human psychopharmacology

Hum Psychopharmacol Clin Exp 2002; 17: 3544.

DOI: 10.1002/hup.352

Acute, dose-dependent cognitive effects of Ginkgo biloba,

Panax ginseng and their combination in healthy young
volunteers: differential interactions with cognitive demand
Andrew B. Scholey* and David O. Kennedy
Human Cognitive Neuroscience Unit, Division of Psychology, University of Northumbria,
Newcastle upon Tyne NE1 8ST, UK

The present paper describes three studies examining the acute effects of single doses of Ginkgo biloba (GK501), Ginseng
(G115) and their combination (Ginkoba M/E, Pharmaton SA) on the performance of healthy young adults (mean age 21
years) during serial arithmetic tasks with differing cognitive load. In each double-blind, placebo-controlled study three different treatment doses and a placebo were administered, according to a balanced crossover design, with a 7-day washout
period between each dose. Participants' scores on two computerised serial subtraction tasks (Serial Threes and Serial
Sevens) were assessed pre-dosing and at 1, 2.5, 4 and 6 h thereafter.
A number of signicant time, dose and task-specic effects were associated with each treatment. There was a dosedependent improvement in speed of responding during Serial Threes following Ginkgo biloba. Different doses of Ginseng
improved accuracy and slowed responses during Serial Sevens. The most striking result, however, was a highly signicant
and sustained increase in the number of Serial Sevens responses following 320 mg of the GinkgoGinseng combination at all
post-treatment testing times. This was accompanied by improved accuracy during Serial Sevens and Serial Threes following
the 640 mg and the 960 mg dose, respectively. The paper concludes with speculation into the possible mechanisms underlying these effects. Copyright # 2002 John Wiley & Sons, Ltd.
key words Ginkgo biloba; Ginseng; cognitive load; demand

INTRODUCTION
Two of the most commonly taken herbal extracts
throughout Europe and the USA are Ginkgo biloba
and Ginseng. While systematic evidence is accumulating regarding psychogenic and other health
benets associated with chronic administration of
these extracts, less is known about their acute
psychopharmacology.
Extracts from the leaves of the Ginkgo biloba tree
have been used in Chinese medical herbalism for millennia (Major, 1967). Chronic administration of
Ginkgo has been shown to ameliorate the cognitive
decline associated with a number of diseases with a

* Correspondence to: Dr A. B. Scholey, Human Cognitive Neuroscience Unit, Division of Psychology, University of Northumbria,
Newcastle upon Tyne NE1 8ST, UK. Tel: 0191 227 4468. Fax:
0191 2273190. E-mail: a.scholey@unn.ac.uk
Copyright # 2002 John Wiley & Sons, Ltd.

vascular component. Examples include effects upon

intermittent claudication (e.g. Pittler and Ernst,
2000), Alzheimer's disease and vascular dementia
(e.g. Le Bars et al., 1997; Kanowski et al., 1996),
and on a number of generalised conditions with a
cerebro-vascular aetiology which have been encompassed within the rather nebulous term `cerebral
insufciency' (Kleijnen and Knipschild, 1992;
Hopfenmuller, 1994). Additionally a large, open label
study into the effects of daily Ginkgo administration
revealed benets in free-living elderly volunteers,
including improvements in measures of quality of life
and sleep parameters (Cockle et al., 2000).
To date there have been few reports of acute
cognitive effects associated with Ginkgo in healthy
volunteers. These have shown rather mixed results,
including a dose-dependent improvement in speed
of performance on the Sternberg numeric working
memory task 1 h following 600 mg of Ginkgo extract
(Hindmarch, 1986). A partial replication revealed
Accepted 2 October 2001

36

a. b. scholey and d. o. kennedy

enhanced performance on the same measure over 2

days following 120 and 300 mg of Ginkgo (Rigney
et al., 1999), with a specic improvement for a
240 mg dose on day 2 only. In a study from our laboratory, Ginkgo administration resulted in dose-specic
improvements in the performance of two of the four
primary cognitive `factors' that can be derived from
the cognitive drug research (CDR) computerised
assessment battery (Kennedy et al., 2000). In the latter
study, the effects of three single doses of Ginkgo
(120, 240, 360 mg) were compared with placebo.
Specic dose-dependent improvements were demonstrated through three of four testing sessions spanning
1 to 6 h post-treatment on a `speed of attention' factor,
comprising reaction time scores from three attentional
tasks. There was also a dose and time-specic
improvement in performance restricted to two sessions for the lowest dose (120 mg) of Ginkgo on a
`quality of memory' factor (incorporating accuracy
data from six memory tasks).
Whilst the specic mechanisms underlying
Ginkgo's efcacy are unknown, they may be related
to the extract's actions as a Platelet Activating
Factor antagonist (Braquet and Hosford, 1991; Engels
and Wirth, 1997), as a free radical scavenger
(Droy-LeFaix, 1997), and as a modulator of cellular
metabolism (e.g. Oberpichler et al., 1988). Additionally Ginkgo is capable of in vitro interaction with a
number of neurotransmitter systems (e.g. White
et al., 1996; Ramassamy et al., 1992). These effects
have also been implicated in the demonstration in
young volunteers of a reversal of cognitive decits
during hypoxia (Schafer and Reeh, 1985) and the
amelioration of acute mountain sickness (Roncin
et al., 1996). Such mechanisms may also underlie
improvements in cerebral blood ow in rodents
(Oberpichler et al., 1988), as well as a number of vascular and haematological parameters in humans
(Koltringer et al., 1993; Jung et al., 1990; Roncin
et al., 1996).
Ginseng is the root of the Araliaceous plant Panax
ginseng, and has been used in oriental medicine for
millennia as a tonic, a restorative and a prophylactic
agent (Bahrke and Morgan, 1994, 2000). The literature regarding Ginseng is often difcult to interpret,
due partly to the lack of available data regarding the
use of standardised extracts in many studiessee
Bahrke and Morgan (2000) for a comprehensive
review. The following section is restricted to consideration of relevant studies where extracts, or components thereof, have been standardised.
The animal literature does suggest a number of benecial effects from Ginseng, including the attenuation
Copyright # 2002 John Wiley & Sons, Ltd.

