Dolphin Communication and Cognition

Dolphins are very intelligent”. Many people believe this but how do we know how intelligent dolphins are? It’s very difficult to objectively measure global intelligence in humans, especially when people come from different cultures, and even more difficult to study what animals know and understand (cognition). Studies must be very well designed in order to be meaningful; and how much more difficult is it to study marine mammals that have evolved to live in the very different environment of the sea where most of their lives are spent underwater?

It’s well known that cetaceans are social animals with wide vocal repertoires: Humpback whales singing long, complex and continually evolving songs, and Killer whale (Orca) pods that seem to communicate using their own ‘family’ dialects are two better known examples. In an often dimly lit, 3-dimensional underwater environment in which sight can be of little use, dolphins and porpoises are masters of using sound to navigate, find and catch prey, and communicate with each other: they can even use their SONAR to find fish hiding for safety under the sandy sea bottom. Although dolphins can see well in air and underwater if the visibility is good, taking instructions from alien humans with strange arms and legs and very expressive faces must be challenging, and yet, many dolphins readily learn this skill.

Research on dolphin behavior and cognition has been hampered by difficulties involved in communicating information between terrestrial scientists and their aquatic subjects, and vice versa. While marine mammals rely heavily on their evolved acoustic senses for navigation, finding prey and for communication, most instructions given by trainers and researchers to dolphins is in the visual mode in the form of hand cues, reflecting a pragmatic human preference. A whistle signal is usually used to signal a correct response to an instruction before a fish reward is given to the animal subject, and, though a very effective training tool, a simple whistle not convey complex information.

Dolphins are adept at using a large range of sounds to obtain information about their environment and to communicate with con-specifics. In order to study dolphin behaviour and cognition in greater depth, we plan to utilise this ability to enable 2-way communication with human researchers in a dolphin-friendly manner. MMRL has developed a novel computerised interface which can transmit (underwater) synthesised sounds similar to dolphin whistles, and also receive and rapidly process the dolphins’ vocal responses. By training the animals that certain sounds are paired with objects or actions, or people and other dolphins, and training them to mimic these sounds, we hope to be able to communicate with our dolphin subjects.

We are also studying the way in which dolphins encode information within their natural vocalisations for transmission to other dolphins, in particular those within their own social group.

Very little scientific work of this nature has been carried out, the most notable being a substantial series of studies by Professor Louis Herman and Dr Adam Pack at The Dolphin Institute, Hawaii, USA which showed that two Atlantic Bottlenose dolphins were able to understand complex instructions given in the form of novel five word sentences either by using hand-signed instructions or acoustic instructions. Our approach is similar to that used by Drs Herman and Pack, but the computerised system we have developed is theoretically capable of taking the work further. In addition, this is the first ever study of it’s kind with Indo-Pacific Humpback dolphins (Sousa chinensis), a species found in Singapore waters, particularly around the Southern Islands. Indo-Pacific Humpback dolphins are distributed in coastal waters from South Africa to southern China and Northern Australia and relatively little is known about their behaviour compared with the Bottlenose dolphin (Tursiops truncatus).

Key Objectives of our Research

  • Investigate whether dolphins can be trained to use artificial dolphin friendly sounds processed via a computerised interface to engage in 2-way communication with humans.
  • Study the way in which dolphins encode information within their natural vocalizations for transmission to con-specifics.
  • Incorporate the findings of these studies into an improved system for 2-way underwater communication with dolphins.
  • Use this approach to study dolphin behaviour and cognition.

  • We predict that such a communication system would enable the following:
  • More rapid and effective training;
  • Faster and longer-range communication when desired;
  • Facilitation of the quantity and quality of information exchanged;
  • Improved husbandry and veterinary work;
  • A sense of empowerment for the dolphins who learn to use the acoustic interface though appropriate reinforcement of the behaviour

  • We are working in collaboration with Underwater World Singapore Pte. Ltd. at their Dolphin Lagoon facility on Sentosa. Our research subjects are two male Sousa chinensis, 'Pet' and 'Splash'. Pet is in his early twenties and Splash is a 5 year old juvenile born at Dolphin Lagoon.



    While our work is still in its early stages, we are pleased to have the opportunity to work with two very keen dolphins who participate readily in our experimental sessions and have surprised us with their capabilities. Since there is no previous background work with which to compare our findings, we have the great pleasure of exploring the unknown potential of this species – a rare privilege!