Highlights d Responses of neurons in monkey V1 are rapidly modified by visual fear conditioning d Fear-related signals in V1 emerge from the outset of neuronal responses (<40 ms) d Fear learning in V1 is specific to the CS orientation and location d The conditioning effect is independent of neurons' orientation preferences

One of the most challenging topics in Natural Language Processing (NLP) is visuallygrounded language understanding and reasoning. Outdoor vision-and-language navigation (VLN) is such a task where an agent follows natural language instructions and navigates a real-life urban environment. Due to the lack of human-annotated instructions that illustrate intricate urban scenes, outdoor VLN remains a challenging task to solve. This paper introduces a Multimodal Text Style Transfer (MTST) learning approach and leverages external multimodal resources to mitigate data scarcity in outdoor navigation tasks. We first enrich the navigation data by transferring the style of the instructions generated by Google Maps API, then pre-train the navigator with the augmented external outdoor navigation dataset. Experimental results show that our MTST learning approach is model-agnostic, and our MTST approach significantly outperforms the baseline models on the outdoor VLN task, improving task completion rate by 8.7% relatively on the test set. 1

Suppose that there are two congestible modes of travel from A to B -road and rail for concreteness -which are imperfect substitutes in demand. Road congestion from A to B is underpriced; this is an unalterable distortion. Compared to the first best, should the transportation planner choose a wider or narrower road, raise or lower the rail fare, and expand or contract rail capacity? This paper provides a synthetic review of the literature on the problem, presents some new results, and discusses directions for future research on this and related second-best problems.

This study used polystyrene latex colloids as model microplastic particles (MPs) and systematically investigated their retention and transport in glass bead-packed columns. Different pore volumes (PVs) of MP influent suspension were first injected into the columns at different ionic strengths (ISs). The breakthrough curves (BTCs) were obtained by measuring the MP concentrations of the effluents. Column dissection was then implemented to obtain retention profiles (RPs) of the MPs by measuring the concentration of attached MPs at different column depths. The results showed that the variation in the concentrations of retained MPs with depth changed from monotonic to non-monotonic with the increase in the PV of the injected influent suspension and solution IS. The non-monotonic retention was attributed to blocking of MPs and transfer of these colloids among collectors in the down-gradient direction. The BTCs were well simulated by the convection-diffusion equation including two types of first-order kinetic deposition (i.e., reversible and irreversible attachment). However, this model could not well simulate the non-monotonic retention profiles due to the fact that the transfer of colloids among collectors was not considered. The results in this study are critical to developing models to simulate the fate and transport of MPs in porous media such as soil.