Dogs Barking and Babies Crying: The Effect of Environmental Noise on Physiological State and Cognitive Performance

Abstract

Objective:

The exposure to some environmental sounds has detrimental effects on health and might affect the performance in cognitive tasks. In this study, we analyze the effect of the neighborhood noises of a baby crying and dogs barking on the autonomic response and cognitive function.

Materials and methods:

Twenty participants were exposed, in separate sessions, to white noise, a baby crying, a small dog barking, and a large dog barking. During each session, heart rate, skin conductance, reaction times, spatial memory, and mathematical processing measures were taken throughout time.

Results:

The sounds of a baby crying and dogs barking led to significantly higher heart rates and skin conductance levels as opposed to exposure to white noise. Results were not as consistent with exposure to barking as they were to the baby. Exposure to the baby crying and dogs barking led to faster reaction times, possibly due to a facilitation by the autonomic system activation. No significant effects on spatial memory were found. Conversely, participants performed worse and slower in a mathematical task when exposed to the dog and baby sounds, than when exposed to control noise.

Conclusion:

Exposure to the sound of crying babies and dogs barking leads to increased sympathetic response and decreased cognitive ability, as compared to exposure to control sounds. Special attention should be paid to the mitigation of exposure to these types of noises.

INTRODUCTION

Hearing can be the source of negative experiences and effects, namely through exposure to noise. In the WHO annual report of 1999, some problems observed in humans due to noise exposure were reported, such as fatigue, nervousness, stress, anxiety, lack of memory, tiredness, irritation, and problems in social relationships.[1]

Results of some studies revealed that sounds characterized as negative were evaluated as annoying and unpleasant, leading to an increase in feelings of alertness and danger, causing a greater physiological change than sounds considered as neutral.[2,3,4] The “type of sound” factor also has an important effect and interacts with the “sound emotional valence” factor. A “negative human” sound is perceived as more bothersome, unpleasant, and threatening, followed by the “negative discontinuous artificial” sound, the “negative animal” sound and, finally, the “negative continuous artificial” sound.[5]

A study by Partala and Surakka[6] showed that emotional sounds are accompanied by greater pupil dilation, which demonstrates that there is a higher level of autonomous activity activated by emotion. It was also found that disturbing environmental sounds capture more attention of subjects, but do not influence the components of perception.[7,8]

According to Larsson and Västfjäll,[8] emotionally disturbing sounds are more easily remembered than neutral sounds in a task of memorization and subsequent identification. In a study by Surprenant,[9] participants identified and recalled nonsense syllables under different noise levels. There were significant decreases in memory, even in conditions where there was no effect of noise on syllable identification. More convincing evidence has shown that noise or distortion has a direct effect on memory. According to this view, adding noise should increase the cognitive load and should degrade memory performance. Boman et al.[10] conducted a study on the effects that noise had on memory and found that only long-term memory was affected by irrelevant speech and road traffic noise. According to Hygge et al.,[11] reading and memory-based speech are also more vulnerable to noise exposure. The influence of noise on cognitive tasks fluctuates in accordance with the type of the cognitive task at hand and with the type of noise that the subjects are exposed to.[11,12] For a recent literature review on the topic of the impact of noise of physiology and cognition, please refer to the article by Grenzebach and Romanus.[13]

Several studies show that noise has a negative influence on the ability to concentrate when performing a task that requires attention. An experiment by Smith[14] found that focused attention tasks are vulnerable to the effects of noise, although the direction of the noise effect is dependent on the specific characteristics of the task. Klatte et al.[15] point out that among the problems caused by noise, lack of concentration, low productivity, interference in communication, and learning difficulties in adolescents are the ones with the greatest emphasis.

When there is exposure to occupational noise, that is, to noise in the work environment, this can have a psychological influence on the worker and on their cognitive performance. There are, in our daily lives, professional contexts where noise exposure levels are not high enough to cause hearing loss, but which can affect workers in cognitive tasks or activities such as concentration, reaction speed, and memory skills. For this reason, the critical aspects regarding occupational risks must be considered.[16] With more people than ever working remotely, exposure to neighborhood noises becomes more prevalent. In this investigation, we were specifically interested in the effect of biological distress noises, negative noises produced by other humans or animals, which are more common in residential areas. The aim was to understand the specific impact of these environmental sounds, as compared to continuous wideband noise, on physiological state and on cognitive performance. Some recent studies point to the possibility that baby crying noises are more disturbing than other noises. Hechler et al.[17] found that exposure to crying led to less positive emotions and to more errors on a working memory task that uses auditory loop resources (n-back task). Sokol and Thompson[18] found that participants made more errors on a mathematical task while listening to babies crying and whines, as opposed to in silence, and found more distractions with those sounds as opposed to with control noise. Dudek et al.[19] analyzed the performance on a Stroop task, which measures attention and inhibitory control, and found that performance was worse when exposed to baby crying noise, as opposed to baby laughing noise. Electroencephalography data further revealed reduced attention (smaller P200) and higher cognitive conflict (larger N450) in the baby crying condition.

