Abstract
Macphail's (1985) famous "Null Hypothesis" stated that there are no qualitative differences in cognitive ability across vertebrate species, and that any differences in performance on tasks can be attributed to the influence of contextual variables. Chimpanzees have shown evidence of linear pattern recognition (Vauclair et al., 1983) and completion (Levere, 1966). However, no studies to date have investigated the ability of pigeons with a linear pattern recognition or completion task. Therefore, the aim of this thesis was to determine if pigeons would be able to complete a linear pattern task, in the form of a tic-tac-toe precursor.
Two experiments were conducted. In Experiment 1, six pigeons were trained on a 3 x 3 grid (tic-tac-toe board) to respond to the correct white dot which would complete a line of three red discs. There were a total of 18 configurations. The pigeons were first trained with four configurations and then transferred to the remaining 14 novel configurations. They were then trained on 4 more configurations and then transferred to the remaining 10. None of the pigeons performed well with the novel configurations on either of the transfer tests, therefore a second experiment was conducted.
In Experiment 2, sixteen pigeons were trained on the same task but with two fixed response (FR) groups, consisting of nine subjects in FR0 and seven in FR3. The experiment also included a change in stimulus size and shape. The change in size was due to pigeons being unable to view the top row of the configuration in the first experiment and the change in shape was to make the linear pattern more obvious. Pigeons in the FR0 group shared identical training and testing to pigeons in Experiment 1. The pigeons in the FR3 group underwent the same procedure, however prior to completing the experiment on FR3 the pigeons had two additional training phases of FR1 and FR2. Three of the seven pigeons in the FR3 group showed strong evidence of transfer to novel configurations, thus showing an understanding of linear pattern completion. The results of the second experiment also demonstrated a stark difference in acquisition trials between the FR0 and FR3 pigeons with the FR0 pigeons requiring substantially more training than the FR3 pigeons.
The impact of FR is well researched in comparative cognition and this thesis provides further evidence of its efficacy. In this study the contextual variable of FR plays an important role in pigeons’ ability to learn the underlying concept and their speed of completing the acquisition trials. Moreover, the results demonstrate that pigeons have no difference in ability to complete a linear pattern task when compared to chimpanzees and as such provides further evidence to support MacPhail's (1985) "Null Hypothesis".