Lionfish (Pterois volitans and P. miles) have become invasive across the western Atlantic and are considered one of the most heavily impacting invasive marine vertebrates. Most lionfish research focuses on consumptive effects in small-scale patch reef studies, with little research on their behaviour, non-consumptive effects or regional impacts. Using binary choice laboratory experiments, I found that lionfish were not attracted to conspecific cues. Instead, habitat preference likely drives aggregation formation, as in situ observations revealed that lionfish avoid fine-scale complexity but are attracted to broad-scale complexity, such as caves and overhangs. Shelter use by lionfish may impact native shelter-using species, such as commercially important spiny lobsters and ecologically important long-spined sea urchins. I found that lionfish did not affect shelter use by either native species in laboratory experiments, indicating that negative impacts of lionfish from competition for shelter may not be as widespread as feared. Interestingly, lionfish reduced their shelter use in the presence of the native species. I also investigated herbivorous fish grazing as a potential non-consumptive effect of lionfish. By comparing the same small areas of reef with and without lionfish, I found that responses to lionfish varied even within a small reef system, indicating that lionfish presence is not the only driver of the observed changes. The highly variable results of small-scale studies like these highlight the urgent need for regional-scale studies to determine the true extent of lionfish impacts in their invaded range. Prior to this thesis, no research had been conducted into the impacts of lionfish on native fish communities that simultaneously considered sites with different invasion timelines across the western Atlantic. I took a novel approach to understanding large-scale impacts using temporally offset additive mixed models centred on year of first invasion, combining 11,000 transects from eight countries with 14 contextual variables known to affect reef fish communities. I found clear shifts in native fish communities between pre-invasion and two or more years post-invasion, indicating that lionfish invasion rapidly results in a stable change in native fish communities. I found pre-invasion increases in native fish abundance, species richness and diversity that continued in the initial years of the invasion, however, I detected simultaneous declines in all metrics at five years post lionfish invasion. This thesis provides new insights into invasive lionfish behaviour and non-consumptive effects and provides the first suggestion that lionfish exert region-wide impacts by using a novel approach for studying the large-scale impacts of invasive species.