In the study here reported, the impact of the baby crying noise was further analyzed and compared to the exposure to noises of a large barking dog, a small barking dog, and control white noise. To assess physiological response, heart rate and galvanic skin responses were tracked. To assess cognitive performance, participants were tested on a reaction times task, on a spatial working memory task, and on a mathematical processing task. To assess the short-term adaptation process, participants were tested on three consecutive moments. The main goal was to assess cognitive performance and physiological responses to the different types of noise and to observe if the effect of the baby crying noise also extends to other distress sounds, namely dogs barking, or if it is exclusive to human distress sounds.

METHODOLOGY

Participants

Twenty people participated in the study, comprising 15 females and 5 males, aged between 18 and 34 years, with a mean age of 21.95 and a standard deviation of 3.63 years. This sample size had to do with the fact that each participant was extensively tested, both physiologically and cognitively, and with many experimental blocks within each of the four sessions. Prior to the experiments, participants were questioned about their vision and hearing, the presence of dogs and babies in their daily lives, and whether they consumed medication, psychoactive substances, and/or coffee.

All participants signed an informed consent, were informed about the procedure and the possibility of withdrawing at any time during the experiment. The methodology of the present study was approved by the Ethics Committee of the University of the Azores.

Experimental Apparatus

The experiment took place in a dark, soundproofed, acoustically treated, and distraction-free room, with the participant seated in front of a screen, and they performed the tests using a mouse and a computer keyboard. The display used as interface was a 27-inch DELL monitor with 1920 × 1080-pixel resolution, and Marshall on ear headphones were used for noise presentation. During the tests, physiological data were recorded using Neulog GSR Sensors (Galvanic Skin Response), Neulog HR Sensors (Heart Rate), Neulog USB Module connected to the computer, and Neulog software for data visualization and recording.

Stimuli and Experimental Task

Auditory stimuli consisted of four sounds, each with a level of 79 dB L_AEQ. This sound level was chosen to be loud enough to ensure effect magnitudes were significant, while remaining within the threshold of safe noise exposure. Two sounds consisted of dogs barking, one small dog and one large dog, one sound consisted of a baby crying, and one sound was white noise. These sounds were extracted from the ESC-10 Dataset for Environmental Sound Classification,[20] which consists of a dataset of validated and normalized recordings of specific classes of environmental sounds. The sounds were presented continuously in a loop and can be consulted in.[21] After each session, participants were asked to report, in a scale of 1 to 10, how distressing the sounds were. All participants considered the sounds produced by the baby and dogs as distressing at a moderate-to-high level (6–10). The most distressing sounds varied across participants. The ratings of the white noise varied significantly from person to person, with ratings ranging from 2 to 9. Most participants rated the white noise <5, thus evidencing that it was, on average, perceived as less distressing than the other sounds.

During the experiment, participants had their heart rate and galvanic response recorded. Each sound session comprised three cognitive tasks: Reaction Times task (Spatial priming), Spatial Memory Task (Corsi Blocks), and Mathematical processing task (Mathproc). These tasks are available in the Psychology Experiment Building Language (PEBL) Test Battery (Mueller and Piper, 2014)[22]. The tests and their respective instructions were translated to the native language of the participants. A demonstration of a participant executing the task can be found in the article by Mendonça.[21]

The Reaction Times task (Spatial priming) involves the subject reacting to the target stimulus (yellow square turning blue) by clicking the mouse as fast as possible. The visual interface is organized in three columns by three lines of yellow squares (3 × 3) and any square, at any time, can change color. Each block of this test comprises 54 trials. This is a task that assesses focused attention, reaction time, and processing speed. Results are expresses in average reaction time, in milliseconds.

The Spatial Memory Task (Corsi blocks) requires the participant to replicate a sequence of stimuli previously presented, with these stimuli being represented by a blue square moving in space. The size of the sequence, and therefore the task difficulty, increases as the subject progresses in the task. The spatial arrangement of the sequences is different in each trial and is randomly determined. The task ends when the participant fails two consecutive trials. Two executive functions are involved in this task: visuospatial working memory and attention. The results are expressed in memory span, consisting of the average number of items that the participant could reproduce correctly.

