participants provided their written informed consent to participate in this study.

LH: conceived the study. LH, EP, and AG-D: designed the initial approach and survey. EG: developed the ethogram and method of video analyses, conducted statistical analyses, and drafted initial manuscript. JC and SC: compiled data from videotapes. EG and LH: edited manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the Center for Companion Animal Health, University of California, Davis (# 2009-54-F/M).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.