AI Can Teach Us a Lot Scientists Say Cats’ Expressions Richer Than Imagined and Aim To Translate Them

SensorSpot/Getty Images

Artificial intelligence being used to unpick meanings behind vocal and physical cues of host of creatures. If an unexpected meow, peculiar pose, or unusual twitch of the whiskers leaves you puzzling over what your cat is trying to tell you, artificial intelligence may soon be able to translate.

Scientists are turning to new technology to unpick the meanings behind the vocal and physical cues of a host of animals. “We could use AI to teach us a lot about what animals are trying to say to us,” said Daniel Mills, a professor of veterinary behavioural medicine at the University of Lincoln.

Previous work, including by Mills, has shown that cats produce a variety of facial expressions when interacting with humans, and this week researchers revealed felines have a range of 276 facial expressions when interacting with other cats….Continue reading….

By: Nicola Davis Science correspondent

Source: ‘AI can teach us a lot’: scientists say cats’ expressions richer than imagined and aim to translate them

Critics:

It is worth distinguishing “animal language” from “animal communication”, although there is some comparative interchange in certain cases (e.g. Cheney & Seyfarth’s vervet monkey call studies). Animal language typically does not include bee dancing, bird song, whale song, dolphin signature whistles, prairie dogs, nor the communicative systems found in most social mammals.

The features of language as listed above are a dated formulation by Hockett in 1960. Through this formulation Hockett made one of the earliest attempts to break down features of human language for the purpose of applying Darwinian gradualism. Although an influence on early animal language efforts (see below), it is no longer considered the key architecture at the core of animal language research.

Animal language results are controversial for several reasons (for a related controversy, see also Clever Hans). In the 1970s, John C. Lilly was attempting to “break the code”: to fully communicate ideas and concepts with wild populations of dolphins so that he could share human and dolphin culture, history, and more. This effort failed. Early chimpanzee work was with chimpanzee infants raised as if they were human; a test of the nature vs. nurture hypothesis.

Chimpanzees have a laryngeal structure very different from that of humans, and it has been suggested that chimpanzees are not capable of voluntary control of their breathing, although better studies are needed to accurately confirm this. This combination is thought to make it very difficult for the chimpanzees to reproduce the vocal intonations required for human language.

Researchers eventually moved towards a gestural (sign language) modality, as well as keyboard devices laden with buttons adorned with symbols (known as “lexigrams”) that the animals could press to produce artificial language. Other chimpanzees learned by observing human subjects performing the task. This latter group of researchers studying chimpanzee communication through symbol recognition (keyboard) as well as through the use of sign language (gestural), are on the forefront of communicative breakthroughs in the study of animal language, and they are familiar with their subjects on a first name basis: Sarah, Lana, Kanzi, Koko, Sherman, Austin and Chantek.

Perhaps the best known critic of animal language is Herbert Terrace. Terrace’s 1979 criticism using his own research with the chimpanzee Nim Chimpsky was scathing and basically spelled the end of animal language research in that era, most of which emphasized the production of language by animals. In short, he accused researchers of over-interpreting their results, especially as it is rarely parsimonious to ascribe true intentional “language production” when other simpler explanations for the behaviors (gestural hand signs) could be put forth.

Also, his animals failed to show generalization of the concept of reference between the modalities of comprehension and production; this generalization is one of many fundamental ones that are trivial for human language use. The simpler explanation according to Terrace was that the animals had learned a sophisticated series of context-based behavioral strategies to obtain either primary (food) or social reinforcement, behaviors that could be over-interpreted as language use.

In 1984 Louis Herman published an account of artificial language in the bottlenosed dolphin in the journal Cognition.^[48] A major difference between Herman’s work and previous research was his emphasis on a method of studying language comprehension only (rather than language comprehension and production by the animal(s)), which enabled rigorous controls and statistical tests, largely because he was limiting his researchers to evaluating the animals’ physical behaviors (in response to sentences) with blinded observers, rather than attempting to interpret possible language utterances or productions.

The dolphins’ names here were Akeakamai and Phoenix. Irene Pepperberg used the vocal modality for language production and comprehension in a grey parrot named Alex in the verbal mode, and Sue Savage-Rumbaugh continues to study bonobos such as Kanzi and Panbanisha. R. Schusterman duplicated many of the dolphin results in his California sea lions (“Rocky”), and came from a more behaviorist tradition than Herman’s cognitive approach. Schusterman’s emphasis is on the importance on a learning structure known as equivalence classes.

However, overall, there has not been any meaningful dialog between the linguistics and animal language spheres, despite capturing the public’s imagination in the popular press. Also, the growing field of language evolution is another source of future interchange between these disciplines. Most primate researchers tend to show a bias toward a shared pre-linguistic ability between humans and chimpanzees, dating back to a common ancestor, while dolphin and parrot researchers stress the general cognitive principles underlying these abilities.

More recent related controversies regarding animal abilities include the closely linked areas of theory of mind, Imitation (e.g. Nehaniv & Dautenhahn, 2002), Animal Culture (e.g. Rendell & Whitehead, 2001),and Language Evolution (e.g. Christiansen & Kirby, 2003). There has been a recent emergence in animal language research which has contested the idea that animal communication is less sophisticated than human communication.

Denise Herzing has done research on dolphins in the Bahamas whereby she created a two-way conversation via a submerged keyboard.The keyboard allows divers to communicate with wild dolphins. By using sounds and symbols on each key the dolphins could either press the key with their nose or mimic the whistling sound emitted in order to ask humans for a specific prop.

This ongoing experiment has shown that in non-linguistic creatures sophisticated and rapid thinking does occur despite our previous conceptions of animal communication. Further research done with Kanzi using lexigrams has strengthened the idea that animal communication is much more complex than once thought.

Related contents:

The Animals Are Talking. What Does It Mean? – Language was long understood as a human-only affair.

The Faculty of Language: What Is It, Who Has It, and How Did It Evolve?”

Unity and diversity in human language”.

Wild Chimpanzees Inform Ignorant Group Members of Danger”.

Speaking Bonobo”. Smithsonian.

Multistability, cross-modal binding and the additivity of conjoined grouping principles”.

The emergence of hierarchical structure in human language”.

The Language Instinct: How the Mind Creates Language.

“Skunks, Skunk Pictures, Skunk Facts – National Geographic”.

Baboons Can Learn to Recognize Words”.

Monkey vocal tracts are speech-ready”.