ArticlePDF AvailableLiterature Review

Abstract

This paper reviews the effects of benzodiazepines (BZs) on the performance of tasks measuring human cognitive abilities. The paper reviews the most common cognitive side effects of BZs: increased sedation, decreased attention, and anterograde amnesia. In particular, this paper focuses on recent findings regarding time course-related effects on BZ-induced deficits in explicit and implicit human memory performance. Specifically, we reviewed recent research indicating that both explicit memory and priming are impaired by BZs if the encoding task takes place near the time of the theoretical peak plasma concentrations of the drug. Although BZs also appear to increase objective and subjective sedation, as well as to impair attentional processing, these other cognitive impairments do not appear to fully account for the widespread memory deficits caused by BZ administration. The theoretical and clinical implications of benzodiazepine-induced memory impairments are discussed.
Current Pharmaceutical Design, 2002, 8, 45-58 45
Cognitive and Sedative Effects of Benzodiazepine Use
S. E. Buffett-Jerrott*
1
and S.H. Stewart
2
1
Department of Psychology, Dalhousie University,
2
Departments of Psychology & Psychiatry,
Dalhousie University, Life Sciences Centre, 1355 Oxford Street, Halifax, Nova Scotia,
Canada B3H 4J1
Abstract: This paper reviews the effects of benzodiazepines (BZs) on the performance of
tasks measuring human cognitive abilities. The paper reviews the most common cognitive
side effects of BZs: increased sedation, decreased attention, and anterograde amnesia. In
particular, this paper focuses on recent findings regarding time course-related effects on
BZ-induced deficits in explicit and implicit human memory performance. Specifically, we
reviewed recent research indicating that both explicit memory and “priming” are impaired by BZs if the
encoding task takes place near the time of the theoretical peak plasma concentrations of the drug. Although BZs
also appear to increase objective and subjective sedation, as well as to impair attentional processing, these
other cognitive impairments do not appear to fully account for the widespread memory deficits caused by BZ
administration. The theoretical and clinical implications of benzodiazepine-induced memory impairments are
discussed.
COGNITIVE AND SEDATIVE EFFECTS OF
BENZODIAZEPINE USE
impairments, and memory problems. It should be noted that
which drug effects are considered “side-effects” depends upon
the reason that the BZ has been prescribed. For example, if a
BZ is prescribed as a sleeping pill, feelings of sedation are
not considered a side-effect, but rather an intended,
therapeutic effect. However, sedation is an unwanted side-
effect for individuals taking BZs as a day-time treatment for
clinical or chronic anxiety. This paper reviews the research
findings and controversies regarding the most common
cognitive changes associated with BZ use in humans:
sedation, inattention, and amnesia. It should be noted that
most of the studies regarding the cognitive effects of BZs
investigate the effects of acute doses of BZs in normal
participants without a history of BZ or other psychotropic
medication use.
The first benzodiazepine (BZ), chloriazepoxide
(Librium), was introduced into clinical practice in 1960,
followed by the introduction of diazepam (Valium) which
quickly became the most commonly-prescribed drug in
North America [1]. Since 1960, over 50 BZs have been
introduced into clinical practice, with alprazolam (Xanax)
presently being the most commonly prescribed [2]. At the
period of their greatest popularity, in 1973, over 87 million
BZ prescriptions were filled world-wide [3]. This figure
represents more than half of the psychotropic prescriptions
filled that year. Although prescription rates are declining,
BZs still remain a very commonly prescribed psychotropic
medication [4] with 10% of the adult population ingesting at
least one BZ in a one-year period [1]. BZs are prescribed for
their sedative, anxiolytic, hypnotic, and muscle-relaxant
effects [5]. Although the number of BZ prescriptions has
declined worldwide since the 1970’s, it is estimated that one
in four current BZ users has been taking these drugs
continuously for one year or longer [6]. The Food and Drug
Administration has cautioned against this long-term use
because there are many clinical problems associated with
chronic administration of BZs, such as tolerance,
withdrawal, and side effects [7].
One of the difficulties in investigating cognitive
impairments associated with BZ administration is that tasks
that are “pure” measures of sedation, attention, or memory
are rare or even nonexistent [9]. This review divides such
cognitive tasks according to the most salient aspect of
cognition thought to be assessed by each. In addition,
whenever possible, this paper describes the procedures that
have been used to determine what cognitive processes are
actually being measured by each task.
SEDATION
Until the 1970s, it was assumed that oral BZs did not
have any significant cognitive side effects [8]. However, it is
now known that there are several consequences of BZ use in
terms of impact on aspects of human cognition. Common
cognitive side effects include sedation, attentional
In the 1960s, anesthesiologists began using BZs as a pre-
operative medicant to calm patients before surgical
procedures. These individuals noted that, in addition to
decreased anxiety, patients were reporting increased sedation
[10]. This sedation was not considered an unwanted side-
effect by anaesthesiologists, as it was believed to combine
with the drug’s anxiolytic effect to keep the patient calm and
compliant before the surgical procedure. However, sedation
*Address for Correspondance to that Author at Susan Buffett-Jerrott,
Ph.D., Department of Psychology, Dalhousie University, Life Sciences
Centre, 1355 Oxford Street, Halifax, Nova Scotia, Canada B3H 4J1. E-
mail: sjerrott@is2.dal.ca; Tel (902) 494-3793; Fax: (902) 494-6585.
1381-6128/02 $28.00+.00 © 2002 Bentham Science Publishers Ltd.
46 Current Pharmaceutical Design, 2002, Vol. 8, No. 1 Buffett-Jerrott
and
Stewart
is an unwelcome side effect for individuals taking BZs as a
day-time treatment for clinical anxiety.
Observer-Rated Measures of Sedation
Most studies of BZs do not include observer-rated
measures of sedation. However, a few studies in the pediatric
anaesthesia literature have used observer-rated sedation
measures, most likely because very young children may not
be able to accurately report on their own sedation levels.
Typically, 4- or 5-point Likert scales are used by examiners
to rate sedation. Studies indicate that children who have
been administered a BZ (midazolam) before a medical
procedure appear significantly more sedated than placebo-
treated children [21,22]. However, these studies are
confounded by the possibility of increased anxiety in
children who had not received the anxiolytic. Obviously, an
unmedicated child who is crying due to pre-operative anxiety
is not going to be observer-rated as calm or asleep, therefore
merely appearing less “sedated” than the BZ-treated child.
When assessing the sedative effects of a drug it is
important to obtain subjective, observer-rated, and objective
measures of sedation. Subjective sedation measures require
the participant to use internal or bodily cues (i.e., feeling
sleepy) to measure sedation. Subjective measures of sedation
require that the participant be aware of changes in their
behavior and feelings, and that they correctly interpret these
changes [11]. It is possible that a participant may not report
being sedated but may show psychomotor slowing on an
observer-rated or objective sedation measure. Alternatively, a
participant may report feeling sleepy, but this subjective
state might not affect their speed at performing cognitive
tasks, or be observable to others. Observer-rated sedation
measures require the experimenter/observer to use
information from a participant’s outward appearance (e.g.,
participant appears lethargic) to measure sedation. Observer-
rated sedation measures can be prejudiced by factors such as
the observers’ training or beliefs about the effects of the drug
[12]. However, if participants do not rate themselves as
significantly sedated, but observer ratings and/or objective
sedation measures suggest sedation, this would indicate that
BZs may be affecting “meta-cognitive processes” or
“reflective” functions [11,13]. In other words, such a
discrepancy would indicate that BZs may impair users’
abilities to evaluate their own competence.
Objective Measures of Sedation (i.e., Psychomotor Speed
Measures)
To objectively measure sedation, researchers usually
present participants with one of many behavioral tasks that
measure their response speed. Researchers investigating the
effects of BZs on objective sedation have used many different
tasks. The most commonly used task is the Digit Symbol
Substitution Task (DSST) from the Wechsler Adult
Intelligence Scale [23]. This has been found to be a reliable
measure of sedation in both drugged and non-drugged
participants [11]. In this task, participants are required to
learn a code in which symbols are paired with the digits 1
through 9. Participants are asked to copy the symbols
associated with each digit, in the order presented, as quickly
and accurately as possible. The DSST is scored as the total
number of symbols correctly substituted for digits in a 90
second period. Although usually given as a paper and pencil
task, the test has also been administered in computerized
form [24].
Subjective Measures of Sedation
To assess subjective sedation in most studies of BZs,
participants are usually presented with a series of visual
analogue scales (VASs; e.g. alert-drowsy; energetic-lethargic)
on which they are asked to rate their current state [14]. Most
commonly, the scales used are derived from a series of VASs
created by Bond and Lader (1974) [15] that have been found
to reliably measure sedation in drugged participants [11].
Studies using diazepam [16], lorazepam [17] and oxazepam
[18] have almost consistently indicated that participants rate
themselves as being more drowsy, more relaxed, and
mentally slower after a single clinical dose of a BZ relative
to both their own pre-drug sedation levels and to placebo-
treated participants. However, one study, using lorazepam,
did not find significant effects, although the means were in
the expected direction [94]. However, in this study, the
sedation ratings were obtained at a time when participants
were aware that they were about to receive a needle. It is
probable, as suggested by the authors, that anticipatory
anxiety regarding the injection may have aroused
participants, causing them to feel less drowsy.
Consistently, results of studies using the DSST indicate
that clinical doses of BZs such as oxazepam [25], lorazepam
[25], and diazepam [26,27] slow performance compared to a
placebo-treated group. When attentional performance (see
next section) is covaried out, BZ-induced impairments still
remain on the DSST [25]. The relationship between
performance on the DSST and subjective measures of
sedation is significant for high doses of triazolam (r = -.73 to
-.54, depending on time (post-drug administration) sedation
is tested), but is non-significant for moderate or low doses
[11]. This finding suggests greater correspondence between
objective and subjective sedation measures at higher BZ
doses.
In another study, healthy participants were asked to
provide subjective ratings of the quality of their sleep, speed
of awakening, and alertness upon waking, after taking a
clinical dose of alprazolam [20]. Results indicated that
participants believed that the quality of their sleep was
improved by BZs compared to placebo-treated participants,
but that there was no difference in speed of awakening or
feelings of alertness once awake [20]. Taken together, these
studies indicate that BZs reliably induce subjective feelings
of sedation.
One difficulty with the DSST is that the task does not
strictly measure sedation. The task also has a large memory
component. In addition to assessing psychomotor speed in
replacing the symbols with numerals, performance on the
task can typically be facilitated when the participant
memorizes the 9 symbol-digit pairs. However, memory is
another cognitive function that is impaired by BZs (see later
section). In contrast, the Symbol Copying Test (SCT)
requires participants to copy a series of symbols, but they do
Effects of Benzodiazepine Use Current Pharmaceutical Design, 2002, Vol. 8, No. 1 47
not have to remember a code. Studies have found that
participants who have taken a clinical dose of lorazepam
[28,29] or oxazepam [25] are significantly slowed compared
to placebo participants on the SCT. However, in another
study, the pattern of results was similar, but the effect was
not statistically significant after a clinical dose of diazepam
[30]. Although the SCT does not have a memory
component, it may still be confounded by BZ-induced
attentional impairments. However, when attentional
performance is covaried out, the BZ-induced differences on
the SCT remain significant [25].
[29,33]. In this task participants use one eye to view a
flickering light. The frequency of the flicker is first increased
until the participant indicates that the flicker has disappeared.
Next, the frequency of the flicker is decreased until the
participant is again able to detect a flicker. Typically,
participants are administered a series of ascending and
descending trials. Next, detection threshold on both the
ascending and descending CFFT trials are combined to
create a total CFFT performance score. Researchers have
found that ability to detect flicker changes on the CFFT
(i.e., total performance score) is impaired by a clinical dose
of lorazepam [29,33]. In addition, long-term users of BZs
also showed impaired performance on the CFFT [34].
