Introduction to Forensic Psychology and Eyewitness Identification
Introduction to Forensic Psychology
Forensic psychology is the intersection of psychology and the legal system, applying psychological principles to legal issues.
The legal system is vast, offering numerous roles for psychologists beyond clinical practice.
Topics covered include:
Eyewitness identification: Psychological factors affecting eyewitness accuracy.
Investigative interviewing: Techniques for effective and ethical interviewing.
Detecting deception: Methods for identifying deceitful behavior.
Predicting recidivism: Assessing the likelihood of reoffending.
Special populations (children, adults, seniors) will be considered for each topic, addressing unique vulnerabilities and legal protections.
The course will discuss actual criminal cases (not too graphic), illustrating the application of forensic psychology principles.
Career paths in forensic psychology are diverse, including consultant, investigator, counselor, researcher, mentor, or teacher.
These careers involve intersecting with the legal system and are broader than just providing counseling in prisons, encompassing roles in law enforcement, courts, and correctional facilities.
Assessments
Two discussion board assignments: review and summarize a paper, provide your opinions, fostering critical thinking and engagement with course material.
Final exam: take-home exam with two essays, 48-hour window, requiring in-depth analysis and synthesis of course concepts.
Eyewitness Identification: Witnessing a Crime
The lecture simulates witnessing a crime, with the instructor role-playing a perpetrator, emphasizing the challenges of accurate recall.
Witness: Someone who saw a perpetrator commit a crime, providing firsthand accounts of events.
Perpetrator: The person who committed the crime, the focus of eyewitness testimony.
Terminology
Eyewitness reports the crime to police, describing the perpetrator, initiating the investigation.
Example: "young white male with short brown hair, clean shaven, wearing a gray t-shirt," illustrating the level of detail provided.
Police search for someone matching the description and may apprehend a suspect, based on initial eyewitness accounts.
Suspect: A person under investigation who may be innocent or guilty, requiring careful examination of evidence.
Police may put the suspect in a lineup, typically a six-person photo lineup, to assess eyewitness recognition.
Lineups consist of a suspect (innocent or guilty) surrounded by fillers, ensuring a degree of similarity among all lineup members.
Fillers: People who match the general features of the suspect, preventing the suspect from standing out.
A fair lineup ensures fillers match the eyewitness's original description, minimizing bias in the identification process.
The eyewitness can either identify someone from the lineup or say "not present," indicating whether anyone matches their memory.
Lineup Results and Errors
Filler Identification: Identifying a filler, which is an error but less serious since police know fillers are innocent, highlighting the importance of lineup composition.
False Identification/Mistaken Identification: Identifying an innocent suspect; a serious error with potentially disastrous consequences, contributing to wrongful convictions.
Prosecutors can use this as evidence to convict an innocent person, leading to imprisonment, underscoring the gravity of eyewitness misidentification.
Correct Rejection: Identifying no one when the perpetrator is absent from the lineup; a desirable outcome, affirming the witness's accuracy.
Target Present/Perpetrator Present Lineup: A lineup where the guilty suspect is present, testing the witness's ability to correctly identify the perpetrator.
Correct Identification: Identifying the guilty person in a target-present lineup; a desirable outcome, validating the eyewitness's memory and identification.
Miss: Failing to identify the guilty person in a target-present lineup, indicating memory failure or poor lineup construction.
Example: Ted Bundy was not identified in a live lineup and was released, leading to additional crimes, illustrating the real-world consequences of eyewitness error.
Types of Lineups
Target-Absent Lineup: Police apprehend an innocent suspect and put them in a lineup; the desired outcome is for the witness to identify no one, protecting the innocent from wrongful accusation.
Target-Present Lineup: Police apprehend the guilty suspect; the desired outcome is for the witness to identify the guilty person, ensuring justice is served.
Goal: Minimize false identifications and maximize correct identifications, optimizing the accuracy of eyewitness evidence.
Improving Lineup Procedures
Ideas for improvement:
Use interactive images or immersive 3D technology, enhancing the realism and memorability of the lineup.
Present images sequentially instead of simultaneously, reducing the likelihood of relative judgment errors.
Carefully select fillers based on their similarity to the suspect, ensuring a fair and unbiased lineup.
A theory is needed to guide decisions about improving the procedure, not just relying on intuition, emphasizing the importance of evidence-based practices.
Multi-Store Model of Memory
Information from the environment enters sensory memory, where visual, auditory, and haptic input are initially processed, highlighting the fleeting nature of sensory impressions.
Most information is forgotten unless it is important or salient, demonstrating the selective nature of attention and memory.
Important information is given attention, moving it to short-term memory (working memory), where it can be actively manipulated and rehearsed.
Through rehearsal, information moves from short-term memory to long-term storage, where it can be saved for days, weeks, months, or years through consolidation.
In long-term storage, information can be saved for days, weeks, months, or years through consolidation, forming stable and enduring memories.
Information can be retrieved from long-term storage back to short-term memory when needed, enabling conscious recall and recognition.
Forgetting can occur at any stage, illustrating the fallibility of memory processes.
