Home>Psychology>Thesis
Auditory-visual matching and exclusion in a California sea lion.
Thesis information
Author:
Lindemann, Kristy L.
Advisor(s):
Wilson, Margaret
Degree:
Ph.D.
School:
University of California, Santa Cruz.
Year:
2007Full Abstract
Problem solving requires that animals be able to integrate information to make sense of their environment. I investigated the ability of an experimentally experienced, female California sea lion, named Rio, to form transitive relations across the sensory modalities in order to understand the mechanisms that underlie the many complex behaviors observed in these amphibious marine mammals both in captivity and in the wild. Prior to the present study, Rio had a long history of discrimination learning using a matching-to-sample paradigm in successful attempts to demonstrate equivalence class formation within the visual modality. In general, equivalence class formation occurs when stimulus members of a class become substitutable.;In the initial experiments of the current study associative transitivity was investigated across sensory modalities. Transitivity is a type of logic rule such that if one has learned that A is related to B and that B is related to C then transitivity would be demonstrated by the emergence of a relation between A and C. To test Rio for the ability to use a transitive logic rule across the senses she was trained by trial and error, using class specific food reinforcement, to relate an acoustic stimulus with one member from each of two previously established visual classes labeled "letters" and "numbers". Correct responses to the "letter" class were reinforced with a capelin fish and correct responses to the "number" class were reinforced with a herring fish. Once the auditory-visual relations were formed Rio was tested to determine if untrained transitive relations would emerge between each of the acoustic cues and the remaining members of each visual class. The results showed that Rio immediately solved novel transfer problems, with new transitive relations emerging between acoustic and visual stimulus members that were separately related to a common visual mediating stimulus. Rio's ability to pass a cross-modal transitivity test allows for a more complete understanding of how this cognitive ability may aid a sea lion in adapting to its environment.;Once it was demonstrated that the sea lion was capable of learning auditory-visual associations by trial and error the next set of experiments were run to investigate the learning strategy of exclusion. Exclusion is demonstrated when an individual, in the presence of an undefined sample, chooses an undefined comparison stimulus as opposed to a familiar defined comparison stimulus. Rio was taught new auditory-visual stimulus pairings with novel auditory stimuli and the visual stimuli from the previously formed classes. Responses to the two stimulus classes continued to be reinforced with differential fish rewards. Rio's acquisition of the new stimulus relations was assessed as well as the ability to demonstrate spontaneous learning outcomes in which novel problems were presented to her that could not be solved by the use of an exclusion strategy. Rio did not demonstrate errorless exclusion performances during the acquisition phase, but her learning outcome performance was significantly higher than expected by chance and generally not different from performance on baseline trials. This finding illustrates that training associations with an auditory-visual exclusion procedure can lead to successful and spontaneous cross modal learning outcome performances.;The last experiment in my study concerned the role of the food reinforcer. Rio's performance on the auditory-visual matching task in the absence of class specific food reinforcement was assessed. Two experimental tests were run with a homogenous food reinforcer rather than the heterogeneous reinforcement used previously. Therefore one experimental set was run with capelin fish serving as the only food reinforcer and then afterwards a second experimental set was run with herring fish serving as the only food reinforcer. This allowed me to investigate Rio's performance with both food types. Based on Sidman's revised (2000) theory of equivalence in which the reinforcer is considered a class member, I predicted a decrement in Rio's performance with the homogenous reinforcement followed by a rapid increase back to baseline performance levels. The initial decrement would indicate the merging of the two classes due to the common stimulus of the reinforcer and the sharp rise back to baseline levels would be indicative of Rio learning that the food type is irrelevant to the class distinction. Contrary to this hypothesis, Rio showed no decrement in performance in either test indicating that although the differential outcomes probably played an important role during the acquisition of associations, it was no longer a relevant factor for classification.;In summary this research demonstrated the ability of a nonhuman to use a transitive logic rule across sensory modalities as well as the flexibility to use different learning strategies and to perform successfully in the presence or absence of differential outcomes. These cognitive capabilities may serve as the foundation for how sea lions respond appropriately in a number of novel situations that are relevant to their social and ecological environment.
Free Downloads