Biomedical subjects

M A MacIver

Publications and source records attributed to M A MacIver.

3 recordsLinked to original sources

PubMed2006

Sensory acquisition in active sensing systems.

A defining feature of active sensing is the use of self-generated energy to probe the environment. Familiar biological examples include echolocation in bats and dolphins and active electrolocation in weakly electric fish. Organisms that utilize active sensing systems can potentially exert control over the characteristics of the probe energy, such as its intensity, direction, timing, and spectral characteristics. This is in contrast to passive sensing systems, which rely on extrinsic energy sources that are not directly controllable by the organism. The ability to control the probe energy adds a new dimension to the task of acquiring relevant information about the environment. Physical and ecological constraints confronted by active sensing systems include issues of signal propagation, attenuation, speed, energetics, and conspicuousness. These constraints influence the type of energy that organisms use to probe the environment, the amount of energy devoted to the process, and the way in which the nervous system integrates sensory and motor functions for optimizing sensory acquisition performance.

Animals↗

PubMed2001

Prey-capture behavior in gymnotid electric fish: motion analysis and effects of water conductivity.

Animals can actively influence the content and quality of sensory information they acquire from the environment through the positioning of peripheral sensory surfaces. This study investigated receptor surface positioning during prey-capture behavior in weakly electric gymnotiform fish of the genus Apteronotus. Infrared video techniques and three-dimensional model-based tracking methods were used to provide quantitative information on body position and conformation as black ghost (A. albifrons) and brown ghost (A. leptorhynchus) knifefish hunted for prey (Daphnia magna) in the dark. We found that detection distance depends on the electrical conductivity of the surrounding water. Best performance was observed at low water conductivity (2.8 cm mean detection distance and 2 % miss rate at 35 microS cm(-)(1), A. albifrons) and poorest performance at high conductivity (1.5 cm mean detection distance and 11 % miss rate at 600 microS cm(-)(1), A. albifrons). The observed conductivity-dependence implies that nonvisual prey detection in Apteronotus is likely to be dominated by the electrosense over the range of water conductivities experienced by the animal in its natural environment. This result provides the first evidence for the involvement of electrosensory cues in the prey-capture behavior of gymnotids, but it leaves open the possibility that both the high-frequency (tuberous) and low-frequency (ampullary) electroreceptors may contribute. We describe an electrosensory orienting response to prey, whereby the fish rolls its body following detection to bring the prey above the dorsum. This orienting response and the spatial distribution of prey at the time of detection highlight the importance of the dorsal surface of the trunk for electrosensory signal acquisition. Finally, quantitative analysis of fish motion demonstrates that Apteronotus can adapt its trajectory to account for post-detection motion of the prey, suggesting that it uses a closed-loop adaptive tracking strategy, rather than an open-loop ballistic strike strategy, to intercept the prey.

Animals↗

PubMed2000

Body modeling and model-based tracking for neuroethology.

The accurate tracking of an animal's movements and postures through time has broad applicability to questions in neuroethology and animal behavior. In this paper we describe methods for precision body modeling and model-based tracking of non-rigid animal movements without the use of external markers. We describe the process of obtaining high-fidelity urethane casts of a model organism, the weakly electric knifefish Apteronotus albifrons, and the use of a stylus-type 3-D digitizer to create a polygonal model of the animal from the cast. We describe the principles behind markerless model-based tracking software that allows the user to translate, rotate, and deform the polygon model to fit it to digitized video images of the animal. As an illustration of these methods, we discuss how we have used model-based tracking in the study of prey capture in nocturnal weakly electric fish to estimate sensory input during behavior. These methods may be useful for bridging between the analytical approaches of quantitative neurobiology and the synthetic approaches of integrative computer simulations and the building of biomimetic robots.

Animals↗