The direct or indirect human artificial selection process made the dog bark as we know, said Csaba Molnar, formerly an ethologist at Hungary’s Eotvos Lorand University. Molnar’s work was inspired by a simple but intriguing fact: Barking is common in domesticated dogs, but infrequent if not downright absent in their wild counterparts. Wild dogs yip and squeal and whine, but rarely produce the repetitive acoustic percussion that is barking. Many people had made that observation, but Molnar and his colleagues were the first to rigorously investigate it.

The results show that dog barks carry a wealth of information, says Molnár. “In the past, scientists thought that dog barks originated as a by-product of domestication and so have no communicative role,” he says. “But we have shown there are contextual differences.” Molnár thinks that selection by humans could explain why an artificial neural network is better at recognising that a dog is fighting than playing. Domestic dogs have always been used for their vigilance, and so it may be important for humans to quickly identify when a dog is barking at a stranger. “In this context, every dog barks singularly,” says Molnár. “But in a play situation, there has been no strict selection for creating a uniform bark among dogs, so each has its own individual style.”

New Scientist, Read more…

While the computer correctly recognised the emotional state of 43% of dogs, humans did almost as well with 40%. But the author of the research – Csaba Molnar, from Eotvos University in Budapest – says the software can be improved, and told the BBC it may have applications for analysis of human communication. “I would say that we proved there are very strong contextual differences between the barks, but that very long further work is needed to determine which emotional states and which characteristics belong to each (different breed). He added: “In the future we can use this software for any other vocal or any other signal categorisation.”

What are dogs really saying when they bark? A team of researchers (led by ethologist Csaba Molnár from Eötvös Loránd University in Budapest) set out to find out. Using specially designed software, the group studied the acoustic qualities of more than 6,000 barks from 14 Hungarian sheepdogs in six different situations: alone, engaged in a dogfight, in an encounter with a stranger, on a leisurely stroll, exercising with a ball, and playing. The team recorded the pooch calls on tape, which was then transferred to and digitalized on a computer, which coded, classified and evaluated the individual barks. The software correctly linked the bark with the situation in 43 percent of cases. The best recognition rates came during dogfights and stranger confrontations; the software had the least success in trying to analyze the pups at play.

Scientific American, Read more…

Can you tell the difference between a dog bark that says “I’m lonely, don’t leave me!” and one that warns “Don’t get close, I might bite!”? Researchers in Hungary have found that we humans have a remarkable ability to categorize various types of dog vocalizations and understand their emotional content—in effect, to “speak canine.” Try it yourself, and learn what this inter-species communication might mean about the evolution of dogs.

Our ability to understand dogs may peak when we’re around 10 years of age, with preadolescents possessing a natural talent for deciphering dog barks, according to new research. The study, published in the journal Applied Animal Behaviour Science, could help to strengthen the theory that there’s a universal animal “language.” The connections allow us to understand rudimentary meanings of many animal calls, especially those made by mammals. Study co-author Csaba Molnár told Discovery News that the new research proves “that basic understanding of barks is an inherited trait in humans, and learning can only slightly improve this ability.”

Discovery Channel News, Read more…

Young children often misinterpret a dog’s snarling face as a smile, which may be one reason why kids suffer a high number of dog bite injuries. But youngsters 6 to 10 years of age have no trouble understanding the threat in a dog’s aggressive bark, researchers report in the current issue of Applied Animal Behaviour Science. The scientists tested 30 children and ten adults, asking them to judge a dog’s emotions by listening to its barks in three different situations: alone, facing a stranger at a gate, and when playing.

Science, Read more…

Scientists at Eötvös Loránd University in Budapest, Hungary, recorded more than 6,000 barks uttered by 14 individual Mudis, or Hungarian sheepdogs, in six different situations — “stranger,” “fight,” “walk,” “alone,” “ball’ and “play.” They then fed the sounds to a computer program, which was able to correctly categorize the barks 43 percent of the time — far from perfect, but much better than the 16.7 percent rate that would have arisen by chance. The computer had the best results with the “fight” and “stranger” contexts, the worst with “play.” It was also able to pick out individual dogs just by the sound of their barks slightly more than half the time — again, much better than chance. “Since we have no reasons to say that Mudis are special among other dog breeds, I am pretty sure that this method for categorizing barks could work in other dog breeds’ barks as well,” study leader Csaba Molnár told London’s Daily Telegraph.

Fox News, Read more…

Ungarische Verhaltensforscher haben ihrem Computer “Hundesprache” beigebracht. Ihr Computerprogramm kann das Gebell einzelner Hunde besser auseinanderhalten als der Mensch und verschiedene Laute einzelnen Situationen zuordnen. So kann der Computer einigermaßen erkennen, ob der bellende Vierbeiner nun eher Gassi gehen, kämpfen oder spielen will, berichten Csaba Molnar und sein Team von der Budapester Eötvös-Universität im Fachblatt “Animal Cognition”

Die Welt, Read more…

Einen solchen Übersetzer von Hunde- in Menschensprache wollen ungarische Verhaltensforscher jetzt entwickelt haben. Die von Csaba Molnar und seinem Team an der Budapester Eötvös-Universität ausgearbeitete Software hat mehr als 6000 Bell-Laute von 14 Mudis analysiert, einer ungarischen Hirtenhund-Rasse. Ihr Computerprogramm könne das Gebell einzelner Hunde besser auseinanderhalten als ein Mensch, schreiben die Forscher im Fachblatt “Animal Cognition”. Die Trefferquote habe 52 Prozent betragen, für Menschen sei die Aufgabe nahezu unlösbar. Zudem könne das Programm in gewissem Grade das Gebell auf sechs für den Alltag der Tiere bestimmende Situationen zurückführen: “Fremder”, “Kampf”, “Gassi gehen”, “allein”, “Ball” und “Spielen”. Bei einem ersten Versuch habe die Software das Gebell in 43 Prozent der Fälle richtig eingeordnet, hieß es. Am besten gelang die Zuordnung in den Situationen “Kampf” und “Fremder”, am schlechtesten in der Situation “Spielen”.

Spiegel, Read more…

De hecho, la investigación sugiere que algunos de estos sonidos son comprensibles para los demás perros, por lo que les servirían para comunicarse entre ellos y quedaría descartada la vieja teoría de que los perros sólo aprendieron a ladrar para imitar el lenguaje de las personas. Sin embargo, también en esto se nota el contacto con los humanos: cuando los perros juegan entre ellos, cada cual tiene su propia personalidad y su ladrido es claramente diferenciable por el ‘software’. Por el contrario, los ladridos emitidos ante la presencia de una persona extraña son casi siempre idénticos, independientemente del perro individual del que procedan. El estudio ha sido realizado por los equipos de Csaba Molnár, de la Universidad Eötvös Loránd (Hungría) y Frédéric Kaplan, de la Escuela Federal de Lausanne (Suiza). Sus resultados se acaban de publicar en la revista ‘Animal Cognition’.

El Mundo, Read more…