Investigative Interviewing – Category and Location Clustering
WRITTEN BY:
Rui Paulo – Liverpool; John Moores University, United Kingdom
Mark Severino – Los Angeles Police Department (Ret.), United States
Science-based techniques are key for obtaining detailed and accurate accounts from witnesses and victims. However, some techniques like the change order mnemonic or a witness-compatible questioning can be too complex or challenging to use and sometimes elicit a limited amount of new information. To address this, Paulo et al. (2021) developed two retrieval strategies that could apply to most crimes/events and are easily adapted/used by detectives with various levels of training: Category Clustering Recall (CCR) and Location Clustering Recall (LCR).
CCR consists of asking eyewitnesses to recall a crime using broad information categories that are present in most crimes to guide and organize their recall (e.g., person-related details, object/ environmental-related details, location-related details, action-related details, and auditory information). This is based on research suggesting that focusing on one specific cluster at a time can trigger relevant associated memories that otherwise might not be recalled.
LCR consists of asking eyewitnesses to recall a crime using locations to guide and organize their recall. Therefore, witnesses are asked to focus and recall information concerning each separate location of the crime scene, one at a time. This was based on research suggesting that locations can be effective cues for enhancing episodic memory.
These techniques were initially intended to be used after a free recall to obtain additional detail from witnesses. In the lab, CCR was found to be more effective in eliciting information after a free recall in comparison with the change order mnemonic (Paulo et al., 2016) and witness-compatible questioning (Paulo et al., 2017). However, recent research suggests these techniques might also be more effective (and possibly replace) a free recall (Paulo et al., 2021; Shahvaroughi et al., 2021; Thorley, 2018).
There are several ways we would use the CCR/LCR concepts.
After the free recall, I would ask the subject to concentrate on PLATO topics individually. The problem I have with the context reinstatement mnemonic is it is a large request (e.g., concentrate on the people, actions, objects, odors, emotions felt, etc.) Too large of a request for a victim/witness who is having problems concentrating and many times too general.
CCR on the other hand allows the subject to concentrate on one topic at a time. The other advantage of CCR is I can include as many general topics as needed specific to the investigation/interview. For example, with the context reinstatement mnemonic, I could ask for objects observed and would receive a general description. With CCR, I would ask to concentrate on objects. Description of said objects, color, size, weight, markings, scratches exact locations, etc. The sketch Plan works hand in hand with CCR. The subject could draw the object, articulate the markings, etc. The Sketch Plan facilitates communication during CCR.
The same holds for Location Clustering. Locations can be different areas of a room the subject went to, and different rooms the subject was placed in (e.g., a kidnapped subject was moved and tortured in different rooms within a large commercial structure. The subject was tortured in each room by different suspects using different means). Locations can be physical locations the subject visited or was forced to travel to. This works very well with Haneen Deeb’s Mapping research. Location Clustering allows the subject to concentrate on one location at a time. This also works well with the Timeline technique, when the event being investigated is spread out over hours/days.
I have used the CCR concept when explaining the interview process (pre-free recall) and advising the subject to think about the different CCR categories. I like front-loading the categories before the free recall. After obtaining the free recall, I would select the topics from the free recall that required further follow-up. After exhausting the subject’s memory recall, I would still ask the subject to think about CCR general categories (PLATO) even if they were not mentioned during free recall. It is common for subjects to remember more when concentrating on one category at a time. (e.g., a witness who observed several people sitting in a vehicle only gave a general description during free recall. However, when asked to only concentrate on the individuals we obtained more fine-grain detail on the occupants as well as the subject remembering another person who had walked by the vehicle. Thus, another witness. This is compatible with laboratory research from Paulo et al. (2016) and Paulo et al., (2017) suggesting CCR can be effective in eliciting detailed information after a free recall, particularly person-related details.
In a tactical situation when debriefing a witness (e.g., active shooter), the CCR concept is useful when we need specific information promptly. There is no time for CI/free recall, that would come much later. The CCR is a very flexible concept and appears to be easy to understand for the subject and flexible for the interviewer. Anytime a subject can concentrate on a single topic, we always receive more information vs. a request that includes many topics.
When we select the topics from the free recall that require further explanation, we use the mini-context reinstatement to focus the subject on only that topic. For example, the subject stated during free recall that a male suspect entered the convenience store, walked down the subject’s aisle, approached the subject, and pointed a gun at the subject, demanding his wallet.
We would break down the event;
- When the subject first observed the suspect enter the store. CCR categories, other people near the suspect (additional witnesses), actions, objects, conversations, etc.
- When the subject observed the suspect walking down the aisle toward him. CCR categories, other people, objects, conversations, etc.
- When the suspect pointed a weapon demanding the subject’s wallet. CCR categories include the suspect (full description), objects (logos on clothing/hat), jewelry worn by the suspect, odors (cologne) conversations (exact words used by the suspect), and the weapon (full description).
Again, anytime we can have the subject concentrate on a specific category or event, we always receive more detailed information. Even with the free recall, we always would break down the free recall into smaller bites. I am an avid believer in CCR and can attest that most events can be broken down into CCR general categories. Sometimes, the subject’s free recall is very general lacking any detail as supported by academic research that found CCR and LCR to elicit more detailed accounts in comparison with a free recall (Paulo et al., 2021; Shahvaroughi et al., 2021; Thorley, 2018). CCR categories are a good way of prompting the subject to extend the free recall.
The opinions expressed in this article, which was prepared in Mark Severino’s capacity, are his own and do not reflect the views of the Los Angeles Police Department.
References
Paulo, R. M., Jones, E., & Mendes, R. (2021). Testing two retrieval strategies to enhance eyewitness memory: Category and location clustering recall. Psychology, Public Policy, and Law, 27, 140-149.
Paulo, R. M., Albuquerque, P. B., Vitorino, F., & Bull, R. (2017). Enhancing the cognitive interview with an alternative procedure to witness-compatible questioning: Category clustering recall. Psychology, Crime & Law, 23, 967-982.
Paulo, R. M., Albuquerque, P. B., & Bull, R. (2016). Improving the enhanced cognitive interview with a new interview strategy: Category clustering recall. Applied Cognitive Psychology, 30, 775-784.
Shahvaroughi, A., Ehsan, H., Hatami, J., Monajem, A., & Paulo, R. M. (2021). Testing a modified cognitive interview with category clustering recall in Iran. Applied Cognitive Psychology, 35, 148-159.
Thorley, C. (2018). Enhancing individual and collaborative eyewitness memory with category clustering recall. Memory, 26(8), 1128-1139.
