Roads bring connectivity and benefits to society; however, they impact wildlife in many ways. One of the most negative effects of road networks is wildlife-vehicle collision (WVC), which directly impacts wild animals and can cause material, psychological and physical damage to people. To minimize this effect, one of the most used mitigation measures is vertical wildlife crossing signs. These signs are used to educate and influence the driver’s behavior and are widely used. However, signals' effectiveness in influencing the behavior of drivers is still a matter of debate. The present work aimed to test the effect of vertical wildlife crossing signage on vehicle speed, with the objective of (1) identifying whether the signs influence the reduction of vehicle speed and (2) identifying which wildlife sign category is responsible for the greater reduction. Our study was carried out along the highway BR-262 in Mato Grosso do Sul, Brazil. This road has one of the highest collision rates with large mammals, impacting endangered and emblematic species of Cerrado and Pantanal biomes. Three signaling categories of wildlife crossing signs were compared: standard, educational, and risk. We selected signs free of confounding factors (i.e., roundabouts, radars, or other signs) within a radius of 1 km. Each sample unit consisted of 2 km of road, with the signal under study in the central position. Vehicle speeds were collected at seven points along the sample unit a mobile radar, 20 minutes at each point, in four different periods of the day (day, dawn, dusk, and night). Surveys were repeated four times per site, collecting data from vehicle categories: cars, trucks and SUVs. Linear mixed models were used to test the effect of the interaction speed sign x distance to sign, for each vehicle type and time period. Our results showed that ‘Alert’ signs cause a speed reduction of trucks in all periods, and of cars and SUVs in all times but daylight period. During nighttime, both ‘Standard’ and ‘Educational’ signs also promoted a significant reduction in car speed. However, we also noted that most vehicles quickly returned to their previous seed, after 1000 m from the sign. As such, the effect of signs seems to be very localized and may not be effective to avoid collisions with animals at higher distances from the sign. Other mitigation measures should be implemented in complement to warning signs.