Spatial sensitivity of midbrain lateral line units of the goldfish, <em>Carassius auratus</em>

Abstract

The mechanosensory lateral line system of fish responds to water motions and pressure gradients caused by biotic factors (e.g. prey, predators or conspecifics) or abiotic factors. Theoretical and neurophysiological data show that a stationary vibrating sphere creates a spatial stimulus pattern along the lateral line. This spatial stimulus pattern contains information about the stimulus, i.e. sphere position and sphere vibration direction, and is represented in primary lateral line afferents. Up to now it is not clear how the excitation patterns of primary lateral line afferents are processed along the ascending lateral line pathway. <br /> In the medulla, the first nucleus of the ascending lateral line pathway, no representation of the position and/or vibration direction of a stationary vibrating sphere was found. The present study examines if lateral line units in the second nucleus of the ascending lateral line pathway, the torus semicircularis, encode the position and/or vibration direction of a stationary vibrating sphere. To do so, a stationary vibrating sphere (dipole, diameter 10 mm, frequency 50 Hz, duration 500 ms) was positioned at different positions along the rostro-caudal axis of a fish. Sphere vibration directions were 0° (parallel to the long axis of the fish), 45°, 90° (perpendicular to the long axis and the dorso-ventral axis of the fish) and 135°. The maximum peak-to-peak sphere displacement was 124--237 µm. The minimum distances between the fish and the surface of the sphere were 5, 10, 15, 20, 30 or 40 mm.<br /> A total of 98 unimodal toral lateral line units were recorded. 16 of these units could be examined with the complete stimulation protocol, that lasted up to four hours. This study examines only the responses of these 16 units. The recording sites of 12 out of these 16 units were verified, they were located in the ventrolateral nucleus of the Torus semicircularis, which is known to receive only lateral line input.<br /> The ongoing activity of 14 units was less or euqal 0,44 Hz (mean plus/minus SD 0,44 plus/minus 0,45 Hz, median 0,32 Hz, range 0,01--1,40 Hz). Two units had mean ongoing activities of 12,37~Hz and 51,72~Hz. The responses of one unit were phasic, nine units responded phasic-tonic and six units showed tonic responses. 15 units responded with an increase in evoked neural activity at all sphere positions. One unit showed an increase in evoked neural activity or a decrease in evoked neural activity inhibition, depending on the sphere position.<br /> None of the examined toral lateral line units was space sensitive in terms of evoked spike rates or phase locking. In some units a small change (5 mm) in sphere position led to a significant change of the evoked frequency, the phase angle of the response and/or the phase-locking. Similar effects were observed for changes in the direction of sphere vibration. <br /> Three units showed siginficant phase-locking at multiple side by side sphere positions. Two of these units responded to only one half of a full wave cycle. One unit, however, responded to both halfs of a full wave cycle. <br /> Spatial excitation patterns of three units were independent from sphere vibration direction. The spatial excitation patterns of four units were sytematically and continuously displaced towards the snout of the fish with increasing sphere vibration angles. The spatial excitation patterns of nine units showed no systematic changes to changes of sphere vibration direction.<br /> In this study no toral lateral line units were found that single-handedly encode the position and/or vibration direction of a stationary vibrating sphere. I therefore assume that lateral line information on the position and vibration direction of a stationary vibrating sphere is encoded in a population code. If space sensitive lateral line units exist, that encode sphere position and vibration direction, they can possibly be found in the optic tectum (the next level of processing of lateral line information in the ascending lateral line pathway), where computed lateral line maps have been proven in other animals with a lateral line.