Edward Fisher of the University of California, Los Angeles, and Ronald Geiselman, of Florida
Edward Fisher of the University of California, Los Angeles, and Ronald Geiselman, of Florida

International University, were investigating ways of turning findings from psychological
International University, were investigating ways of turning findings from psychological

research, such as the context reinstatement effect, into practical techniques that could
research, such as the context reinstatement effect, into practical techniques that could

be used in real police investigations. They developed a procedure that they called the
be used in real police investigations. They developed a procedure that they called the

cognitive interview.
cognitive interview.

The way that police traditionally interview eyewitnesses involves frequent interruptions
The way that police traditionally interview eyewitnesses involves frequent interruptions

by the interviewer, a focus on details relevant to the investigation, and the use of the question-answer
by the interviewer, a focus on details relevant to the investigation, and the use of the question-answer

format. As we’ve already seen, one problem with a question-focussed approach is that
format. As we’ve already seen, one problem with a question-focussed approach is that

there is the opportunity for false memories to be created by leading questions. Edward
there is the opportunity for false memories to be created by leading questions. Edward

Fisher and Ronald Geiselman raided the cognitive psychology literature, including the work
Fisher and Ronald Geiselman raided the cognitive psychology literature, including the work

on context reinstatement, to create the cognitive interview.
on context reinstatement, to create the cognitive interview.

Here are some of the key ideas behind it. First, memory research tells us that memory
Here are some of the key ideas behind it. First, memory research tells us that memory

traces are usually complex and contain different sorts of information. Second, research
traces are usually complex and contain different sorts of information. Second, research

indicates that the ease with which a memory is recalled depends on how much "informational
indicates that the ease with which a memory is recalled depends on how much "informational

overlap" it has with the cues being used to retrieve it. This refers to work on what’s
overlap" it has with the cues being used to retrieve it. This refers to work on what’s

known as the “encoding specificity principle” by Endel Tulving of the University of Toronto,
known as the “encoding specificity principle” by Endel Tulving of the University of Toronto,

amongst others.
amongst others.

When something is remembered, it is encoded with respect to the context in which it
When something is remembered, it is encoded with respect to the context in which it

is studied, producing a unique trace which incorporates information from both target
is studied, producing a unique trace which incorporates information from both target

and context. This means that the probability of successfully remembering something depends
and context. This means that the probability of successfully remembering something depends

on the degree of overlap between the information present at retrieval and the information stored
on the degree of overlap between the information present at retrieval and the information stored

in memory. That is, the more overlap you can get between a memory of something that
in memory. That is, the more overlap you can get between a memory of something that

you’ve already recalled and something you want to recall, then the better chance of
you’ve already recalled and something you want to recall, then the better chance of

retrieving the desired memory. Another finding is that any given memory can be accessed by
retrieving the desired memory. Another finding is that any given memory can be accessed by

a number of different retrieval cues. That is, if you can’t access the memory you want
a number of different retrieval cues. That is, if you can’t access the memory you want

via one particular cue, then try another.
via one particular cue, then try another.

The cognitive interview uses these principles of memory retrieval to help witnesses maximise
The cognitive interview uses these principles of memory retrieval to help witnesses maximise

their chances of recalling the events of a crime.
their chances of recalling the events of a crime.

The question is, what’s the evidence that it actually does result in better eyewitness
The question is, what’s the evidence that it actually does result in better eyewitness

memory recall than more traditional police interviews? Fisher and Geiselmen tested the
memory recall than more traditional police interviews? Fisher and Geiselmen tested the

original version of the cognitive interview by recruiting 17 highly experienced interviewers,
original version of the cognitive interview by recruiting 17 highly experienced interviewers,

all of whom had been trained in forensic hypnosis. This included police detectives, CIA investigators,
all of whom had been trained in forensic hypnosis. This included police detectives, CIA investigators,

and private detectives.
and private detectives.

They were randomly assigned to either use the cognitive interview or to use standard police interviews
They were randomly assigned to either use the cognitive interview or to use standard police interviews

– as well as other options. Then the researchers got 89 undergraduates
– as well as other options. Then the researchers got 89 undergraduates

to watch one of four films depicting a violent murder – borrowed from the LAPD training
to watch one of four films depicting a violent murder – borrowed from the LAPD training

academy. Forty-eight hours after watching the crime, the students were interviewed by
academy. Forty-eight hours after watching the crime, the students were interviewed by

the professional investigators. These interviews were transcribed and scored.
the professional investigators. These interviews were transcribed and scored.

An exhaustive list of all the details of the crimes that were mentioned was drawn up and
An exhaustive list of all the details of the crimes that were mentioned was drawn up and

then researchers calculated the proportion of those items mentioned by each of the students.
then researchers calculated the proportion of those items mentioned by each of the students.