The Mathematical Processing task (Mathproc) presents a mathematical stimulus that involves numerical addition and subtraction, with an increase in the degree of difficulty as the test progresses. This task is divided into three levels: the first being mathematical operations involving two values with a duration of 1.5 seconds, the second level involving three values in 2 seconds, and the last level involving four values in 4 seconds. Each level comprised 60 trials. This task evaluates mathematical reasoning, which encompasses a variety of cognitive functions. Results are expressed in terms of response time and rate of correct answers.

Experimental Design

Before the beginning of the experiment, participants read an information sheet. Then, they signed the informed consent and completed the questionnaire. The instructions given to the subjects were that they could not use the cell phone during the experiment, that the instructions for carrying out the tasks would be on the screen, and that as soon as they completed each task they would call the collaborators to proceed with the following tasks. Before carrying out the tasks, the participants were asked to wet their hands, since when they are wet, they promote a better reading made by the sensors. There were four separate sessions, one for each sound exposure (baby, large dog, small dog, noise). Within each session, participants performed all three of the cognitive tasks and respective blocks. The order of tasks and sound sessions was counterbalanced across participants. The Reaction Times task and Spatial Memory task were performed three times for each sound, henceforth referred to as block 1, block 2, and block 3, to evaluate possible adaptation processes. The Math Processing task was performed only once for each sound, considering that it was the most exhausting and extensive activity.

Statistical Analysis

After confirming the normality of distributions, all data were analyzed with Repeated Measures analyses of variance (ANOVAs), defining the block (1,2,3) and experimental condition (Baby, Small Dog, Large Dog, Noise) as a within-subject variables. When sphericity could not be assumed, the Greenhouse–Geisser correction was applied to the degrees of freedom. Post-hoc Tuckey tests were applied to identify which experimental condition pairs were significantly different at the P = 0.05 level. Response Times in the Math Processing task were tested with a Friedman test, since this variable was not normally distributed.

The dependent variables were, for the physiological recordings, heart rate in beats per minute and galvanic skin response as microsiemens. In the reaction times task, the dependent variable was the reaction times, expressed in milliseconds. In the spatial memory tasks, the dependent variable was the memory span, expressed as the number of items. In the mathematical reasoning task, the dependent variables were response accuracy, expressed as the percentage of correct answers, and response times, expressed in milliseconds.

RESULTS

Reaction Times Task

Some effects were observed regarding heart rate in the different sound conditions throughout the three blocks [Figure 1]. The sound condition with the highest heart rate was Baby, Noise yielded the lowest mean bpm levels, and conditions Small dog and Large dog had intermediate results.

Figure 1.

Heart rate (bpm) per sound in each of the three experimental blocks. CI = confidence interval.

As can be observed in Table 1, differences between blocks were not significant. A significant and large effect of sound condition was obtained, particularly due to the differences between conditions Baby and Noise.

Table 1.

Results of the Statistical Tests Regarding Heart Rate in the Reaction Times Task.

Test	F	P	η 2	Significant Post-hoc (P ≤ 0.05)
Differences between blocks	1.696	0.196	n.s.	n.s.
Block × sound condition interactions	0.959	0.372	n.s.	n.s.
Block × participant interaction	1.184	0.315	n.s.	n.s.
Differences between sound conditions	4.478	0.009	0.272	Baby vs noise

Galvanic skin response data also showed clear effects of sound condition [Figure 2]. Skin conductance was highest in the Baby condition and lowest in the Noise condition. These levels were at an intermediate level in sound condition Small Dog and Large Dog.

Figure 2.

Galvanic skin response in each sound condition and each block. CI = confidence interval.

The statistical tests [Table 2] revealed no effect of block over the skin conductance levels. A significant and strong effect of sound condition was obtained, namely due to the Noise and Baby and between Noise and Large Dog conditions.

Table 2.

Results of the Statistical Tests Regarding Galvanic Skin Response in the Reaction Times Task.

Test	F	P	η 2	Significant Post-hoc (P ≤ 0.05)
Differences between blocks	0.786	0.434	n.s.	n.s.
Block × sound condition interactions	0.495	0.569	n.s.	n.s.
Differences between sound conditions	7.409	<0.001	0.382	Noise—babyNoise—large dog

Regarding the task response data, all participants responded correctly in all trials. Reaction times varied with sound condition but remained similar throughout the three experimental blocks [Figure 3]. The lowest reaction times were obtained in the condition Baby and the highest reaction times were obtained in Noise.

