The independence of eye movements in a stomatopod crustacean is task dependent

Abstract

Stomatopods have an extraordinary visual system, incorporating independent movement of their eyes in all three degrees of rotational freedom. In this work, we demonstrate that in the peacock mantis shrimp, Odontodactylus scyllarus, the level of ocular independence is task dependent. During gaze stabilization in the context of optokinesis, there is weak but significant correlation between the left and right eyes in the yaw degree of rotational freedom, but not in pitch and torsion. When one eye is completely occluded, the uncovered eye does not drive the covered eye during gaze stabilization. However, occluding one eye does significantly affect the uncovered eye, lowering its gaze stabilization performance. There is a lateral asymmetry, with the magnitude of the effect depending on the eye (left or right) combined with the direction of motion of the visual field. In contrast, during a startle saccade, the uncovered eye does drive a covered eye. Such disparate levels of independence between the two eyes suggest that responses to individual visual tasks are likely to follow different neural pathways.

Keywords: Gaze stabilization; Independent eyes; Mantis shrimp; Neural connections; Optokinesis; Saccade; Visual system.

Fig. 1.

A frontal view of the colourful eyes of Odontodactylus scyllarus. (A) Each eye is capable of >90 deg independent rotation in the yaw (green arrow), pitch (yellow arrow) and torsion (blue arrow) directions. Image courtesy of Mike Bok. (B,C) A view of an experimental animal from (B) the front and (C) above during the covered phase of the experiment, in which one eye was completely occluded with black nail varnish. This animal had its right eye occluded.

Fig. 2.

Experimental set-up. (A) The set-up for the optokinesis experiments, displaying a cutaway of the rotating drum to show the experimental aquarium and location of the stomatopod. The inner face of the drum was covered with a high-contrast grating, providing the animal with a horizontal field of motion when the drum was in motion. The movement of the eyes was recorded with a calibrated stereoscopic camera pair and tracked using the method described by Daly et al. (2016). (B) The experimental set-up for the looming experiment (modified from Daly et al., 2016), which involved presentations of a black circle that rapidly appeared on a CRT monitor. Eye tracking as for A.

Fig. 3.

Rotational responses of the left and right eyes to the horizontal motion of the visual field. (A) The yaw rotation of the left (blue) and right (red) eyes of an individual during rotation of the drum showing stereotypical yaw optokinesis. The dotted lines indicate the progress of points on the surface of the drum, rotating in the yaw plane, but these lines do not necessarily represent specific stripe boundaries. (B) The pitch rotation and (C) torsion rotation of the same eyes during the same stimulus presentation. Note that the drum has no component of rotation in either pitch or torsion. (D) Boxplots displaying the range of Spearman's rank correlation coefficients between the left and right eyes of O. scyllarus in the yaw (green), pitch (yellow) and torsion (blue) degrees of rotational freedom (n=6).

Fig. 4.

The effect of covering one eye on the gaze stabilization response. (A) The average distribution of the relative velocity ratio during both the slow and fast phases of yaw optokinesis for the eye from all six individuals in the ‘uncovered’ group during all trials (clockwise and anticlockwise) before occlusion. (B) The average distribution of the relative velocity ratio during both the slow and fast phases of yaw optokinesis for the eye from all six individuals in the ‘covered’ group before occlusion. Note that before occlusion, neither eye is occluded but they are grouped according to their treatment after occlusion. (C,D) The average distribution of the relative velocity ratio for the eyes of all six individuals in (C) the ‘uncovered’ and (D) the ‘covered’ groups after occlusion (i.e. as for A and B, respectively, but after occlusion). ‘Perfect’ gaze stabilization (relative velocity ratio=1) is indicated by the dashed vertical line. A–D show data from six repeated trials from a single eye from six individuals (n=6); red error bars are the standard deviation of the data in each abscissa interval (width 0.5). (E) When one eye is covered (in this case, the left eye), the contralateral eye (cyan line) continues to perform yaw optokinesis, while the covered eye (black line) remains stationary. (F) Alternatively, the covered eye (the right eye in this case; black) performs a yaw rotation that does not fit the optokinetic profile.

Fig. 5.

The average distribution of the relative velocity ratio of the uncovered eye whilst the contralateral eye is occluded shows lateralization. After occlusion, three individuals had their right eye covered and three had their left eye covered. All six animals were shown six presentations of the drum: three clockwise trials and three anticlockwise. The average distribution of the relative velocity ratio of the left eyes of all three individuals in the ‘uncovered’ group after occlusion differs significantly between (A) clockwise and (B) anticlockwise rotation of the drum. Similar results were obtained for the three right eyes in the ‘uncovered’ group after occlusion during (C) clockwise and (D) anticlockwise drum rotation. A–D show data from three repeated trials for a single eye from three individuals (n=3); red error bars are the standard deviation of the data in each abscissa interval (width 0.2).

Fig. 6.

The effect of covering one eye on the response to a looming stimulus. (A) An example of a startle saccade in the yaw degree of rotational freedom involving both the left (blue line) and right (red line) eyes in response to a high-contrast looming stimulus (onset at brown line) when both eyes were uncovered. (B) A startle saccade to the same looming stimulus involving both eyes when the right eye (black line) was completely occluded and the left eye (cyan line) was uncovered. (C) The proportion of responses across all six animals before (light grey) and after (dark grey) occlusion involving both eyes, just the uncovered, just the covered or neither eye. As previously, though both eyes were uncovered before occlusion, the eyes are grouped by their treatment after occlusion (n=6).