The response of the hovering hawk moth Macroglossum stellatarum to translatory pattern motion

Abstract

• 1. The European hawk moth Macroglossum stellatarum, while collecting nectar in hovering flight in front of flowers, follows moving stripe patterns in the lateral visual field. This response counteracts a second one, that is the animals' effort to stabilize their distance from dummy flowers. We investigated the response to motion stimuli in the lateral visual field using sinusoidally oscillating stripe patterns (Fig. 1), as well as its interaction with the distance stabilizing response.

• 2. In both responses moths attempt to compensate for image speed. The balance between the two depends on the number of elementary motion detectors stimulated by the dummy flower and the stripe pattern, respectively. Increasing the diameter of the dummy flower (Figs. 2 to 4) or the spatial frequency of the stripe pattern (Fig. 7) shifts the balance in favour of distance stabilization. The reverse is true when the length of the stripes in the pattern (Fig. 5) or their number is increased (Fig. 6). It does not matter whether the stripe pattern is presented in the lateral (Fig. 4A) or in the dorsal and ventral visual field (Fig. 4B).

• 3. The gain-frequency relations of the response to the lateral stripe pattern obtained with dummies in two different positions within the drum have their maxima around 3 Hz and decline rapidly towards lower and higher frequencies like the response of a bandpass filter. The distance stabilizing response also has bandpass properties, but with a broad plateau between 0.15 and 5 Hz (Fig. 8). The most likely explanation for this difference is that there is a regional or direction-dependent variation of motion detector properties.

• 4. The responses to ramp-like stimuli are phasic in accordance with the amplitude frequency characteristics, but the responses to progressive (front to back) and regressive motion of the pattern differ (Figs 9, 10).

• 5. The response appears to depend on the azimuthal position of the stripe pattern within the visual field (Fig. 11). It is strongest when the pattern covers equally large parts of the frontal and caudal visual fields. The optomotor sensitivity to translational pattern motion is higher in the frontal than in the caudal visual field (Fig. 12, Table 1).

• 6. When the stripe pattern on one side is removed, the response amplitude is halved. There is no detectable turning response around the vertical axis to the oscillation of the stripe pattern (Fig. 13, Table 2).

• 7. The possible role of the response to pattern movements parallel to the longitudinal body axis under natural conditions is discussed.