Regulated protein synthesis, or translation, is essential for life, and allows the cell to flexibly respond to external stimuli and stress. Conversely, dysregulated translation is a hallmark of diseases including cancer, viral infection, and developmental disorders. Translation is regulated principally through its initiation phase, where a crucial regulatory function of the initiation machinery is to ensure selection of the correct translation start site on messenger RNA. Failure to do so compromises the proteome by permitting synthesis of elongated, truncated, or nonsense proteins. In eukaryotes, start-site selection requires a directional search beginning at the 5’ end of the message. This search must move the megadalton ribosomal pre-initiation complex (PIC) efficiently through mRNA leader sequences that can span tens, hundreds, or even over a thousand nucleotides, and then halt this motion at exactly the correct start site. A linear “scanning” mechanism was first proposed over 40 years ago for this remarkable biophysical feat. However, fundamental properties of scanning have never been directly validated experimentally, and alternative mechanisms have been proposed. The importance of motion through the mRNA leader in translational control has also been brought into renewed sharp focus in recent years with the discovery that many mRNAs contain upstream open reading frames in their leaders that control translation of the main open reading frame; the scanning mechanism lies at the heart of how these are utilized. A critical barrier to progress is the remarkable molecular complexity and dynamism of the scanning machinery, whose numerous transient intermediates have made it challenging to characterize experimentally. Directly visualizing scanning in real time would allow many key unsolved questions to be addressed. Single-molecule methods are uniquely positioned to do this with the molecular resolution required to dissect mechanism. We have developed a single-molecule fluorescence assay for scanning of a reconstituted yeast PIC on full-length mRNAs. Here we will apply this assay to address the scanning mechanism. In Aim 1 we will directly determine the physical mechanism of motion in scanning, establishing the contributions of mRNA sequence and structure to the scanning rate. In Aim 2, we will elucidate how scanning directionality is established and maintained, focusing on the central translational helicase, eIF4A. We will distinguish between proposed mechanisms for how eIF4A transduces the chemical potential of ATP to bias scanning direction. In Aim 3, we will define the roles of pre- initiation complex components in scanning, with experiments that isolate their contribution to scanning specifically, rather than their aggregate functions throughout initiation. These studies will establish a physical- mechanistic model for scanning that will deepen understanding of translational control in health, and inform ongoing efforts to understand and rever...