Social crowding is everywhere. Although many empirical studies on social crowding have been carried out, there is no research to explore social crowding from a sensory perspective. This paper argues that social crowding increases consumers' preference for strong (vs. weak) sensory stimulation (hypothesis 1), and this effect is driven by perceived loss of control (hypothesis 2). In addition, this study also proposes that the influence of social crowding on consumers' sensory stimulation preference is moderated by the sense of power (hypothesis 3). In order to verify the above hypotheses, 3 experiments were carried out in this paper. The purpose of experiment 1 (N = 125) was to examine hypothesis 1. Experiments adopted social crowding (crowded group vs. non-crowded group) single-factor between-subjects design. The purpose of experiment 2 (N = 223) was to examine hypothesis 2. This experiment also adopted social crowding (crowded group vs. non-crowded group) single-factor between-subjects design. The purpose of experiment 3 (N = 193) was to examine hypothesis 3. This experiment was a 2 (social crowding: crowded group vs. non-crowded group) × 2 (sense of power: high vs. low) between-subjects factorial design. The results of experiment 1 found that social crowding increased consumers' preference for intense sensory stimulation compared to non-crowding (M crowded = 5.70 vs. M uncrowded = 5.15，F(1，120) = 4.46，p = 0.037，η2 = 0.02 ). Hypothesis 1 has been verified. Experiment 2 not only found that social crowding increased consumers' preference for sensory stimulation (M crowded = 4.23 vs. M uncrowded = 3.76，t(221) = -2.09，p = 0.038，Cohen’s d = 0.28), It is also found that this effect is driven by perceived loss of control (LLCI = -0.60，ULCI = -0.03), in other words, social crowding will reduce the sense of control, leading to individuals seeking for strong sensory stimulation to restore the sense of control. Hypothesis 2 is verified. The results of experiment 3 show that the main effect of social crowding on consumers' sensory stimulation preference is still significant (M crowded = 3.58 vs. M uncrowded = 3.04，t(191) = -1.99，p = 0.048，Cohen’s d = 0.29), and the sense of power can moderate the main effect of this paper(F(1，189) = 8.20，p = 0.005，η2 = 0.04). For high-power individuals, social crowding will increase consumers' preference for intense sensory stimulation (M crowded = 4.06 vs. M uncrowded = 2.75，F(1，189) = 11.93，p = 0.001); for low-power individuals, the effect disappears(M crowded = 3.10 vs. M uncrowded = 3.33，F(1，189) = 0.35，p = 0.557). Theoretically, this study found that social crowding increases consumers' preference for strong sensory stimulation, and perceived loss of control plays a mediating role in this effect. This builds a bridge for the independent research of social crowding and sensory stimulation preferences. The construction of this bridge provides new antecedent variables for sensory stimulation preferences and new outcome variables for social crowding. At the same time, this paper also makes a contribution to the literature of the sense of power. In practice, this study has a certain reference value for companies to formulate marketing strategies, such as the need to consider the crowdedness of consumers' environment when recommending products to consumers.