Figure 3.

Reaction times per sound condition per block [±95% confidence interval (CI)].

There were no statistically significant differences between blocks regarding reaction times [Table 3]. There were significant differences between the reaction times in the different sound conditions, where the condition Noise differed from all other sound conditions.

Table 3.

Results of the Statistical Tests Regarding the Reactions Times in the Reaction Times Task.

Test	F	P	η 2	Significant Post-hoc (P ≤ 0.05)
Differences between blocks	0.601	0.552	n.s.	n.s.
Block × sound condition interactions	1.016	0.370	n.s.	n.s.
Differences between sound conditions	11.741	<0.001	0.557	Noise—babyNoise—large dogNoise—small dog

Spatial Memory Task

The heart rate values in the Spatial Memory task across sound condition and block can be observed in Figure 4. Heart rate levels in this task were most elevated in the Baby and Small dog conditions. They were lowest in Large dog and Noise.

Figure 4.

Heart rate in the spatial memory task: results per sound condition (left) and per sound condition per block (right).

There were significant differences between all blocks [Table 4], which occurred in the Baby and Small dog conditions, revealing an increase in physiological response, which could point to a sensitization effect, with increased stress experienced throughout time. There was also a large effect of sound condition on heart rate, namely between the Baby and the Noise conditions.

Table 4.

Results of the Statistical Tests Regarding Heart Rate in the Spatial Memory Task.

Test	F	P	η 2	Significant Post-hoc (P ≤ 0.05)
Differences between blocks	5.069	0.013	0.118	Block 1–Block 2–Block 3
Block × sound condition interactions	2.834	0.075	0.069	n.s.
Differences between sound conditions	4.128	0.013	0.412	Noise—Baby

Regarding galvanic skin responses, as can be seen in Figure 5, the highest skin conductance levels were obtained in the Baby condition, followed closely by the values obtained in the Small dog condition. The values were lower in the Large dog condition and lowest in the Noise condition.

Figure 5.

Galvanic skin response in the spatial memory task: mean values by sound condition (left) and by sound condition per block (right). CI = confidence interval.

There was no statistically significant effect of block of the skin conductivity data. The effect of sound condition across all blocks was significant [Table 5].

Table 5.

Results of the Statistical Tests Regarding Galvanic Skin Response in the Spatial Memory Task.

Test	F	P	η 2	Significant Post-hoc (P ≤ 0.05)
Differences between blocks	1.634	0.202	n.s.	n.s.
Block × sound condition interactions	1.261	0.294	n.s.	n.s.
Differences between sound conditions	3.305	0.031	0.216	Noise—baby

The memory span results were analyzed in terms of the maximum number of spatial event locations which were correctly evoked in two consecutive trials. Similar results were obtained in the Baby, Large dog, and Small dog conditions [Figure 6].

Figure 6.

Average memory span per sound condition (left) and per block per sound condition (right). CI = confidence interval.

It was observed that average values of memory span increased throughout blocks in most sound conditions; however, there was no statistically significant effect of block on the memory span [Table 6]. There were also no significant effects of sound condition on memory span.

Table 6.

Results of the Statistical Tests Regarding Memory Span in the Spatial Memory Task.

Test	F	P	η 2	Significant Post-hoc (P ≤ 0.05)
Differences between blocks	0.076	0.927	n.s.	n.s.
Block × sound condition interactions	0.161	0.851	n.s.	n.s.
Differences between sound conditions	0.385	0.764	n.s.	n.s.

Math Processing Task

In this task, there was only one block of testing, but there were three types of mathematical task (two numbers, three numbers, and four numbers). Heart rates and galvanic skin response values correspond to averages obtained per participant per sound condition across all tasks. Heart rate levels were, on average, less elevated in the Noise condition than in the remaining conditions [Figure 7]. These heart rate levels were in the same ballpark as those obtained in the Reaction Times and Spatial Memory tasks.

Figure 7.

Heart rate (bpm) and galvanic skin response (arb) in the math processing task per sound condition.

A significant and large effect of sound condition was observed (F₍₃₎ = 5.096, P = 0.005, η² = 0.298). Post-hoc tests revealed significant differences between Noise and Baby, Noise and Small dog, and Noise and Large dog.

