9. Echolocation in Dolphins
403
presented underwater were dipped into a box 1.1m wide by O.6m deep by
1.0m high, with the front of the box facing the center of the pool consisting of O.32-cm thick black plexiglass. After several years of training, the
dolphin could do a visual-to-echoic and an echoic-to-visual matching to
sample correctly on the first trial for many novel pairs of objects. Pack and
Herman concluded that shape information registers directly in dolphin
perception through vision or through echolocation. Harley et al.'s (1996)
vision-echoic cross-modal results with familiar targets and both visionechoic and echoic-vision results with novel targets were not consistent with
those of Pack and Herman. Unfortunately, the study of Pack and Herman
did not include any acoustic measurements. The dolphin's echolocation
signals were not measured. The echoes from the various objects were not
measured. The effects of the surrounding "target" boxes were not measured. Although they claimed that the boxes were anechoic, they provided
no data to support that claim. Therefore, it is extremely difficult to evaluate the experiment of Pack and Herman. Target recognition cues may have
been entirely different than what the investigators suggested. Certain features of the target echoes may have allowed the dolphin to make accurate
cross-modal choices that may not have anything to do with the shape of
the targets. Therefore, until acoustic data of the targets are collected, this
experiment cannot be properly evaluated.
6. Summary
The echolocation system of dolphins has undergone millions of years of
refinement under evolutionary pressures and has emerged as a sophisticated and highly sensitive sensory mechanism. It is superior to any humanmade system for short ranges (around 100m) and in shallow waters
containing bottom clutter and false targets. The dolphin's ability to discriminate and recognize features of targets by echolocating is a characteristic that human-made sonar systems do not possess. Although we have
gained considerable knowledge of the capabilities and properties of the
dolphin echolocation system, there is much we still do not know and we
should pursue these gaps in our knowledge, making use of the most current
technology in computers and electronics in order to probe deeper into the
processes and mechanisms involved with echolocation.
Our understanding of dolphin echolocation has come about mainly
through experiments using targets that are totally alien to these animals,
in environments that may also be very unnatural, and that often require
animals to be constrained in a relatively fixed location by a hoop or some
other stationing device. The past approaches are justifiable given the difficulty of working with free-ranging dolphins. However, our knowledge of
how dolphins utilize echolocation in the wild is extremely poor. There are
many questions concerning how dolphins use their echolocation abilities in
403
presented underwater were dipped into a box 1.1m wide by O.6m deep by
1.0m high, with the front of the box facing the center of the pool consisting of O.32-cm thick black plexiglass. After several years of training, the
dolphin could do a visual-to-echoic and an echoic-to-visual matching to
sample correctly on the first trial for many novel pairs of objects. Pack and
Herman concluded that shape information registers directly in dolphin
perception through vision or through echolocation. Harley et al.'s (1996)
vision-echoic cross-modal results with familiar targets and both visionechoic and echoic-vision results with novel targets were not consistent with
those of Pack and Herman. Unfortunately, the study of Pack and Herman
did not include any acoustic measurements. The dolphin's echolocation
signals were not measured. The echoes from the various objects were not
measured. The effects of the surrounding "target" boxes were not measured. Although they claimed that the boxes were anechoic, they provided
no data to support that claim. Therefore, it is extremely difficult to evaluate the experiment of Pack and Herman. Target recognition cues may have
been entirely different than what the investigators suggested. Certain features of the target echoes may have allowed the dolphin to make accurate
cross-modal choices that may not have anything to do with the shape of
the targets. Therefore, until acoustic data of the targets are collected, this
experiment cannot be properly evaluated.
6. Summary
The echolocation system of dolphins has undergone millions of years of
refinement under evolutionary pressures and has emerged as a sophisticated and highly sensitive sensory mechanism. It is superior to any humanmade system for short ranges (around 100m) and in shallow waters
containing bottom clutter and false targets. The dolphin's ability to discriminate and recognize features of targets by echolocating is a characteristic that human-made sonar systems do not possess. Although we have
gained considerable knowledge of the capabilities and properties of the
dolphin echolocation system, there is much we still do not know and we
should pursue these gaps in our knowledge, making use of the most current
technology in computers and electronics in order to probe deeper into the
processes and mechanisms involved with echolocation.
Our understanding of dolphin echolocation has come about mainly
through experiments using targets that are totally alien to these animals,
in environments that may also be very unnatural, and that often require
animals to be constrained in a relatively fixed location by a hoop or some
other stationing device. The past approaches are justifiable given the difficulty of working with free-ranging dolphins. However, our knowledge of
how dolphins utilize echolocation in the wild is extremely poor. There are
many questions concerning how dolphins use their echolocation abilities in