of experimentally induced stress (Nguyen et al., 1995;

Kim et al., 1998) and fatigue (Filaretov et al., 1988;
Wang and Lee, 1998), and improvements in learning
in both young and old rodents (Petkov and Mosharrof,
1987; Petkov et al., 1993).
The effects of Ginseng in humans are less clear.
Evidence of ergogenic benets is at best equivocal
(Allen et al., 1998; Bahrke and Morgan, 1994, 2000;
Engels and Wirth, 1997). Although a number of studies involving chronic administration of Ginseng alone
or in combination with vitamins have demonstrated
improvements in subjective ratings of `well being' or
`quality of life' (Neri et al., 1995; Wiklund et al.,
1994; Marasco et al., 1996; Sotaniemi et al., 1995;
Hallstrom et al., 1982). Where studies have focused
on cognitive performance, improvements have been
found, following an extended chronic Ginseng regime,
on a mental arithmetic test (D'Angelo et al., 1986),
and a computerised version of the Wisconsin card sort
test (Soerensen and Sonne, 1996).
To the best of our knowledge only one study has
examined the acute cognitive effects of Ginseng
(Kennedy et al., 2001). Employing the same methodology as the Ginkgo study outlined above (Kennedy
et al., 2000), the effects of each of three single doses
of Ginseng (200, 400 and 600 mg) were compared
with placebo. We found dose-dependent improvements in performance on the `secondary memory'
factor across all post-dose time points, with this
effect restricted to the middle dose under investigation (400 mg). There was, however, also a dose
and time dependent slowing of performance on the
`speed of attention' factor for the lowest (200 mg)
and highest (600 mg) doses of Ginseng at the
later two post-dose testing sessions (4 and 6 hs).
This decrement was accompanied by a signicant
reduction in subjective ratings of alertness as
assessed by the Bond-Lader visual analogue scales
(Bond and Lader, 1974) for the 200 mg and 400 mg
these doses at the 6 h time point.
This demonstration of both cognitive costs and
benets associated with Ginseng reects the literature
on putative mechanisms, where evidence from in vitro
and animal studies suggests a plethora of physiological
consequences. Ginseng shares with Ginkgo an antagonistic inuence on in vitro platelet aggregation
(Shi et al., 1990; Jung et al., 1998). On the other hand,
Ginseng has also been shown to have opposing
effects on a number of parameters including blood
pressure (Wood et al., 1964), vasoconstriction (Lee
et al., 1981; Lei and Chiou, 1986) and measures of
hypothalamicpituitaryadrenal axis activity (Luo
et al., 1993).
Hum Psychopharmacol Clin Exp 2002; 17: 3544.

acute effects of ginkgo biloba and panax ginseng

Regarding GinkgoGinseng in combination, previous research has shown that chronic administration
of the standardised commercial 60 : 100 Ginkgo:
Ginseng extract (Ginkoba M/E, Pharmaton Natural
Health Products, Switzerland), can improve memory
performance in rodents (Petkov et al., 1993), in sufferers from neurasthenia (Wesnes et al., 1997), and
in healthy middle aged humans (Wesnes et al.,
2000). Data from our own laboratory suggest that
the cognitive effects of single doses of the Ginkgo
Ginseng combination are not unlike those observed
with Ginseng. Specically the `secondary memory'
factor was improved by 960 mg of the combination,
while `speed of attention' was slowed by the 320 mg
dose (Kennedy et al., 2001b).
These results regarding the acute effects of Ginkgo
and Ginseng provide valuable insights into the cognitive domains targeted by these extracts. However, the
tasks utilised would not be described as being heavily
cognitively loaded. One focus of work in our laboratory relates to the relationship between putative cognitive enhancers and tasks with a relatively high
cognitive demand. In particular we have been developing a simple model which suggests that a reciprocal relationship exists between the delivery and use
of blood-borne metabolic substrates (glucose and
oxygen) and the impact of these substrates on cognitive performance (Moss and Scholey, 1996; Scholey
et al., 1999; Kennedy and Scholey, 2000; Scholey,
2001; Scholey et al., 2001). Task performance appears
to be `fuel-limited' under conditions of cognitive
demand, and may be facilitated by mechanisms which
serve to aid the neural delivery of glucose and oxygen.
While this model may not hold true for all situations,
it does appear to have potential heuristic value. Thus
administration of oxygen or glucose improves cognitive performance and, particularly in the case of
glucose, this effect appears to be more marked
under conditions of cognitive demand (Kennedy and
Scholey, 2000; Scholey et al., 2001). Moreover, cognitive demand itself can lead to measurable reductions
in blood levels of oxygen (Scholey et al., 1999) and
glucose (Scholey, 2001; Scholey et al., 2001).
Typically these investigations have used arithmetic
tasks where cognitive demand can be systematically
titrated, for example using serial subtractionswhere
a particular number (usually three or seven) is
subtracted from a starting number, then from the
resulting number and so on. We have previously
reported that, compared with Serial Threes (repeated
subtraction of three), Serial Sevens performance was
rated as more demanding, engendered a bigger change
in heart rate, was more susceptible to the enhancing
Copyright # 2002 John Wiley & Sons, Ltd.

