I just posted the entire text instead of linking to it. This is an interesting article on how to extract information from witness. I believe it is a bit dated as it does not include waterboarding.
by Margo Bennett and John E. Hess
Margo Bennett, M.Ed.
John E. Hess, M.Ed.
Special Agents and Instructors
When interviewing crime victims, few investigators begin with questions
such as: How tall was the subject? What color was his hair? Did he have
any scars? Common sense, experience, and fundamental training lead
investigators to the conclusion that such specific questions give
witnesses little opportunity to tell what they know. Instead, open-ended
questions tend to produce the best results. A question like, "What did
he look like?" eliminates the need for investigators to anticipate every
detail of description victims may have noted. Investigators can always
follow up the witness' statements with specific, direct questions to
fill in gaps. At least, that is what many interview textbooks suggest.
But what happens when even these direct questions fail to produce the
details needed from witnesses? The cognitive interview method is a
proven technique, effective because it provides interviewers with a
structured approach to help retrieve such details from the memories of
Consider the following scenario: At a robbery scene, a uniformed officer
briefs the investigating detective. Hoping to obtain additional
information, the detective approaches the clerk, introduces himself, and
sensing her anxiety, takes some time to assure her that she has nothing
to worry about. He tells her he understands the trauma she has just
undergone, gets her a cup of coffee, and delays asking any questions
until she has regained her composure. He then tells her that he needs
her help and asks that she start at the beginning and tell him exactly
what happened. She replies:
"I was behind the counter when all of a sudden, I heard a voice telling
me to give him all the money, and I would not get hurt. I looked up and
saw a man wearing a ski mask pointing a gun right at me. I just froze
and stared at the gun. He told me to get a move on or there would be
trouble. I opened the cash register and handed him all of the bills.
There was just under a hundred dollars in the register. He then told me
to lie on the floor and not move. I did as he told me and waited until I
was sure he was gone. I yelled to Joe, the manager, who was in the
office, who asked me if I was okay. He then ran to the phone and called
the police. The next thing I knew, the police officer arrived, and I
told him the same thing I just told you. I don't know what the guy
looked like, where he came from, or how he got away. I'm sorry I can't
be more help."
The detective tells her that she has been very helpful and that now he
would like to go over the story again, and this time, if she doesn't
mind, he will interrupt her with questions as she goes along. As she
retells her story, he constantly probes for additional details, such as
the possibility of additional witnesses, more descriptive data regarding
the subject and his weapon, words he may have used, noticeable accent,
and the means of his escape. However, except for a bit more descriptive
data, the victim was correct; she had told the responding officer
everything she could remember.
THE PROBLEM: INABILITY TO REMEMBER
The above scenario illustrates a problem encountered by many
investigators. That problem results not from investigators being unable
to ask good questions but simply from witnesses who are unable to
provide the answers. Responses such as, "I don't remember," "That's all
I saw," or "I can't recall" frustrate many interviewers on a regular
basis. In the past, this led investigators to try hypnosis as a means of
enhancing witness recall. Improved results verified what many
investigators suspected--an inability of witnesses to remember, not a
lack of observations, was the main problem. (1) Although investigators
achieved some success through hypnosis, those successes did not last
long. Courts, on a regular basis, began ruling in favor of defense
attorneys who alleged that hypnotically elicited information may contain
flaws and that hypnosis as a means of refreshing recall lacks scientific
acceptance. (2) Therefore, investigators now primarily reserve hypnosis
for situations where the need for lead information supersedes all other
considerations. They know full well that using hypnosis will probably
disqualify a witness from testifying.
SOLVING THE PROBLEM: THE COGNITIVE INTERVIEW
To enhance witness recall without the stigma attached to hypnosis,
Ronald P. Fisher and Edward Geiselman, professors at Florida
International University and UCLA respectively, have developed a system
they call the cognitive interview. Although their process contains few,
if any, new ideas, they have systematized some techniques which have,
for the most part, been used by investigators only in a sporadic,
piecemeal fashion. Research indicates that the cognitive approach to
interviewing witnesses increases the quantity of information obtained
(3) and does not jeopardize the witness' credibility in court, as
This article compares the traditional interview with the cognitive
interview. Specifically, this article deals with the cognitive interview
technique as it assists witness memory retrieval by: 1) Reinstating the
context of the event, 2) recalling the event in a different sequence,
and 3) looking at the event from different perspectives. It also deals
with specific retrieval techniques and time factors that affect the
Reinstate the Context
Traditional interviews of victims and witnesses, similar to the one
described above, usually begin with interviewers first taking the time
to make introductions and putting witnesses at ease before asking, "What
happened?" or "What can you tell me about...?" Then, specific questions
follow that are geared to fill in the gaps inadvertently left by
witnesses. Proponents of the cognitive interview suggest this will not
usually produce optimum results. Asking people to isolate an event in
their minds and then to verbalize that event requires them to operate in
a vacuum. Even without the trauma that often results from involvement in
a crime, common sense says that human memory functions better in
context. The cognitive interview process takes this into account.
