Interspecies communication

Humans are not the only species with complex communication systems. Animals like prairie dogs, whales, and bats use sounds, visual signals, touch, and chemical cues (pheromones) to convey information. For example, prairie dogs emit distinct yip sequences to warn others about different predators such as coyotes or hawks. Con Slobodchikoff, founder of Zoolingua, argues that decoding animal communication could reveal that animals are sentient beings with thoughts, hopes, and dreams. This research aims to bridge the gap between human and animal understanding, showing that other species may perceive the world in ways we are only beginning to grasp. The goal is not just to translate their sounds but to uncover the inner lives of animals, which could reshape how society views their rights and treatment.

Deciphering animal communication is a formidable task due to several obstacles. First, many animals use multiple forms of communication beyond sound, including body language, pheromones, and touch, which humans may not easily perceive. Second, identifying which animal made a sound in a group requires precise tools and methods. For instance, studying deep-diving whales or subterranean mole rats is difficult because they are hard to observe directly. Additionally, without a multispecies Rosetta stone—a universal key to translate all animal languages—researchers risk misinterpreting sounds. Yossi Yovel, a neuroecologist, warns that AI models might confuse similar sounds with different meanings, leading to inaccurate translations. These challenges highlight why progress in this field is slow and complex.

Over the past 50 years, technological advancements have significantly improved scientists' ability to study animal communication. Devices like hydrophones, which are underwater microphones, capture whale songs in the ocean, while biologgers—tiny wearable devices—record the vocalizations of bats and other animals. In the last decade, artificial intelligence (AI) has further enhanced this research. AI can analyze vast amounts of audio data, detect different calls, identify which animal made each sound, and identify patterns in communication. For example, researchers at the Earth Species Project (ESP), a nonprofit founded in 2018, used machine learning to analyze the vocalizations of carrion crows and determine which bird produced each call. ESP is now developing NatureLM, a large language model trained on human speech, music, and animal sounds, with the goal of interpreting various species' vocalizations.

While AI shows promise in decoding animal communication, significant uncertainties remain. Sara Keen, a behavioral ecologist at ESP, suggests that advanced AI models like NatureLM could eventually translate animal chatter by finding relationships between sounds and meanings, similar to how large language models like ChatGPT process human language. However, Yossi Yovel cautions that these models might rely too heavily on acoustic similarities, leading to errors such as grouping together sounds that have different meanings. Additionally, achieving a two-way conversation between humans and animals is still theoretical and may never happen. Con Slobodchikoff remains optimistic, arguing that many animals have brain structures similar enough to humans that their communication could eventually be understood. For now, the field is still in its early stages, with much work left to do.