Forensic Art Fine Art


A facial composite is an image created by an artist based on a crime witness/victim description of an offender. The aim is not to create a polished kind of portrait but to achieve a resemblance to the subject being described from memory. An integral part of this process is an investigative interview called a cognitive interview. The cognitive interview sets a guideline for a successful interaction between the interviewer and the interviewee, but since every case and interview is different, there is no strict script to follow. The main concern is to let the witness describe the subject at his or her own pace, with the artist providing simple, unbiased prompts.


Interviewing an individual who holds a mental image of a person needing identification is a challenge. People's memory and verbal abilities vary considerably, which is one of the reasons why the accuracy of composites is also variable. The arrest of a suspect cannot therefore be made from a facial composite alone. It can however aid police investigations by providing a lead to the identity of a suspect.

The facial composite can be a sketch or a computerized image (e.g. Efit). I have used both techniques in my studies and have developed strong sketching skills during and after the course.

I received intensive training in essential draftsmanship and interview techniques during the facial composite module led by Gregory Mahoney, within the MSc Forensic Art course. However, a personal drawing and interviewing style was always encouraged and accepted.



MSc dissertation project at the University of Dundee

My research project, during summer 2010, aimed to discover whether it was feasible to conduct a Cognitive Interview via the internet, using a live web camera, to create a facial composite. The participants were shown an image of an unknown individual for one minute, without knowing the study's purpose at this stage. They returned to be interviewed 24 hours later. I did not see the images prior to the interview and was located in a different room to the volunteers. The primary aim was to find out about the interaction via the participants' surveys and my own researcher analysis. The accuracy and draftsmanship of the composites was a secondary aim. The participant survey results were encouraging and indicated that interaction for this purpose was effective in laboratory settings.

Here are a few examples from the study. The subject images are courtesy of the City of Sydney.


Composite A and B: these composites highlight the concentration required from the witness during the interview, in order to achieve a successful likeness. Maintaining a mental image of a subject is a difficult task, complicated further by a need to provide clear and concise descriptions to an artist.

Composite C: In this interview the volunteer was not happy with the proportions of the sketch towards the end of the interview. When the drawing pad was accidentally titled forwards, the altered proportions matched better with the volunteer's mental image of the subject face.This was initially considered a risky method of displaying a composite, but ended up providing a favourable result when compared to the suspect photograph.

Abilities vary between people and it is impossible to recall all the fine detail. Some features can also be recalled wrongly like in Composite D with respect to the ears, but enough accurate information can result in a close enough likeness to the subject.


Composite E: the witness confidence level doesn't necessarily tell us what the outcome is going to be like. The volunteer in this interview didn't believe in her own abilities, yet ended up eliciting more information than she expected.

Composite F and G: These interviews had technical problems that interfered with the fluidity of the interview process. The composites ended up being much more accurate than I had however expected.

The time it takes to perform an interview is dependant on the witness/victim. During this study, the interview times varied between 40 minutes for Composite H and 2 hours 30 minutes for Composite I.

I made a short video (less than ten minutes) of my project for the Masters Show 2010 at the University of Dundee. If you are interested in seeing it, please email me.


Further research

I have continued to research my dissertation subject and firstly wanted to compare interviewing via Skype to face-to-face interviews. I worked with Dr Robert Nash from the University of Surrey and conducted more interviews there. This time the participants (42) were mainly undergraduate students and were aware of the full task before viewing a still image of an unknown person for one minute. The participants returned to be interviewed the next day. According to a rating and a sorting task, the composites from the face-to-face interviews were seen to be better. The difference wasn't big, however enough to be statistically significant. This project raised more research questions, which I am planning to explore further.

An academic paper of this research was accepted for publication:

Kuivaniemi-Smith, H. J., Nash, R. A., Brodie, E. R., Mahoney, G., & Rynn, C. (2014). Producing facial composite sketches in remote Cognitive Interviews: A preliminary investigation. Psychology, Crime & Law, 20, 389-406.

See some composites from the study by clicking on the images to the right hand side. Subject images are from this source:

Minear, M. & Park, D.C.(2004). A lifespan database of adult facial stimuli. Behavior Research Methods, Instruments, & Computers. 36, 630-633. See specific stimuli sets for additional citation information. - See more at:


      Skype interview composites



      Face-to-face interview composites

Bottom Left

All rights reserved © 2014 Heidi Kuivaniemi-Smith            

Bottom Right

Valid CSS! LinkedIn qrcode Valid XHTML 1.0 Transitional!

Traditional portraits Age progression pastels Windsor Forensic Art Digital portraits Facial Reconstruction Family portraits Composites tablet drawing

Buckinghamshire portrait artist Marlow Cognitive interview London Portrait Artist Finland Pencil drawings Efit alternative portfolio UK portrait artist Photo restoration Bucks Open Studios