What is meant by context and how do interviewers establish it? Simply
put, interviewers make efforts to reestablish the environment, mood,
setting, and experiences by asking witnesses to relive mentally the
events prior to, during, and after the crime.
Let's return to the robbery scene described above with the detective who
had already introduced himself to the victim and asked for her help.
Instead of asking her what happened during the crime, using the
cognitive interview approach, he proceeds as follows: "It's only about
10:00, and it's already been a pretty full day for you. How about
telling me how your day started. Tell me what time you got up, the
chores you did, the errands you ran and anything else that happened
before you came to work."
As she recounts her activities, he joins the conversation, discussing
events with her, including the problems of a working mother, what she
fixed for breakfast, and any other details that she mentions. Only when
they have developed a clear picture of those events does the detective
next suggest that the victim describe her travel to work. He handles
this portion of the conversation in the same way. He does not ask
perfunctory questions geared to getting her quickly to the crime scene,
but rather, he discusses her commute to work in depth. They discuss the
route she took, weather and traffic conditions she encountered, events
she may have noticed, and finally, where she parked her car and what she
noticed at that time. He wants her not only to just describe her day in
general but also to relive it.
He uses the same interview technique regarding her arrival at work. By
the time they finally get to the discussion of the robbery, they have
put the event into context. In many instances, this process enhances
measurably a person's retrieval of stored information. Thus, witnesses
can see details of the robbery in their proper sequence and context.
Concentration is more focused than during any previous interviews, which
may have only consisted of isolated questions and answers. The response,
"I can't remember," will occur less frequently.
To continue the interview and further develop the witness' recall,
another phase of the cognitive interview follows next in sequence.
Initially, retrieving information from witnesses occurs in a normal,
chronological flow of events. However, when recounting from memory,
people tend to edit as memory playback occurs. This results in a summary
based upon what witnesses regard as important. Therefore, interviewers
should address this problem by prompting witnesses not to hold back even
the most insignificant detail. Even so, most interviewers can cite
experiences where valuable information went unmentioned because
witnesses chose to omit it.
By changing the sequence of recall, witnesses can look at each stage of
the event as a separate entity much akin to looking at individual frames
from a film. Reverse or out-of-order recall also encourages an overly
zealous witness to stick to the facts. Witnesses find it more difficult
to embellish the event when they separate themselves from the natural
flow of events and independently deal with each activity.
Returning to the eye-witness interview in the opening scenario, the
detective might continue using the cognitive interview technique.
Accordingly, he would discuss the conversation the victim had with the
responding officer and ask where she was when the officer arrived. He
wants to know exactly what she was doing at that time. What did she do
immediately before that? Through this line of questioning, he gradually
arrives back at the time of the robbery and before hand. Thus, he leads
her through a second recounting of the crime, only in reverse sequence.
This time, her information is a collection of pieces, each viewed
independently. Just as looking at a portion of the landscape may reveal
details missed while taking in the panoramic view, looking at stages of
an event may enable witnesses to "see" previously unnoticed items.
To further stimulate witness memory recovery, Fisher and Geiselman also
suggest changing the perspective. (4) Witnesses experience an event one
time; however, they may perceive it from various views. During initial
recollection, witnesses articulate from their personal perspectives and
rarely vary from their point of view. By prompting witnesses to
physically change the positioning in their memories, interviewers give
them the opportunity to recall more of their experiences. (5)
Interviewers can change perspective by asking witnesses to consider the
view of another witness, victim, or an invisible eye on the wall.
Using the technique of changing the perspective of witnesses, the
detective in the opening scenario might say: "You know those
surveillance cameras they have in banks and some stores? Too bad there
wasn't one on the wall over there. I wonder just what it would have
recorded; it certainly would have had a different vantage point than you
did." Through this opening statement, he can draw the victim into a
discussion of what might have been recorded on the nonexistent camera.
