Formal Presentations - Tuesday
Abstracts

​Watch Video ​(Login required)

KSHAMENK, CAN YOU CHOOSE?

Virginia Cardozo Uz
Natalia Díaz Buldurini

Mundo Marino

Full Abstract

During the last decade, we have witnessed advances in cognitive behavior conditioning. Additionally, we have been amazed by the cleverness of different trainers around the world who have taught their animals to choose their own reinforcement or select the trainers with whom they wish to interact with. We have seen various dog breeds communicate using audio buttons to indicate commands such as "walk," "cuddles", or "food." This inspired us to try to raise the idea of "consent training" to a higher level.

Kshamenk, the only Orcinus orca under human care in Latin America, presented us with a challenge. Among the variety of sessions in his daily routine, we observed on many occasions, that he did not respond with much enthusiasm. Sometimes, if we provided a relationship session, he seemed too energetic for a low-intensity proposal, and the other way around. Cognitive proposals were sometimes highly successful, while at others times, he seemed less participative, it was similar with environmental enrichment interactions. As a result, we had to adjust the proposals during the sessions, adapting them to his behavior, which led us to ask: Could Kshamenk previously choose his own sessions? If he could somehow indicate his preference, at the moment we will interact with him, it would improve his motivation, guarantee the success of the proposal, and would strengthen our relationship.

The method of communication we chose to achieve this goal was visual association. We assigned a letter and a color to his five main interactions, choosing a green "E" for environmental enrichment sessions, a pink "V" for relationship sessions, a yellow "L" for exercise sessions, a black "CO" for cognitive sessions, and a red "S" for gating to the largest environment of the circuit. We began with visual associations and immediate response: the
indicated letter was held in front of him for several seconds, and then the corresponding session was carried out. This visual association was repeated with each letter for two months. When we noticed that Kshamenk, upon observing the letter, anticipated the S.D of the session he would have, we understood that the association period had concluded, and we proceeded to the first step in the selection process.

We started by presenting him two different letters. Initially, he touched one of them, and he immediately received the session represented by that letter. This process was repeated in different sessions over three months, changing the letters' positions and observing the type of response he showed during the chosen interaction. When we detected that his choice was not schematic, meaning he did not always choose the same place or letter, we added a third, and then a fourth. At first, we detected that Kshamenk did approach a letter and receive that session, but we could not asseverate that he was truly remembering the meaning of each letter, nor that he was “choosing”. However, one day, we noticed that he began avoiding a particular type of session represented with “CO” letter, and if it was proposed outside the "Choose" sessions, his response was of low quality as well. This was the first step in reinforcing the idea of "choice" because not choosing something is, precisely, the first step in making a choice. We used that particular letter to encourage Kshamenk to make deliberate choices, clearly showing that he avoided it, until in one session he finally "chose" it, and his response was very satisfactory, indicating that his choice was not random.

Although it is very difficult to determine with certainty whether he is truly expressing a desire, or accessing a subsequent proposal, we assumed that his attitude would provide the answer. As we progressed with the "Choose" sessions, we noticed an improvement in Kshamenk's quality of response. His motivation is a key indicator that the beginning of the path to “Choose” is halfway completed.

“SOLVING SEELEY”:
A DIVE INTO PROBLEM SOLVING STEREOTYPIC BEHAVIORS

Sierra N. Spears

Grizzly & Wolf
Discovery Center

Full Abstract

Grizzly bears (Ursus arctos horribilis) removed from the wild due to food conditioned behaviors are assessed by multiple government agencies for if they’ll be a good fit for a captive placement. Many young individuals are considered to be the best to be placed based on their age and resiliency they can build at that young age while growing up in a captive environment. Keeping age in mind, there have been instances of individuals developing stereotypic behaviors within the first five years of their lives at the Grizzly & Wolf Discovery Center (GWDC).

This abstract takes a closer look at Seeley, a five year old Grizzly bear, who developed a unique jumping and head swinging stereotypic behavior and what steps we took to mitigate the behavior. We started by assessing Seeley’s individual history paired with her natural history and included the history we had access to in regards to her entire lifetime at the GWDC. We immediately began collecting data in a way that was realistic for our staff and in return gave us the best representation of what Seeley’s day-to-day looks like during the working day. A behavior bank was created and her current observed behavior was recorded once every hour. Once a baseline of data was established (see graph “Seeley Stereotypic Behavior Baseline”), we were able to reassess her behavior as a team and implement changes based on when and how often her
stereotypic behavior showed up. This included: changing her shifting schedule, changing what kinds of enrichment we were offering, introducing additional training sessions, and training an incompatible behavior. As a team, we went back to the drawing board multiple times. By continuing to collect data throughout the entire experience, we gained a better knowledge of
exactly when we needed to provide Seeley with something stimulating or a different experience that will allow her to work through her stereotypic behavior (see graph “Seeley Stereotypic Behavior July 2024). We identified areas that we were insufficient in meeting her needs and were able to rectify some of this through consistent behavior change of not only encouraging appropriate behaviors from Seeley herself, but also the staff who work directly with her.
Through working as a team and implementing consistent changes, we were able to see a significant decrease in Seeley’s stereotypic behavior during the times in which it had previously been most prevalent. This paper will discuss the challenges and mistakes our team made, the data we collected, and the welfare improvements we were able to make based on data collected by
keepers. Data collection is ongoing and our team continues to make necessary changes based on the information we are able to collect.

