Recording and processing a survey using an eye tracker The eye-tracker used is a Pupil Core from Pupil Labs. The basic eye tracker configuration, i.e. a fixation time of 80 ms to 200 ms, is kept for this experiment. The aim of the experiment is to understand what a person looks at to find their way around a multi-scale map and to understand the different strategies used. To do this, the user will be free to use the map as he wishes, i.e. he can use pan and zoom at will. Four types of tasks will be asked in order to have a maximum of types of use of multi-scale map.The first task is to simulate that a user is using an application like map or Google map and is looking for a specific address. The map application will then zoom in very strongly on the address. The user has little spatial context and it often takes some time to find his way around. To simulate the application, a point is placed on Paris or its surroundings and the display is very zoomed (Paris was chosen because most people have a more or less detailed mental map of Paris). The user is then asked to interact with the map (zooming and panning) until he feels he is sufficiently located, as he would if he had to search for a place on his mobile phone. When he is located, he just needs to move on to the next stage without asking for validation. This stage is carried out in four locations. The four points are located near Montmartre, at the entrance to the catacombs of Paris, in Vincennes and finally at Porte d'Asnières The second task is to find a place from an aerial image. The aerial image of a specific area is displayed and the map is zoomed out to the city where the location is located. The user must then try to find the location in the image. Unlike the first task, the user must request validation before proceeding to the next stage. This task is repeated in two different cities. The two images are the tête d'or park in Lyon and a building block next to a railway in Dijon. The third task also consists of finding a precise location using textual indications. The user still has to ask for validation to go to the next stage . This task is repeated in two different cities.The first was "to find the town hall which is just south of the town centre and next to the library" and the second was "to find the stadium east of the town centre and north of the river Vilaine with a north/south orientation. The last task builds on tasks 2 and 3. The map is again zoomed out, an aerial image appears and textual indications are given. This task is repeated on two different cities. The first image is of a building in beauvais with the indication: "the building is in the north west of sqare next to the SNCF station". The second one is a picture of a stadium in lyon with the indication: "the stadium is west of the confluence of lyon". data format : Coord_fixation_on_map_x_y: geolocated fixation point with x the survey type 1 or 2 and y the candidate number (id_fixation,x,y,zoom,etape) Pan: pan on the map during the survey Pan_fixation_on_map : fixation during a pan zoom: zoom on the map during the survey zoom_fixation_on_map: fixation during a zoom stat: number of zoom, pan and fixation per step result_map_x = map status every 100 ms during the survey x 00x : export file of the eye-tracker pupil Lab