Learning by Association in Plants
A total of 83 seedlings (Light, n = 28; Light-Dark, n = 30; Dark, n = 25) were individually trained inside a Y-maze, where the fan [F] and the blue LED light [L] were always positioned together on one of the arms of the maze (as in the [F + L] group of the first experiment). The conditioning protocol was the same as that used in the first experiment. It consisted of three training days of three training sessions per 24 hours, and a subsequent testing day, all delivered between 09:00 and 17:00. Accordingly in this experiment, seedlings received their training and tested sessions at times based on their entrained temperature and photoperiod cycles and corresponding to their “subjective” day (Light group), or were partly (Light-Dark group) or entirely outside their subjective day (Extended Data Fig. 2). For example, plants from the Light group were trained and tested at a time of day matching their entrained light period, while plants from the Dark group were trained and tested at a time of day, which corresponded to their entrained dark period.
During each training session, the fan exposure as the CS always preceded by 1 hour and overlapped by 30 min with the delivery of blue light as the US (Fig. 1Bi). Fan and light were always presented in the same position relative to each other, but their position in each maze was re-assigned for each training session to ensure the source of the light remained unpredictable. Additionally to ensure that each pea was exposed to the blue light in an equal number of ‘right’ and ‘left’ sides prior to testing, we used a simple randomization procedure (as described above) to assign individual peas to one of two alternative 3-day training schedules. Schedule A always started on the right arm and followed the pattern: day 1, R/R/L; day 2, R/L/L; day 3, R/L/R (and tested on the L). Schedule B always started on the left and followed the pattern: day 1, L/L/R; day 2, L/R/R; day 3, L/R/L (and tested on the R). This approach was applied to all peas across experimental groups. As in the first experiment, each training day consisted of three 2-h training sessions separated by 1-h intervals, during which all seedlings were kept in darkness and undisturbed in their individual mazes. During the training phase, seedlings grew and approached the Y-bifurcation of the maze. As described above, all seedlings were checked at the end of the training phase to ensure that they had not yet grown into one or the other arm and had grown to be ca. 1 cm away from the bifurcation of the maze.
At the end of the training phase, seedlings within each group were further split into an experimental sub-group (Fig. 1Bii) and a control sub-group (Fig. 1Biii). Control seedlings (Light, n = 10; Light-Dark, n = 13; Dark, n = 10) were left in darkness and undisturbed during the testing day. On the testing day, experimental subgroups of seedlings (Light, n = 18; Light-Dark, n = 17; Dark, n = 15) were kept at their treatment-specific temperature cycles in the dark, while being exposed to the fan blowing for three 90-min sessions separated by a 90-min interval. The fan was positioned on the opposite arm of the maze where each pea experienced the light (and the fan) during its last training session. The position of the fan was then maintained fixed for all testing sessions. The morning after the testing day, we inspected all seedlings and recorded which arm of the maze they grew into, scoring whether they made the correct (i.e. towards the fan) or incorrect choice (i.e. away from the fan). Once scored, seedlings were transferred from the controlled-environment room, transplanted into larger pots and later replanted in a community garden.
The seedlings were initially entrained by a light-dark cycle in conjunction with a temperature cycle. The latter was maintained as sole Zeitgeber during the training and testing phase. Temperature cycle has been previously shown to effectively synchronize circadian rhythms by inducing circadian-type oscillations in the level of RNAs in pea seedlings16. The application of three one-hour light pulses during the training days at different phases of the circadian cycle may have induced phase shifts4041. Therefore, the phase of the circadian rhythm may have somewhat deviated from the Zeitgeber cycle.
Statistics were based on comparisons between experimental groups (exposure to fan) and control groups (no fan). Since the behaviour of the controls of Experiment 1 and of the ‘Light Group’ of Experiment 2 was determined by last light exposure (phototropism), the question was whether the fan-exposed experimental group could counteract the phototropic effect. The two-tailed Fisher’s Exact Test of independence was used to assess significance.
How to cite this article: Gagliano, M. et al. Learning by Association in Plants. Sci. Rep. 6, 38427; doi: 10.1038/srep38427 (2016).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.