Item Description

Show full item record

Title: Reporting on relationships between symbolically-named objects by a dolphin (Tursiops truncatus) 
Author: Forestell, Paul H
Date: 1988
Abstract: Three experiments were carried out with an Atlantic bottlenosed dolphin (Tursiops truncatus, Montague, 1821) named Akeakamai (Ake). Ake was a subject. in a series of long-term language studies, and had been shown capable of reporting on the presence or absence of named objects in response to both imperative and interrogative sentences. The present studies investigated Ake's definition of presence and absence, the relative effects of number of objects and length of delay on memory for the presence or absence of named objects, and the ability to report on relationships between objects. Testing took place in one of two circular, outdoor, sea-water tanks each 15.2-m in diameter, and 1.5-m in depth. The outside of the tank walls rose .9-m above the laboratory's concrete deck. The two tanks were connected by two parallel channels, separated by a 25 cm thick wall level with the top of the outside walls. Ake lived and was tested in either of the two tanks, along with a second dolphin named Phoenix. In Experiment 1, objects were shown to Ake and then a) either thrown into the tank or returned to a tankside assistant outside the tank wall, b) thrown behind the wall of the channel between the tanks where Ake could not see them, or c) thrown completely across the two channels so they landed out of the tank on the other side. Ake was then asked, through a gestural language, whether or not one of six named objects was present in the tank. The critical determinant of whether or not Ake reported an ,object as present appeared to be whether she had seen it pass over her head, as if being thrown into the tank. The results demonstrated that Ake did not need to see or hear the objects at the time of the sentence in order to make a judgment about their presence or absence. In Experiment 2, a delay was introduced between presentation of the objects, and the gestural sentence. Ake was shown either one, two, or three objects, and then given a sentence after either 5, 10, or 15 seconds. The results showed that, within the delays tested, increasing the number of objects produced a marked decrement in accuracy of responding, while increasing the delay had no significant effect. A strong recency effect was found in both the 2-object and 3-object conditions, with more recently introduced items more accurately classified correctly as "present" than earlier-introduced objects. Performance decrement appeared to be a result of processes operating at the time each object was shown, and not as a result of processes operating during the retention interval. In Experiment 3, Ake was instructed to respond to two objects, either by fetching one object (the Direct Object) to another (the Indirect Object), or by putting one on top of the other. On some trials, both the Direct Object and Indirect Object were present in the tank when the sentence was given. On other trials, either the Direct Object, or the Indirect Object, or both were missing. When both named objects were present, Ake sometimes took the Direct Object to the Indirect Object, and sometimes took the Direct Object to the Yes Paddle. When the Indirect Object was missing, but the Direct Object was present, Ake took the Direct Object to the No Paddle. When the Direct Object was missing, or both objects were missing, Ake pressed the No Paddle in response to the signer's instruction. Transporting the DO to the Yes or No paddles in most cases accurately reflected the state of the 10, and was an untaught behavior developed spontaneously by Ake. The overall results demonstrated the dolphin's ability to maintain an inventory of item representations in memory, to make judgments about the status of named objects and report that status using arbitrary responses, and to report on the relationships between objects named in complex instructions.
Description: Typescript. Thesis (Ph. D.)--University of Hawaii at Manoa, 1988. Includes bibliographical references. Microfiche. xi, 122 leaves, bound ill. (some col.) 29 cm
URI: http://hdl.handle.net/10125/10205
Rights: All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner.

Item File(s)

Description Files Size Format View
Restricted for viewing only uhm_phd_8912131_r.pdf 3.061Mb PDF View/Open
For UH users only uhm_phd_8912131_uh.pdf 3.025Mb PDF View/Open

This item appears in the following Collection(s)

Search


Advanced Search

Browse

My Account

Statistics

About