Another task that measures psychomotor speed is the
Discriminant Reaction Time Task (DRRT). In this test,
participants look at a computer screen on which a series of
digits are flashed [14]. The participants are instructed to
press a button whenever a target digit (e.g., 3) is flashed.
The flashing of the digits slows or accelerates depending on
the correctness and speed of the participant’s responses. The
mean time between digits being flashed is used as a measure
of the participant’s response speed. Results indicate that after
a clinical dose of alprazolam or lorazepam, BZs slow
response speed, compared to placebo-treated participants on
the DRRT [14]. However, this task has an attentional
component and the results of the [14] study might be due to
attentional impairments, as opposed to increased sedation.
Summary of Effects of BZs on Sedation Measures
It appears that BZs induce high levels of subjective
sedation, as rated by participants and observers, after a
single, acute dose. After administration of a BZ, participants
are slowed on objective sedation measures (cognitive
processing speed and psychomotor speed tasks). These
effects are found across various BZs and when confounding
variables of memory and attentional requirements of the
paradigms are controlled. Although objective and subjective
measures of sedation are significantly correlated, the
correlations are not strong enough to recommend using only
one measure exclusively [12]. In fact, one study indicated
that after administration of different doses of triazolam,
participants showed dose-dependent impairment on objective
measures of sedation, but the same participants were not able
to differentiate between the doses on subjective measures of
sedation [11].
Other objective measures of sedation have assessed
psychomotor speed, as opposed to cognitive processing
speed. The Finger Tapping Test uses a finger tapping
mechanism to measure how many times a participant can tap
in a specific amount of time (usually 60 seconds). Although
most studies indicate that finger tapping speed is slowed
compared to placebo-treated participants after administration
of a clinical dose of diazepam [27], oxazepam [31], or
lorazepam [28,29,31,32], another study using lorazepam [33]
failed to find a significant difference. It should be cautioned,
however, that this objective measure of sedation is usually
considered to be influenced by motivational factors as well as
arousal, and thus the Finger Tapping Test cannot be
considered a “pure” measure of psychomotor speed.
In a recent study, participants who had taken a sedative
dose (2.5 mg) of lorazepam were compared to sleep deprived
individuals (problem snorers and on-call doctors). It was
found that the BZ group became sleepier as the cognitive
testing progressed, while the sleep-deprived participants
became more alert [9]. Therefore, it appears that it is
difficult, if not impossible, to overcome the sedation
associated with an acute dose of a BZ [9]. The sedative
effects of BZs have been found to contribute to traffic
accidents and workplace injuries. In addition, BZs have been
found to contribute to increased risk of falls and hip fractures
in the elderly [35]. Therefore, it is essential that patients be
advised of this impairment before they begin taking BZs [5].
This is especially true for elderly patients, for whom BZ-
induced sedative effects last longer and are more severe [5]. It
should be noted that the sedative effects of BZs are subject to
tolerance with repeated dose administration. Tolerance to the
various cognitive effects of BZs is discussed at the end of
this paper.
As noted above, most measures of objective sedation are
confounded by other factors such as attention, subjective
sedation, motivational and memory requirements [12]. To
deal with these problems, measures of saccadic eye
movements have been used as an alternative objective
measure of sedation. Saccadic eye movements are rapid eye
movements that occur when a participant quickly shifts his
or her gaze from one target to a second target. Research has
shown that the velocity of these movements cannot be
consciously changed [12]. Therefore, this may be a more
“pure” measure of objective sedation [12]. Although rarely
used, research indicates that peak saccadic velocity is
significantly slowed after administration of clinical doses (1
or 2 mg) of lorazepam [12]. This objective measure has been
shown to be significantly correlated with a subjective
measure of sedation (VAS: r = -.36) [12].
ATTENTION
The effects of BZs on attentional processes are not as
widely-studied as the effects of BZs in inducing sedation.
Obviously, it is important to study the effects of BZs on
attention, as attentional abilities are important for many
aspects of day-to-day functioning [18]. A difficulty with the
assessment of attention is that the concept of attention is
An additional physiological measure of objective
sedation that has been used in studies of benzodiazepines is
the Critical Flicker Fusion Threshold (CFFT) which is
believed to be a measure of central nervous system arousal
48 Current Pharmaceutical Design, 2002, Vol. 8, No. 1 Buffett-Jerrott
and
Stewart
multidimensional [36]. Attention cannot be reduced to a
unitary description, and cannot be assessed using a single
test [37]. In addition, there are differences in the
nomenclature used by theorists when describing the
components of attention that are being measured by a certain
task [36,38] conducted a principal components analysis on
many commonly used attention tasks in an attempt to create
an empirically-derived model for conceptualizing the
components of attention. Their research indicates a four-factor
model of attention composed of focus/execute, sustain,
encode, and shift elements. The empirical distinction
between these four elements of attention appears to be
supported by parallel activation of different regions of the
brain [38]. The focus/execute element measures a
participant’s ability to direct attention to stimuli, or to
select target stimuli from an array. The sustain element
measures a participant’s ability to attend to stimuli, or to
maintain focus over time. The encode element measures a
participant’s ability to perform mental manipulation tasks.
The element of shifting attention is the ability to attend to
one task, switch to another, and then back to the first task.
In addition to Mirsky et al.’s four elements, researchers also
measure divided attention, or the ability to perform two or
more tasks at the same time [39]. Although it is helpful to
investigate these various elements of attention separately,
attentional tasks usually measure more than one attentional
process. For review purposes, we have attempted to divide
various attention tasks based on their most salient
attentional demands.
paradigms, these studies have also found performance to be
impaired following BZ administration. The List Repetition
Task requires participants to listen to a list of categorically
related words, half of which are presented twice. Participants
are instructed to indicate whenever they hear a repeated word.
Performance on this task is impaired by administration of a
clinical dose of triazolam [43], diazepam [44] and alprazolam
[45], relative to placebo treated participants.
3) Encode Attention Tasks
Again, using a variety of different BZs and paradigms,
studies assessing the encoding function of attention further
support attentional impairments with BZs. In the Paced
Auditory Serial Addition Task, participants are presented
with pairs of single digit numbers read at increasing speed.
Participants are required to add the digits and respond before
the next numbers are read. Results indicate that participants
who have taken a sedative dose (2.5 mg) of lorazepam are
impaired, relative to placebo-treated participants, when
responding on this task [9].
In another encode task, the participant is asked to listen
to groups of “bleeps” from a computer. At the end of each
set, the participant is asked how many bleeps they counted.
Results indicate that participants who have taken a clinical
dose of lorazepam display more errors on this task relative to
placebo-treated participants [33].
1) Focus/Execute Attention Tasks
In our own research [46], participants were administered
the Digit Span Test from the Wechsler Adult Intelligence
Scale – Revised [23]. First, they were asked to repeat
number lists of increasing size. Next, they were asked to
repeat the lists backwards (e.g., if the experimenter said “8-
1-2” the participant is required to say “2-1-8”). Lezak (1995)
has suggested that if a participant performs well on the
forward test [47], but not the backward portion, it is
attention, rather than short-term memory that is impaired.
Our results indicated that participants who had received a
clinical dose of oxazepam did not show impairment, relative
to placebo-treated controls, on the Digit Forward task, but
were significantly impaired, relative to placebo-treated
controls, on the Digit Backward task [46]. A study using
lorazepam found similar results [9]. In addition, when
performance on other attentional measures was used as a
covariate, no impairment in digit span performance
remained, providing further evidence that it is attention, not
short-term span memory, that is impaired [9].
Results of investigations using several different types of
paradigms, and a variety of BZs, converge in finding that
BZs interfere with the focus/execute function of attention.
The most commonly used paradigms are symbol, digit, or
letter cancellation tasks. In these tasks, participants are
required to search for and cross out targets (e.g. the number
4) from a sheet of random symbols, digits or letters. Results
indicate that ability to find the target items is slowed,
compared to placebo-treated participants, by administration
of a clinical dose of diazepam [27,40,41] or lorazepam
[40,41]. In our own research, we found only a marginally-
significant effect on a symbol cancellation task, after a
clinical dose of lorazepam or oxazepam [18]. However our
pattern of results was in the same direction as that noted in
previous research [27,40,41].
In a different type of focus/execute task [42], participants
were presented with target words on a computer screen. The
targets were the words “left” and “right” and the participant
had to respond by pressing a specific key with the left hand
for “left” and another key with the right hand for “right”.
Distractor words also appeared on the screen with the target
words. Results indicated that a clinical dose of diazepam
slowed responding on this task compared to placebo-treated
participants [42].
4) Shift Attention Tasks
Although few studies investigating attentional shifting
have been conducted, two studies, using triazolam, indicate
that BZs do have an effect on an individual’s ability to shift
attention. Attention researchers have noted that shifts of
attention can be controlled either voluntarily (i.e.,
endogenously) or involuntarily (i.e., exogenously) [48].
These two processes have also been referred to as
“controlled” and “automatic” attention allocation processes,
respectively. With respect to involuntary or automatic
attentional orienting, visual attention can be “captured” by
2) Sustain Attention Tasks
This attentional element has typically been assessed
using the List Repetition Task. Like focused attention
Effects of Benzodiazepine Use Current Pharmaceutical Design, 2002, Vol. 8, No. 1 49
the sudden onset of a stimulus (i.e., an “exogenous”
stimulus) in the periphery. In contrast, voluntary or
controlled attentional orienting involves the individual
making a voluntary, conscious decision to shift attention to
a particular position in space using internal (“endogenous”)
cognitive processes (see review [49]). Both types of
attentional shifts function to enhance processing of visual
information at that specific position in space [50-52]. These
two types of attentional orienting are typically studied using
Posner’s cued visual search paradigm [51].
triazolam) vs. placebo on visual attentional orienting [59].
Two types of cues were utilized (central and peripheral) to
assess the effects of triazolam on controlled and automatic
attention allocation mechanisms, respectively. In addition,
cues were either valid or invalid with respect to the future
location of the target. A main effect of drug condition was
observed such that RT was slowed overall for triazolam
compared to placebo. As described previously, this drug
group main effect might reflect sedation rather than
attentional impairment, per se. Drug condition did not
interact with cue type (central vs. peripheral). Rather, RTs
following both exogenous and endogenous cues were slowed
by triazolam compared to placebo. This finding suggests
that triazolam does not selectively impair only controlled or
automatic attention allocation mechanisms. However, drug
condition did interact with cue validity. Specifically, the
costs plus benefits were significantly greater for triazolam
than for placebo: the RT slope for valid vs. invalid cues was
steeper for triazolam than for placebo. The authors interpreted
this finding to suggest that triazolam selectively impairs
attentional disengagement and/or attention shifting
mechanisms. However, their failure to include a neutral cue
condition complicates the interpretation of these findings
with respect to which particular aspects of visual attentional
orienting are impaired by triazolam, as the authors were
unable to calculate separate “costs” and “benefits”.
In this paradigm, participants are often asked to search for
and respond to a visual “target” (e.g., a letter). Their task
may involve simple target detection or a choice speeded
response. On each trial, a visual “cue” is presented prior to
the target. This cue can vary in terms of its “validity” and
“type”. In terms of validity, the cue provides one of three
kinds of information about the future location of the visual
target. Valid cues provide correct information, invalid cues
provide incorrect information, and neutral cues provide no
information, as to the future location of the upcoming target.
Valid and invalid cues typically consist of arrows pointing
to a particular spatial location, whereas neutral cues typically
consist of a plus sign giving no location information. In
terms of cue type, the cue may appear at a central location
(i.e., endogenous cue) requiring the participant to make a
voluntary decision to shift attention to the cued location.
