Figure 1. (a) Apparatus used to measure classical conditioning in D. melanogaster. Training phase: groups of flies (~150) are placed into the training tube (A), which is lined with an electrifiable grid. Vacuum-generated air currents are used to deliver the concentrated sucrose CS+ and CS– odours (D) to the training tube, one at a time. The CS+ odour is presented with a shock, while the CS– is not. Testing phase: the flies are then moved down to the testing choice point, via an elevator (C). During testing, the flies have the choice of moving into one of two test tubes (B); one contains the CS+ odour and the other the CS– odour. The tubes are then removed and the number of flies in each is counted to assess whether learning (to avoid the odour paired with the shock) has occurred. (Adapted from Tully, 1991.); (b) Putative stages of memory in D. melanogaster. The four stages: short-term memory (STM), middle-term memory (MTM), anaesthesia-resistant memory (ARM) and long-term memory (LTM) are functionally independent and act additively to produce the observed learning curves. (Modified from DeZazzo and Tully, 1995.)

Figure 2. Apparatus used to assess habituation in C. elegans. Individual worms are placed on to a Petri dish, held in a plastic attachment that is connected to a micromanipulator. A mechanical tapper driven by an electromagnetic relay is used to deliver the taps in a consistent and uniform manner to the Petri dish from a Grass S88 stimulator. A video camera attached to the microscope and connected to a video monitor and video cassette recorder is used to monitor and record the behaviour of the worm. To allow for precision of the timing of events, a time–date generator is used to superimpose the time (in thousandths of a second) and date of the experiment onto both the monitor and the recorder. After habituation training, the magnitude of the responses to tap are quantified using stop-frame video analysis by tracing each response on to an acetate, which is then scanned into a microcomputer to measure the length of each response. (Adapted from Peters et al. 1999).

Figure 3. (a) Comparison of reversal magnitude using 24 h long-term memory protocol for control (open bars) and trained (filled bars) wild-type and glr-1(n2461) strains of C. elegans. Habituation is assessed by measuring the distance the worms travel backwards in response to tap. Trained wild-type worms show smaller reversals compared with untrained worms indicating long-term memory for habituation. glr-1 worms show no long-term memory for habituation; (b) The quantification of GLR-1::GFP protein expression along the ventral nerve cord in trained (filled bars) and control (open bars) C. elegans. Confocal images were used to determine the size of GLR-1::GFP clusters along the ventral nerve cord. Trained worms expressed significantly smaller GLR-1::GFP clusters along than do control worms.