On dairy farms, milking practices can temporarily restrict cows from accessing drinking water. This study examined how different types of temporary water restriction, similar to those commonly encountered on farms, may affect the individual behavior and physiological dehydration status of cows and whether these effects could differ according to the cow's position in the social hierarchy of the group. Four treatments were applied (CTRL = free access to the drinker and feed fence, HL = 2 h of headlock at the feed fence after morning and evening milking, D2H = free access to the feed fence and no access to water for 2 h after milking and D4H = free access to the feed fence and no access to water for 4 h after milking) to 4 groups of 10 lactating dairy cows using a 4 × 4 Latin square design. Each treatment period lasted 1 week. The cows were housed within a free-stall barn, and each group was in a pen equipped with 1 electronic drinker, 10 individual feeders, and 2 cameras. Cow social dominance hierarchies were characterized using normalized David's scores, which were calculated from video-detected replacements at each pen's drinker. During the last 4 d of each treatment period, drinking behavior was measured using interconnected drinkers, and eating, idle, and lying times were measured using accelerometers. On d 6 of each treatment period, during the morning and evening milkings, blood and then milk samples were collected to analyze bioindicators of dehydration and milk composition, respectively. The daily water intake of cows was unaffected by either social dominance or treatment type. Drinking rate and drinking bout frequency were higher during the HL, D2H, and D4H treatments than during the CTRL treatment, and drinking time was lower during the D4H treatment than during the CTRL treatment. Hourly lying time decreased at the end of the water restriction periods, i.e., at 1000 h in HL, at 1100 h in D2H, and at 1200 and 1300 h in D4H, which was not observed in CTRL, which could be interpreted as a sign of cow discomfort. For blood collected before the morning milking, hematocrit levels were lower in the D4H treatment than in the HL treatment. For blood collected before the evening milking, creatinine concentrations were lower in the D4H treatment than in the D2H treatment, with a greater difference seen for dominant cows than for subordinate cows. For both milkings, milk freezing point was higher in the HL, D2H, and D4H treatments than in the CTRL treatment, indicating dilution. In contrast to the other social categories, subordinate cows made more daily visits to the drinker during the D4H treatment than during the CTRL treatment but displayed no differences in drinking bout duration or the duration of drinking interruptions within bouts in the D2H treatment compared with the CTRL treatment. Overall, cows were able to maintain their water intake despite the temporary water restrictions when they had access to the number of drinkers recommended by animal welfare guidelines. That said, their idle and lying time appeared to be disturbed.