Alternatively, the cue may appear at a peripheral location
(i.e., exogenous cue), automatically capturing the
participant’s visual attention and drawing it toward the cued
location. These two cue types are designed to assess
endogenous and exogenous shifts of attention, respectively.
A later study by Carter et al. (1998) [60] was designed to
replicate and extend the findings of [59]. Like Johnson et al.
(1995), Carter et al. (1998) also used versions of Posner’s
visual search paradigm to examine the effects of triazolam on
controlled and automatic attentional allocation mechanisms
[59,60]. However, Carter et al. (1998) [60] felt that there
were a number of methodological limitations to the [59]
study and they made alterations to the paradigm accordingly.
For example, by including a neutral cue condition in
addition to the valid and invalid cue trials, Carter et al.
(1998) were able to separately examine drug effects on each
aspect of visual attentional orienting[60]. In addition, Carter
et al. (1998) examined participants’ performance at two
separate cue-to-target intervals (i.e., stimulus onset
asynchronies; SOAs) [60], in contrast to [59] who examined
only the shorter of these SOAs. Attention researchers
typically employ more than one SOA within the Posner
paradigm [60]. In this study, triazolam selectively modified
performance on automatic orienting to exogenous cues at the
shorter SOA. Specifically, the benefits of valid cueing were
greater for triazolam than for placebo. The authors interpreted
this finding to suggest that triazolam might lead to an
increase in facilitation and/or a reduction in inhibition for
automatic attentional orienting mechanisms. With respect to
increased facilitation, the findings suggest that the attentional
grasp reflex may be enhanced with triazolam, as has been
shown with aging and with Alzheimer’s disease [55].
In both the endogenous and exogenous cueing
paradigms, the effects of informational cues are typically
measured by conducting “cost - benefit” analyses involving
three types of comparisons. First, the difference between
reaction time (RT) in the valid and neutral conditions is
termed “benefits”. Second, the difference between RT in the
invalid and neutral conditions is termed “costs”. Finally,
the differences between RT in the valid and invalid cue
conditions is termed “costs plus benefits” [52]. Benefits refer
to the RT advantage or facilitation conferred by having
available correct information as to target location prior to
target presentation. This allows target stimuli to be detected
more quickly at the attended location relative to non-
attended locations [53]. Benefits are thought to result from
the shift and engagement of attention to the correct spatial
location prior to target onset [54]. The automatic facilitation
effect or benefit seen with exogenous cueing has been referred
to as an “attentional grasp reflex” [55]. Costs refer to the
reaction time disadvantage conferred by having one’s
attention shifted to an incorrect spatial location prior to
target presentation. Costs are thought to result from the shift
and engagement of attention to an incorrect spatial location,
and then having to disengage attention from the incorrect
location, shift and re-engage attention to the correct target
location [56]. Costs plus benefits [57,58] represent a
composite measure of the time required for all of these
various components involved in the orienting of visual
attention.
Thus, there are some consistencies and some
inconsistencies between the results of the [59,60] studies. In
terms of similarities, both found attentional changes with
triazolam at the shorter SOA. Both found attentional changes
with triazolam for the exogenous cueing paradigm. Carter et
al.’s (1998) [60]findings of greater RT benefits with
triazolam are consistent with [59] greater costs plus benefits
for triazolam (i.e., the greater costs plus benefits in the latter
Johnson et al. (1995) used a version of Posner’s visual
search paradigm to examine the effects of a BZ (i.e.,
50 Current Pharmaceutical Design, 2002, Vol. 8, No. 1 Buffett-Jerrott
and
Stewart
study may have been entirely attributable to increased
benefits). However, the studies differed in terms of whether
drug induced attentional changes were specific to automatic
attentional mechanisms [60] or more general to both
automatic and controlled attentional processes [59]. This
discrepancy may be due to changes in the paradigm (e.g., %
of pre-cues that were valid) and/or to differences in the dose
of triazolam used (i.e., a lower dose was used [60]) across
the studies. Finally, Carter et al.’s claim that triazolam leads
to greater facilitation and/or decreased inhibition (i.e.,
increased benefits) on the exogenous cueing paradigm must
be interpreted cautiously. A closer examination of their data
indicates that their neutral cue condition did not always
“behave” as would be expected. In fact, at the shorter SOA
for the exogenous cue paradigm, the placebo group did not
just show less RT benefits than triazolam. In fact, placebo-
treated subjects showed no benefits whatsoever since neutral
pre-cues were associated with faster (rather than slower) RT
compared with valid cues. Many researchers have noted
similar problems with the neutral cue condition leading
some to suggest abandoning its use altogether [57] as was
the case in the [59] study.
of one study indicate that divided attention is not impaired
by BZs [61]. However, divided attention is a very complex
cognitive function that is also very difficult for non-drugged
participants. Therefore, it is possible that a floor effect may
have occurred on the task [61]. Further studies are necessary
in this area. Findings of impairment across various cognitive
paradigms used to assess different aspects of attention
highlight the robustness of this effect. However, it should be
noted that the attentional effects of BZs are subject to
tolerance with repeated dose administration. Tolerance to the
various cognitive effects of BZs is discussed at the end of the
paper.
MEMORY
A large number of studies have been conducted which
investigate the effects of BZs on memory. The amnesia
associated with BZ use is an important area of study for
many reasons. First, there is little doubt that memory is the
most important human cognitive ability [62]. Obviously,
memory impairments would be highly detrimental to a
person’s day-to-day functioning. Therefore, it is important to
determine what aspects of memory are affected by BZs, and
how long lasting these effects are.
Taken together, these two studies clearly suggest that
BZs do affect attentional shift functions. However, which
particular aspects of visual attention allocation mechanisms
are implicated (e.g., shift, engage, disengage, re-engage)
remains unclear. Further research is also needed on whether
BZ effects on this component of attention are limited to
automatic attention allocation processes (as suggested by the
work of Carter et al. [60]) or whether they are generalizable
to both automatic and controlled attention allocation
processes (as suggested by the work of Johnson et al. [59]).
In addition to understanding the effects of BZs on an
individual’s memory functioning, researchers are also
attempting to determine if BZ-induced amnesia is similar to
organic amnesia [62]. If the two forms of amnesia are indeed
similar, this could lead to significant advances in the
understanding of organic amnesia. Research in this area
could proceed much more quickly since researchers could use
BZs to induce a temporary amnesia rather than having to
recruit clinical participants with organic amnesia. Moreover,
by using BZs to create the experimental amnesia group, it
would be much simpler for researchers to obtain comparable
groups of amnesiacs and controls. In fact, participants would
be able to act as their own controls [63].
5) Divided Attention Tasks
Few studies have investigated BZ effects on divided
attention. A recent study examined the effect of dividing
attention of participants who had taken a clinical dose of
diazepam [61]. Participants were required to learn word pairs
in a single task condition, and then to learn word pairs at the
same time as performing a visual discrimination task.
Overall, dividing attention did reduce the learning of
unrelated word pairs, and the BZ group was impaired overall
in the learning of unrelated word pairs relative to placebo-
treated participants. However, the learning in the BZ group
was not disproportionately reduced compared to the placebo
group under the divided attention condition, suggesting that
BZs do not impair divided attention [61].
Researchers also study the effects of BZ treatment on
memory with the purpose of determining if BZs differentially
affect the various types of memory processes suggested by
memory theorists. This research can assist in determining if
theoretically distinct memory systems are, in fact,
empirically separable memory processes [63].
Finally, approximately 60% of panic disorder patients
that seek treatment from a psychotherapist are taking an
anxiolytic, most often BZs [64]. If these drugs are affecting
memory, it is possible that these individuals may not
remember the material they are learning in therapy. Much
research indirectly supports this hypothesis, and indicates
that BZs taken concurrently with psychotherapy has a
detrimental effect on cognitive behavioral therapy outcome in
the treatment of anxiety disorders [65,66].
SUMMARY OF BZ EFFECTS ON ATTENTIONAL
TASKS
Overall, it appears that BZs impair most aspects of
attention, with the possible exception of divided attention.
The research on shifting attention is not conclusive [59,60].
Johnson et al. (1995) suggested that both automatic and
voluntary attentional shifting is affected by triazolam [59].
However, Carter et al. (1998) [60], suggest that only the
automatic attentional shift reflex is affected by triazolam.
Obviously, more research is needed in this area. The results
As reviewed by Ghoneim and Mewaldt (1990) [62],
research indicates that memory is not a unitary concept.
Although many different models of memory functioning
exist, some memory theorists have described three basic
types of memory: sensory memory, short-term memory, and
long-term memory [67].
Effects of Benzodiazepine Use Current Pharmaceutical Design, 2002, Vol. 8, No. 1 51
Sensory memory contains brief sensory impressions that
remain after a stimulus is no longer present [68]. This
memory process occurs automatically, without voluntary
control. According to a review by Ghoneim and Mewaldt
(1990) [62], there are no studies regarding the effects of BZs
on sensory memory.
processes: implicit and explicit memory [73]. Explicit
memory occurs when a participant is aware that his or her
memory is being tested, and is consciously and effortfully
attempting to remember a previous experience. This is the
most commonly studied aspect of memory. In contrast,
implicit memory occurs when a participant’s ability to
perform a task is facilitated by a previous experience. Unlike
explicit memory, implicit memory occurs without conscious
awareness; the individual does not know that their memory
is being tested and does not consciously or effortfully
attempt to remember their previous experience [73].
Short-term memory occurs when an individual actively
attends to a stimulus. The capacity of short-term memory is
small, and if the memory is not actively rehearsed, it will be
quickly forgotten [69]. Studies using the digit span test have
found that short-term memory is not impaired, relative to
placebo-treated participants, by a clinical dose of oxazepam
[25], lorazepam [25,32,70] or diazepam [70], although there
are differences between forward and backward performance on
this task (see section on encode attention tasks). In addition,
participants who have taken a BZ still show a “recency
effect” (better memory for words shown at the end of an
encoding list) when recalling a previously-presented word
list, providing further evidence that short-term memory
remains intact [25].
Implicit and explicit memory processes are dissociated
by a number of factors including organic amnesia [74],
Alzheimer’s disease [75], anticholinergic drugs [76], divided
attention [77] and alcohol [78]. In all of these cases, explicit
memory is impaired while implicit memory is spared.
Although it is rare to find instances where implicit memory
is impaired while explicit memory remains intact, a
literature review [79] indicates that altering certain physical
aspects (e.g., typography) of the memory stimuli between
the encoding and memory testing phases impairs only
implicit memory. This double dissociation lends support to
the hypothesis that implicit and explicit memory are
qualitatively distinct memory processes that are supported
by separate memory systems [80].
Long-term memory contains large amounts of
information. This information remains in our long-term
memory indefinitely without active effort [67]. This is the
most commonly studied type of memory. Although much
research suggests that long-term memory is impaired by BZ
administration, studies have consistently indicated that BZs
only impair memory for information acquired after
administration of the drug (i.e., anterograde amnesia). No
well-designed study has found evidence of amnesia for
information acquired before BZ administration (i.e.,
retrograde amnesia) [71]. Information that was presented after
BZ administration is not well remembered, even when the
drug has been cleared from the body. However, once the drug
has been eliminated from the body, there is no amnesia for
new information [71].
Explicit Memory
Examples of paradigms used to assess explicit memory
include free recall, cued recall, and recognition memory
tasks. The most common method of assessing explicit
memory is to present participants with a list of words (after
the BZ has been administered) and to later test their memory
for items from the list. Consistently, free recall studies
indicate that participants who have received a clinical dose of
diazepam [81], lorazepam [81] or triazolam [82] are
impaired, relative to placebo-treated participants, in their
ability to remember word lists. In addition, studies using
cued recall tasks which “cue” subject’s memory of the
encoded stimuli (e.g., presenting the first three letters of an
encoded word) have found that individuals who have
received a dose of lorazepam [18,17] or oxazepam [18,46] are
impaired, relative to placebo-treated participants, in their
ability to remember the encoded word list. Some studies
have also found impaired recognition, relative to placebo-
treated participants, after BZ administration [82]. However,
other studies have found no impairment in recognition
memory [83]. Another study only noted recognition
impairments for higher doses of BZs [43]. This pattern of
results is most likely explained by the fact that recognition
tasks place fewer cognitive demands on a participant and are,
therefore, less sensitive to drug-induced memory deficits than
relatively more taxing free recall tasks [71].
