Modality-specific long-term memory enhancement in Heliconius butterflies

Abstract

How animals perceive, process and respond to environmental cues is tightly tuned to species-specific ecological demands and reflected by the structure of neural systems. In the Neotropical butterflies, Heliconius, the mushroom bodies, insect learning and memory centres are significantly expanded compared with their closest relatives. This expansion coincided with the evolution of a novel diet of pollen and the ability to learn spatial foraging routes. Previous research has shown that Heliconius have more accurate long-term visual memory than other Heliconiini. Here, we test whether this enhanced memory stability is specific to visual contexts by conducting a long-term olfactory memory assay in two Heliconius species and two outgroup species. We found no difference in long-term olfactory memory between Heliconius and outgroup Heliconiini, and combining data from olfactory and visual memory trials confirmed a modality-specific improvement in memory recall in Heliconius. Tests of how Heliconiini species prioritize olfactory and visual cues when presented in conflict show no consistent pattern, suggesting that variation in memory stability is not explained by inter-specific differences in attentiveness to sensory cues. Our data provide a rare case where memory performance has been compared across species and sensory modalities to identify evidence of a modality-specific shift.This article is part of the Theo Murphy meeting issue 'Selection shapes diverse animal minds'.

Keywords: Heliconiini; associative memory; learning; mushroom body; olfaction; vision.

Figure 1.

Experimental set-up (A) and protocol timeline (B) for the olfactory LTM experiment. (A) Feeder stand set-up: during the training, the artificial feeders (orange stars) are positively reinforced with sugar solution or negatively reinforced with quinine solution. Individuals were trained in groups, either inducing a positive preference for lemongrass or citrus, contra to their naive preference, i.e. the experiment was balanced, with two separate training cages where each odour was positively reinforced during training to different individuals. During the preference tests, the feeders were empty. Odour wells (white dots) between the two feeders are filled exclusively with either citrus odour or lemongrass odour solutions. (B) The bold lines indicate the training period and dashed lines indicate waiting periods when odours are not reinforced. Circles indicate the recorded preference tests.

Figure 2.

Experimental set-up for the training (A) and conflict test (B), and the timeline for the full conflict experiment (C). (A) Feeder stand set-up for the training: the artificial feeders (yellow or purple stars) were positively reinforced with sugar solution or negatively reinforced with quinine solutions, contra the naive preference test of the individual, i.e. the experiment was balanced, with two separate training cages so that each odour/colour combination could be positively reinforced during training to different individuals. The odour wells were filled with either lemongrass odour (squares) or citrus odour (triangles) solutions. (B) Feeder stand set-up for the conflict test: the artificial feeders were empty during the preference tests and the positively reinforced colour was paired with the negatively reinforced odour and vice versa. (C) Solid bold lines indicate the training periods and dashed lines indicate waiting periods when odours were not reinforced. Circles indicate the recorded preference tests.

Figure 3.

(A) Phylogeny of species used in the behavioural assays (pollen feeders: H. erato, H. melpomene; non-pollen feeders: A. vanillae and D. iulia). (B/C) Performance across long-term olfactory (B) and visual (C) memory experiments, colour-coded by trial: white = naive preference, dark orange = recall test, light orange = LTM test after 8 days of no reinforcement of the trained cues. Data for the visual trials are taken from Young et al. [39].

Figure 4.

(A–D) Differences in performance across the three trials (naive, recall and LTM) in the visual (magenta) and olfactory (green) experiments for each species. Values, including standard errors, extracted from coefficients of a GLMM including a significant Species × Trial × Experiment interaction. Asterisks indicate significant pairwise differences in performance between the visual and olfactory experiments. Heliconius species are shown with solid lines and outgroup Heliconiini species with dashed lines. (E) 3D models of the mushroom body calyx showing the visual processing areas (magenta) and olfactory processing areas (green) in a representative Heliconius and outgroup, with their relative proportions shown in pie charts. Scale bar = 100 μm. (F) Pooled Heliconius species performance (circles with solid line) compared to pooled outgroup Heliconiini species (triangles with dashed line) between the visual (magenta) and olfactory (green) experiments, showing the shift in memory fidelity in Heliconius.

Figure 5.

Proportion of feeding attempts made on the trained colour by H. erato (A), H. melpomene (B) and D. iulia (C) across the different trials. The graphs are colour-coded by test: white for naive preference, dark orange for the recall test, light pink for the conflict test and pink for the colour-only test. In (D) colour-coding is by species, orange for D. iulia, light red for H. erato and red for H. melpomene, to compare the performance of the three species across trials.