37

effect of glucose and was associated with a greater fall

in blood glucose (Kennedy and Scholey, 2000). A version of the computerised Serial Sevens task used in the
present study was shown to signicantly reduce blood
glucose levels compared with somatically matched
key-pressing control (Scholey et al., 2001).
Thus it appears that more heavily loaded cognitive
tasks draw upon blood-borne glucose (and oxygen),
and when levels of these substrates are high, facilitation of performance occurs. Ginkgo and, possibly to a
lesser extent, Ginseng affect physiological parameters
which may inuence the delivery of glucose and oxygen (e.g. blood viscosity, heart rate, blood oxygenation, hypothalamicpituitaryadrenal axis activity).
It therefore seems reasonable to investigate the possibility that these extracts alone and in combination
may have differential inuences on tasks of different
cognitive load.
In the current study we report the results from three
separate randomised, double blind, counterbalanced,
placebo-controlled experiments investigating the
effects of three doses of Ginkgo, three doses of
Ginseng and three doses of their combination on
performance of novel computerised versions of the
Serial Threes and Serial Sevens tasks.
MATERIALS AND METHODS
These experiments formed part of a comprehensive
assessment of the acute cognitive effects of Ginkgo
and Ginseng. All three studies utilised similar
placebo-controlled, double blind methodology. All
used multi-dose, multiple-testing time regimes with
a 7 day washout between doses.
Participants
Prior to participation in each study participants signed
an informed consent form and completed a medical
health questionnaire which had been approved by
the Joint Ethics Committee of Newcastle and North
Tyneside Health Authority. All participants reported
that they were in good health, and were taking no
medication with the exception, for some female
volunteers, of the contraceptive pill. Heavy smokers
(>10 cigarettes/day) were excluded from the studies.
All participants abstained from caffeine containing
products and alcohol throughout each study day. Eighteen female and two male undergraduate volunteers
(mean age 19.9 years, SD 1.47) took part in study 1
(Ginkgo). In study 2 (Ginseng) there were 14 female
and six male undergraduate volunteers (mean age 21.3
years, SD 2.64). In study 3 which examined the effects
Hum Psychopharmacol Clin Exp 2002; 17: 3544.

38

a. b. scholey and d. o. kennedy

of a GinkgoGinseng combination, there were 10

male and 10 females (mean age 20.6 years, SD 4.20).
Treatments
Study 1Ginkgo biloba. Treatments comprised six
capsules in which placebos and capsules containing
60 mg Ginkgo biloba extract, standardised to a content
of 24% Ginkgo avone glycosides and 6% terpene
lactones (GK501, Pharmaton SA), were combined to
produce a placebo and doses corresponding to 120,
240 and 360 mg of the extract.
Study 2Ginseng. On each study day participants
received six similar capsules, each containing either
100 mg of Ginseng extract (G115, Pharmaton SA),
standardised to contain 4% triterpenoid glycosides, or
an inert placebo. Depending on the condition to which
they were allocated on that particular day the
combination corresponded to a dose of either 0
(placebo), 200, 400 or 600 mg of Ginseng extract.
Study 3GinkgoGinseng combination. Treatments
were in the form of six capsules of identical appearance, each contained either placebo or a combination
(Ginkoba M/E, Pharmaton SA) comprising 60 mg of
the standardised Ginkgo biloba extract and 100 mg of
the standardised Ginseng extract. Depending on the
condition to which they were allocated on that
particular day the treatment corresponded to a dose
of either 0 (placebo), 320, 640 or 960 mg of the
combined extracts.
Cognitive measures
A modied computerised version of the Serial Sevens
test was utilised. The original verbal Serial Sevens test
(Hayman, 1942) has appeared in a number of forms,
including as part of the mini-mental state examination
(Folstein et al., 1975). It has been used to assess cognitive impairment during hypoglycaemia (e.g. Hale
et al., 1982; Taylor and Rachman, 1987), and has also
been used to investigate the relationship between
increased blood glucose levels and cognitive performance (Kennedy and Scholey, 2000; Scholey et al.,
2001; Scholey, 2001).
In the current studies computerised versions of
Serial Subtractions were implemented (see Scholey
et al., 2001 for details), here using tests of 2 min duration. For the Serial Sevens task a standard instruction
screen informed the participant to count backwards in
sevens from the given number, as quickly and accurately as possible, using the numeric keypad to enter
Copyright # 2002 John Wiley & Sons, Ltd.

each response. Participants were also instructed verbally that if they were to make a mistake they should
carry on subtracting from the new incorrect number.
A random starting number between 800 and 999
was presented on the computer screen, which was
cleared by the entry of the rst response. Each
three-digit response was entered via the numeric keypad with each digit being represented on screen by an
asterisk. Pressing the enter key signalled the end of
each response and cleared the three asterisks from
the screen. The task was scored for total number of
subtraction and number of errors. In the case of incorrect responses, subsequent responses were scored as
positive if they were correct in relation to the new
number.
The Serial Threes task was identical to Serial
Sevens, except that it involved serial subtraction of
threes.
Procedure
Each participant was required to attend a total of 5
study days that were conducted 7 days apart, to ensure
an appropriate wash-out between conditions. Testing
took place in a suite of laboratories with participants
visually isolated from each other.
On arrival at their rst session participants were
randomly allocated to a treatment regime using a
Latin square which counterbalanced the order of treatments across the four active days of the study. The rst
day of the study was identical to the following four,
but with no treatment (active or placebo). This
allowed familiarisation with the tasks and procedure
and controlled for practice effects. Data from the rst
day were not included in the analysis.
Each study day comprised ve identical testing sessions. The rst was a pre-dose testing session which
established baseline performance for that day, and
was immediately followed by the day's treatment.
Further testing sessions took place 1, 2.5, 4 and 6 h
following administration of the day's treatment. Each
testing session included the completion of both 2 min
computerised subtraction tasksSerial Threes followed by Serial Sevens.
Statistics
For each treatment the `total subtractions' and `number of errors' scores from the four post-dose testing
sessions were transformed into `change from baseline' scores. The resulting data were analysed using
two-way repeated measures ANOVAs (dose  time).
Planned comparisons were then made between the
Hum Psychopharmacol Clin Exp 2002; 17: 3544.

