It is known that, when humans are subject to au- ditory and visual stimuli which are not in con- gruence with another, they tend to localize sound sources with more errors, than in a setting where the audiovisual stimuli is congruent. Further- more when it comes to binaural reproduction, it is known that humans have better ability to lo- calize given sound sources via playback, when the stimuli that is listened to is a reproduction us- ing their own physical characteristics - either by the use of binaural recordings or binaural Synthe- sis. Research was done regarding how subjects are able to localize a given sound source in Real- Life (RL), Virtual-Reality (VR) and in a setting with No-Visual (NV), with the two latter made, to make scenarios where audiovisual stimuli is not congruent. Scientific tests are conducted to estab- lish how well five subjects aged between 23 and 31 (mean = 26) perform localization in a RL settings. Followed by recordings of all subjects’ heads along with that of a Head And Torso Simulator, for later use in the VR and NV settings. The performance of the subjects in all three settings for two types of stimuli (Speech and Noise), for both individual and nonindividual recordings was observed and collected. Statistical analysis was applied to the the answers accumulated throughout the experi- ments. The analysis showed that the subjects per- formed best in a RL scenario, as opposed to when listening to binaural recordings. However when listening to ones own binaural recordings as op- posed to others’ recordings, the subjects showed an increase in performance for speech as stimuli - it is speculated that performance for noise is stimuli due to choosing Pink Noise. It was also shown that the between all error categories, the subjects had the most di culties trying to dis- tinguish between the loudspeakers in the Median Plane - especially for close neighbors.