To retain information, or memory, is a crucially adaptive cognitive ability in animals (Menzel, 1999). The adaptive value of memory is related to the ability to make quick and accurate decisions when faced with a situation similar to one previously experienced (Menzel and Benjamin, 2013). Memory allows animals to avoid harmful situations, to remember important locations or specific information, and to avoid energy loss by not responding to irrelevant stimuli; in other words, memory contributes to overall fitness (Couto et al., 2023). At the same time, memory formation and maintenance have different costs (Niven and Laughlin, 2008). As the brain is metabolically expensive, the resources allocated to encode, consolidate, and access information generate important expenditures (Kandel, 2001). Different types of memory coexist, defined by their duration and the physiological processes involved in their development. They end up being adaptative or not depending on the context. For instance, in stable environments, where the probability of encountering a certain situation again is high, it may be adaptative to invest in long-term memory. In a rapidly changing environment, however, it may be better to prioritise short-term memory (Pull et al., 2022).

The properties and the physiological mechanisms underlying the different types of memory have been studied in many invertebrate models, notably in the fruit fly Drosophila melanogaster (Tully 1994) and the honey bee Apis mellifera (Menzel 2001a). In addition, habituation to visual stimuli and memory has been well characterised in the mudflat crab Neohelice granulata (Tomsic and Silva, 2023). These experiments provided insights about the ecological relevance of memory duration according to the habitat. In a study by Tomsic et al. (1993), the authors compared the habituation of two related semi-terrestrial crabs that occupy different habitats, Neohelice granulata and Pachygrapsus marmoratus. By analysing the influence of diverse parameters on visual habituation performances (e.g., individual size, number of trials), the authors showed that habituation is species-dependent and that contextual cues are memorised differently. Tomsic et al. (1993) concluded that ecology played a major role in the origin of these differences. Indeed, Neohelice granulata crabs live in self-dug burrows, closed to the mud substrate and surrounded by conspecifics and halophyte vegetation. On the other hand, Pachygrapsus marmoratus live on rocky outcrops, close to the sea and without vegetation. So, a shadow passing over Neohelice crabs would induces stronger and longer habituation because it represents an ambiguous signal (e.g., grass undulation), whereas for Pachygrapsus crabs, the probability of being an actual flying predator would be higher in their environment which is poor in objects passing overhead (Tomsic et al., 1993), resulting in a weak habituation response in the latter.

A key parameter for habituation and the mesic mark it can generate, is the inter-trial interval (Giurfa et al., 2009). Short inter-trial intervals (e.g., from few seconds to few minutes) are more likely to reinforce short-term memory, which relies on neural facilitation (i.e., increase in synaptic strength) and reversible changes (Hemmi and Tomsic, 2012), but not long retention. In contrast, long inter-trial intervals will lead to the formation of long-term memory, which depends on the activation of specific genes leading to new protein synthesis and structural changes in neural circuits (Tomsic et al., 1996; reviewed in Margulies et al. 2005 in Drosophila). In between, intermediate inter-trial intervals produce intermediate memory, which involves synaptic consolidation through the activation of specific kinases (e.g., cAMP-dependent protein kinase PKA) and early gene expression (Tomsic and Romano, 2013). While the duration of inter-trial intervals has been empirically tested, these types of memory have also been described in several taxa (Tully 1994; Izquierdo et al., 1998, Menzel, 2001b).

In this work, we investigated the ability to develop memory after learning in an aquatic insect, the mosquito larva (Aedes aegypti). Mosquito larvae spend most of their time hanging from the water surface. When a stimulus is perceived as a potential danger, larvae dive (Clements, 1999). If the stimulus turns out to be innocuous upon repeated occurrences, larvae no longer respond to further stimulation due to habituation, a form of non-associative learning, potentially forming a mnesic trace (Baglan et al., 2017, Dessart et al., 2023).

Although much attention has been paid to cognition in adult mosquitoes, this is the first study to investigate the memory of mosquito larvae. In freshwater ecosystems, mosquito larvae are part of the neuston (i.e., organisms living at the water surface). They are therefore surrounded by unpredictable aquatic and aerial predators such as dragonfly larvae or water striders (for review see: Vinogradov et al., 2022). In this type of environment, a shadow repeatedly casting over the water surface in a short period of time is likely to be projected by the same object, whereas a shadow projected over the water hours later could be produced by a different moving body. In this situation, we could expect that mosquito larvae stop to respond to the repetition of an aversive stimulation in the short term, while resetting their responsiveness in the long term, i.e., not to remember, would be a more adaptive strategy.

In addition to very-well studied aquatic invertebrates such as the sea hare Aplysia californica (Glanzman, 2009) or the crab Neohelice granulata (Tomsic et al., 2017) which exhibit remarkable forms of long-term memory, other freshwater organisms also showed consistent long-term memories, as for example crayfish Procambarus cubensis up to 24 h (Abramson et al., 2005), great pond snails Lymnaea stagnalis up to 3 days (Lukowiak et al., 2003), and water fleas Daphnia sp. up to 6 days (Ringelberg and Gool, 1995). Since long-term memory has been demonstrated in several aquatic species, the possibility of long-term memory in mosquito larvae cannot be ruled out without experimental evidence.

On the one hand, the highly unpredictable environment could prioritise the formation of a short-term memory in mosquito larvae. On the other hand, other organisms from similar environments show robust long-term memory. To distinguish between these two hypotheses, we conducted a series of experiments with A. aegypti mosquito larvae to investigate (1) how long mosquito larvae could retain information after habituation, and (2) whether the duration of inter-trial intervals would play any role in memory formation.