Cue-Dependent Retrieval: When information is encoded into long-term memory, cues (context, state) are attached to it; reinstating these cues can help retrieve the memory, emphasizing the importance of context in memory retrieval.
Eyewitness Identification and Memory
Eyewitnesses attempt to encode the perpetrator's face and save it in long-term storage, creating a lasting memory representation.
Environmental and internal cues are attached to the episodic memory, influencing subsequent recall.
When presented with a lineup, eyewitnesses use the faces as retrieval cues, triggering memory matching processes.
The memory system generates a memory match signal based on how well the faces match those in long-term memory, determining the likelihood of recognition.
Successful Match: Results in a correct identification if the perpetrator's face was encoded well and good retrieval cues are present, leading to accurate identification.
Retrieval Failures: Can occur if the face was not encoded well or the retrieval cues are poor, resulting in a miss, highlighting the challenges of memory retrieval.
The memory matching process is noisy, introducing variability and potential errors.
Even if the perpetrator is absent, a witness's memory system can provide a weak memory match signal for every face in the lineup, leading to potential false identifications.
Witnesses may identify a face they've never seen before if the perpetrator's face was not encoded well, cues are poor, or someone resembles the perpetrator, emphasizing the susceptibility of memory to error.
The lecture introduces the concept of memory strength: weak, moderate, and strong signals, influencing the likelihood of accurate recognition.
Signal Detection Theory
It is a mathematical model that describes the ability to discriminate between signals and noise, providing a framework for understanding decision-making under uncertainty.
Originally developed in radar research during World War II, it has broad applications beyond its initial context.
Serves as the theoretical framework of recognition memory since the 1950s, guiding research and understanding of memory processes.
It is widely used in eyewitness identification research, providing insights into factors affecting accuracy.
The model considers the strength of a signal (d-prime ) and the strategy used to respond to that signal (criterion C), allowing for a nuanced analysis of decision-making.
Radar Operator Task
Operators use radar to monitor and respond to enemy aircraft, making critical decisions under pressure.
Decisions: identify a dot as a friend or an enemy, requiring accurate assessment of available information.
Outcomes:
Identifying a dot as a friend: can lead to a correct rejection if the dot is truly a friend or a miss if it's an enemy aircraft, highlighting the risks of misclassification.
Identifying a dot as a foe: can lead to a correct identification if it is an enemy aircraft or a false identification if it's a friend, emphasizing the trade-offs between different types of errors.
The goal is to maximize correct identifications and minimize false identifications, optimizing the effectiveness of radar monitoring.
Visualizing the Signal Detection Model
The model uses a continuum of signal strength: weak, moderate, to strong, reflecting the range of possible signal intensities.
Weak signals: appear clearly as friendly aircraft, allowing for easy differentiation.
Strong signals: appear clearly as enemy aircraft, facilitating accurate identification.
Signals near the middle are confusing, requiring careful evaluation.
Response Criterion (C): A line in the sand where any signal surpassing the criterion is identified as an enemy, and any signal falling below is identified as a friend, reflecting the operator's decision threshold.
Friendly aircraft can behave suspiciously, and enemy aircraft can conceal their identity, introducing uncertainty and the need for vigilance.
Two distributions of signals: one for friendly aircraft (light gray) and one for enemy aircraft (darker gray), representing the statistical distribution of signal strengths.
D-prime (d'): Represents the separation of these distributions; a large d-prime means the operator can clearly discern signals, while a small d-prime means they often confuse signals, reflecting the operator's ability to discriminate between signals.
Operators can have different d-prime levels and place their internal response criterion at different locations, influencing their decision-making strategies.
Analyzing Decision Outcomes
If 76% of the enemy aircraft distribution falls above the response criterion, the correct identification rate is 76%, reflecting the proportion of correctly identified enemy aircraft.
The miss rate (failure to identify enemy aircraft) is 24% (100% - 76%), indicating the proportion of missed enemy aircraft.
The area where the friend distribution exceeds the set criteria represents the false identification (False ID) rate, reflecting the proportion of friendly aircraft misidentified as enemy aircraft.
The lecturer continues to explain and emphasize visualizing the model and the relations between rates, and model parameters, reinforcing understanding of the model's application.
Applying Signal Detection Theory to Eyewitness Identification
Instead of radar screens, we're talking about memory signals, shifting the context to eyewitness testimony.
Signals come from innocent suspects (distribution on the left) and guilty suspects (distribution on the right), reflecting the range of potential sources of memory signals.
The continuum is memory strength, representing the intensity of the memory signal.
It can be applied to other tasks such as:
Radiologists diagnosing cancer,
Weather forecasting,
Machine learning,
Lie detection,
Risk assessment.
Minimizing False IDs and Maximizing Correct IDs
Static vs. Interactive Lineups
Static lineup: Shown in the lecture; may generate a lot of noise because of its limited number of retrieval cues.
Interactive lineup: Allows witnesses to rotate the images, enhancing retrieval cues.
Interactive lineups generate bigger d-primes, meaning witnesses perform better due to richer retrieval cues.