UTILIZING BEHAVIORAL CONDITIONING
TO ASSESS HEARING SENSITIVITY OVER TIME IN
KILLER WHALES (ORCINUS ORCA) UNDER MANAGED CARE

Kayla Nease

Seaworld San Diego

Full Abstract

Killer whales (Orcinus orca) are one of the most cosmopolitan marine mammal species, and they experience exposure to many anthropogenic noise sources. Noise impact estimates rely on audiograms, which plot a listener’s hearing sensitivity as a function of sound frequency (Branstetter, 2017). Previous audiograms have been published on this species using both physiological and behavioral methods (Branstetter, 2017), but few individuals have complete behavioral audiograms.

Some of the killer whales from the study by Branstetter et al. (2017) have undergone repeated behavioral hearing tests over nearly a decade. Some of these individuals had long gaps between initially learning this task and repeating it years later, while others were engaged in hearing tests more consistently. All mammals, including marine mammals, lose hearing sensitivity with age (presbycusis), particularly at higher frequencies. We therefore investigated hearing sensitivity across variable periods of time within a group of killer whales to determine if changes in hearing could be detected.

The trained behaviors for conducting the hearing test involved stationing precisely with the rostrum and waiting for a sound stimulus to be played before either emitting a vocal “raspberry” response or leaving the station to touch a paddle with the rostrum. The required behaviors have been trained using a variety of techniques including target pole recognition, operant conditioning, and the use of variable ratio reinforcement with variety (VRRV), least reinforcing scenario (LRS), and multiple trials with intermittent feedback.

This presentation will focus on the use of several training techniques involved in the behavioral training required for the hearing tests. By comparing ongoing data collection with the previously gathered data from the killer whale study group in Branstetter et al. (2017), the presentation will showcase how behavioral audiometry has been invaluable in understanding the sensory capabilities of killer whales under managed care and is useful in determining how hearing sensitivity changes with age. These changes are important for maintaining effective training tools/techniques for individual whales, and overall welfare quality, as well as benefiting the conservation of wild killer whales.

Let’s Get Physical… Therapy:
Training a Geriatric Pacific Walrus, (Odobenus rosmarus divergens)
to Voluntarily Participate in Physical Therapy Sessions

Benjamin M. Matsunaga

SeaWorld Orlando

Full Abstract

​Physical therapy has long been utilized to address and alleviate musculoskeletal ailments in both human and animal populations. Recently, its application has become more prevalent within the zoological community. At SeaWorld Orlando, a diverse range of physical therapy modalities, including: cold laser treatment, stretching, and massage therapy, are utilized to enhance the wellbeing of the animals in care. This practice is well established at our facility, where species such as Sea Lions (Zalophus californianus) and Walruses (Odobenus rosmarus) regularly participate in these therapeutic regimes. Slowpoke, a Pacic Walrus (Odobenus rosmarus divergens), is the oldest residing walrus not just at SeaWorld Orlando’s Sea Lion and Otter Stadium, but in human care. Staff initially observed a refusal by Slowpoke to offer certain behaviors, which was believed to be attributed to her being of geriatric age as well as her arthritis. Consequently, a specialized team was formed to administer a targeted physical therapy program aimed at maintaining consistency in husbandry behaviors, preventing further regression, and improving welfare scores. The experience with Slowpoke's physical therapy program underscores the critical role of various training techniques in improving the quality of life for animals under human care. As the zoological community continues to advance in animal welfare, the lessons learned from Slowpoke's results offer promising insights for future endeavors in animal husbandry, inspiring further exploration and refinement of physical therapy techniques.

​​IT'S A GAMER:
ADAPTING ENCLOSURE VIDEO ENRICHMENT FOR NAVY DOLPHINS

Maximillian Du
Nicole Sill

Stanford University
Navy Marine Mammal
Program

Full Abstract

The Enclosure Video Enrichment (EVE) system at the U.S. Navy Marine Mammal Program is successfully providing cognitive enrichment and welfare enhancement to the Navy’s California sea lions (Winship et al., 2024). Introduction of EVE to Navy dolphins required adjustments to the controller due to anatomical and physiological differences between species. Our initial design combined a joystick with a detachable bite plate—a typical piece of Navy dolphin equipment. But a safer controller was needed for unsupervised enrichment sessions. Our next prototype controller used pairs of infrared sensors to create a playing field within the dolphin’s enclosure such that cursor movement was linked to the dolphin breaking different light beams. Six months of working with this controller established the behavioral foundation for the dolphins to operate “non-contact” sensors, but the design presented both technical and training hurdles that were hard to overcome. We shifted to a much better option: a computer vision controller. A standard web-cam focused on a pre-determined location in the enclosure was combined with object-tracking software. The dolphins’ previous training with the infrared sensors generalized easily to the computer vision paradigm. Like before, the dolphin is required to station in front of the web-cam, and the software translates the location and movement of the dolphin in the area to create cursor movement on a large screen. However, the computer vision software’s robust image processing tools (Bradski, 2000) more reliably tracks the position of the dolphin’s head with less latency, providing more consistent gameplay. This controller made our training goals achievable for Navy dolphins in the exposed, bay environment where EVE sessions occur. The computer vision controller is supporting new training milestones for them, and its reliability and low-cost hold promise for enhancing cognitive wellbeing more broadly for dolphins in zoos and aquariums.

ENHANCING COGNITIVE TRAINING IN PACIFIC BOTTLENOSE DOLPHINS
THROUGH AUDITORY STIMULI

Jose Luis Dorantes
Enrique Parra

Cabo Dolphin Center

Full Abstract

Objective: This project aims to enhance the cognitive behaviors of Pacific bottlenose dolphins (Tursiops truncatus) by utilizing auditory stimuli in training programs, particularly focusing on improving motivation and mental stimulation.