Heat stress augments cool-seeking behaviour during exercise and recovery but does not modify the relative contributions of afferent signals driving this behaviour

INTRODUCTION: Cool-seeking behaviour (CSB) aims to mitigate the impacts of heat stress, yet our understanding of the afferent signals (i.e., skin temperature (Tsk), core temperature (Tcore), and skin wettedness (Wsk)) that contribute to decisions to engage in CSB during exercise and recovery is limited to moderate environments. PURPOSE: To test the hypotheses that during exercise 1) CSB is initiated earlier, more frequently and for longer total duration in hot (HS) vs moderate (MOD) environments and 2) the relative contribution of Tsk and Wsk to CSB would increase and Tcore would decrease in HS vs MOD environments. METHOD: Twelve healthy adults (6F, 24±5 y, 177±7 cm, 75±8 kg, 38±8 ml/kg/min estimated VO2max) completed a MOD (23.8±0.8°C; 35.1± 4.6% RH; 7.6±1.2 g/m³) and HS (33.8 ±0.6°C, p< 0.01; 23.8 ± 3.0% RH, p< 0.01; 8.8±1.2 g/m³, p=0.02) trial in a counterbalanced design. Each involved 45 min of cycling at ~50% VO2max (EX) and 45 min seated recovery (REC). Participants maintained neck thermal comfort via a dual-channel fluid-perfused device (34°C) on the dorsal neck, activating cold flow (-20°C) as required. Device temperature (Tdevice) was measured continuously to provide an index of CSB. Tcore (rectal), mean Tsk and mean Wsk (10-site iButtons) were measured continuously. Heart rate (HR, Polar) and mean arterial pressure (MAP, Tango M2) were measured at 10 min intervals. Initiation, frequency and duration of CSB were analysed by paired t-tests. Physiological and relative contribution data were analysed by two-way mixed-effects ANOVAs with Sidak post hoc tests. Relative contributions were determined by multiple linear regression using 1 min data for Tdevice as the dependent and log-transformed Tcore, mean Tsk and mean Wsk as independent variables. Standardised β-coefficients were used to calculate percent relative contribution of Tcore, Tsk and Wsk to Tdevice. RESULTS: CSB was first initiated earlier in HS (7±4 min) vs MOD (14±8 min, p< 0.01). Frequency was greater (HS: 7±3; MOD: 5±2 initiations, p< 0.01) and duration longer (HS: 33.0±22.0; MOD: 17.7±12.6min, p< 0.01) in HS vs MOD. At end-baseline, physiological variables did not differ between conditions (all p>0.05). By end EX, HS elicited a higher Tsk (HS: 35.4±0.5; MOD: 33.2±0.5ºC, p< 0.01) and HR (HS: 150±10; MOD: 138±7bpm, p< 0.01) vs MOD, but Tcore and Wsk did not differ between conditions (p>0.05). By end REC, Tcore (HS: 37.5±0.3; MOD: 37.1±0.3ºC, p=0.01), Tsk (HS: 35.0±0.6; MOD: 32.6±0.4ºC, p< 0.01), Wsk (HS: 0.5±0.1; MOD: 0.3±0.2AU, p=0.03) and HR (HS 86±11; MOD: 73±10bpm, p< 0.01) remained higher in HS vs MOD. MAP did not differ between conditions at any timepoint (p>0.05). There were no differences in the relative contributions of Tcore, Tsk and Wsk to CSB between HS (Tcore: 26±12%, Tsk: 42±22%, Wsk: 31±27%) and MOD (Tcore: 28±19%, Tsk: 39±20%, Wsk: 32±22% p>0.13). CONCLUSION: Exercise and recovery in the heat augments CSB compared to a more moderate environment. Despite this, no significant differences in the relative contributions of the afferent physiological signals driving this behaviour were observed. Funding Sources: American Physiological Society Summer Undergraduate Research Fellowship. This abstract was presented at the American Physiology Summit 2026 and is only available in HTML format. There is no downloadable file or PDF version. The Physiology editorial board was not involved in the peer review process.