Interestingly, several researchers have noted an effect
opposite to retrograde amnesia on explicit memory tasks
assessing memory for material presented prior to BZ
administration. BZ-treated participants appear to remember
more words from a list presented before drug administration,
when compared to placebo-treated participants. This
“retrograde facilitation” effect is only apparent if a different
word list is presented after drug administration than the list
presented prior to drug administration [71]. In one study
[11], the results indicated that the placebo participants
remembered 60% of the word list, regardless of whether it
was presented pre- or post- placebo administration. In
contrast, participants who had received a clinical dose of
triazolam remembered approximately 80% of the pre-drug
word list and only 30% of the post-drug list. This retrograde
facilitation is most likely due to the fact that the BZ-treated
participants’ learning of the post-drug list is less likely to
interfere with memory for the pre-drug list due to post-drug
memory impairments in learning of the post-drug list [72].
Paradigms using recall of short stories or news items
have also demonstrated that participants show impaired
ability to remember the details of a story, relative to the
placebo group, after receiving a clinical dose of lorazepam
[25,32], oxazepam [25] or diazepam [84]. Similarly,
performance on a task using a series of pictures as stimuli is
impaired, relative to placebo-treated participants, by a
clinical dose of diazepam or lorazepam [81].
Long-Term Memory: Explicit and Implicit Memory
Within the construct of long-term memory, researchers
commonly distinguish between two main memory
52 Current Pharmaceutical Design, 2002, Vol. 8, No. 1 Buffett-Jerrott
and
Stewart
Explicit Memory: Episodic Versus Semantic Memory placebo group. Overall, the results are supportive of explicit
(i.e., episodic) memory deficits following BZ
administration. In addition, the onset of the memory
impairment appears to be partially determined by the
absorption rate of the drugs. Lorazepam, which has a peak
absorption time of 120 minutes post drug, began to show
impairment at Time Block 2 (64-111 min). However,
oxazepam, with a theoretical peak absorption time of 170
minutes post-drug, did not show impairment until Time
Block 3 (112-160 min).
Within the domain of explicit memory, researchers often
distinguish between episodic and semantic explicit memory
[85]. Semantic memory is concerned with stored knowledge
about language, rules and the world and does not have to be
remembered with reference to a particular context [86]. In
contrast, episodic memory is the memory for a sequence of
occurrences [8], or personally experienced events [86].
Episodic memory is the most commonly assessed type of
explicit memory. Episodic memory tasks include all tasks
described above (recall and recognition of words, stories, and
pictures); performance on these tasks appears to be
consistently impaired by BZs. To test semantic memory,
participants are usually asked to generate lists of words that
fall into a semantic category (e.g., large animals).
Consistently, results indicate that BZs do not impair
semantic memory [9].
In summary, results of studies using different BZs and
different paradigms consistently indicate that long-term
explicit memory is impaired by BZs. In addition, this
memory impairment remains when a memory task is used
which may more closely mimic real world memory
requirements (i.e., a task which may be more ecologically-
valid).
One criticism of the lab-based explicit tasks of episodic
memory, which are described above, is that they may lack
ecological validity [71]. For example, memory for word lists
may not reflect “everyday memory demands” and BZ-
induced impairments on such tasks may not be as apparent
in the real world. A person who is asked to remember a list
of relatively “meaningless” words in a lab may not be
sufficiently motivated to remember the list. However, the
same individual may be quite motivated to remember what
groceries they need to buy. Lending support to this
hypothesis, one study found that BZ-treated participants
were less likely to show impairments on memory tasks that
more closely simulate real-world situations [20].
Interestingly, studies of meta-memory in participants
who have taken a BZ indicate that participants are relatively
unaware of these memory impairments. In fact, when asked
to rate their “feeling of knowing” regarding their answers on
an explicit memory test, participants who had taken a
clinical dose of lorazepam performed at chance levels [13].
Analysis of their performance on the memory tasks indicated
that their performance was impaired. The results of this
study indicate that BZ-treated individuals may not be aware
that their performance is impaired on a test of explicit
memory. It appears that if an individual doesn’t remember
anything, they may conclude that nothing worth
remembering has occurred [8]. Lending support to this
explanation, most BZ-treated patients do not complain about
memory problems [87]. This lack of awareness of memory
impairments has serious implications for individuals who are
regularly taking BZs for the treatment of clinical anxiety, as
they may not be aware of the detrimental effect of the drug on
their daily functioning and may therefore not bring it to the
attention of their prescribing physicians. However, another
study found that when participants who had taken a clinical
dose of lorazepam or oxazepam were asked to remember the
details of a story, as opposed to a word list, they were aware
that their memory was impaired [25]. Perhaps, BZ-treated
individuals are more aware of memory deficits when the
memory task being used more closely simulates real-world
memory demands.
To further investigate this problem, Buffett-Jerrott and
associates created the Movie Memory Task [18]. In this
task, participants who had taken a clinical dose of lorazepam,
oxazepam or a placebo watched a movie for 160 minutes after
drug administration. Every 15 minutes the movie was
stopped and the participants were asked 5 questions about
their memory for details of the movie (e.g., “what color
jacket was …[the principal character] wearing?”). It was
hypothesized that participants would find the movie more
interesting than traditional memory task stimuli (i.e. word
lists). In addition, it is important for most people to
remember the details of a story, so that the plot can be
followed. For this reason, participants might be more
motivated to remember on the Movie Memory Task.
Research has also been conducted to determine what
stage(s) of memory processing is/are impaired by BZs.
Researchers talk about three stages in remembering
information: acquisition, storage, and retrieval (see review
[88]). First an individual must acquire or encode the
information. Secondly, the individual must store this
information for later use. Finally, when needed, the
individual must be able to retrieve the memory from storage.
If BZs interfere with any of these three stages, information
will not be remembered [62]. The present view is that it is
the acquisition of new information, rather than storage or
retrieval, that is affected by BZs [84]. However, it is still
unclear which encoding operations involved in memory
acquisition are impaired by BZs [61].
Buffett-Jerrott and associates compared performance of
lorazepam, oxazepam and placebo-treated participants on the
Movie Memory Task [18]. To interpret the data, the
questions were combined into three 45-minute Time Blocks.
In the first Time Block (16-63 min. post-drug), there was no
difference between the drug groups and the placebo
participants for the number of questions correctly answered.
In the second Time Block (64-111 min. post-drug), the
lorazepam participants were marginally impaired, compared
to the placebo group. However, the oxazepam participants
did not show impairment relative to placebo participants at
the second time block. For the final Time Block (112-160
min. post-drug), both the lorazepam and oxazepam
participants were significantly impaired relative to the
Effects of Benzodiazepine Use Current Pharmaceutical Design, 2002, Vol. 8, No. 1 53
Implicit Memory the time of peak blood concentration of a BZ for impairment
to be observed. To test this hypothesis, Legrand and
colleagues examined the effects of clinical doses of diazepam
and lorazepam at three time points (50, 130, and 300 min.
post-drug) [81]. On a picture fragment completion task,
results indicated that diazepam impaired priming at 50 min.
post-drug, while lorazepam did not impair priming until 130
minutes post-drug. As noted by the researchers, these
impairment times are similar to the time of peak plasma
concentration for each drug (61 min for diazepam; 120
minutes for lorazepam), lending support to the time course
hypothesis.
The paradigms that are most commonly used to
investigate implicit memory are “priming” tasks such as the
word-stem completion task and the picture-fragment
completion task (see review [79]). In the word-stem
completion task, participants encode a word list and are later
asked to complete a series of three letter word stems with
words that come to mind. Typically, participants are more
likely to complete these stems with words they have seen
previously in the encoding list (i.e., the “primes”) than with
equally frequent alternatives. This effect is called “priming”
[79]. Another commonly used priming task is the picture-
fragment completion task. In this task, participants encode a
series of line drawings. Later they are presented with
fragmented pictures and asked to identify them. They are
presented with increasingly complete pictures until the
picture is identified. Priming is said to occur when
participants are quicker to recognize the fragmented pictures
that they have seen earlier than novel pictures [70,79].
There were several problems with [81] study that made it
difficult to conclusively accept the time dependence
hypothesis. First, the results indicated contamination of the
implicit memory task from the explicit task due to task
repetition. Implicit memory tasks should always be
presented before explicit tasks, to ensure that the participant
does not become aware of the true purpose of the implicit
task and consequently use conscious memory retrieval
strategies to complete the implicit task. If a BZ-treated
participant attempts to use their impaired explicit memory
on an implicit test, their performance may be decreased on
the implicit task [91]. With multiple testing times, a
participant will have had previous experience with the
explicit task administered at the first testing time before
completing the other implicit tasks, increasing the chances of
contamination of the later-administered implicit tasks with
explicit memory strategies. For this reason, it is
recommended that implicit memory tasks should never be
used as repeated measures [79].
Early studies indicated that BZs such as diazepam [27]
and oxazepam [31] did not impair implicit memory relative
to placebo participants on the word-stem completion task.
However, further research indicated that a clinical dose of one
BZ -- lorazepam -- impaired priming on both word-stem
completion and picture fragment completion tasks, relative
to the placebo-treated participants [32,70,89]. In 1993,
Curran and Gorenstein directly compared the effects of
clinical doses of oxazepam and lorazepam on an implicit
memory task. Participants encoded a word list at 120
minutes post-drug and were later asked to complete a word-
stem completion task. The results indicated that lorazepam
impaired priming, while oxazepam did not. Until very
recently, it was believed that only lorazepam impaired
priming. The finding that lorazepam impaired implicit
memory was a very intriguing finding because implicit
memory is not affected by most other variables. Possible
reasons given for this finding include a different cortical
distribution of lorazepam or a different population of BZ
receptors uniquely affected by lorazepam [31,89].
A second problem with [81] study was that the implicit
and explicit tasks did not satisfy the “retrieval intentionality
criterion” [92]. That is, the implicit and explicit tasks
should be identical, except for the instructions to the
participants, if they are to be directly comparable. If explicit
and implicit tasks are not similar, differences in BZ effects
might be explained by other confounding factors. For
example, a free recall task may have very different cognitive
demands than a word-stem completion task.
In 1994, Vidailhet and colleagues compared the effects of
clinical doses of lorazepam and diazepam on picture fragment
and word-stem completion. Encoding of the words and
pictures took place at approximately two hours post-drug. In
contrast to previous studies, both lorazepam and diazepam
impaired implicit memory. As noted by the researchers [40],
the time elapsed between drug administration and encoding
was longer in this study than in the previous studies which
found no impairment of priming by diazepam (previous
encoding occurred at 45-60 min. post-drug). The researchers
suggested that diazepam might impair priming if enough
time is allowed for the drug to take effect. Another study also
found impaired priming after alprazolam administration,
lending further support to the idea that a number of BZs may
impair implicit memory [45].
Stewart and associates (1996) used tasks which satisfied
the “retrieval intentionality criterion” to investigate the
potential time-dependent effects of clinical doses of lorazepam
and oxazepam on priming [17]. Encoding took place at two
time points (100 and 170 minutes post drug). The implicit
task used was a word-stem completion task. This task was
followed by a cued-recall explicit memory task, in which
participants were again presented with word stems but were
asked to complete them with words that they remembered
from the encoding list. The results of this study indicated
that at 100 minutes post-drug, similar to the results of [31],
only lorazepam impaired priming, while both drugs
impaired explicit memory. However, at 170 minutes post
drug, both drugs impaired implicit and explicit memory.