The effects of sound condition on the galvanic skin responses during the mathematical processing task are presented in Figure 7. Again, levels were lower in the Noise condition, when compared to the remaining conditions. A large effect of sound condition was observed on galvanic skin response (F₍₃₎ = 4.713, P = 0.007, η² = 0.446). Post-hoc tests revealed that there were significant differences between the Noise condition and the Baby, the Small dog, and the Large dog conditions.

Regarding the behavioral data, the accuracy levels were highest in the Noise condition and lowest in the Small dog and in the Baby condition [Figure 8]. A significant effect of noise condition was obtained (F₍₃₎= 4.149, P = 0.01, η² = 0.179). Post-hoc tests found a significant difference between the Noise and Baby conditions.

Figure 8.

Response accuracy and decision times in the mathematical task, by sound condition. CI = confidence interval.

Decision times were lowest in the Noise condition, followed by the Large dog condition, the Baby condition, and the Small dog condition [Figure 8]. Decision times were not normally distributed. There was a significant effect of sound condition over the decision times (χ²₍₃₎ = 22.6, P ≤ 0.001) on a Friedman test. Pairwise Wilcoxon tests found that differences were significant between Noise and Baby, between Noise and Small dog, and between Noise and Large dog, but not between the remaining conditions.

DISCUSSION

This study intended to clarify if exposure to certain environmental biological noises affect the physiological activation levels and/or cognitive performance, as compared to exposure to static white noise of equal level. The study focused particularly on the noises of a baby crying, a large dog, and a small dog barking, which were compared with the effects of exposure to white noise. Before discussing the results, we stress that the findings reported are directly related to the specific tasks used, the specific stimuli type and level, and the context of data collection. For instance, exposure to lower-level sounds might have led to lower overall physiological activations or smaller magnitudes of cognitive effects. The results reported revealed a relation between exposure to certain noise types, physiological activation, and the performance in certain cognitive tasks, which we discuss below.

While performing the Reaction Times task, the heart rate of participants when exposed to the sound of the baby crying was higher compared to exposure to the remaining sounds. The lowest heart rates were obtained when exposed to white noise. The galvanic skin responses followed a similar pattern. The physiological responses did not change significantly across the three blocks, despite a slight upward trend in the sound condition Baby. These results indicate that exposure to these noises produces a relatively stable autonomic response. The task at hand was a relatively stable task that did not lead to enhanced cognitive load requiring only sustained attention and adequate response, which might also explain the constancy in physiological activation.

Regarding behavioral data in the Reaction Times task, all participants responded correctly in all trials and reaction times varied across sound condition but, once again, not throughout time. Baby was the sound condition that yielded lower reaction times, but reaction times were significantly lower in all conditions, as compared to the Noise condition. We observe a general trend where the sounds associated with the highest levels of physiological activation were also associated with quicker responses. This would be in line with a high arousal state, where the sympathetic response leads to greater readiness for action.[23] According to Evans and Maxwell,[24] exposure to noise has the potential to improve the performance in simpler tasks, and that is what was observed here, namely with the biological distress sounds.

Regarding the Spatial Memory task, the physiological responses were more elevated in the sound condition Baby and were the least elevated in the condition Noise. More specifically, the heart rate levels were higher in the Baby condition, followed by Small dog, Large dog, and were smallest in Noise. The same order of effects was observed in the galvanic skin responses. However, differences were only statistically significant between the Baby and Noise condition. There was an increase in heart rate throughout time in the most arousing sound conditions (Baby and Small dog), but not in the others. What distinguished this task from the others was that the cognitive load and attentional demands were higher. Therefore, exposure to more complex sounds such as a baby crying or a dog barking might have led to a higher stress response with time, and consequently to higher heart rates.[13]

The behavioral results in the Spatial Memory task revealed that the memory spans were higher in the biological sound conditions, as compared to the Noise condition, but these differences did not reach significance. Therefore, it might be concluded that exposure to sounds of barking dogs or crying babies will change the physiological state of the listener, but that change might not be enough to affect the performance in tasks involving visuospatial memory. It is important to note that previous studies had found that exposure to environmental sounds reduced the working memory ability,^[19–11] but in those studies the task at hand required the activation of the phonological loop, the auditory component of working memory. Our visuospatial memory task did not require phonological loop resources, but visual sketchpad resources instead. Therefore, we hypothesize that exposure to this type of environmental noises might impact the auditory components of the working memory, but not its visuospatial components. According to the duplex-mechanism account for the effect of auditory distraction on cognitive processes[25,26] one of the two ways that the sound can affect working memory is by competition in concurrent cognitive processes. Since out task presented less competition than other traditional syllable or word-based working memory tasks that would explain the reduced distraction effect obtained. Further studies directly comparing both components of working memory are needed.

