QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online August 30, 2006
Journal of Experimental Biology 209, 3677-3684 (2006)
Published by The Company of Biologists 2006
doi: 10.1242/jeb.02414

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Graham, P. Articles by Collett, T. S. Search for Related Content PubMed PubMed Citation Articles by Graham, P. Articles by Collett, T. S.

Bi-directional route learning in wood ants

Paul Graham and Thomas S. Collett^*

School of Life Sciences, University of Sussex, Brighton, BN1 9QG, UK

View larger version (42K): [in a new window]   Fig. 1. The experimental arena. (A,C) Experimental arena and training layout. Ants travelled along an open-topped, narrow channel and under a large cylinder before heading over an open arena to the feeder. The feeder was at the bottom of a small pit (shown enlarged in A) and for the first experiment, a small barrier was placed around the pit to prevent the ants from viewing the landmark and its surroundings from the pit (B,D). In tests ants were taken directly from the nest and placed on a feeder (indicated by F). Two landmarks (large black circles) were positioned symmetrically in the arena. Cumulative density distributions of all return trajectories from ants trained with the channel on the left or right, respectively. Dark areas represent areas where ants spent the most time. Values are given as proportion of the total time spent in the arena. Because of the large amount of time spent at or near the feeder, all values above 2.5% are represented as black squares.

View larger version (42K): [in a new window]   Fig. 2. The ants' first and second trips home after many food-bound runs. (A) Trajectories of ants after leaving the feeder (F). Trajectories ended after 6 min or before if the camera `lost' the ant. Trajectories are shown from when the ant has reached at least 10 cm from the feeder and end-points are marked by small black circles; large black circles indicate the landmarks. Trajectories from ants trained with the channel on the right hand side of the direct route to the food have been mirrored. (B,C,D) Circumferential positions of ants at 20 cm, 40 cm and 60 cm from the feeder. Grey lines represent the direct trajectory to the normal position of the cylinder. Arrow and range represent the mean heading and 95% confidence interval.

View larger version (29K): [in a new window]   Fig. 3. The development of trajectories. Trajectories are shown grouped by run number. Ants are only included if they survived to perform at least 10 routes. The end of the channel and the food are at (0,0) and (50,50), respectively. Low barriers that were used to constrain ants within the half of the arena containing the feeder influenced the paths on initial runs. (A) Food-bound trajectories of ants (N=14) that were carried back to the nest after feeding without performing homeward trajectories. (B,C) The food-bound and homeward trajectories, respectively, of ants that performed round trips (N=12).

View larger version (49K): [in a new window]   Fig. 4. The straightness and consistency of trajectories. (A-C) Measure of the straightness of food-bound and homeward trajectories plotted against run number. (A) Ants that only performed food-bound routes. (B,C) Ants that performed both food-bound and homeward routes. (D-F) Consistency of trajectories plotted against run number for ants grouped as in A-C. Thin lines plot the straightness or consistency of individual ants over time. The thick black line and the grey area show the overall mean and the 95% confidence interval, respectively. Both straightness (repeated measures ANOVA; F=39.7; d.f.=2,48; P<0.005) and consistency (repeated measures ANOVA; F=10.2; d.f.=2,46; P<0.005) improved with experience. There was a significant difference between the groups in their straightness (one-way ANOVA, F=28.8; d.f.=2,24; P<0.005), but not in their consistency (one-way ANOVA, F=1.2; d.f.=2,23; P=0.313). The two types of food-bound paths were significantly straighter than homeward paths, but were not significantly different from each other.

View larger version (28K): [in a new window]   Fig. 5. Temporal and spatial distribution of reversals on food-bound and homeward routes. (A,D) Examples of reversals on food-bound (A) and homeward (D) routes. (B,E) The proportion of ants whose food-bound (B) and homeward (E) trajectories contain reversals is plotted against run number. (C,F) The distribution of reversals along the food-bound (C) and homeward (F) paths is shown relative to the end of the channel. (G) Starting loops from the feeder. The mean number of loops per trajectory is plotted against run number over three successive days. Loops are defined as round trips from, and back to, the feeder in which the ant travels at least 5 cm from the feeder. Data are from 12 ants performing a total of 259 homeward trajectories. The length of each x axis indicates the total number of training runs on that day.