Again, this study lends support to the time-dependence
hypothesis: that is, implicit memory is only impaired if it is
tested around the theoretical peak blood concentration of the
drug. Lorazepam’s peak concentration occurs at 120
minutes, while oxazepam is not maximally absorbed until
170 minutes. However, as noted previously for the [81]
In 1995, Legrand and associates noted that the novel
findings of [40] could be explained by the different
absorption rates of BZs. BZs are absorbed at different rates,
and these speeds are important predictors of their clinical
effects [90]. Therefore, encoding may need to take place near
54 Current Pharmaceutical Design, 2002, Vol. 8, No. 1 Buffett-Jerrott
and
Stewart
study, the repeated measures design used in this study might
have led to explicit contamination of the implicit task at the
second testing time (170 min).
Overall, these findings of a differential time course of BZ-
induced impairments in explicit vs. implicit memory task
performance are supportive of the hypothesis that implicit
and explicit tasks are subserved by distinct memory systems
[80].
Buffett-Jerrott and associates (1998b) [18] attempted to
replicate the Stewart and associates’ (1996) study after
omitting the first cycle of memory testing [17]. Therefore, if
oxazepam still impaired priming at 170 minutes post-drug,
it was less likely that the implicit task had been
contaminated by the explicit task, as prior exposure to the
explicit task had been eliminated. The results of this study
indicated that both oxazepam and lorazepam impaired
implicit memory when encoding took place at 170 minutes
post-drug. This finding lends stronger support to the time-
dependence hypothesis. It appears that when oxazepam is
nearing its peak blood concentration [90] it begins to cause
implicit memory impairments.
Implicit Memory: Conceptual Versus Perceptual Priming
More recently, researchers have distinguished between
two types of priming in implicit memory paradigms [93]:
perceptual and conceptual priming. Perceptual priming is the
most commonly studied type of priming. It is said to occur
when prior exposure to a stimulus facilitates the participant’s
ability to later identify the same stimulus perceptually when
relevant cues are presented [93,94]. Examples of tasks
tapping perceptual priming include word stem completion
and degraded pictures. In contrast, conceptual priming occurs
when a participant is presented with a cue that is
conceptually related to a previously- encoded stimulus [79].
For example, a participant might encode a word list that
includes the word “elephant”. Later, when asked to name an
animal, he/she might be more likely to say “elephant” than
to name an animal to which they had not been previously
exposed. Therefore, semantic processing must have occurred,
as there is no perceptual similarity between the original
stimulus (“elephant”) and the cue presented (“animal”) [94].
Researchers suggest that conceptual priming is based on the
semantic memory system, which assesses knowledge,
whereas perceptual priming is based on the perceptual
representation system (PRS) that assesses the structure and
form of stimuli [93]. Studies indicate that conceptual and
perceptual priming can be dissociated by factors such as age
[95], supporting the hypothesis that they are distinct
memory processes subserved by dissociable memory
systems.
Interestingly, the data appear to indicate that the time
course of BZ-induced implicit and explicit memory
impairments may differ. For example, Stewart and associates
(1996) found that explicit memory was impaired by
oxazepam at both testing times (pre-peak and peak) [17], but
priming was not impaired until the second testing time
(peak). Also, findings with alprazolam indicate that explicit
memory impairments may last for a longer time than
implicit impairments, following BZ administration [45].
Buffett-Jerrott and associates (1998a) [46] directly
investigated these hypothesized time course differences by
examining the effects of a clinical dose of oxazepam at three
different time points (100, 170 and 240 minutes post-drug),
on directly comparable implicit and explicit memory tasks
(word-stem completion and cued recall). Due to the potential
for explicit contamination of implicit memory task
performance with a repeated measures design, separate groups
of participants were used at each time point. Very few well-designed studies have examined the effects
of BZs on conceptual priming. However, one study of
conceptual priming indicated that it was not impaired by
administration of a clinical dose of lorazepam (examined at
two hours post-drug) [94]. This is in contrast to other
studies that have found impairments in perceptual priming
assessed at the same time point with the same BZ [32]. A
recent study compared the effects of a clinical dose of
lorazepam on a conceptual (word association) and a
perceptual (word identification) task. Results indicated that
perceptual priming was impaired by lorazepam, relative to
placebo-treated participants. However, the lorazepam-treated
participants did not show impaired performance, relative to
the placebo group, on the conceptual priming task [96]. This
differential impairment of conceptual and perceptual priming
lends further support to the view that perceptual and
conceptual priming processes are subserved by two different
memory systems [93].
Overall, the results of this study supported the time-
dependence hypothesis regarding implicit memory
impairments. At the first testing time, before the peak blood
concentration of oxazepam would be reached, implicit
memory was not impaired, relative to the placebo-treated
participants. However, at 170 minutes, which is the time of
theoretical peak blood concentration for oxazepam, priming
was impaired. Finally, at 240 minutes, after the peak blood
concentration of oxazepam, oxazepam-induced priming
impairments were only marginally significant compared to
placebo, suggesting that the priming impairments were
beginning to wane. In contrast to the implicit memory
findings, explicit memory impairment was apparent at each
of the three testing times. This lends support to the idea of
separate time courses of implicit and explicit memory
impairments. It appears that explicit memory impairment
begins at an earlier time point than implicit memory
impairments. In addition, these impairments appear to
persist for longer periods of time. It is possible that BZ-
induced implicit memory impairments require relatively
high levels of the drug in the bloodstream, while explicit
memory impairments are apparent at lower levels.
Alternatively, a separate population of BZ receptors which
are uniquely affected by high concentrations of BZs, may be
responsible for priming, but not for explicit memory [46].
In summary, it appears that implicit memory is impaired
by BZs if participants are tested at the time of theoretical
peak plasma concentration of the drug. Thus far, four BZs
have been found to impair priming: lorazepam [31],
alprazolam [45], oxazepam [46], and diazepam [81]. Each of
these studies is supportive of the time-dependence
hypothesis. Priming is an ever-present occurrence in our
lives and underlies our ability to complete many tasks. The
Effects of Benzodiazepine Use Current Pharmaceutical Design, 2002, Vol. 8, No. 1 55
impairment of this process has serious implications for
individuals to adequately function after taking a BZ [93]. It
appears that BZ-induced implicit memory impairments may
be evident over a smaller window of time around drug
administration relative to explicit memory impairments.
Moreover, some preliminary work suggests that BZ-induced
implicit memory impairments may be specific to perceptual
as opposed to conceptual priming processes.
induced inattention and sedation, but insufficient to
completely block the amnestic effects. This hypothesis
would help explain why [97] found no reversal of amnesia
using .50 mg of flumazenil while the [98] found partial
reversal using an average dose of .73 mg. Lending further
support to this hypothesis is a recent study that used a
relatively high dose of flumazenil (1.0 mg) and found that
the explicit amnesic effects of triazolam were completely
reversed [102]. However, the receptor occupancy hypothesis
is not consistent with the results of [44] who used a much
larger dose of flumazenil (.035 mg/kg) than [102] and did
not find reversal of amnesia (i.e., the Hommer et al. dose
would be about 2.4 mg for a 68 kg participant). Clearly,
further research is needed in this area.
Separating the BZ-Induced Amnestic Effects from
Sedation and Attentional Impairments
As noted earlier, BZs produce profound increases in
subjective sedation, slow psychomotor and cognitive
processing speed, and lead to attentional impairments. It is
possible that the memory deficits associated with BZ
administration are simply by-products of these other
cognitive impairments. Obviously, if a participant is too
sedated or inattentive to encode a word list, it will appear as
though their memory is impaired. Sedation and/or
inattention may contribute to memory encoding difficulties,
rather than these effects proving to be primary BZ-induced
memory impairments. Researchers have attempted to
dissociate the amnestic effects of BZs from sedation and
inattention in several ways. In general, these findings
suggest that even when sedation and attentional impairments
are controlled, memory impairments remain.
The occurrence of sedation without amnesia also lends
indirect support to the idea of BZ-induced primary memory
deficits. In one study, lorazepam was compared with the
anti-psychotic medication, chlorpromazine. The results
indicated that the two drugs induced equal levels of sedation,
but that only lorazepam impaired memory [12]. In a second
study, BZ-treated participants showed similar levels of
sedation to individuals suffering from sleep deprivation.
However, only the participants who had taken a BZ were
impaired on the memory tasks [9]. The fact that certain
manipulations (i.e., sleep deprivation and chlorpromazine
administration) induce equal levels of sedation to BZs, but
fail to induce memory impairments, provides further support
that BZ-induced amnesia is not merely a by-product of
sedation.
First, most studies indicate that when performance scores
on tasks measuring attention or sedation are treated as
covariates, the significance levels of the BZ-induced memory
effects may be reduced, but the effects still remain significant
or marginally significant [17,18,25,32,46]. These findings
suggest that inattention and sedation cannot fully account for
BZ-induced memory impairments.
Finally, studies using repeated dose administrations
indicate that tolerance to the various cognitive side-effects
associated with BZ use increases differentially. One study
indicated that tolerance to the psychomotor and sedative
effects of BZs developed within three weeks. However,
memory performance remained impaired over this same
interval [103]. Overall, it appears that tolerance develops to
the sedative effects of BZs, but does not occur for memory
impairment [104]. In a study of individuals who had taken
BZs regularly for an average of 10 years, the results indicated
that attention and objective sedation were not impaired after
administration of an acute dose of BZ. However, some
aspects of memory performance remained significantly
impaired [104].
Lending further support to these findings, the majority of
studies indicate that the use of flumazenil, a BZ-receptor
antagonist, reverses the sedative and attentional effects of
BZs while leaving the memory impairment effects intact
[44,97]. An additional study found a complete reversal of
sedative and attentional BZ effects and a partial reversal of
BZ-induced amnestic effects [98]. If the BZ-induced memory
deficits are still apparent when the sedation and attentional
impairments are eliminated pharmacologically, it is likely
that the amnestic effects of BZs are not entirely determined
by these other factors. The reason that a BZ-receptor
antagonist fails to block the amnestic effect of BZs is not
known. One possibility is that BZ-induced sedative and
attentional effects versus amnestic effects are subserved by
different populations of BZ receptors, which in turn are
differentially blocked by flumazenil. Indeed, different
subtypes of BZ receptors have been identified [99,100].
However, this explanation appears unlikely given that
flumazenil has been shown to show equal affinity for different
subtypes of BZ receptors [101]. An alternative explanation is
that different percentages of BZ receptors need to be occupied
to produce the various cognitive effects of BZs [44]. Perhaps
the attentional and sedative effects require higher percentage
of receptor occupation, while the amnestic effects occur at
lower occupation. Therefore, the concentration of flumazenil
used in previous studies might be sufficient to block BZ-
OVERALL EFFECTS OF BZS ON HUMAN
COGNITION
Overall, studies using a wide variety of BZs, and
numerous experimental paradigms, indicate that sedation is
increased, and attentional processes are impaired after an
acute dose of a BZ. However, these impairments are subject
to tolerance, and are not a major problem for long-term BZ
users. Memory deficits, however, appear to be severe and
long-lasting. BZs impair long-term memory in an
anterograde fashion and these memory deficits are apparent
even on memory tasks that simulate real-life memory
requirements. Overall, it appears that BZs impair
performance on both implicit and explicit memory tasks if
these memory abilities are tested close to the time of peak
blood concentration of the drug. In addition, these memory
56 Current Pharmaceutical Design, 2002, Vol. 8, No. 1 Buffett-Jerrott
and
Stewart
deficits are not simply by-products of the detrimental effects
of BZs on sedation and attention.
[2] Medina, J.H., Paladini, A.C., Izquierdo, I. Behavioral
Brain Research, 1993, 58, 1-8.
