Magnetosensation during re-learning walks in desert ants (Cataglyphis nodus)

Abstract

At the beginning of their foraging careers, Cataglyphis desert ants calibrate their compass systems and learn the visual panorama surrounding the nest entrance. For that, they perform well-structured initial learning walks. During rotational body movements (pirouettes), naïve ants (novices) gaze back to the nest entrance to memorize their way back to the nest. To align their gaze directions, they rely on the geomagnetic field as a compass cue. In contrast, experienced ants (foragers) use celestial compass cues for path integration during food search. If the panorama at the nest entrance is changed, foragers perform re-learning walks prior to heading out on new foraging excursions. Here, we show that initial learning walks and re-learning walks are structurally different. During re-learning walks, foragers circle around the nest entrance before leaving the nest area to search for food. During pirouettes, they do not gaze back to the nest entrance. In addition, foragers do not use the magnetic field as a compass cue to align their gaze directions during re-learning walk pirouettes. Nevertheless, magnetic alterations during re-learning walks under manipulated panoramic conditions induce changes in nest-directed views indicating that foragers are still magnetosensitive in a cue conflict situation.

Keywords: Landmark panorama; Learning and memory; Magnetic compass; Navigation; Path integration.

Fig. 1

Setup from the ant’s perspective. a Helmholtz coil and camera setup as it was installed during training. b Setup with additional landmark in the north. c Setup with new panorama represented by three large sheets (black—north, green—east, white—west). The observer was always sitting in the south–east.

Fig. 2

Characteristics of re-learning walks (reLWs). a Individual path of a reLW (black line) performed by a forager leaving the nest (N) shows that after the introduction of an artificial landmark (LM) the ant moves towards the opposite direction (north) following a meandering path before leaving the recording area along a more straight path towards south (black arrow). The dotted rectangle marks the area that is shown in more detail in b. b Both traces of the mandibles (gray) and of the thorax (black) reveal the fine structure of the reLW. The forager performed pirouettes (indicated by stars *), tight turns about the body axis. During meandering it moved laterally (gray and black dots lie parallel to each other, white arrows indicate the direction of movement). The reLWs also includes straight path segments (gray and black dots lie above each other, dotted arrows indicate the direction of movement) during which the forager moved faster (dots in the path represent the positions of the mandibles (gray) and the thorax (black) every 20 ms). The triangles point towards north. Scale bars = 5 cm

Fig. 3

Comparison of the structures of initial learning walks (iLWs) and of re-learning walks (reLWs). Examples of paths from one individually marked ant at different stages along its lifetime. a, b First (30.6 s) and second (44.2 s) iLWs at the beginning of the ant’s foraging career. The ant was marked 1 day before reappearing and starting to perform short learning walks during which it explored different sectors around the nest entrance (first south sector, then east sector). Before leaving the nest entrance (open circle) for the learning walks, the ant looked out from the nest entrance several times (two times before the first and two times before the second learning walk). The time between the first and second learning walk was 15 min. The inset above a and b shows the ant’s paths at higher magnification. Positions of the mandibles (gray) and of the thorax (black) are marked separately every 20 ms. c–f Outbound (black) and inbound (gray) path of the ant’s foraging trips. c First foraging trip performed 2 days after the first learning walks. d As an experienced forager, the ant left the nest in a straight line and returned very fast (we detected the returning ant just shortly before entering the nest; therefore, the inbound path was only recorded during the last few centimeters). e, f First and second reLW after setup of the landmark. The paths document the typical characteristics of reLWs. The ant left the recording area to pursue a foraging trip after circling around the nest entrance. It turned several times and meandered during both outbound and inbound trip. Note that in f the outbound path is missing a short part of the path, because the ants run below the camera’s tripod leg. The ant’s position (thorax) was marked every 20 ms. The triangles point towards north. Scale bars = 5 cm

Fig. 4

Differences between initial learning walks (iLWs) and re-learning walks (reLWs). a, b Gaze directions of novices during iLW pirouettes (a, n = 14) and of foragers during reLW pirouettes (b, n = 10). Data is plotted in gray and the corresponding statistics in red. The bins include 10°. The red circle indicates the significance level of the Rayleigh uniformity test (α = 0.05). The red arrow is the r-vector pointing to the mean gaze direction. If the vector exceeds the red circle, data is directed significantly. In that case, the 95% confidence (95% CI) interval is shown as a red line outside the circle. If the expected direction (nest = 180°) lies within the 95% CI limits, data is directed towards the nest entrance. The outer circle indicates tic 3. c Proportion of novices (n = 15) and foragers (n = 10) that returned to the nest (black) or left the recording area (white). Fisher’s Exact Test: p < 0.001. For further details, see text

Fig. 5

Gaze directions of foragers before and after 180° rotation of the horizontal component of the magnetic field. For figure conventions, see Fig. 4. a–b′ Gaze directions of foragers during reLW pirouettes when one additional landmark had been installed a n = 14, b, b′ n = 12. c–d′ Gaze directions of foragers during reLW pirouettes when they were confronted with a new panorama (n = 15. a, c Gaze directions relative to the nest entrance before magnetic alteration (under natural geomagnetic field, GMF). b, d Gaze directions relative to the nest entrance after magnetic alteration. b′, d′ Same data as in b, d) plotted relative to the fictive nest entrance