Allowing several path steps per 20-second bin adds 4% to 51% to five-minute reachability in six contact records

We tested how a path rule changes short-term reachability in six public SocioPatterns contact records. Each bin contains 20 seconds of contacts. Strict paths use at most one contact per bin. Non-strict paths can use several contacts in the same bin. Both rules allow waiting and assume that every transfer succeeds. We calculated exact counts in 3,988 five-minute windows. The relative increase in reachable ordered pairs was 4.1% and 15.3% in two workplace records, 10.9% in a hospital, 32.2% in a primary school, 51.3% in a conference, and 26.0% in a high school. The conference increase was 48.3% to 51.3% under three window alignments. In the primary school, the increase was 105.6% for one-minute windows. An independent set method checked 180 real windows and 253 small test networks. Exact one-bin checks passed for all 47,215 nonempty bins. A second full run gave identical numerical files. The package preserves code, figures, aggregate results, source hashes, and retrieval instructions. Raw data stay in retained storage under their source licence. These results describe a model choice in the selected records. They do not measure disease risk or actual information transfer.

temporal networks · contact networks · reachability · time resolution · reproducible methods