[3] Alexander, P.E. The clinical and scientific rationale for
tapering benzodiazepines. Ontario: The Upjohn
Company. 1988.
In the future, researchers should strive to use paradigms
that are more similar to real-world memory requirements.
Likewise, studies of implicit and explicit memory should
satisfy the retrieval intentionality criteria (i.e., the tasks
should be directly comparable [79]). It is important that
other BZs be studied at the time of their peak blood
concentration, to determine if the time-dependence
hypothesis continues to be supported across a variety of BZs
in addition to lorazepam. In addition, it would be interesting
to examine the effects of these drugs at pre- and post-
theoretical-peak testing times, to determine if implicit and
explicit memory impairments begin to appear and disappear
at different time points, as has been suggested by one study
[46]. More research on the conceptual/perceptual priming
distinction within the implicit memory literature should be
conducted to determine whether initial results suggesting
that perceptual priming is impaired, while conceptual
priming is spared, are replicable across a variety of BZs and
priming tasks. Finally, it is important to note that the
majority of studies regarding the cognitive effects of BZ
administration have used normal participants without a
history of BZ/medication use and/or anxiety disorders.
Further study with anxious populations, and individuals
with chronic BZ use histories, is needed to determine if the
deleterious cognitive effects of BZs generalize to these
important clinical populations.
[4] Leonard, B.E. Journal of Psychiatric Research, 1993,
27, 193-207.
[5] Ashton, H. Psychiatric Annals, 1995, 25, 158-165.
[6] Woods, J.H., Winger. Psychopharmacology, 1995, 118,
107-115.
[7] Greenblatt, D.J., Shader, R.I., Abernaethy, D.R. New
England Journal of Medicine, 1983, 309, 410-416.
[8] Hommer, D.W. In P.P. Roy-Byrne and D.S. Cowley
(Eds.), Benzodiazepines in clinical practice: Risks and
benefits (pp. 113-130). Washington, D.C.: American
Psychiatric Press. 1991.
[9] Fluck, E., File, S.E., Springett, J., Kopelman, M.D., Rees,
J., Orgill, J. Pharmacology, Biochemistry and Behavior,
1998, 59, 909-915.
[10] Haslett, W.H.K., Dundee, J.W. British Journal of
Anaesthesia, 1968, 40, 250-258.
[11] Weingartner, H.J, Sirocco, K., Rawlings, R., Joyce, E.,
Hommer, D. Psychopharmacology, 1995, 119, 27-33.
[12] Green, J.F., McElholm, A. & King, D.J.
Psychopharmacology, 1996, 128, 67-73.
Knowledge of when implicit and explicit memory
impairment can be expected to be maximal will be helpful
for anesthetists who wish to use BZs as surgical pre-
medicants [97]. Future research should be conducted to
directly investigate the influence of time-course of BZ-
induced memory impairments in a group of anxious
individuals who are receiving concurrent psychotherapy.
This knowledge could aid therapists who wish to minimize
the detrimental impact of these amnestic effects for clients
who are receiving both psychotherapy and BZ therapy for the
treatment of anxiety disorders [65].
[13] Bacon, E., Danion, J.M., Kauffmann-Muller, F.,
Schelstraete, M.A., Bruant, A., Sellal, F., Grange, D.
Psychopharmacology, 1998, 138, 318-325.
[14] Block, R.I., Berchou, R. Pharmacology, Biochemistry
and Behavior, 1984, 20, 233-241.
[15] Bond, A., Lader, M. British Journal of Medical
Psychology, 1974, 47, 211-218.
[16] Danion, J.M., Zimmermann, M.A., Willard-Schroeder, D.,
Grangé, D., Singer, L. Psychopharmacology, 1989, 99,
238-243.
ACKNOWLEDGEMENT
[17] Stewart, S.H., Rioux, G. F., Connolly, J.F., Dunphy, S.,
Teehan, M.D. Psychopharmacology, 1996, 128, 139-
149.
Preparation of this manuscript was supported, in part, by
grants from the Natural Sciences and Engineering Research
Council of Canada (NSERC) (to the second author) and the
IWK Grace Health Centre Research Services (to the first and
second author). A Doctoral Fellowship from the Natural
Sciences and Engineering Research Council of Canada
(NSERC) funded the first author’s graduate studies. We
thank Dr. Raymond Klein, Dr. Allen Finley and Dr. Valerie
Curran for their advice on an earlier version of this
manuscript.
[18] Buffett-Jerrott, S.E., Stewart, S.H., Teehan, M.D.
Psychopharmacology, 1998b, 138, 344-353.
[19] Curran, H.V., Bond, A., O’Sullivan, G., Bruce, M., Marks,
I, Lelliot, P., Shine, P., Lader, M. Psychological
Medicine, 1994, 24, 969-976.
[20] Allen, D., Curran, H.V., Lader, M. European Journal of
Clinical Pharmacology, 1991, 40, 355-362.
REFERENCES
[21] Wilton, N.C.T., Leigh, J., Rosen, D.R., Pandit, U.A.
Anesthesiology, 1988, 69, 972-975.
[1] Gold, M. S., Miller, N.S., Stennie, K., Populla-Vardi, K.
Psychiatric Annals, 1995, 25, 146-148.
[22] Weldon, B.C., Watcha, M.F., White, P.F. Anesthesia and
Analgesia, 1992, 75, 51-55.
Effects of Benzodiazepine Use Current Pharmaceutical Design, 2002, Vol. 8, No. 1 57
[23] Wechsler, D. Manual for the Wechsler Adult
Intelligence Scale – Revised (WAIS-R). New York: The
Psychological Corporation. 1981.
[43] Weingartner, H.J., Hommer, D., Lister, R.G., Thompson,
K., Wolkowitz, O. Psychopharmacology, 1992, 106,
341-345.
[24] Ellinwood, E.H., Heatherly, D.G., Nikaido, A.M.,
Bjornsson, T.D., Kilts, C. Psychopharmacology, 1985,
86, 392-399.
[44] Hommer, D.W., Weingartner, H., Breier, A.
Psychopharmacology, 1993, 112, 455-460.
[45] Fleishaker, J.C., Garzone, P.D., Chambers, J.H., Sirocco,
K., Weingartner, H. Psychopharmacology, 1995, 120,
169-176.
[25] Curran, H.V., Schiwy, W., Lader, M.
Psychopharmacology, 1987, 92, 358-364.
[26] Danion, J.M., Zimmermann, M.A., Willard-Schroeder, D.,
Grangé, D., Welsch, M., Imbs, J.L., Singer, L.
Psychopharmacology, 1990, 102, 422-424.
[46] Buffett-Jerrott, S.E., Stewart, S.H., Bird, S., Teehan, M.D.
Journal of Psychopharmacology, 1998a, 12, 338-347.
[47] Lezak, M.D. Neuropsychological assessment, 3
rd
ed.
Oxford, UK: Oxford University Press. 1995.[27] Fang, J.C., Hinrichs, J.V., Ghoneim, M.M. Pharmacology,
Biochemistry and Behavior, 1987, 28, 347-352.
[48] Wright, R.D., Ward, L.M. Canadian Journal of
Experimental Psychology, 1994, 48, 151-166.[28] Curran, H.V., Schifano, F., Lader, M.
Psychopharmacology, 1991, 103, 83-90.
[49] Klein, R., Kingstone, A., Pontefarct, A. In K. Rayner
(Ed.), Eye movements and visual cognition: Scene
perception and reading (pp. 46-65). New York:
Springer-Verlag. 1992.
[29] Curran, H.V., Pooviboonsuk, P., Dalton, J.A., Lader,
M.H. Psychopharmacology, 1998, 135, 27-36.
[30] Boulenger, J.P., Squillace, K., Simon, P., Herrou, M.,
Leymarie, P., Zarifan, E. Neuropsychobiology, 1989, 22,
83-89.
[50] Posner, M. I. Quarterly Journal of Experimental
Psychology, 1980, 32, 3-25.
[31] Curran, H.V., Gorenstein, C. International Clinical
Psychopharmacology, 1993, 8, 37-42.
[51] Posner, M., Nissen, M.J., Ogden, W.C. In H.L. Pick and
E. Saltzman (Eds.), Modes of pereciving and processing
information (pp. 137-157).New York: Erlbaum. 1978.
[32] Bishop, K.I., Curran, H.V., Lader, M., Experimental
Clinical Psychopharmacology, 1996, 4, 292-299. [52] Posner, M. I., Snyder, C.R., Davidson, B.J. Journal of
Experimental Psychology: General, 1980, 109, 160-
174.
[33] Preston, G.C., Ward, C.E., Broks, P., Traub, M., Stahl,
S.M. Psychopharmacology, 1989, 97, 222-227.
[53] Posner, M. I., Snyder, C. In R.J. Solso (Ed.), Information
processing and cognition: The Loyola symposium (pp.
55-85). Hillsdale, NJ: Erlbaum. 1975.
[34] Lucki, I., Rickels, K., Geller, A.M.
Psychopharmacology, 1986, 88, 426-433.
[35] Kirby, M., Denihan, A., Bruce, I., Radic, A., Coakley, D.,
Lawlor, B.A. International Journal of Geriatric
Psychiatry, 1999, 14, 280-284.
[54] Wright, M.J., Burns, R.J., Geffen, G.M., Geffen, L. B.
Neuropsychologica, 1990, 28, 151-159.
[55] Maruff, P., Currie, J. Neuropsychologia, 1995, 33, 689-
702.
[36] Ewing-Cobbs, L., Prasad, M., Fletcher, J.M., Levin, H.S.,
Miner, M.E., Eisenberg, H.M. Child Neuropsychology,
1998, 4, 35-48. [56] Clark, C.R., Geffen, G.M., Geffen, L.B. Neuroscience and
Biobehavioral Reviews, 1987, 11, 341-352.
[37] VanZomeren, A.H. & Brouwer, W.H. Clinical
neuropsychology of attention. New York: Oxford
University Press. 1994.
[57] Jonides, J. & Mack, R. Psychological Bulletin, 1984, 96,
29-44.
[38] Mirsky, A.F., Anthony, B.J., Duncan, C.C., Ahearn, M.B.,
Kellam, S.G. Neuropsychology Review, 1991, 2, 109-
145.
[58] Posner, M.I. In M.I. Posner (Ed.), Chronometric
explorations of mind: The third Paul M. Fitts lectures
(pp. 185-216). New York: Oxford University Press.
1986.
[39] Stankov, L. Journal of Educational Psychology, 1983,
75, 471-490. [59] Johnson, D.N., Weingartner, H.J., Andreason, P., George,
D.T. Psychopharmacology, 1995, 121, 145-149.
[40] Vidailhet, P., Danion, J.M., Kauffmann-Muller, F.,
Grangé, D., Giersch, A., Van Der Linden, M., Imbs, J.L.
Psychopharmacology, 1994, 115, 397-406.
[60] Carter, C.S., Maddock, R., Chaderjian, M., Post, R.
Progressive Neuro-Psychopharmacology and
Biological Psychiatry, 1998, 22, 279-292.
[41] Vidailhet, P., Kazès, M., Danion, J.M., Kauffmann-Muller,
F., Grangé, D. Psychopharmacology, 1996, 127, 63-72. [61] Gorissen, M.E.E. & Eling, P.A.T.M.
Psychopharmacology, 1998, 138, 354-361.
[42] Coull, J.T., Sahakian, B.J., Middleton, H.C., Young, A.H.,
Park, S.B., McShane, R.H., Cowen, P.J., Robbins, T.W.
Psychopharmacology, 1995, 121, 222-230.
[62] Ghoneim, M.M., Mewaldt, S.P. Anesthesiology, 1990,
72, 926-938.
