Over the course of the last two decades, DNA profiling has grown to be
one of the most widely known and applied techniques for the
identification of biological samples in forensic science. Arguably,
the “fame” of DNA profiling is responsible for the courts’ current
interest in the raising of standards for scientific examination of all
forms of physical evidence (e.g. toolmarks, soils, dust, questioned
documents, shoe prints, fire debris, fingerprints, gunshot residue,
tire tracks, hairs, fibers, etc.). It is undoubtedly the clear
applicability of simple statistical methods to DNA profiling
techniques which lies at the core of their success. Our laboratory
and our collaborators at R.I.T., Cedar Crest College and in Law
Enforcement are interested in addressing the need for a sound
scientific basis for studying and comparing all categories of physical
The 1993 Daubert decision1 by the U.S. Supreme court stipulates that
a scientific technique before the court should satisfy one or more of
• Has empirical testing for the technique been done?
• Are the conclusions of the technique falsifiable?
• Has the science been subject to peer review?
• Are there known error rates for the technique?
• Is there general acceptance of the technique within the relevant
Examination of physical evidence is frequently challenged in court on
the basis that claims of uniqueness are typically determined via
examiner experience rather than with empirical data and use of
statistics. The 2009 National Academy of Sciences report on the
raising of standards in forensic science highlights this fact and the
need for objective comparison measures and error rates.
Our approach recognizes that all of the above classes of physical
evidence can be studied as patterns. In order to treat pattern
comparisons as objectively and rigorously as possible we look to the
mathematics of statistical learning and algorithmic complexity
theories. The practical implementation of these ideas forms the basis
of modern machine learning. By using these statistical methods to
analyze empirical “physical evidence data” one can validate and lend
objectivity to a field seen largely as subjective.
We are implementing progressively more sophisticated computer
programs to carry out pattern recognition on physical evidence data we
collect. As we collect data it is being stored in an ever expanding
database to test and refine the pattern recognition algorithms we
implement and to observe changes (if any) in the pattern data over
time. Also, we are developing protocols for the use of multiple
machine learning methods to analyze the degree of similarity between
arbitrary patterns in order to quantitatively define the term
Our research on toolmark comparisons has recently been profiled in
the New York Times, Popular Mechanics and The New Scientist. We are
also seeking major funding for the expansion of our database, in
particular for toolmarks.
1In the 1993 Daubert decision, the U.S. Supreme Court elevated the
importance of federal rule of evidence 702 and rejected the Frye
“general acceptance rule” governing the admissibility of scientific
evidence. The Federal Government and about half of the States (and
counting) abide by the “Daubert Rules” of some form thereof.
Previous Featured Scientist
Dr. Peter R. De Forest is Professor of Criminalistics at the John Jay College of Criminal Justice, City University of New York where he has taught for 37 years. Prior to joining the faculty and helping to found the Forensic Science BS, MS, and Ph.D. Programs at John Jay and the City University of New York, he worked in several laboratories. He began his career in forensic science at the Ventura County Sheriff’s Crime Laboratory, Ventura , California in 1960. He earned a Bachelor of Science Degree (1964) in Criminalistics and a Doctor of Criminology Degree in Criminalistics (1969) from the University of California at Berkeley , under Dr. Paul L. Kirk. In addition to his university teaching and research activities, he also serves as a scientific consultant and expert witness for police departments, prosecutor’s offices, municipal law departments, public defender agencies, and private attorneys in criminal and civil casework in the US and Canada. The consultation has included advisory panel membership on major case investigations in the UK . He is the author or co-author of several book chapters, a textbook, and numerous articles in scientific journals. In addition to membership in several scientific societies, he served as a member of the editorial boards of journals including the Journal of Forensic Sciences. For over ten years, dating from the inception of the American Board of Criminalistics (ABC), Professor De Forest served as the chairman of ABC Examination Committee, which was responsible for designing and administering certification examinations in a range of forensic science specialties. He has presented lectures and workshops for several professional societies and in other universities and has served as Visiting Professor at the University of Strathclyde , Glasgow , Scotland . During the fall 1997 semester he served as Exchange Professor with the National Crime Faculty at the Police Staff College , Bramshill , England and also delivered the Founders Lecture for the California Association of Criminalists. Dr. De Forest is a past commissioner with the Forensic Science Program Accreditation Commission (FEPAC) of the American Academy of Forensic Sciences (AAFS) and is Criminalistics Section Chairman of the AAFS for 2006-2007. Awards received include the Paul L. Kirk Award (1999) of the Criminalistics Section of the AAFS.
Saul Kassin (Distinguished Professor) received his Ph.D. in personality and social psychology at the University of Connecticut . In 1984, he was awarded a U. S. Supreme Court Judicial Fellowship, and spent the year at the Federal Judicial Center . In 1985 he was a postdoctoral fellow and visiting professor in the Psychology and Law Program at Stanford University . Dr. Kassin has conducted research on police interviewing, interrogation, and the elicitation of confessions, and on the psychology of eyewitness identifications and testimony. He has also studied the impact of these and other types of evidence on jurors and jury decision-making. Dr. Kassin is a Fellow of the American Psychological Association and the American Psychological Society. He has served on the editorial board of Law and Human Behavior since 1986. He lectures frequently to judges, lawyers, psychologists, and law enforcement groups. He has worked as an analyst for various news media and as a consultant and expert witness in federal, military, and state courts. He has also co-authored or edited a number of scholarly books, including: Confessions in the Courtroom, The Psychology of Evidence and Trial Procedure, The American Jury on Trial: Psychological Perspectives, and Developmental Social Psychology
A Short Course in Crime Scene Analysis for Trial Lawyers
in Criminal Cases
September 17-18, 2009
Cohosted with the Crime Scene Academy, and sponsored in conjunction with The Legal Aid Society of New York, The Bronx Defenders, The New York Office of the Chief Medical Examiner and The Northwestern University School of Law.
This course is being held at
Office of the Chief Medical Examiner
Forensic Biology Building
421 East 26th Street New York, NY
National Public Radio's "On Point"
The Center will be convening scientists and criminal justice practitioners of a wide-ranging discussion of the practical impact of the National Academy of Science's recent analysis of the state of forensice science and its future. The issue is discussed by the Center Director and others on National Public Radio.
Center Wins Grant for Ground-Breaking Arson Screening Program
Dr. Nicholas Petraco
The Center has won a grant of $248,000 from the JEHT Foundation for an innovative Arson Screening Project, designed to assess the damage done by generations of “bad science” arson convictions. The Arson Screening Project will be the first program to address systematically the roles played by improved science in revealing mistaken convictions in a non-DNA context.