Ligand-gated ion channels propagate electrochemical signals in the nervous system by opening ion-selective pores in response to neurotransmitter release. A diverse set of allosteric modulators including neurosteroids, anesthetics, and lipids modulate their function in myriad ways, suggesting a complex conformational landscape. However, structures of ligand-bound ion-channel complexes can be difficult to capture by experimental techniques like cryogenic electron microscopy, particularly when binding is dynamic or transient. Here, we used computational methods to identify a possible bound state of a modulatory stimulant derivative (4-bromoamphetamine) in a cryptic vestibular pocket of a mammalian serotonin-3A receptor. Starting from an experimental activated structure containing a closed pocket, we first applied a molecular dynamics (MD) simulations-based goal-oriented adaptive sampling method to identify possible open-pocket conformations. To find plausible ligand-binding poses, we performed ensemble docking of the newly identified modulator, and reweighted docking scores by the Boltzmann energy function derived from Markov state model analysis of our trajectories. We then performed replicates of unbiased MD simulations of representative complexes in two forcefields to estimate ligand stability, and screened the most stable complexes for accessibility to the aqueous environment. For one relatively stable and accessible site, mutations predicted to disrupt lig- and binding were validated by electrophysiology recordings inXenopus laevisoocytes, and provided a mechanistic rationale for allosteric stabilization of an activated state. Given the pharmaceutical relevance of serotonin-3 receptors in emesis, pain, psychiatric and gastrointestinal diseases, characterizing relatively unexplored modulatory sites in these proteins could open valuable avenues to understanding conformational cycling and designing state-dependent drugs.