58 Current Pharmaceutical Design, 2002, Vol. 8, No. 1 Buffett-Jerrott
and
Stewart
[63] Hirshman, E., Passanante, A., Arndt, J. Cognitive Brain
Research, 1999, 7, 473-479.
[83] Eves, F.F., Curran, H.V., Shine, P., Lader, M.H.
Psychopharmacology, 1988, 95, 386-389.
[64] Otto, M.W., Pollack, M.H., Penava, S.J. & Zucker, B.G.
Behaviour Research and Therapy, 1999, 37, 763-770.
[84] Gorissen, M.E.E., Curran, H.V., Eling, P.A.T.M.
Psychopharmacology, 1998, 138, 334-343.
[65] Westra, H.A., Stewart, S.H. Clinical Psychology Review,
1998, 18, 307-340.
[85] Tulving, E. Elements of episodic memory. Cambridge,
UK: Oxford University Press. 1983.
[66] Westra, H.A., Stewart, S.H., Conrad, B.E. Journal of
Anxiety Disorders. (in press).
[86] Curran, H.V. Psychopharmacology, 1991, 105, 1-8.
[87] Barbee, J.G. Journal of Clinical Psychiatry, 1993, 54,
86-97.46[67] Atkinson, R.C., Shiffrin, R.M. The psychology of
learning and motivation (vol.2) (pp. 89-105). New
York: Academic Press. 1968.
[88] Murdock, B.B., Anderson, R.E. In R.J. Solso (Ed.),
Information processing and cognition: The Loyola
symposium (pp. 145-194). Hillsdale, NJ: Erlbaum. 1975.[68] Atkinson, R.C., Shiffrin, R.M. Scientific American, 1971,
225, 82-90.
[89] Brown, M.W., Brown, J., Bowes, J.B. Quarterly Journal
of Experimental Psychology, 1989, 41A, 599-617.[69] Murdock, B.B. The psychology of learning and
motivation (Vol 5, pp. 67-127). New York: Academic
Press. 1971.
[90] Greenblatt, D.J., Shader, R.I., Divoll, M., Harmatz, J.S.
British Journal of Clinical Pharmacology, 1981, 11,
11S-16S.[70] Sellal, F., Danion, J.M., Kauffmann-Muller, F., Grange, D.,
Imbs, J.L., Van Der Linden, M., Singer, L.
Psychopharmacology, 1992, 108, 371-379.
[91] Schacter, D.L., Chiu, C.Y., Ochsner, K.N. Annual Review
of Neuroscience, 1993, 16, 159-182.
[71] Curran, H.V. Biological Psychology, 1986, 23, 179-213.
[92] Schacter, D.L., Bowers, J., Booker, J. In S. Lewandowsky,
J.C. Dunn, & K. Kirsner (eds.), Implicit Memory:
Theoretical Issues (pp. 47-65). Hillsdale, NJ: Erlbaum.
1989.
[72] Hinrichs, J.V., Mewaldt, S.P., Ghoneim, M.M., Berie, J.L.
Pharmacology, Biochemistry and Behavior, 1982, 17,
165-170.
[73] Graf, P., Schacter, D.L. Journal of Experimental
Psychology, 1985, 11, 501-518.
[93] Tulving, E., Schacter, D.L. Science, 1990, 247, 301-306.
[94] Bishop, K. I., Curran, H.V. Psychopharmacology, 1998,
140, 345-353.[74] Squire, L.R., Shmamura, A.P., Graf, P. Neuropsychologia,
1987, 25, 195-210.
[95] Jelicic, M., Craik, F.I., Moscovitch, M. European
Journal of Cognitive Psychology, 1996, 8, 225-234.[75] Keane, M.M., Gabrieli, D.E., Fennema, A.C., Growdon,
J.H., Corkin, S. Behavioral Neuroscience, 1991, 105,
326-342.
[96] Thompson, C.E., Stewart, S.H., MacPherson, P.S.
Canadian Psychology, 1999, 40, 5.
[76] Kopelman, M.D., Corn, T.H. Brain, 1988, 111, 1079-
1110.
[97] Curran, H.V., Birch. Psychopharmacology, 1991, 103,
519-523.
[77] Rabinowitz, J.C., Craik, F.I., Ackerman, B.P. Canadian
Journal of Psychology, 1982, 36, 325-344.
[98] Flumazenil Study Group. Clinical Therapeutics, 1992,
14, 895-909.
[78] Lister, R.G., Gorenstein, C., Risher-Flowers, D.,
Weingartner, H.J., Eckardt, M.J. Neuropsychology, 1991,
29, 1205-1212.
[99] Montaldo, S., Serra, M., Concas, A., Corda, M. G., Mele,
S., Biggio, G. Neuroscience Letters, 1984, 52, 263-268.
[100] Podhorna, J. Current Pharmaceutical Design, 2001.
[79] Roediger, H.L. III, & McDermott, K.B. In H. Spinnler &
F. Boller (Eds.), Handbook of Neuropsychology (pp. 63-
131). Amsterdam: Elsevier. 1993
[101] Hantraye, P., Kaijima, M., Prenant, C., Guibert, B., Sastre,
J., Crouzel, M., Naquet, R., Comar, D., Maziere, M.
Neuroscience Letters, 1984, 48, 115-120.
[80] Schacter, D.L. In D.L. Schacter, & E. Tulving (Eds.),
Memory systems (pp. 235-268). Cambridge, MA:
Massachusetts Institute of Technology Press. 1994.
[102] Curran, H.V. Journal of Psychopharmacology, 1992, 6,
69-75.
[81] Legrand, F., Vidailhet, P., Danion, J.M., Grangé, D.,
Giersch, A., Van Der Linden, M., Imbs, J.L.
Psychopharmacology, 1995, 118, 475-479.
[103] Wesenten, N.J., Balkin, T.J., Davis, H.Q., Belenky, G.L.
Psychopharmacology, 1995, 121, 242-249.
[104] Ghoneim, M.M., Mewaldt, S.P., Berie, J.L., Hinrichs, J.V.
Psychopharmacology, 1983, 73, 147-151.
[82] Weingartner, H.J., Rawlings, R., George, D.T., Eckardt,
M. Psychopharmacology, 1998, 138, 311-317.
... Benzodiazepines (BZDs) are a class of medications used as anxiolytics, sedatives, and, most significantly, anticonvulsants (Griffin et al., 2013). They are more recognized for clinical concerns like cognition and neuropsychomotor deficiency and their role in long-term memory impairment (Buffett-Jerrott and Stewart, 2002). Curcumin reduced the levels of oxidative stress in the blood and the hippocampus and decreased the extracellular signalregulated kinase (ERK 1/2)/nuclear transcription factor-NF-κB/ pNF-κB pathway in a diazepam-induced cognition impairment model, according to a recent study (Sevastre-Berghian et al., 2017). ...
Article
Full-text available
Recurrent spontaneous seizures with an extended epileptic discharge are the hallmarks of epilepsy. At present, there are several available anti-epileptic drugs (AEDs) in the market. Still no adequate treatment for epilepsy treatment is available. The main disadvantages of AEDs are their associated adverse effects. It is a challenge to develop new therapies that can reduce seizures by modulating the underlying mechanisms with no adverse effects. In the last decade, the neuromodulatory potential of phytoconstituents has sparked their usage in the treatment of central nervous system disorders. Curcumin is an active polyphenolic component that interacts at cellular and molecular levels. Curcumin’s neuroprotective properties have been discovered in recent preclinical and clinical studies due to its immunomodulatory effects. Curcumin has the propensity to modulate signaling pathways involved in cell survival and manage oxidative stress, apoptosis, and inflammatory mechanisms. Further, curcumin can persuade epigenetic alterations, including histone modifications (acetylation/deacetylation), which are the changes responsible for the altered expression of genes facilitating the process of epileptogenesis. The bioavailability of curcumin in the brain is a concern that needs to be tackled. Therefore, nanonization has emerged as a novel drug delivery system to enhance the pharmacokinetics of curcumin. In the present review, we reviewed curcumin’s modulatory effects on potential biomarkers involved in epileptogenesis including dendritic cells, T cell subsets, cytokines, chemokines, apoptosis mediators, antioxidant mechanisms, and cognition impairment. Also, we have discussed the nanocarrier systems for encapsulating curcumin, offering a promising approach to enhance bioavailability of curcumin.
... These include inadequate knowledge of adverse effects, a perceived lack of skills to manage discontinuation symptoms and the perception that discontinuation is not appropriate in the older adults with a limited life expectancy [4]. In particular, these medications have been associated with impaired learning, memory and executive functioning [5], especially when taken by individuals already suffering from neurocognitive disorders [6]. In addition, they are associated with the risk of addiction, falls with fractures, traffic accidents and all-cause mortality [7]. ...
... Anxious and fearful children experience the pain of a higher amount and extended duration [31] . Hence, to attain this goal, several behavior management methods, both nonpharmacological and pharmacological, have been used [32] . Pharmaceutical anxiolytic medications, such as benzodiazepines, have been correlated to unpleasant sedative and withdrawal effects; the danger of addiction is a further adverse impact [33][34] . ...
... Benzodiazepines (BZDs) were the first drugs used as anxiolytics; however, the long-term use of BZDs was associated with memory impairments, sedation, and dependency -hence the frequency decay of BZD use as the first option of treatment [42]. BZDs act by binding to allosteric sites in the GABA A receptor complexwhich enhances the response to GABA, allowing the influx of chloride, leading to decreased neuronal excitability. ...
Chapter
Anxiety and depressive disorders are two distinct mental health conditions that can significantly impact a person’s well-being and daily functioning. According to the Global Burden of Diseases, in 2020, before pandemics, major depressive disorder (MDD) affected around 193 million people (2471 cases per 100,000 population), while 298 million people (3825 cases per 100,000 population) were affected by anxiety disorder. An increment of 28% in MDD (a total of 246 million cases or 3153 cases per 100,000 population) and 26% (a total of 374 million cases or 4802 per 100,000 population) were reported during the COVID-19 pandemic. Anxiety disorders encompass a range of conditions characterized by excessive and persistent feelings of fear, worry, and unease. Common types of anxiety disorders include generalized anxiety disorder (GAD), panic disorder, social anxiety disorder (SAD), and specific phobias. Individuals with anxiety disorders may experience a variety of symptoms such as restlessness, irritability, difficulty concentrating, muscle tension, and sleep disturbances. The symptoms of depression often coincide with those seen in various psychiatric disorders, but a few symptoms are particularly characteristic of depression, such as anhedonia and increased feelings of guilt. This work aims at presenting novel molecular findings on anxiety and depression disorders using proteomics, but, at first, a neurobiological characterization of these disorders will be presented.
... Behavioral (e.g., cognitive behavioral therapy, acceptance, and commitment therapy) and pharmacological (e.g., selective serotonin reuptake inhibitors, benzodiazepines) treatments are common interventions for anxiety. However, behavioral interventions can be expensive and difficult to access (Andrade et al. 2014;Mojtabai et al. 2011), and available pharmacological interventions are associated with negative side effects such as sedation and weight gain (Alonso-Pedrero et al. 2019;Buffett-Jerrott and Stewart 2002;Lucki et al. 1986). Moreover, a substantial portion of individuals do not respond to existing pharmacological and behavioral interventions (Loerinc et al. 2015;Rutherford et al. 2009). ...
