This project aimed to produce a method of visualizing long-term perceptual memories of faces, to visualize the self-face representation. It aimed to produce the method, test it via the use of an ‘ideal theoretical observer’ using computer vision as initial proof of concept and then pilot with human subjects to assess feasibility. Results were initially promising with images produced being significantly more similar to targets than non-targets. Unfortunately, this finding was not mimicked in participant data leading to alternative approaches to data collection being trialled as well as test-retest validity measures to be run. Both initial test-retest of the original method as well as the alternative approach devised produced only poor non-significant correlations between participants. It remains an open question as to whether the images produced contain a valid ‘self’ like quality to participants.