acute effects of ginkgo biloba and panax ginseng

placebo and the three active treatment conditions at
each time point utilising t-tests (incorporating mean
squares for dose  time  subjects as an error term).
To ensure the overall protection level, only probabilities associated with pre-planned comparisons were
considered.
RESULTS
Study 1Ginkgo biloba
There were number of signicant time and dosespecic changes following each active dose of Ginkgo
(Figure 1). Planned comparisons of the change from
baseline data revealed that ingestion of all three doses
of Ginkgo resulted in a signicant increase in the
number of subtractions during Serial Threes in comparison with placebo at the 4 h testing session. This
effect was evident following 120 mg [t(171) 2.21,
p < 0.05], 240 mg [t(171) 4.5, p < 0.001] and
360 mg [t(171) 2.8, p < 0.01], with a single increase
at the 6 h testing session for the 240 mg dose
[t(171) 2.37, p < 0.05]. Signicantly more errors
were made following ingestion of 120 mg of Ginkgo
at the 4 h testing session [t(171) 2.8, p < 0.01].
For Serial Sevens, whilst there was no signicant
reduction in the total number of subtractions for any
of the doses of Ginkgo, there was a signicant
improvement in the number of errors in comparison

39

with the placebo for all doses at the 2.5 h time point
following 120 mg [t(171) 2.16, p < 0.05], 240 mg
[t(171) 1.98, p < 0.05] and 360 mg [t(171) 1.98,
p < 0.05].
Study 2Ginseng
There were a number of signicant changes following
Ginseng administration which were restricted to performance of the Serial Sevens task (Figure 2). Planned
comparisons of the change from baseline data from
the Serial Threes sessions revealed that there were
no signicant differences in comparison with the placebo in either the total number of subtractions or the
number of errors for any of the doses of Ginseng. In
contrast, analysis of Serial Sevens performance
revealed a signicant decrement in performance for
the 200 mg dose of Ginseng with participants making
fewer subtractions compared with placebo during the
1 h session [t(171) 2.07, p < 0.05], 2.5 h session
[t(171) 2.01, p < 0.05] and the 6 h session
[t(171) 2.59, p < 0.05]. However, there was a significant improvement in accuracy following the 400 mg
dose of Ginseng with a reduction in errors, in comparison with placebo, at the 4 h session [t(171) 2.46,
p < 0.05], and the 6 h session [t(171) 2.12,
p < 0.05], with a similar improvement for the
200 mg dose at the 4 h time point [t(171) 2.01,
p < 0.05].

Figure 1. Effects of Ginkgo biloba on Serial Subtractions performance. Graphs depict mean change from baseline scores 1, 2.5, 4 and 6 h
post-treatment. Effect on number of responses (top) and errors (bottom) are shown for both Serial Threes (left) and Serial Sevens (right).
Scores following Ginkgo biloba extract GK501 administered at doses of 120 mg (&), 240 mg (~) and 360 mg (!) were compared with
placebo (*). *p < 0.05; **p < 0.01; ****p < 0.001 compared with corresponding placebo score

Copyright # 2002 John Wiley & Sons, Ltd.

Hum Psychopharmacol Clin Exp 2002; 17: 3544.

40

a. b. scholey and d. o. kennedy

Figure 2. Effects of Ginseng on Serial Subtractions. Graphs depict mean change from baseline scores 1, 2.5, 4 and 6 h post-treatment.
Effect on number of responses (top) and errors (bottom) are shown for both Serial Threes (left) and Serial Sevens (right). Scores following
Ginseng extract G115 administered at doses of 200 mg (&), 400 mg (~) and 600 mg (!) were compared with placebo (*). *p < 0.05
compared with corresponding placebo score

Study 3GinkgoGinseng combination

There were a number of signicant and sustained
improvements in performance following the Ginkgo
Ginseng combination (Figure 3). Planned comparisons of the change from baseline data revealed
that participants made more subtractions in com-

parison with the placebo during the during the Serial

Threes task at the 4 h testing session following ingestion of 320 mg of the GinkgoGinseng combination
[t(171) 2.29, p < 0.05].
Participants also performed more accurately on
Serial Threes. Signicantly fewer errors were made

Figure 3. Effects of a GinkgoGinseng combination on Serial Subtractions. Graphs depict mean change from baseline scores 1, 2.5, 4 and
6 h post-treatment. Effect on number of responses (top) and errors (bottom) are shown for both Serial Threes (left) and Serial Sevens (right).
Scores following a 60 : 100 Ginkgo:Ginseng preparation administered at doses of 320 mg (&), 640 mg (~) and 960 mg (!) were compared
with placebo (*). p < 0.05; **p < 0.01; ****p < 0.001 compared with corresponding placebo score

Copyright # 2002 John Wiley & Sons, Ltd.

Hum Psychopharmacol Clin Exp 2002; 17: 3544.

