The Psychology of Criminal Justice: Criminal Profiling

Introduction to Criminal Profiling and the Washington Sniper Case

  • Foundational Question in Investigation: The central challenge in any criminal investigation is determining how to find the perpetrator. While forensic tests and eyewitnesses are standard, they have significant limitations:

    • Forensic Tests: These take considerable time and do not always provide a conclusive or complete result.
    • Eyewitness Accounts: These are not always available and, as seen in many cases, can be conflicting and unreliable.

  • Offender Profiling Definition: Offender profiling is the process of trying to identify a suspect based on what is known about the crime itself, rather than relying solely on external evidence like DNA or testimony.

  • Case Study: The Washington Snipers:

    • Timeline and Scope: Between October 2 and October 22, 2002, 10 people were murdered and 3 wounded in Maryland, Virginia, and Washington, DC.
    • Modus Operandi (MO): Victims were shot from a distance with a single rifle shot. The killers disappeared in the resulting confusion.
    • Victimology: The attacks seemed random. Locations included grocery store car parks, petrol stations, and victims performing mundane tasks like mowing lawns or reading. No clear link existed between victims.
    • Community Impact: The randomness caused widespread fear and panic, creating immense pressure on investigators despite having very little information.

  • Public Profiles and Inconsistencies:

    • Many individuals offered opinions, including forensic psychologists, criminologists, homicide detectives, and both self-taught and FBI-trained profilers.
    • Pat Brown (Self-taught profiler/author): On October 11, 2002 (99 days and 99 shootings into the spree), she stated on CNN that it was likely a single individual.
    • Candice DeLong (Retired FBI agent/field profiler): On October 16, 2002, quoted in the New York Times, she also suggested a single perpetrator and hinted at specific occupations.
    • Gregg McCrary (Former FBI profiler/instructor): On October 23, 2002, he suggested the perpetrator was White, drawing on the demographic makeup of the Washington region.
    • Robert Ressler (Former FBI profiler): On October 18, 2002, he argued there was more than one person involved. This directly contradicted the assumptions of Brown, DeLong, and McCrary.
    • Pat Brown's Rebuttal: On October 22, 2002, Brown publicly disagreed with Ressler's multi-offender theory on CBS.

  • The Reality of the Case:

    • Convicted Offenders: John Muhammad (4141-year-old Gulf War veteran) and John Lee Malvo (1717-year-old Jamaican immigrant).
    • Mismatch with Profiles: The offenders were not White, not firemen, not construction workers, and did not fit the "Rambo movie fan" stereotype.
    • Capture: The suspects were not caught via profiling. Instead, a phone call from one of the shooters linked the crimes to an earlier shooting. A fingerprint found at that earlier scene led to their identification.

Criminal Profiling Methods

  • Core Principle: The fundamental belief is that a careful examination of the offense can reveal characteristics of the offender.

  • General Process:

    • Gather all information about the crime (details of the act, location, victim info).
    • Draw up a profile of the person most likely to have committed the crime.
    • Profile includes: Demographics (age, gender, ethnicity) and motivation (the "why" behind the behavior and choice of time/location).

  • Primary Application: Serious serial crimes like murder, especially when standard methods fail.

  • Major Approaches:

    1. Criminal Investigative Analysis (CIA): Developed by the FBI.
    2. Investigative Psychology: Associated with David Canter.
    3. Diagnostic Evaluation.
    4. Crime Action Profiling.
    5. Geographical Profiling: Focuses on locating the offender's base.

  • The FBI's Criminal Investigative Analysis (CIA):

    • Development: Developed at the FBI Academy in Virginia and the Behavioral Science Unit.
    • Methodology: Based on officer experience and in-depth interviews with a small number of volunteer convicted sexual murderers.
    • Caveats: The memories of these murderers may be imperfect, and volunteers may not represent the broader category of offenders.

  • FBI Typologies: Organized vs. Disorganized:

    • Organized Offender:
      • Leads an orderly life, reflected in the crime.
      • Average to high intelligence and socially competent.
      • Likely to have skilled employment.
      • Plans offenses, uses victim restraints, brings and removes the weapon from the scene.
      • Offense often precipitated by situational stress.
    • Disorganized Offender:
      • Shows little preparation or planning; crime scene appears random.
      • Uses whatever is at hand as a weapon and often leaves it at the scene.
      • Little attempt to conceal evidence.
      • Socially incompetent with below-average intelligence.
      • Likely to live alone and work/live near the crime scene.
    • Mixed Offender:
      • Elements of both organized and disorganized behavior.
      • Causes: Surprises during the crime (e.g., victim fights back harder than expected, extra people present), evolving methods, or a failure to translate a perfect fantasy into reality (Hazelwood and Warren, 2001).

  • Investigation Stages (What, Why, Who):

    • The WHAT: Behavior present in the crime (victimology, scene reconstruction, interpolsenal transaction, forensics).
    • The WHY: Determining the motivation for each crime scene detail.
    • The WHO: Resulting demographic/physical hypotheses (race, age, occupation, marital status, vehicle type).

