Agrowing body of research on the intersection of schizophrenia and memory
illuminates both the nature of the disease and practical issues for
"Subtle memory deficits in schizophrenia can lead to difficulty in
functioning independently in real life, even more than the severity of the
hallucinations and delusions that affect patients," said Deana Barch,
Ph.D., an associate professor of psychology, radiology, and psychiatry at
Washington University in St. Louis and coauthor of a new study on episodic
memory and cortical activity. Study results are published in the July 1
The researchers recruited 17 patients with schizophrenia and 26 healthy
controls. They were shown anonymous faces and a series of printed words in the
first phase of the study and then quizzed on their ability to recall or answer
questions about what they had seen.
Their brains were scanned with a functional magnetic resonance imaging
(fMRI) system while they completed these tasks. All but one of the patients in
the schizophrenia group were taking medications.
"Past research has shown that [people with schizophrenia] have
deficits in memory, but the question is whether these are permanent, or if
some strategies might improve their memory," said lead author Aaron
Bonner-Jackson, a graduate student in psychology at Washington University, in
an interview with Psychiatric News.
The researchers did not describe the study to the subjects as a memory
test, but the questions asked and the way they were presented allowed them to
test how well the subjects learned and recognized information.
"They were not explicitly memorizing each word, but the deep semantic
processing they were doing helped them to recognize a good percentage of those
words later on," said Bonner-Jackson. "I believe that knowing they
were going to be tested later would probably affect their strategy in some
way, which in turn probably would have affected their brain activity and
subsequent recognition of the words. However, it was not necessary for them to
know about the later memory test in order to obtain consent, since the brief
deception was minor and necessary for our experiment."
The researchers asked two kinds of questions. "Deep" questions
required the subjects to think about the semantic meaning of the words, that
is, if they were abstract or concrete. "House" is concrete, while"
love" is abstract, for instance.
"The point is, you've asked them to think about the word," said
Barch. "If you think about what words mean, you remember them better
"Shallow" processing meant simply asking about the arrangement
of letters in the word: Did the first or last letter of the word come earlier
in the alphabet?
Study subjects weren't specifically trained in techniques for remembering
words or images or for answering test questions, but the structure of the
study demanded certain strategies to process the information, said
"The advantage of using deep versus shallow processing is that
patients don't have to generate a strategy for memory," commented J.
Daniel Ragland, Ph.D., an associate professor of neuropsychology in psychiatry
at the University of Pennsylvania School of Medicine, in an interview."
And it's gratifying to see research taking strategy into account during
Subjects with schizophrenia in the Washington University study performed
worse than controls on both the shallow (alphabetical judgments) and deep
(abstract/concrete) encoding tasks. Members of both groups did better on
shallow than deep tasks.
In the recognition phase of the trial, subjects with schizophrenia were
less accurate than were control subjects, but the difference was not
Compared with shallow encoding, deep encoding of words resulted in
significant memory performance benefits among the schizophrenia subjects.
Those subjects "benefit from deep encoding, just like controls
do," said Bonner-Jackson. "They don't have a generalized memory
deficit; they can do what controls do."
The fMRI imaging delineated areas of the brain in both subject groups that
showed more activity during deep than shallow encoding. Both groups showed
effects in regions typically active during deep semantic encoding. However,
schizophrenia subjects showed greater activity for deep versus shallow
encoding in three other regions: the left inferior frontal, right inferior
frontal, and left middle frontal. These regions were not usually engaged by
Two hypotheses might explain the involvement of these added brain areas
seen in the fMRI.
"If greater activity appeared in better-performing subjects, it may
mean that these regions are compensating for some deficit," said
Bonner-Jackson. "Or, if the increased activity appeared in
lower-performing subjects, it could simply represent wasted effort and
indicate greater pathology."
Seeking evidence for either of these two alternatives, the researchers
compared high- and low-performing subgroups among the schizophrenia patients.
The effect size for the deep versus shallow contrast was greater in the
low-performing group. Thus, people with schizophrenia who perform worse on the
encoding tasks use more (and bilateral) brain regions, while high performers
require a smaller set of regions to complete the same tasks, a set more like
those activated in control subjects.
The "pathological" explanation seems more likely, given the
results of this study, said Bonner-Jackson. Schizophrenia is associated with a
deficit in the use of effective strategies to influence memory performance,
and those abnormalities are not fully overcome by use of those strategies,
said the authors.
Nevertheless, the study outcome also suggests avenues for clinical
improvement of patients, Bonner-Jackson said: "People with schizophrenia
have more memory capability than we thought."
Such insights derived from the current state of basic research might
influence ways to improve their lives.
"Patients with schizophrenia in a new learning situation tend to take
a rather passive approach to processing stimuli, not actively organizing
information," said Ragland, whose related research is now in press with
the American Journal of Psychiatry. "With the shallow/deep
paradigm you can compensate by giving them a strategy, but the real challenge
is teaching patients to do it for themselves, without a person in a white coat
standing at their shoulder."
"It won't help to tell people with schizophrenia just to work harder
or emphasize rote rehearsal of information," said Barch. "Rather
than practice the same inefficient methods, it would be better to help them
understand how they're learning by stressing the meanings and relationships of
the things they encounter."
An abstract of "The Influence of Encoding Strategy on Episodic
Memory and Cortical Activity in Schizophrenia" can be accessed at<www.elsevier.com/locate/biopsychiat>
under the July 1 issue. ▪