This technique not only provides her with an opportunity to "replay" the
event from a different perspective but it also serves to further
detraumatize the situation. Reviewing a film is much less traumatic than
reliving an armed robbery.
Interviewers can use additional techniques to promote memory retrieval,
depending on the facts of the crime and witness information. After
witnesses have recounted an event in its natural sequence, reverse
sequence, and from different perspectives, the interviewer can induce
specific retrieval by asking direct questions. One technique of specific
retrieval includes associating witness recollection of physical
appearance, clothing, and sound with something or someone familiar to
them. Other areas of recall, such as remembering names and numbers, may
be enhanced by dealing with individual components of the item, such as
the first letter or number. Once established, interviewers direct
concentration to the next letter or number and build the response.
Using this technique, the detective in the robbery scenario might have
first reviewed the details obtained thus far. At certain points, he
might have stopped to ask questions such as: "You say he had a scary
voice. How so? Does it remind you of anybody you know, or perhaps
somebody you've seen in a movie?" "The coveralls he was wearing--ever
seen that type before? Where? Were they like a pilot's suit, or more
like a carpenter's?"
This context-enhancing technique stems from realizing that the victim
did not experience this event as a clean slate. She had a lifetime of
experiences that preceded this activity. Therefore, when getting a
description of the subject, a detective's questions, "Does this person
remind you of anyone you know? In what way?" likewise provide a context
from which the victim can make comparisons. This removes her need to
create, thus enabling her to draw on information with which she is
The cognitive interview encourages a witness' in-depth retrieval of
memory. Success with this technique, although a time-consuming process,
forces interviewers to avoid some traps normally associated with police
interviews, specifically, rushing the recall of witnesses and
interrupting their narratives.
Witnesses must feel confident that they have time to think, speak,
reflect, and speak again as often as they need. Interviewers can instill
this confidence by allowing sufficient time for the interview and by
refraining from interrupting witnesses. (6) All too often, interviewers
say, "Tell me what happened," but before witnesses speak for 30 seconds,
interviewers begin interrupting with specific questions. Those specific
questions should be asked after witnesses have had the opportunity to
recount the event fully. Allowing time to respond also applies when
witnesses answer specific retrieval questions. Rushing witnesses sends a
message to them that their information is trivial. This results in
witness retrieval shutdown. If interviewers don't give them the time,
witnesses cannot concentrate or remember.
The cognitive interview technique not only enhances witness recall but
also addresses another common problem among interviewers--their
inability to sustain the interview. Interviewers, particularly
inexperienced ones, are often reduced to saying, "I can't think of
anything else to ask. Is there anything you're leaving out?" If a
witness responds in the negative, the interview is over. Using the
cognitive technique can help interviewers avoid prematurely reaching
this point. Experience demonstrates that the cognitive interview
technique allows interviewers to continue discussing events without
sounding redundant. Indeed, continued conversation in a constructive,
helpful direction often prompts additional information.
Despite significant advances in various forensic fields, most crimes are
solved by information furnished by people. The interview remains the
foremost investigative tool for gaining information.
Although most victims and witnesses try to cooperate, their inability to
recall vital details can be discouraging, and they need help in
remembering. This help must come from investigators. Merely asking the
right questions does not suffice; enhancing someone's memory requires
active involvement. The cognitive approach to interviewing has proven
more effective than the traditional one by increasing the quality and
quantity of information obtained from witnesses and victims.
(1) John C. Yuille and N. Hope McEwan, "Use of Hypnosis as an Aid to
Eyewitness Memory," Journal of Applied Psychology, 1985, vol. 70, No. 2,
(2) Martin T. Orne, David F. Dinges, and Emily C. Orne, "The Forensic
Use of Hypnosis," National Institute of Justice, December 1984, p. 1.
(3) R. Edward Geiselman, Ronald P. Fisher, David P. MacKinnon, and Heidi
L. Holland, "Eyewitness Memory Enhancement in the Police Interview:
Cognitive Retrieval Mnemonics Versus Hypnosis," Journal of Applied
Psychology, 1985, vol. 70, No. 2, p. 403.
(4) R. Edward Geiselman, and Michael Nielsen, "Cognitive Memory
Retrieval Techniques," The Police Chief, March 1986, p. 70.
(6) R. Edward Geiselman, Ronald P. Fisher, David S. Raymond, Lynn M.
Jurkevich, and Monica L. Warhaftig, "Enhancing Eyewitness Memory:
Refining the Cognitive Interview," Journal of Police Science and
Administration, December 1987, vol. 15, No. 4, p. 292.