acute effects of ginkgo biloba and panax ginseng

during the 2.5 h session for the 640 mg dose
[t(171) 2.02, p < 0.05]. Additionally error scores
were signicantly reduced at all time points following
the 960 mg dose of the combination. This effect was
evident by 1 h post-treatment [t(171) 3.86,
p < 0.001], and maintained at 2.5 h [t(171) 2.41,
p < 0.05], 4 h [t(171) 2.32, p < 0.05] and 6 h
[t(171) 2.61, p < 0.01].
Moving to Serial Sevens, analysis of the change
from baseline data revealed that participants generated signicantly more subtractions in comparison
with the placebo at all time points following the
320 mg dose of the combination. Again this effect
was evident by 1 h post-treatment [t(171) 3.42,
p < 0.001], and was sustained over the sessions taking
place at 2.5 h [t(171) 3.36, p < 0.001], 4 h
[t(171) 3.17, p < 0.01] and 6 h [t(171) 2.91,
p < 0.01]). Participants also made more subtractions
at the 4 h session following ingestion of 640 mg of
the combination [t(171) 2.91, p < 0.01].
There were also marked improvements in accuracy
for all doses of the combination at both the
2.5 h (320 mg [t(171) 2.22, p < 0.05], 640 mg
[t(171) 3.46, p < 0.001], 960 mg [t(171) 3.43,
p < 0.001]) and the 6 h session (320 mg [t(171)
2.02, p < 0.05], 640 mg [t(171) 3.36, p < 0.001],
960 mg [t(171) 2.42, p < 0.05]), with a single significant reduction at the 4 to 4.5 h session following the
640 mg dose [t(171) 2.75, p < 0.01].
DISCUSSION
Each of the three treatments under investigation signicantly affected performance on computerised
Serial Subtractions in a dose, time and task specic
manner. The effects of single doses of Ginkgo and
Ginseng were reasonably consistent with previous
ndings. The most striking (and unexpected) result,
however, was a marked and sustained improvement
in Serial Sevens performance following the Ginkgo
Ginseng combination.
The time and dose dependent increases in the total
number of subtractions performed on the Serial
Threes task following single doses of Ginkgo is
broadly in line with previous ndings. Using the more
comprehensive CDR test battery, faster `speed of
attention' was associated with a 360 mg dose of
Ginkgo (Kennedy et al., 2000). Any intervention
which produces faster reaction times on attentional
measures might reasonably be expected to also result
in more responses during a relatively underloaded task
such as Serial Threes. On the other hand it was the
240 mg, rather than the 360 mg, dose that was the
Copyright # 2002 John Wiley & Sons, Ltd.

41

most effective here. Although all three active doses

resulted in the generation of more Serial Three
responses at the 4 h time point, inspection of the data
suggests that this result may be attributable to poor
performance at that time point by the placebo group
(Figure 1).
Ginseng had no effect on Serial Threes, but was
associated with a signicant dose specic (200 mg)
decrement in the total number of subtractions on the
Serial Sevens task (which was evident at all posttreatment assessments except the 4 h session). There
was also a reduction in errors for the same dose, which
reached signicance at the 4 h session. This pattern
raises the possibility of a time-specic `speed/accuracy trade-off' for the 200 mg dose alone, with greater
accuracy associated with slower performance. Interestingly the 400 mg dose produced a specic benecial effect, evincing a signicant reduction in errors
at both the 4 and 6 h testing sessions on Serial Sevens,
with a total subtraction performance that remained
virtually indistinguishable from placebo (Figure 2).
Again this pattern is not unlike that gleaned using
other measures (Kennedy et al., 2001a) where
400 mg Ginseng was associated with an improved
`quality of memory' while the 200 mg and 600 mg
doses resulted in a slowing of attentional measures
coupled with reduced self-rated alertness at later time
points.
While these results do roughly mirror earlier ndings (Kennedy et al., 2000, 2001a), it should be
emphasised that we previously found no effect on a
`working memory' factor (comprising accuracy
scores from a numeric working memory and a spatial
working memory task). Serial subtractions combines
features of both attention and working memory, as
well as a procedural learning component for Serial
Threes (see later). Additionally, unlike the CDR battery tasks used previously, the present tasks were
deliberately designed to manipulate cognitive
demand. Therefore as well as similarities between
the present study and those which have used the
CDR factors, there are also important differences.
Moving to the GinkgoGinseng combination, performance on the less demanding Serial Threes task
was improved by all three doses utilised (Figure 3).
For the lowest dose (320 mg) this improvement took
the form of a signicant increase in the number of subtractions during the 4 h testing session. Additionally,
signicantly improved accuracy was observed at
2.5 h only for the 640 mg dose, and at all post-dose
testing sessions for the 960 mg dose. A similar, but
stronger, pattern of results was evinced for the more
demanding Serial Sevens task. In this case the
Hum Psychopharmacol Clin Exp 2002; 17: 3544.

42

a. b. scholey and d. o. kennedy

320 mg dose was associated with signicantly faster

speed of performance sustained across all time points,
with a similar (though largely not signicant) pattern
of improvements for the 640 mg dose. Remarkably, all
three doses were also associated with signicantly
improved accuracy of performance, with this effect
being most consistent for the 640 mg dose. Thus the
benecial effects on number of responses (320 mg)
and error reduction (640 mg) were not associated with
cognitive costs in terms of accuracy and speed respectively.
It has been proposed that the performance of cognitively demanding tasks may be `fuel limited', to some
degree. That is during periods of intense processing,
performance may be impaired or enhanced respectively by low or high levels of blood-borne glucose
and oxygen (Kennedy and Scholey, 2000; Scholey,
2001; Scholey et al., 2001).
To what extent might the current ndings be accommodated within such a model? The mechanisms
underlying these effects are not known, although they
may be related to the extracts' effects on various physiological as well as neurotransmitter systems. In the
case of Ginkgo, there is evidence of benecial modulatory effects on several vascular and haematological
parameters (Koltringer et al., 1993; Jung et al., 1990;
Roncin et al., 1996), as well as cerebral glucose consumption (Rapin et al., 1986). Such processes may
contribute to facilitated performance on serial subtraction tasks which appear to be sensitive to increases in
the availability and/or delivery of glucose and oxygen
(Kennedy and Scholey, 2000; Scholey et al., 2001).
Ginseng has been shown to reduce blood glucose
levels in both Type II diabetes (Sotaniemi et al.,
1995) and in non-pathological populations (Vuksan
et al., 2000), although it apparently exerts negligible
or even opposite effects on other, related parameters
(Wood et al., 1964; Lee et al., 1981; Lei and Chiou,
1986).
The above effects may go some way to explaining
why Serial Sevens performance was largely unaffected by Ginkgo, and to why Ginseng engendered
slower performance of Serial Sevens. On the rst of
these points it may be that the implementation of a
computerised version of the two tasks has subtly
changed their nature. In practice, compared with the
verbal version, Serial Sevens appears largely
unchanged, with an inherent cognitive demand due
to the relatively high working memory and attentional
load. On the other hand, compared with the verbal
form, computerised Serial Threes performance
benets from the standard `three-by-three' layout of
the computer keyboard numeric keypad. As the task
Copyright # 2002 John Wiley & Sons, Ltd.

