dataset content:
experiment videos:
- two videos, files: FILM_1_to_start_no_et_data and FILM_2_for_ET_data (mp4 and ogg) prepared by researchers and used during the experiment.
dynamic AOI:
- an image (jpg) showing how all dynamic AOI was set on videos.
data on participants (ods, xlxs):
- demographic and individual data:
- gender
- age
- experience
- education
- vision impairments
- results of the health and safety (H&S) knowledge test.
eye-tracking data from 2nd film (in ods and xlsx):
- TTFF (Time To First Fixation): Time to the first fixation in the examined Areas of Interest (AOI).
- TFD (Total Fixation Duration): Total duration of fixation in the same AOIs.
basic description of data collection conditions:
participants:
Participants were adult Poles with proven knowledge of health and safety (H&S) on construction sites, verified through a test. They had no visual impairments or had corrected vision (allowing clear sight at a distance of 75–85 cm) and were able to recognize yellow, red, blue, and green colors.
A total of 201 individuals were invited, of whom 169 datasets passed every verification stage successfully.
procedure:
stage 1: completion of consent forms and questionnaires.
stage 2: vision tests (color naming/Ishihara + FrACT for distances of 70–80 cm - FrACT Homepage).
stage 3: knowledge test consisting of 10 questions about health and safety.
stage 4: recording visual behaviors during an H&S inspection.
Participants wore one of three prepared goggles (UVEX construction goggles with transparent (T), red (R), or yellow (Y) frames, all with the same shape and transparent lenses). The timing began immediately after putting on the goggles. Participants sat in front of a monitor (approximately 75 cm from their eyes) and rested their heads on a medical chinrest (with the head tilted slightly downward). The workstation (table height) was adjusted to meet the participants’ needs. Participants were tasked with inspecting a construction site based on two specially prepared videos (two 100s films). They were required to visually identify irregularities and signal their detection by pressing a button. Only data from 2nd film (file: FILMHALAC14.mp4) was analyzed. The first film was designed to familiarize the subjects with the situation. Visual behaviors were tracked using the Tobii Pro Spectrum 600Hz, positioned directly below a 22-inch monitor (configured per the manufacturer's recommendations), alongside the Tobii Pro Lab software.
laboratory conditions: Silence, temperature maintained at 21–23°C, constant lighting conditions (only artificial light, more than 300lux / less than 400lux, and no screen glare), gray walls, cream simple furniture, and no significant visual distractions.
stage 5: completion of the timing and disclosure of the study's true purpose.
models of construction goggles used in the study:
1. 9301-906 Ultravision Ventilated Goggles / UVEX (transparent goggles, matte finish)
2. 9301-613 Ultravision Yellow Gas Tight Goggles / UVEX (RAL 1023 goggles, glossy finish)
3. 9301-603 Red Gastight Safety Goggles / UVEX (RAL 3020 goggles, glossy finish)
data processing: Tobii Pro Lab software was used to analyze the data, focusing on dynamic AOI (Areas of Interest - rectangles corresponding to red and yellow objects visible in the videos).
participant groups division:
- Y - Wearing yellow goggles
- R - Wearing red goggles
- T - Wearing transparent goggles
- K - Excluded participants for various reasons
abbreviations:
AOI – Area of Interest
TTFF - Time to First Fixation
TFD - Total Fixation Duration
T - Transparent
R - Red
Y - Yellow
K - elimination / no data
F - female
M - male
O - other
