Object representation in the bottlenosed dolphin (Tursiops truncatus): Integration and association of visual and echoic information

Representational processes were investigated in a young Atlantic bottlenosed dolphin (Tursiops truncatus) trained to perform a 3-alternative matching-to-sample task in which the stimuli were presented in different modality conditions. In one condition (Visual/Echoic), stimuli were presented in the water, accessible to vision and echolocation. In the second condition (Visual), the stimuli were presented in air, allowing visual, but preventing echoic access. In the third condition (Echoic), stimuli were presented in water, but the dolphin's eyes were covered with soft removable latex eye cups, preventing visual access.; After the dolphin was trained in the Visual/Echoic Condition, she immediately transferred to the single modality conditions without further training. Her choice accuracy was significantly above-chance in all conditions. In Experiment 2, the dolphin's ability to integrate visual and echoic information was assessed. Unfamiliar objects were presented to the dolphin in one or both of the single modality conditions until performance was about 70%. Then the dolphin was allowed to use both vision and echolocation to discriminate among the objects. Performance accuracy immediately rose to about 95% when she could use both mdalities, thereby indicating that the subject could simultaneously use visual and echoic information for object recognition.; In a third experiment, the conditions varied between the initial presentation of the sample and the presentation of the alternatives. The dolphin was able to match familiar objects in the cross-modal conditions, although her performance accuracy varied depending on the nature of the stimuli. Her ability to match cross-modally suggests that a dolphin can associate stored object information received through different modalities.