progresses, a distinct spatial pattern emerges, with

the nal digit of each response being made on adjacent keys on the number pad columns. In essence,
Serial Threes could best be described as an attentional
task with a procedural learning element. As such the
task requires concentration rather than the central
executive resources which may be drawn upon during
Serial Sevens. This would further support the contention that these results are broadly consistent with an
improved `speed of attention' following Ginkgo
(Kennedy et al., 2000).
Similarly, Kennedy et al. (2001a) demonstrated that
single doses of Ginseng were associated with a significant slowing of performance on the same attentional
factor (in addition to improved memory performance).
This effect was accompanied by a decrease in selfrated alertness at the later sessions following the same
200 mg dose that slowed Serial Sevens performance in
the current study. It is not clear why Ginseng apparently targets the more demanding task. Given the
plethora of putative physiological effects attributable
to Ginseng administration, the mechanism, or more
probably mechanisms, underlying these effects
remain unknown. However, it seems unlikely that
the opposite effects demonstrated both in the current
Ginkgo and Ginseng experiments and on the `speed
of attention' factor in the previous studies (Kennedy
et al., 2000, 2001a) are due to opposing inuences
on the same neural mechanism.
As well as the vascular parameters outlined previously, administration of Ginkgo is known to
modulate a number of neurotransmitter systems
(White et al., 1996; Ramassamy et al., 1992) including the cholinergic system. Acetylcholine is critically
involved in the modulation of attention (Rusted
and Warburton, 1989) and acetylcholine synthesis
may itself be inuenced by availability of glucose
(e.g. Wenk, 1989; Messier and Gagnon, 1996;
Kennedy and Scholey, 2000) and oxygen (Moss and
Scholey, 1996; Moss et al., 1998; Scholey et al.,
1998, 1999). Thus the differential effects of Ginkgo
and the GinkgoGinseng combination may be mediated either via cholinergic mechanisms (Serial Threes),
with associated affects on attentional capacity, or
through the delivery of metabolic substrates particularly during high local neuronal demand (Serial
Sevens).
Clearly these putative mechanisms are merely
conjecture at present. However, such a model may
also account for the striking pattern of results following administration of the combination product. In this
instance total subtraction performance on the Serial
Threes task was similar to that following Ginkgo
Hum Psychopharmacol Clin Exp 2002; 17: 3544.

acute effects of ginkgo biloba and panax ginseng

(compare Figures 1 and 3). In contrast, on the more
demanding Serial Sevens task the 320 mg dose was
associated with a marked and substantial increase in
subtractions across all time points. It remains a possibility that the Serial Threes task was improved within
the bounds of increased acetylcholine synthesis,
whilst only the more demanding Serial Sevens task
engendered enough cognitive demand to take advantage of an increased supply of metabolic substrates
(see Scholey et al., 2001).
It is clear that this effect of the GinkgoGinseng
combination could not be predicted by simple extrapolation of the results from the two extracts in isolation, either on the tasks used here or previously
(Kennedy et al., 2000, 2001a). Additionally, recent
ndings suggest that the cognitive effects of the
GinkgoGinseng combination on the CDR factors
described earlier, are similar to those of Ginseng alone
(Kennedy et al., 2001b), whereas here on Serial Subtraction tasks the combination gave very different
results to Ginseng. Interestingly, medical herbalism
emphasises synergy between components of plant
extracts. It would appear that the comprehensive
improvements in performance associated with the
GinkgoGinseng combination represent a synergistic
behavioural effect of the two extracts interacting with
cognitive demand.
The possibility that these results are the consequence of increased cerebral blood ow and metabolism coupled with neurotransmitter effects is
necessarily highly speculative. Nevertheless it is clear
that a number of neurotransmitter systems, metabolic
processes and other physiological responses are
affected by both Ginkgo and Ginseng. It seems possible that the cognitively loaded Serial Sevens task may
have beneted from a serendipitous combination of
these factors interacting with task demands. The extent
to which this nding may generalise to other tasks and
other populations merits further investigation.
REFERENCES
Allen JD, McLung J, Nelson AG, Welsch M. 1998. Ginseng
supplementation does not enhance healthy young adults'
peak aerobic exercise performance. J Am College Nutr 17:
462466.
Bahrke MS, Morgan WP. 1994. Evaluation of the ergogenic properties of Ginseng. Sports Med 18: 229248.
Bahrke MS, Morgan WP. 2000. Evaluation of the ergogenic properties of Ginseng: an update. Sports Med 29: 113133.
Bond A, Lader M. 1974. The use of analogue scales in rating subjective feelings. Br J Psychol 47: 211218.
Braquet P, Hosford D. 1991. Ethnopharmacology and the development of natural PAF antagonists as therapeutic agents. J Ethnopharmacol 32: 135139.

Copyright # 2002 John Wiley & Sons, Ltd.

