Ocellar spatial vision in Myrmecia ants

Abstract

In addition to compound eyes, insects possess simple eyes known as ocelli. Input from the ocelli modulates optomotor responses, flight-time initiation, and phototactic responses - behaviours that are mediated predominantly by the compound eyes. In this study, using pattern electroretinography (pERG), we investigated the contribution of the compound eyes to ocellar spatial vision in the diurnal Australian bull ant Myrmecia tarsata by measuring the contrast sensitivity and spatial resolving power of the ocellar second-order neurons under various occlusion conditions. Furthermore, in four species of Myrmecia ants active at different times of the day, and in European honeybee Apis mellifera, we characterized the ocellar visual properties when both visual systems were available. Among the ants, we found that the time of activity had no significant effect on ocellar spatial vision. Comparing day-active ants and the honeybee, we did not find any significant effect of locomotion on ocellar spatial vision. In M. tarsata, when the compound eyes were occluded, the amplitude of the pERG signal from the ocelli was reduced 3 times compared with conditions when the compound eyes were available. The signal from the compound eyes maintained the maximum contrast sensitivity of the ocelli as 13 (7.7%), and the spatial resolving power as 0.29 cycles deg-1. We conclude that ocellar spatial vison improves significantly with input from the compound eyes, with a noticeably larger improvement in contrast sensitivity than in spatial resolving power.

Keywords: Bull ants; Contrast sensitivity; Flying; Honeybees; Pattern electroretinography; Spatial resolving power; Walking.

Fig. 1.

Electroretinograms (ERGs) from median ocelli of Myrmecia tarsata for visual systems intact (E⁺O⁺) and compound eyes occluded (E⁻O⁺) individuals. Mean ERGs from ocelli in response to ON–OFF LED light stimulus are shown for E⁺O⁺ individuals (n=5; pink) and for E⁻O⁺ individuals (n=5; black). The shaded regions show the s.e.m. for the respective conditions. Solid arrows indicate the presence of Component 3. The dashed arrow shows that Component 3 is absent in E⁻O⁺ individuals. Stimulus onset and offset are represented at the bottom. (A) ERGs for E⁺O⁺ and E⁻O⁺ individuals. (B,C) ERGS from A on a magnified time scale, as indicated by the green (B) and orange bars (C). Pictograms in A represent the treatment condition: ovals, compound eyes; circles, ocelli; filled symbols, occlusion; open symbols, no occlusion.

Fig. 2.

Amplitude of pattern ERG (pERG) response signal from the ocellar second-order neurons of M. tarsata under different treatment conditions. (A) Visual systems intact (E⁺O⁺) individuals, (B) compound eyes occluded (E⁻O⁺) individuals, (C) ocelli occluded (E⁺O⁻) individuals and (D) visual systems intact individuals with incisions made in the lamina (E⁺O⁺L⁻) (see Materials and Methods for details). Each coloured data point is the mean amplitude of the signal of all individuals at the corresponding spatial frequency at 95% contrast. Individual significant data points (see Materials and Methods) are shown as grey circles. Individual non-significant data points (see Materials and Methods) are shown as grey triangles (n=5 for E⁺O⁺, n=4 for E⁺O⁻, n=3 for E⁻O⁺ and E⁺O⁺L⁻). Pictograms represent treatment condition: ovals, compound eyes; circles, ocelli; filled symbols, occlusion; open symbols, no occlusion. Note, only the median ocellus was stimulated in all experiments.

Fig. 3.

Spatial properties of ocellar second-order neurons of M. tarsata for E⁺O⁺ and E⁺O⁻ individuals. (A) Contrast sensitivity function and (B) spatial resolving power for the two treatment conditions. In A, each data point is the mean contrast sensitivity of all individuals of M. tarsata for a particular treatment condition at the corresponding spatial frequency. The error bars show 95% confidence intervals. Data points for each condition were shifted to the right or left of the recorded spatial frequency to improve visualisation. In B, each coloured data point is the mean spatial resolving power of all individuals of M. tarsata for a particular condition at 95% contrast. Error bars show s.e.m. Individual data points are shown in grey. Pictogram descriptions as in Fig. 2 (n=5 for E⁺O⁺, n=4 for E⁺O⁻).

Fig. 4.

Spatial properties of ocellar second-order neurons of four Myrmecia species and Apis mellifera for E⁺O⁺ individuals. (A) Contrast sensitivity and (B) spatial resolving power for each species. In A, each data point is the mean contrast sensitivity of all individuals of a particular species at the corresponding spatial frequency. The error bars show 95% confidence intervals. Data points for each species were shifted to the right or left of the recorded spatial frequency to improve visualisation. In B, each coloured data point is the mean spatial resolving power of all individuals of a particular species at 95% contrast. Error bars show s.e.m. Individual data points are shown in grey. Triangles indicate nocturnal ant species, circles indicate diurnal-crepuscular ant species. Pictogram descriptions as in Fig. 2 (n=4 for M. gulosa, n=5 for remaining species).