  • Modus Operandi (MO) vs. Signature:

    • MO: The functional way the crime is committed.
    • Signature: An individualized, specific behavior that serves no obvious functional purpose (like keys being taken) but fulfills a psychological or emotional desire for the offender.

Professional Profilers and Empirical Evaluation

  • The Validity Gap: FBI methods rely heavily on field experience, but published empirical analyses are rare and controversial.

  • Pinizotto and Finkel Study (1990):

    • Goal: Determine if professional profilers are more accurate than non-profilers and if their thought processes differ qualitatively.
    • Participants (N=24N=24 analyzed):
      1. Profilers: 6 police detectives trained by the FBI (1-year program, 1-6 years experience).
      2. Detectives: 6 homicide/sex offense detectives without profiling training.
      3. Psychologists: 6 participants.
      4. Students: 6 participants.
    • Materials: Data from two closed cases: one homicide and one sexual offense. Information included victim reports, crime scene photos, autopsies, and toxicology reports. Maps were excluded for anonymity.
    • Tasks: Recall details, identify important scene variables, write/audio record a profile, answer 20 MCQs about the suspect, and rank five suspects in a line-up.

  • Results:

    • Report Quality: Profilers provided "richer" and longer reports with more predictions.
    • Accuracy: Profilers were significantly more accurate in their written predictions for both cases.
    • Sexual Offense Case: Profilers scored significantly better on MCQs and the line-up task compared to the other groups.
    • Homicide Case: Profilers performed no better than detectives, psychologists, or students on MCQs or line-up tasks.
    • Qualitative Process: There was no evidence that profilers processed information in a fundamentally different way than non-profilers.

  • Discussion on Homicide Discrepancy: The homicide offender might not have fit usual "base rates," or critical missing information noted by the profilers might have hindered them. It highlights the risk when an offender is not "typical."

Investigative Psychology

  • Alternative Model: David Canter’s Investigative Psychology focuses on actions rather than the "why" (psychological mechanisms).

  • Methodology: Statistical analysis of large datasets rather than clinical intuition. Results are peer-reviewed.

  • Salfati and Canter Study (1999):

    • Analyzed 82 solved British single perpetrator-single victim homicide cases.
    • Three Broad Themes of Aggression:
      1. Expressive (Impulsive): Aim is to hurt. Multiple stab wounds, weapon brought to or found at the scene. Often involves prior violent/sexual/property offenses.
      2. Instrumental (Opportunistic): Murder as a byproduct of other motives.
      3. Instrumental (Cognitive): Planned; evidence removed, body hidden outside. Often involves military service or prior prison sentences.

  • David Canter’s Five-Factor Model:

    1. Interpersonal Coherence: People interact with victims in ways consistent with their everyday social style.
    2. Time and Space: Choice of location is rarely random; offenders select familiar areas where they feel in control.
    3. Forensic Awareness: Level of effort to hide evidence indicates criminal experience or lack thereof.
    4. Criminal Career: Criminal behavior is relatively consistent; changes may reflect reactions to past failures (e.g., starting to drug victims after a previous struggle).
    5. Criminal Characteristics: Aligning specific offense traits with broader types of crime to narrow search categories.

  • CIA vs. Investigative Psychology (IP):

    • CIA: Focuses on motivation and demographics; uses clinical experience/intuition.
    • IP: Focuses on actions and location; uses statistical methods and peer review.
    • Risk of IP: Over-reliance on statistics can miss "statistical outliers" or unique offenders.

Geographic Profiling

  • Core Objective: Identify where an offender is likely to be based (home or work) to prioritize search areas.

  • Basic Principles:

    • Least Effort Principle: People choose the option requiring the least effort (closest proximity).
    • Distance Decay: Offense frequency decreases as distance from home increases.
    • Buffer Zone: Offenders avoid committing crimes too close to home where they are well-known.
    • Mental Schemata/Maps: Internal representations of an area based on interaction (e.g., a driver's map vs. a bus rider's map).

  • Offender Types (Canter and Larkin, 1993):

    • Marauder: Uses home as a base; offenses occur around the home location. The home is often inside a circle defined by the two furthest offenses.
    • Commuter: Travels to a different region to offend.

  • Statistics on Marauding:

    • Original 1993 study: 87%87\% of offenders were marauders.
    • Meaney (2004) Australian study: Only 35%35\% of burglars were marauders, compared to 93%93\% of sex offenders.

  • Complications:

    • Contextual features (e.g., specific victim types only found in specific areas).
    • Linkage blindness: Ensuring all analyzed crimes were actually committed by the same person.
    • Spatial bias: Analysis is only as good as the input of captured crimes.

  • Practical Tools: Computer programs (e.g., Rossmo, 2000) prioritize geographical areas based on distribution. Rossmo suggests a minimum of 55 accurate locations for reliable output, assuming the offender hasn't moved.