Article
Full-text available
Rationale Evidence suggests cannabidiol (CBD) displays broad therapeutic potential in the context of anxiety; however, no study has examined the effects of CBD on worry, a defining, cognitive feature of anxiety. Additionally, no study has examined the effects of an acute, single dose of CBD compared to repeated CBD administration. Objectives Within a sample of 63 individuals with elevated trait worry, the current study aimed to assess the effects of an empirically-derived high dose of CBD (i.e., 300mg) compared to a commercially-derived dose of CBD (i.e., 50mg) versus placebo on worry severity and anxiety symptoms after an acute dose and after a 2-week administration period. Results Results indicated no effect of acute CBD dosing on worry severity or anxiety symptoms. Repeated CBD administration similarly did not impact worry severity; however, 300mg of CBD reduced anxiety symptoms across the 2-week administration period compared to placebo. Conclusions Taken together, these findings suggest 300mg of oral CBD does not attenuate cognitive symptoms of anxiety (i.e., worry), following both acute and repeated administration. Some evidence for repeated administration of 300mg on physical symptoms of anxiety was obtained. Findings from the current study suggest CBD’s modest anxiolytic effects may be specific to the physical aspects of anxious arousal.
... However, in healthy participants, EEG power enhancement with basmisanil occurs more so in the theta band (6-9 Hz) 83 . By the same token, benzodiazepines (i.e., GABA A positive allosteric modulators) generally have the opposite EEG effects-EEG suppression at delta/theta frequencies [83][84][85][86] (but see an exception for delta here 77 ) and enhancement at beta frequencies 77,83,84,87,88 -accompanied by partial or full sedation 89 . This demonstrates a very different PPD in which depressant and hypnotic effects occur without delta oscillations and are instead marked by high frequency activity more typical of cortical activation. ...
Preprint
Full-text available
Delta (1 – 4 Hz) EEG/MEG activity is generally indicative of loss of consciousness and cortical down states, particularly when it is diffuse and high amplitude. Remarkably, however, drug challenge studies of several diverse classes of pharmacological agents—including antiepileptics, GABA-B-ergics, anticholinergics, and psychedelic tryptamines—demonstrate that participants appear to be neurophysiologically “down” (EEG activity resembling cortical down states) even when they are not psychologically “out” (unconscious). Of those substances that are safe to use in healthy volunteers, some may be highly valuable research tools for investigating which neural activity patterns are sufficient for consciousness or its absence.
Article
Full-text available
Benzodiazepines (BZDs) are drugs commonly used for treating insomnia and anxiety. Although they are known to induce cognitive and psychomotor impairments, their effect on the risk of causing accidents at work remains understudied. The objective of this study is to estimate this risk by differentiating between the recommended use and overuse of these drugs (i.e., uninterrupted use for four months). The data come from the French National Health Data System, which provide a population composed of French people who had at least one work accident (WA) from 2017 to 2019 (approximately 2.5 million people). A linear probability model with two-way fixed effects is used to deal with time-constant heterogeneity and the time effect independent of individuals. The results show a reduction in the risk of WA after a short period of BZD use (one month) compared with no use at all, but the risk of WA increases when treatment exceeds the recommended duration. The intensity of use results in a greater risk of WAs: a 1% increase in BZD use (expressed as the amount reimbursed) leads to a 4.4% (p<0.001) increase in the monthly risk of WAs. Moreover, we see an increase in risk in the month following the treatment discontinuation (+3.6%, p<0.001), which could be due to rebounding and catch-up effects. Health professionals and BZD users should be made aware of the WA risk induced by the use of BZDs, particularly after prolonged use and after discontinuation of treatment. This study provides more evidence for the need to limit the duration of BZD treatment.
Article
PURPOSE Providers treating adults with advanced cancer increasingly seek to engage patients and surrogates in advance care planning (ACP) and end-of-life (EOL) decision making; however, anxiety and depression may interfere with engagement. The intersection of these two key phenomena is examined among patients with metastatic cancer and their surrogates: the need to prepare for and engage in ACP and EOL decision making and the high prevalence of anxiety and depression. METHODS Using a critical review framework, we examine the specific ways that anxiety and depression are likely to affect both ACP and EOL decision making. RESULTS The review indicates that depression is associated with reduced compliance with treatment recommendations, and high anxiety may result in avoidance of difficult discussions involved in ACP and EOL decision making. Depression and anxiety are associated with increased decisional regret in the context of cancer treatment decision making, as well as a preference for passive (not active) decision making in an intensive care unit setting. Anxiety about death in patients with advanced cancer is associated with lower rates of completion of an advance directive or discussion of EOL wishes with the oncologist. Patients with advanced cancer and elevated anxiety report higher discordance between wanted versus received life-sustaining treatments, less trust in their physicians, and less comprehension of the information communicated by their physicians. CONCLUSION Anxiety and depression are commonly elevated among adults with advanced cancer and health care surrogates, and can result in less engagement and satisfaction with ACP, cancer treatment, and EOL decisions. We offer practical strategies and sample scripts for oncology care providers to use to reduce the effects of anxiety and depression in these contexts.
Article
Full-text available
An experiment was conducted to investigate the effect of ageing on two implicit and two explicit memory tasks. Within each memory category (implicit or explicit), the tasks differed in their reliance on perceptual or conceptual processing. Large age-related differences were found on the two explicit memory tasks, regardless of the perceptual-conceptual difference. Age-related differences were much smaller on the implicit tasks; no differences were found on the perceptual version (word-fragment completion) but older subjects showed less priming on the conceptual version (category production). It is suggested that the dissociation between perceptual and conceptual priming reflects selective age-related impairments of different cortical regions.
Article
Full-text available
Reviews empirical evidence to show that 2 concepts—attention and intelligence—are related at the empirical level. Data are presented to demonstrate that intelligence correlates with different types of attention, including sustained attention, search, attention switching, and vigilance. The most significant findings are derived from work on divided attention (dual or competing tasks) in which 2 markers for primary factors were given simultaneously on different channels. Under these conditions, the correlation between the components increases, relative to the single-task condition, implying an increase in the general factor loading. The theoretical implications of this finding are discussed in terms of attentional resources and working memory. These results are considered in relation to theories of the structure of abilities and particularly in relation to the theory of fluid and crystallized intelligence. (67 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Presents a historical overview of the study of visual attention shifts. Contemporary research on this problem is outlined, and models of attention shift mechanisms are briefly described. In addition, several methodological variables are described that warrant consideration when evaluating claims about shifts of visual attention. These factors are cue type, temporal parameters, task and response type, trial presentation order, S expertise, and neutral trials and cost/benefit analyses. (French abstract) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Hypothesized that age deficits in recall are due to a reduction in available processing resource. It is also argued that the formation of a distinct encoding, in which unique aspects of the context are integrated with the target item, requires a substantial amount of attentional resource but that the core semantic features of words are encoded relatively automatically. Thus, under conditions of reduced processing resource, a general, stereotyped encoding is predicted. The effectiveness of general, categorical retrieval cues was compared to the effectiveness of contextually specific retrieval cues in 3 experiments with 84 undergraduates and 60 elderly (approximately 66–69 yrs old) Ss. Young adults recalled more than older adults when they were cued with specific retrieval cues, but no age differences were observed when general retrieval cues were used. A similar pattern of results was obtained when the amount of processing resource was experimentally reduced by requiring young adults to perform a concurrent task during encoding. (22 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Cost–benefit analysis of RTs has become a popular chronometric tool in the study of cognitive processes. The present authors review the technique, assumptions underlying its application, and pitfalls that are encountered in actually implementing it in various experimental contexts. It is suggested that the unthoughtful application of the technique may cause one to draw improper conclusions about the underlying mechanisms that produce costs and benefits. (56 ref) (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Full-text available
Scopolamine (SP) and lorazepam have different pharmacological actions but exert very similar effects on people's performance on a wide range of cognitive and psychomotor tasks. Recent research suggests that their effects may be dissociable on perceptual priming. The present study compared the acute effects of SP (0.3 and 0.6 mg sc), lorazepam (1.0 and 2.0 mg orally), and placebo across the range of E. Tulving's memory "systems" (1985; E. Tulving & D. L. Schacter, 1990) and on indices of arousal and attention. The authors obtained a similar, dose-dependent time course of sedation and attentional impairment following active drugs. The effects of SP and lorazepam were qualitatively indistinguishable on tasks tapping working, episodic, semantic, and procedural memory. However, they had a differential effect on word-stem completion: Whereas SP did not affect this, lorazepam produced a deleterious effect that was not dose dependent. This difference between SP and lorazepam on word-stem completion appears to be over and above the drugs' effects on sedation and attention. The authors' findings are discussed in terms of possible distinct neurobiological systems mediating different memory systems. (PsycINFO Database Record (c) 2012 APA, all rights reserved)
Article
Flumazenil is a competitive benzodiazepine antagonist that rapidly reverses the residual effects of benzodiazepines following intravenous conscious sedation. In a double-blind, multicenter study, postoperative patients who had been sedated with intravenous diazepam were randomly allocated to receive intravenous doses of flumazenil (0.4 mg to 1 mg) or placebo. Levels of sedation and psychomotor impairment were evaluated prestudy, at baseline, and at 6 intervals from 5 to 180 minutes posttreatment. A global evaluation of effectiveness was made at the 5-minute assessment, and memory was assessed at the 180-minute assessment. Flumazenil (mean dose: 0.73 mg [7.3 ml]) was significantly more effective than placebo (mean dose: 8.9 ml) in reversing sedation, psychomotor impairment, and amnesia within 5 minutes after the start of administration. At the 5 minute posttreatment assessment, 84% of 102 flumazenil-treated patients (compared with 42% of 52 placebo-treated patients) experienced complete reversal of sedation. Ninety-two percent of 93 flumazenil-treated patients (compared with 41% of 46 placebo-treated patients) had normal psychomotor function. Reversal of amnesia at the 5-minute assessment was achieved in 75% of 101 flumazenil-treated patients and in 20% of 51 placebo-treated patients. Statistically significant differences between flumazenil and placebo were also observed at the 15-minute assessment. Thereafter there were no significant differences between the two treatment groups. Most (70%) flumazenil-treated patients exhibited no recurrence of sedation during the 180 minute assessment period. The most frequent adverse reaction in the flumazenil group was dizziness (6%). There were no serious adverse experiences related to the test drug. Flumazenil provided prompt, controlled reversal of residual effects, especially sedation, in the majority (84%) of patients recovering from intravenous conscious sedation induced by diazepam. For most (70%) of these flumazenil-treated patients, the reversal was maintained throughout the 180-minute assessment.
Article
Attentional disturbance is a common complaint after TBI in children and adolescents, however, few studies have assessed post-traumatic attentional functions. The attentional constructs proposed by Mirsky, Anthony, Duncan, Ahearn, and Kellam (1991) provided a multidimensional framework for prospective assessment of late attentional disturbance after mild-moderate (n = 34) and severe (n = 57) traumatic brain injury (TBI). Attention was evaluated from 5 to 8 years after TBI in children ages 0 to 15 years at the time of injury. Children with severe TBI performed more poorly than children with mild-moderate TBI on tests comprising the focus/execute and shift constructs. Younger children scored below older children irrespective of injury severity on the Digit Span subtest and interstimulus interval scores from an adaptive rate continuous performance test reflecting the encode and sustain constructs of Mirsky et al. (1991). Age × Severity interaction effects were found for speeded perceptual-motor tests; scores were reduced following mild-moderate and severe TBI in younger children and following severe TBI in older patients. Results are discussed in terms of the vulnerability of skills in a rapid stage of development to disruption by acquired brain injury.
Article
The human amnestic syndrome associated with lesions of the hippocampus and amygdala is characterized by a selective impairment of recent (explicit, episodic) memory. Benzodiazepine (BZ) treated normal subjects demonstrate similar, marked impairments in episodic memory, but in addition, BZ also induces sedation and inattention. Thus, the amnestic effects of BZ may be secondary to drug-induced sedation. However, when subjects were pretreated with the specific BZ receptor antagonist, flumazenil, the sedative and attentional effects of diazepam were blocked, but a marked impairment in episodic memory still occurred. This demonstrates that, using neuropharmacological methods, it is possible to produce a dissociation of memory impairment from inattention and sedation. Such distinct patterns of cognitive dysfunction may serve as models for clinical cognitive syndromes.