Human deception detection accuracy is notoriously low generally only reaching levels slightly greater than chance. However, in the advent of recent technological developments in the last 10-20 years, Computer Vision has emerged as a field at the forefront deception studies. Computational methods have undoubtedly provided a means to identify deceptive behaviours at a significantly greater accuracy level than humans. However, the real-world practicalities of these methods mean large-scale implementation is as-of-yet unlikely. Similarly, questions remain surrounding whether technology should entirely replace human detection or instead look to enhance it. This dissertation explores how visualisations may affect and potentially enhance human deception detection accuracy levels. A novel, lightweight software - relying on the opensource dlib and imutils Python libraries for face detection and point identification (which in-turn use a HOG approach for face detection and an ensemble of regression trees for landmark prediction) – used geometric calculations to form point, landmark, and statistic visualisations. An interactive online survey was distributed to assess the extent to which these visualisations effected participant’s deception detection accuracy. 38 responses were collected in total across a 3-week period. Statistical analysis generated some interesting discussion points surrounding how landmark visualisations achieved a 12-14% higher accuracy level and how visualisations may produce conditions to form an accuracy-confidence correlation. Ultimately, this research acts as preconceptual foundation for future research exploring the effects of visualisation on human deception detection accuracy.