Sleep macrostructure is modulated by positive and negative social experience in adult pet dogs.

The effects of emotionally valenced events on sleep physiology are well studied in humans and laboratory rodents. However, little is known about these effects in other species, despite the fact that several sleep characteristics differ across species and thus limit the generalizability of such findings. Here we studied the effect of positive and negative social experiences on sleep macrostructure in dogs, a species proven to be a good model of human social cognition. A non-invasive polysomnography method was used to collect data from pet dogs (n = 16) participating in 3-hour-long sleep occasions. Before sleep, dogs were exposed to emotionally positive or negative social interactions (PSI or NSI) in a within-subject design. PSI consisted of petting and ball play, while NSI was a mixture of separation, threatening approach and still face test. Sleep macrostructure was markedly different between pre-treatment conditions, with a shorter sleep latency after NSI and a redistribution of the time spent in the different sleep stages. Dogs' behaviour during pre-treatments was related to the macrostructural difference between the two occasions, and was further modulated by individual variability in personality. This result provides the first direct evidence that emotional stimuli affect subsequent sleep physiology in dogs.

Reliability of Family Dogs' Sleep Structure Scoring Based on Manual and Automated Sleep Stage Identification.

Non-invasive polysomnography recording on dogs has been claimed to produce data comparable to those for humans regarding sleep macrostructure, EEG spectra and sleep spindles. While functional parallels have been described relating to both affective (e.g., emotion processing) and cognitive (e.g., memory consolidation) domains, methodologically relevant questions about the reliability of sleep stage scoring still need to be addressed. In Study 1, we analyzed the effects of different coders and different numbers of visible EEG channels on the visual scoring of the same polysomnography recordings. The lowest agreement was found between independent coders with different scoring experience using full (3 h-long) recordings of the whole dataset, and the highest agreement within-coder, using only a fraction of the original dataset (randomly selected 100 epochs (i.e., 100 × 20 s long segments)). The identification of drowsiness was found to be the least reliable, while that of non-REM (rapid eye movement, NREM) was the most reliable. Disagreements resulted in no or only moderate differences in macrostructural and spectral variables. Study 2 targeted the task of automated sleep EEG time series classification. Supervised machine learning (ML) models were used to help the manual annotation process by reliably predicting if the dog was sleeping or awake. Logistic regression models (LogREG), gradient boosted trees (GBT) and convolutional neural networks (CNN) were set up and trained for sleep state prediction from already collected and manually annotated EEG data. The evaluation of the individual models suggests that their combination results in the best performance: ~0.9 AUC test scores.

Heart Rate and Heart Rate Variability during Sleep in Family Dogs (Canis familiaris). Moderate Effect of Pre-Sleep Emotions.

The domestic dog (Canis familiaris) has been shown to both excel in recognising human emotions and produce emotion-related vocalisations and postures that humans can easily recognise. However, little is known about the effect of emotional experiences on subsequent sleep physiology, a set of phenomena heavily interrelated with emotions in the case of humans. The present paper examines heart rate (HR) and heart rate variability (HRV) during dogs' sleep, measures that are influenced by both positive and negative emotions in awake dogs. In Study I, descriptive HR and HRV data is provided on N = 12 dogs about the different sleep stages (wake, drowsiness, non-rapid eye movement (non-REM), REM; scoring based on electroencephalogram (EEG) data). We conclude that wakefulness is characterised by higher HR and lower HRV compared to all sleep stages. Furthermore, drowsiness is characterised by higher HR and lower HRV than non-REM and REM, but only if the electrocardiogram (ECG) samples are taken from the first occurrence of a given sleep stage, not when the longest periods of each sleep stage are analysed. Non-REM and REM sleep were not found to be different from each other in either HR or HRV parameters. In Study II, sleep HR and HRV measures are compared in N = 16 dogs after a positive versus negative social interaction (within-subject design). The positive social interaction consisted of petting and ball play, while the negative social interaction was a mixture of separation, threatening approach and still face test. Results are consistent with the two-dimensional emotion hypothesis in that following the intense positive interaction more elevated HR and decreased HRV is found compared to the mildly negative (lower intensity) interaction. However, although this trend can be observed in all sleep stages except for REM, the results only reach significance in the wake stage. In sum, the present findings suggest that HR and HRV are possible to measure during dogs' sleep, and can potentially be used to study the effect of emotions not only during but also after such interactions.

Visual Fixation Patterns During Viewing of Half-Face Stimuli in Adults: An Eye-Tracking Study.

Human faces play a special role in social cognition, since as a core signal of interpersonal communication, they convey various kinds of information (e.g., about sex, age, race, emotions, intentions). Study 1 aimed to explore how this specialization manifests itself in eye movements when looking at neutral, static, female faces. We monitored the gaze pattern of 23 adult participants using eye-tracking method. To test if template-driven processes are involved in face perception, and to see how inversion affects fixations on special facial stimuli, we presented vertically cut half-faces in upright and inverted positions (so half of each stimulus represented a half-face, whereas the other half was left blank). Our results corroborate prior findings consistently demonstrating the dominance of the triangular area marked by the eyes and the mouth, measured by the number and duration of fixations. In addition, we found evidence for so-called complementary fixations, targeted at the non-informative parts (i.e., the half that does not contain any facial information) of the pictures, suggesting that other mechanisms beyond purely stimulus-driven ones might drive looking behavior when scanning faces. Study 2 was intended to test if these systematic eye movements are face-specific or occur in case of other visual objects as well.

Differential effects of speech situations on mothers' and fathers' infant-directed and dog-directed speech: An acoustic analysis.

There is growing evidence that dog-directed and infant-directed speech have similar acoustic characteristics, like high overall pitch, wide pitch range, and attention-getting devices. However, it is still unclear whether dog- and infant-directed speech have gender or context-dependent acoustic features. In the present study, we collected comparable infant-, dog-, and adult directed speech samples (IDS, DDS, and ADS) in four different speech situations (Storytelling, Task solving, Teaching, and Fixed sentences situations); we obtained the samples from parents whose infants were younger than 30 months of age and also had pet dog at home. We found that ADS was different from IDS and DDS, independently of the speakers' gender and the given situation. Higher overall pitch in DDS than in IDS during free situations was also found. Our results show that both parents hyperarticulate their vowels when talking to children but not when addressing dogs: this result is consistent with the goal of hyperspeech in language tutoring. Mothers, however, exaggerate their vowels for their infants under 18 months more than fathers do. Our findings suggest that IDS and DDS have context-dependent features and support the notion that people adapt their prosodic features to the acoustic preferences and emotional needs of their audience.