43

Cockle SM, Kimber S, Hindmarch I. 2000. The effects of Ginkgo

biloba extract (LI 1370) supplementation on activities of daily
living in free living older volunteers: a questionnaire survey.
Hum Psychopharmacol Clin Exp 15: 227235.
D'Angelo L, Grimaldi R, Caravaggi M et al. 1986. A double-blind,
placebo-controlled clinical study on the effect of a standardized
Ginseng extract on psychomotor performance in healthy volunteers. J Ethnopharmacol 16: 1522.
Droy-Lefaix MT. 1997. Effect of the antioxidant action of Ginkgo
biloba extract (Egb 761) on aging and oxidative stress. Age 20:
141149.
Engels HJ, Wirth JC. 1997. No ergogenic effects of Ginseng (Panax
ginseng C.A. Meyer) during graded maximal aerobic exercise.
J Am Dietetic Assoc 97: 11101115.
Filaretov AA, Bogdanova TS, Podvigina TT, Bodganov AI. 1988.
Role of pituitaryadrenocortical system in body adaptation abilities. Exp Clin Endocrinol 92: 129136.
Folstein M, Folstein SE, McHugh PR. 1975. `Mini Mental State':
a practical method of grading the cognitive state of patients for
the clinician. J Psychiatr Resources 12: 189.
Hale F, Margen S, Rabak D. 1982. Postprandial hypoglycaemia and
psychological symptoms. Biol Psychiatry 17: 125130.
Hallstrom C, Fulder S, Carruthers M. 1982. Effects of Ginseng on
the performance of nurses on night duty. Comp Med East West 6:
277282.
Hayman M. 1942. Two minute clinical test for measurement of
intellectual impairment in psychiatric disorders. Arch Neurol
Psychiatry 47: 454464.
Hindmarch I. 1986. Activity of Ginkgo biloba extract on short-term
memory. Presse Med 15: 15921594.
Hopfenmuller W. 1994. Evidence for a therapeutic effect of Ginkgo
biloba special extract. Meta-analysis of 11 clinical studies in
patients with cerebro-vascular insufciency in old age. Arzneimittelforschung 44: 10051013
Jung F, Morowietz C, Kiesewetter H, Wenzel E. 1990. Effect of
Ginkgo biloba on uidity of blood and peripheral microcirculation in volunteers. Arzneimittelforschung 40: 589593.
Jung KY, Kim DS, Oh SR et al. 1998. Platelet activating factor
antagonist activity of ginsenosides. Biol Pharmaceut Bull 21:
7980.
Kanowski S, Herrmann WM, Stephan K, Wierich W, Horr R. 1996.
Proof of efcacy of the Ginkgo biloba special extract EGb 761 in
outpatients suffering from mild to moderate primary degenerative dementia of the Alzheimer type or multi-infarct dementia.
Pharmacopsychiatry 29: 4756.
Kennedy DO, Scholey AB. 2000. Glucose administration, heart rate
and cognitive performance: effects of increasing mental effort.
Psychopharmacology 149: 6371.
Kennedy DO, Scholey AB, Wesnes AB. 2001a. Differential, dose
dependent changes in cognitive performance and mood following
acute administration of Ginseng to healthy young volunteers.
Nutr Neurosci 4: 295310.
Kennedy DO, Scholey AB, Wesnes KA. 2001b. Differential, dose
dependent changes in cognitive performance following acute
administration of a Ginkgo biloba/Panax ginseng combination
to healthy young volunteers. Nutr Neurosci 4: 399412.
Kennedy DO, Scholey AB, Wesnes KA. 2000. The dose dependent
cognitive effects of acute administration of Ginkgo biloba to
healthy young volunteers. Psychopharmacology 151: 416423.
Kim DH, Jung J, Suh H, et al. 1998. Inhibition of stress-induced
plasma corticosterone levels by ginsenosides in mice: involvement of nitric oxide. Neuroreport 9: 22612264.
Kleijnen J, Knipschild P. 1992. Ginkgo biloba for cerebral insufciency. Br J Clin Pharmacol 34: 352358.

Hum Psychopharmacol Clin Exp 2002; 17: 3544.

44

a. b. scholey and d. o. kennedy

Koltringer P, Langsteger W, Klima G, Reisecker F, Eber O. 1993.

Hemorheologic effects of Ginkgo biloba extract EGb 761.
Dose-dependent effect of EGb 761 on microcirculation and viscoelasticity of blood. Fortschr Med 10: 170172.
Le Bars PL, Katz MM, Berman N, Itil TM, Freedman AM,
Schatzberg AF. 1997. A placebo-controlled, double-blind,
randomized trial of an extract of Ginkgo biloba for dementia.
North American EGb Study Group. JAMA 278: 13271332.
Lee MO, Kim CY, Clifford DH. 1981. Effect of ether, ethanol and
aqueous extracts of Ginseng on cardiovascular function in dogs.
Canad J Com Med 45: 182187.
Lei XL, Chiou GC. 1986. Cardiovascular pharmacology of Panax
notoginseng (Burk) F. H. Chen and Salvia miltiorrhiza. Am J
Chin Med 14: 145152.
Luo Y, Cheng XJ, Yuan WX. 1993. Effects of ginseng root saponins
and ginsenoside Rbl on immunity in cold water swim stress mice
and rats. Chung Kuo Yao Li Hsueh Pao 14: 401404.
Major RT. 1967. The ginkgo, the most ancient living tree. Science
15: 12701273.
Marasco A, Vargas Ruiz R, Salas Villagomez A, Begona Infante C.
1996. Double-blind study of a multivitamin complex supplemented with Ginseng extract. Drugs Under Exp Clin Res 22:
323329.
Messier C, Gagnon M. 1996. Glucose regulation and cognitive functions: relation to Alzheimer's disease and diabetes. Behav Brain
Res 75: 111.
Moss MC, Scholey AB. 1996. Oxygen administration enhances
memory formation in healthy young adults. Psychopharmacology 124: 255260.
Moss MC, Scholey AB, Wesnes KA. 1998. Oxygen administration
selectively enhances cognitive performance in healthy young
adults: a placebo-controlled double-blind crossover study.
Psychopharmacology 138: 2733.
Neri M, Andermarcher E, Pradelli JM, Salvioli G. 1995. Inuence
of a double blind pharmacological trial on two domains of well
being in subjects with age associated memory impairment. Arch
Gerontol Geriat 21: 241252.
Nguyen TT, Matsumoto K, Yamasaki K, Nguyen MD, Nguyen TN,
Watanabe H. 1995. Crude saponin extracted from Vietnamese Ginseng and its major constituent majonoside-R2
attenuate the psychological stress- and foot-shock stress-induced
antinociception in mice. Pharmacol Biochem Behav 52:
427432.
Oberpichler H, Beck T, Abdel-Rahman MM, Bielenberg GW,
Krieglstein J. 1988. Effects of Ginkgo biloba constituents related
to protection against brain damage caused by hypoxia. Pharmacol Res Commun 20: 349368.
Petkov VD, Kehayov R, Belcheva S, et al. 1993. Memory effects of
standardized extracts of Panax ginseng (G115), Ginkgo biloba
(GK 501) and their combination Gincosan (PHL-00701). Planta
Med 59: 106114.
Petkov VD, Mosharrof AH. 1987. Effects of standardized Ginseng
extract on learning, memory and physical capabilities. Am J Chin
Med 15: 1929.
Pittler MH, Ernst E. 2000. Ginkgo biloba extract for the treatment of
intermittent claudication: a meta-analysis of randomised trials.
Am J Med 108: 276281.
Ramassamy C, Christen Y, Clostre F, Costentin J. 1992. The Ginkgo
biloba extract, EGb761, increases synaptosomal uptake of 5hydroxytryptamine: in-vitro and ex-vivo studies. J Pharm
Pharmacol 44: 943945.

