We investigated single and combined effects of 2 diet energy densities [lactation diet (NORM; Net Energy for Lactation-NEL=6.59 MJ/kg of dry matter (DM)) vs. lactation diet diluted with 30% straw (REDU; NEL=4.96 MJ/kg DM), both offered for ad libitum intake] and 2 daily milking frequencies (2× vs. 1×) during 7 d before the dry-off day (D0) on automatically monitored feeding behaviour of 119 (72 primi-, 47 multiparous) clinically healthy, loose-housed, pregnant, lactating Holstein cows from the resident herd of Aarhus University, Denmark. All experimental procedures involving cows were approved by the Danish Animal Experiments Inspectorate (Permit No. 2017-15-0201-01230). Biweekly, batches of 1-6 cows (milk yield≥15 kg/d) were allocated to 1 of the 4 treatments on D-14, and moved to an experimental pen in the same barn on D-7. Cows were individually assigned to computerised feed bins and feed was allocated 4× daily. Dry matter intake (kg/d), feeding time (min/d), feeding rate (kg/min), visits to assigned feed bin (no./d) and visit attempts to unassigned feed bins (no./d) were recorded on D-6, D-4, D-3 and D-1. Data were analysed in R using GLMM including as fixed effects: day, diet, milking frequency and possible interactions, and parity group. Number of cows in the pen and baseline value (averaged D-14 to D-8 for each of the variables) were included as covariates. Batch and cow were included as random effects. Cows fed the REDU diet ingested less feed (LSM±SE: 13±0.4 vs. 18±0.4 kg/d; P