In the Math Processing task, the heart rate responses were significantly more elevated in all the biological sound conditions compared to the Noise condition. The galvanic skin responses followed a similar pattern, where Baby, Small dog, and Large dog led to higher responses than Noise. Response accuracy was highest in the Noise condition and lowest in the Baby condition. Therefore, the sound condition with lower physiological activation were associated with better mathematical reasoning levels. These results do not meet the behavioral results in the other two tasks: higher physiological activation led to quicker reaction times and no significant memory effects, but worse mathematical reasoning outcomes. It might be that these differences in results are due to task complexity. Indeed, Evans and Maxwell[24] found that the noise exposure significantly deteriorated the performance in more complex tasks, which demanded more sustained attention and memory. They proposed that noise exposure affects most the cognitive tasks demanding sustained attention and reasoning. Indeed, while the Math processing task demanded computation and mental manipulation of information, the other two tasks merely required reaction and memorization.

According to the duplex-mechanism account for the effect of auditory distraction on cognitive tasks,[25,26] attention capture would be one of the mechanisms that would cause degradation in performance. Attention capture should be an effect mediated by task difficulty, where most difficult tasks would no longer be affected by noise, as all attention-control resources would be allocated to the main task.[27] In this report, the easiest task was the Reaction Times task. In that task, the most complex biological noises led to better performance than the control static white noise. The spatial memory task was very demanding in terms of visual working memory and was not affected significantly be the more complex noises. The Math Processing task was very demanding in terms of working memory, numerical computation, and reasoning. This task typically consumes some resources of the phonological loop of the working memory by activation of the verbal component of number processing. In that way, this would be the type of task that would share cognitive processes with the processing of environmental auditory distractors. Despite its high difficulty, we did not observe a reduction in the effect of sound over cognitive performance in this task. Therefore, our data do not seem to support the existence of attention capture mechanisms mediated by task difficulty. Further tests should directly manipulate the difficulty of each of these tasks, to better assert the presence of this effect. Another factor worth mentioning is that working memory capacity has been argued to often mediate the effect of auditory distraction, where people with higher working memory would be less affected by concurrent noise.[27,28] We did not account for this effect in our data analysis, and possibly this would be able to explain the high intersubject variability in the spatial memory task.

This study revealed that, on average, there is a higher heart rate and more skin conductivity while exposed to noises produced by high-alert living creatures (the incessant cry of the baby and the barking of a dog), which seems to highlight that humans are more sensitive to these biological sounds than to control artificial sounds such as white noise. However, the exposure to these sounds was only compared to the exposure to white noise. While the white noise was at an equivalent level, this was a steady-state stimulus, as opposed to the time-varying biological noises. Also, while these biological sounds consisted of sounds often associated with distress or warning, they were not compared to other nonbiological sounds conveying distress or warning, such as alarm systems of sirens. In fact, none of the previous studies revealing the striking effect of the baby crying sound performed such comparisons, and the control conditions are often silence or white noise.[18] One notable study compared the effect of exposure to a baby crying to the exposure to a baby laughing and found significant differences, where the baby crying was more detrimental to the cognitive task than the baby laughing.[19] In future studies, it would certainly be relevant to compare more sounds, of positive and negative valence, from living and nonliving creatures.

One limitation of this study is the small sample size, which consisted of only 20 participants, which was a consequence of the extensive testing protocol implemented for each person. Future studies should expand the sample size. This would be particularly relevant for the memory task, where significant effects might have been achieved with larger statistical power. Other statistical analyses, integrating all variables into an explanation model, could also have revealed different findings. Additionally, it would be relevant to evaluate mediating effects such as being a parent, being a dog owner, being afraid of dogs, and experience in exposure to these sounds. Another limitation is that no direct comparison could be established to other similar, nonbiological sounds, such as alarm sounds or sirens. Therefore, it is at this point not possible to assert if the effects we report are due to the fact that the sounds were from living creatures, due to the fact that they communicated distress, or due to the fact that the sounds were less continuous in nature than the control white noise.

CONCLUSION

In conclusion, exposure to sounds of a baby crying or a braking dog may increase the level of alertness, which seems to facilitate reaction times, but hinder performance on mathematical tasks.

Financial support and sponsorship

This work was supported by national funding from the Portuguese Foundation for Science and Technology (UIDB/00050/2020).

Conflicts of interest

There are no conflicts of interest.