Leading Pedestrian Interval

St. Petersburg, Florida
Source: Pedestrian and Bicycle Information Center (PBIC)


At signalized intersections, right- and left-turning vehicles present a danger to pedestrians crossing during the WALK interval, and crash statistics show that pedestrians are especially vulnerable to crashes involving left-turning vehicles.


Pedestrians are given a WALK signal three seconds before parallel traffic is given a green light.

One of the most effective ways to decrease crashes that involve motor vehicles and pedestrians is to separate them in time. One practical alternative is to program traffic signals to allow the pedestrian to begin crossing before the vehicle traffic on the parallel street is given the green light. This is commonly referred to as a leading pedestrian interval (LPI).

Research has shown that this treatment is associated with a decrease in pedestrian/motor vehicle conflicts and an increase in the percentage of motorists that yield right-of-way to pedestrians. This study examined the influence of a three-second LPI on pedestrian behavior and conflicts with turning vehicles (Van Houten, Retting, Farmer, Van Houten, Malenfant, 2000).


An LPI was created for study at three signalized intersections in downtown St. Petersburg, FL, where pedestrian crossings occurred at the average rate of 60 per hour. To ensure unbiased results, no public outreach or awareness was conducted prior to execution of the study. The Insurance Institute for Highway Safety funded the study at cost of $30,000.

In order to collect baseline data, prior to the installation of the LPI, each intersection was configured to provide simultaneous onset of the WALK signal and GREEN phase for turning vehicles. During the experiment, the LPI was installed to release pedestrians three seconds ahead of turning vehicles by extending the duration of the all red signal phase by three seconds. Sites one and two were each at intersections where one street carried four lanes of one-way traffic and the other two-way traffic (two lanes in each direction); while site three was an intersection where both streets carried two-way traffic (each street had a total of four lanes). These streets had 30 mi/h (48 km/h) posted speed limits and carried high volumes of traffic.

Observers collected data on three items: a) pedestrian/motor vehicle conflicts, b) pedestrians beginning to cross during the five second period at the start of the WALK interval, and c) pedestrians starting to cross during the remainder of the WALK interval. They also noted the percentage of pedestrians yielding right of way to turning vehicles and the number of half-lanes traversed by the lead pedestrian during the three seconds the LPI was in effect. Data were collected separately for pedestrians 65 and older at all three sites.


Following the introduction of the LPI, conflicts were virtually eliminated for pedestrians departing during the start of the WALK interval. There were 44 total pre-treatment observation periods at all three sites. During each of these sessions, the sites averaged between two and three conflicts per 100 pedestrians, with some periods having up to five conflicts per 100 pedestrians. After the LPI was installed, 34 of the 41 sessions had no conflicts, and no session had more than two conflicts per 100 pedestrians.

This effect held up for senior citizens and non-seniors alike. There was also a smaller reduction in conflicts during the remainder of the WALK interval. This reduction was likely the result of pedestrians claiming the right-of-way during the earlier portion of the WALK interval. The percentage of pedestrians yielding to vehicles also declined following the introduction of the LPI, and data showed that pedestrians tended to cross more lanes during the three-second LPI the longer the intervention was in effect. This was likely the result of regular users discerning the presence of the LPI and modifying their behavior to utilize it to the fullest extent possible.

Over a period of four months at these three sites, no reduction in intersection effectiveness for motor vehicles was detected. Moreover, local authorities opted to retain the LPI in places where the range of permitted turning movements governed by the signal cycles allows safe use of the LPI. This intervention was shown to increase pedestrian safety and improve pedestrian comfort and perceived safety levels as well.


Dr. Ron Van Houten
Western Michigan University
Department of Psychology


Van Houten, R., Retting, R.A., Farmer, C.M., Van Houten, J., Malenfant, J.E.L. Field evaluation of a leading pedestrian interval signal phase at three urban Intersections. Transportation Research Record. No 1734, 2000, p. 86-91.

Back to Search Results