Doing this, they found that the cognitive interview elicited between 25% and 35% more
Doing this, they found that the cognitive interview elicited between 25% and 35% more

information than the standard police interview. So – that’s very promising. As a result
information than the standard police interview. So – that’s very promising. As a result

of this early work, Fisher and Geiselmen went on to refine their methods and also to test
of this early work, Fisher and Geiselmen went on to refine their methods and also to test

them in more realistic situations.
them in more realistic situations.

Here’s one example, where they tested the cognitive interview for real police detectives
Here’s one example, where they tested the cognitive interview for real police detectives

solving genuine crimes. This time they recruited 16 police detectives based in Florida. In
solving genuine crimes. This time they recruited 16 police detectives based in Florida. In

the initial stage of the study, they recorded and analysed 5-7 interviews by each detective
the initial stage of the study, they recorded and analysed 5-7 interviews by each detective

over 4 months. Then they randomly chose half the detectives and trained them in how to
over 4 months. Then they randomly chose half the detectives and trained them in how to

give the cognitive interview. Then they monitored and analysed interviews conducted by both
give the cognitive interview. Then they monitored and analysed interviews conducted by both

the trained and untrained detectives for 7 months afterwards. They eventually amassed
the trained and untrained detectives for 7 months afterwards. They eventually amassed

between 2 and 7 appropriate interviews per detective.
between 2 and 7 appropriate interviews per detective.

All the interviews were transcribed and analysed as before. Of course, one big disadvantage
All the interviews were transcribed and analysed as before. Of course, one big disadvantage

of studying real crime investigations is that it is much harder to determine what is actually
of studying real crime investigations is that it is much harder to determine what is actually

true or not for the purposes of determining whether the interviews were yielding real
true or not for the purposes of determining whether the interviews were yielding real

or false memories. The way they did this in this particular study was by comparing what
or false memories. The way they did this in this particular study was by comparing what

the witnesses said with another reliable source of information, such as another witness, a
the witnesses said with another reliable source of information, such as another witness, a

confession by criminal, or from hidden camera footage.
confession by criminal, or from hidden camera footage.

So they only looked at corroborated facts, and this is what they found. So the cognitive
So they only looked at corroborated facts, and this is what they found. So the cognitive

interview, in this real world study looking at real crimes, led to a 47 percent increase
interview, in this real world study looking at real crimes, led to a 47 percent increase

in the number of facts recalled compared with a standard police interview.
in the number of facts recalled compared with a standard police interview.

There’s since been a large body of research looking at these cognitive interview techniques.
There’s since been a large body of research looking at these cognitive interview techniques.

A meta-analysis of some of this work, by Gunter Kohnken of the University of Kiel and his
A meta-analysis of some of this work, by Gunter Kohnken of the University of Kiel and his

colleagues confirmed that, over 42 studies, the cognitive interview reliably generates
colleagues confirmed that, over 42 studies, the cognitive interview reliably generates

more correct details than the standard interview. And this effect was greater the more realistic
more correct details than the standard interview. And this effect was greater the more realistic

the context of the study.
the context of the study.

The cognitive interview is now routinely taught to police officers in Britain and also widely
The cognitive interview is now routinely taught to police officers in Britain and also widely

used in the US, Australia, New Zealand, the UK, Canada, Singapore, and Hong Kong.
used in the US, Australia, New Zealand, the UK, Canada, Singapore, and Hong Kong.

Of course, as we’re taking a scientific approach in this course, we also need to talk about the
Of course, as we’re taking a scientific approach in this course, we also need to talk about the

limitations of the cognitive interview. Like everything else in life, it’s not perfect.
limitations of the cognitive interview. Like everything else in life, it’s not perfect.

The cognitive interview is more difficult to conduct than a standard interview, it takes
The cognitive interview is more difficult to conduct than a standard interview, it takes

longer and it requires more effort on the part of the interviewer.
longer and it requires more effort on the part of the interviewer.

Also, it’s only useful for co-operative eyewitnesses.
Also, it’s only useful for co-operative eyewitnesses.

It’s very much about improving recall memory, it doesn’t help with recognition memory
It’s very much about improving recall memory, it doesn’t help with recognition memory

- for example recognising a suspect in a line-up. However, taking all that on board, overall
- for example recognising a suspect in a line-up. However, taking all that on board, overall

it’s still considered a valuable tool for the investigator – and it’s a great example
it’s still considered a valuable tool for the investigator – and it’s a great example

of how cognitive psychology research can be successfully applied to the real world.
of how cognitive psychology research can be successfully applied to the real world.