Аннотация:Associative learning is known in many groups of insects, but little is known about the learning abilities and memory of microinsects. Miniaturization causes many changes in the neural system, including reductions in the number of neurons. Nevertheless, associative learning has been shown in some groups of microinsects, mostly parasitic wasps; among non-parasitic microinsects, only beetles were also tested (Polilov et al., 2019); thrips (Thysanoptera), which are at most 2 mm long, have never been tested. We designed an experimental setup for microinsects, based on the idea from Ofstad et al. (2011), where it was used for testing the visual place learning in Drosophila melanogaster. Our setup includes a thermal maze and an LED screen surrounding the maze. The temperature in the maze is too high for thrips, but the maze has four cooler spots (which the insect is supposed to find); the spots switch one by one together with the rotating visual stimuli on the screen. Thripswere experimented one at a time. Every experiment consisted of ten trials of searching, each followed by 1 minute of the thrips staying in the cool spot to memorize the visual stimulus. In addition, there were four tests of sector preference for each thrips: before the experiment, immediately after it, in 1 hour, and in 24 hours after the experiment. In these tests, the cool spots were switched off, while the visual stimuli on the screen were random. We checked whether the insect choses the target section (matching the stimulus) according to the visual learning it underwent. Pairwise comparison of the lengths of the way in every trial shows that significant difference between test and control groups can be seen after 7 trials (t-test, p < 0.05 ). This may indicate that after seven trials the thrips form an association between the visual stimulus and the target spot; this association helps them find the spot sooner. Tests of sector preference performed immediately after all ten trials show that a majority of the insects follow the visual stimulus associated with the comfort spot in the maze (ANOVA Tukey test, p < 0.05). In the tests performed 1 hour later, thenumber of insects who preferred the target sector (supposed to match the cool spot) had decreased, but remained significantly different from the number of insects who preferred either of the three other spots (p < 0.05). The number of insects who follow the visual stimulus was also significantly higher in the 1-hour test than in the test before trials or in the control group (p < 0.05).Thus, we have shown that thrips are not only capable of learning, but also have long-term memory that can be retained for at least one hour. Our study is the first to show such abilities in thrips . Learning capacity and retained principal cognitive abilities in these microinsects make them suitable models for cognitive and neurobiological experiments. This study was supported by the Russian Science Foundation (project no. 19-74-10019).