Copyright # 2002 John Wiley & Sons, Ltd.

Rapin JR, Le Poncin M, Latte M. 1986. Cerebral glucose consumption. The effect of Ginkgo biloba extract. Presse Med 25:
14941497.
Rigney U, Kimber S, Hindmarch I. 1999. The effects of acute doses
of standardized Ginkgo biloba extract on memory and psychomotor performance in volunteers. Phytother Res 13: 408415.
Roncin JP, Schwartz F, D'Arbigny P. 1996. EGb761 in control of
acute mountain sickness and vascular reactivity to cold exposure.
Aviat Space Env Med 67: 445452.
Rusted JM, Warburton DM. 1989. Cognitive models and cholinergic
drugs. Neuropsychobiology 21: 3136.
Schafer K, Reeh PW. 1985. Double-blind study of the hypoxia protective effect of a standardized Ginkgo biloba preparation after
repeated administration in healthy subjects. Arzneimittelforschung 35: 12831286.
Scholey AB. 2001. Fuel for thought. Psychologist 14: 196201.
Scholey AB, Harper S, Kennedy DO. 2001. Cognitive demand and
blood glucose. Physiol Behav 73: 585592.
Scholey AB, Moss MC, Neave N, Wesnes KA. 1999. Cognitive
performance, hyperoxia and heart rate following oxygen administration in healthy young adults. Physiol Behav 67: 783789.
Scholey AB, Moss MC, Wesnes KA. 1998. Oxygen and cognitive
performance: the temporal relationship between hyperoxia and
enhanced memory. Psychopharmacology 140: 123126.
Shi L, Fan PS, Wu L, Fang JX, Han ZX. 1990. Effects of total saponins of Panax notoginseng on increasing PGI2 in carotid artery
and decreasing TXA2 in blood platelets. Chung-Kuo Yao Li
Hsueh PaoActa Pharmacologica Sinica 11: 2932.
Soerensen H, Sonne J. 1996. A double masked study of the
effects of Ginseng on cognitive functions. Curr Therapeut Res
57: 959968.
Sotaniemi EA, Haapakoski E, Rautio A. 1995. Ginseng therapy in
non-insulin-dependent diabetic patients. Diabetes Care 18:
13731375.
Taylor LA, Rachman SJ. 1987. The effects of blood sugar level
changes on cognitive function, affective state and somatic symptoms. J Behav Med 11: 279291.
Vuksan V, Sievenpiper JL, Koo VYY, et al. 2000. American
Ginseng (Panax quinquefolius L) reduces postprandial glycemia
in nondiabetic subjects and subjects with type 2 diabetes
mellitus. Arch Intern Med 160: 10091013.
Wang LC, Lee TF. 1998. Effect of Ginseng saponins on exercise
performance in non-trained rats. Planta Med 64: 130133.
Wenk GL. 1989. An hypothesis of the role of glucose in the mechanism of cognitive enhancers. Psychopharmacology 99: 431438.
Wesnes KA, Faleni RA, Hefting NR, et al. 1997. The cognitive, subjective, and physical effects of a Ginkgo biloba/Panax ginseng
combination in healthy volunteers with neurasthenic complaints.
Psychopharmacol Bull 33: 677683.
Wesnes KA, Ward T, Petrini O, McGinty A. 2000. The memory
enhancing effects of a Ginkgo biloba /Panax ginseng combination in healthy middle aged volunteers. Psychopharmacology
152: 353361.
White HL, Scates PW, Cooper BR. 1996. Extracts of Ginkgo biloba
leaves inhibit monoamine oxidase. Life Sci 58: 13151321.
Wiklund I, Karlberg J, Lund B. 1994. A double-blind comparison of
the effects on quality of life of a combination of vital substances
including standardised Ginseng G115 and placebo. Curr Therapeut Res 55: 3242.
Wood WB, Roh BL, White RP. 1964. Cardiovascular actions of
Panax ginseng in dogs. Jpn J Pharmacol 14: 284294.

Hum Psychopharmacol Clin Exp 2002; 17: 3544.

Copyright of Human Psychopharmacology: Clinical & Experimental is the property of John Wiley & Sons, Inc.
and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for individual use.