An exploratory study on dogs’ vocalizations towards their owner and food in an unsolvable task

Automatic classification of dog barking using deep learning

Barking and other dog vocalizations have acoustic properties related to emotions, physiological reactions, attitudes, or some particular internal states. In the field of intelligent audio analysis, researchers use methods based on signal processing and machine learning to analyze the digitized acoustic signals' properties and obtain relevant information. The present work describes a method to classify the identity, breed, age, sex, and context associated with each bark. This information can support the decisions of people who regularly interact with animals, such as dog trainers, veterinarians, rescuers, police, people with visual impairment. Our approach uses deep neural networks to generate trained models for each classification task. We worked with 19,643 barks recorded from 113 dogs of different breeds, ages and sexes. Our methodology consists of three stages. First, the pre-processing stage prepares the data and transforms it into the appropriate format for each classification model. Second, the characterization stage evaluates different representation models to identify the most suitable for each task. Third, the classification stage trains each classification model and selects the best hyperparameters. After tuning and training each model, we evaluated its performance. We analyzed the most relevant features extracted from the audio and the most appropriate deep neural network architecture for that feature type. Even if the application of our method is not ready for being used in ethological practice, our evaluation showed an outstanding performance of the proposed method, surpassing previous research results on this topic, providing the basis for further technological development.

Relationship between engagement with the impossible task, cognitive testing, and cognitive questionnaires in a population of aging dogs

Introduction

The aim of this study was to evaluate the engagement of aging dogs with a cognitively challenging and potentially frustrating task (the impossible task). Based on previous observations, we predicted that dogs showing signs of cognitive impairment in other cognitive tests and owner-completed questionnaires would show reduced engagement with the task.

Methods

In this task, dogs were shown a piece of food in a clear container that they could not open; time spent interacting with the container and the experimenter was measured. While the impossible task has not been used as a measure of frustration, the parameters of the test design creates a potential frustrate state, making this assessment appropriate. Thirty-two dogs enrolled in a longitudinal aging study participated in the study. Owners were asked to complete two cognitive dysfunction screening questionnaires (Canine Dementia Scale [CADES] and Canine Cognitive Dysfunction Rating Scale [CCDR]) as well a questionnaire assessing general frustration. Dogs participated in multiple measures of cognitive function as well the impossible task.

Results

Latency to disengage from the impossible task was faster for dogs with higher total (more impaired) CADES (p = 0.02) and total CCDR (p = 0.04) scores. Latency to disengage also correlated with decreased performance in cognitive tests observing social cues (p = 0.01), working memory (p ≤ 0.001), spatial reasoning and reversal learning (p = 0.02), and sustained attention (p = 0.02).

Discussion

The high correlation with several cognitive measures and the ease of administration of this test makes it a useful tool in evaluating canine cognitive dysfunction syndrome, however it is unclear if increased frustration or other cognitive processes are contributing to the observed changes.