Dexamphetamine enhances explicitnew word learning for novel objects

Emma Whiting1, Helen Chenery1, Jonathan Chalk2, Ross Darnell3 and David Copland1

1 Division of Speech Pathology, 2 Centre for Magnetic Resonance, 3 School of Health and Rehabilitation Sciences,

The University of Queensland, St Lucia, Queensland, Australia

Abstract

Past research suggests that dexamphetamine (Dex) can facilitate learning and memory in healthy

individuals and after a neurological lesion. This study investigated the effects of Dex on the learning of

names for new objects in young healthy adults (n=37) within an explicit learning paradigm by using a

double-blind, placebo-controlled between- subjects design. Participants received 10 mg Dex or a placebo

each morning over five consecutive days before viewing 100 novel objects with non-word names plus

matched fillers. Compared to the placebo, Dex enhanced both the rate of learning and the retention of the

words 1 wk and 1 month later. The improved word learning correlated with baseline attention and

memory scores for participants in the Dex group only. No correlations were observed between word-

learning success and sustained attention, mood or cardiovascular arousal. It was concluded that the

improved explicit word learning may have reflected dexamphetamine-induced changes in short-term

memory and/or memory consolidation.

Received 23 August 2006 ; Reviewed 14 October 2006 ; Revised 2 November 2006 ; Accepted 23 November 2006 ;

First published online 25 January 2007

Key words : Dexamphetamine, dopamine, word learning.

Introduction

Converging evidence from animal and human para-

digms shows that learning and memory can be modu-

lated by pharmacological agents (Breitenstein et al.,

2004 ; Izquierdo, 1994 ; Knecht et al., 2001 ; Kumari

et al., 1997 ; Martinez et al., 1991 ; Riedel et al., 1995). Of

these agents, dexamphetamine (Dex) has been found

to enhance associative and list-based word learning in

healthy individuals (Breitenstein et al., 2004 ; Soetens,

et al., 1993, 1995), and linguistic and motor recovery

after a cerebrovascular accident (Walker-Batson et al.,

1995, 2001). To date, no studies have investigated the

effects of Dex on the learning of names for new objects

in healthy adults. Further, it has been recently pro-

posed that new word learning may provide a model

for word re-learning after neurological damage (Basso

et al., 2001 ; Breitenstein and Knecht, 2002), providing

additional motivation for this study.

One of the earliest studies to investigate the effects

of Dex on word learning was conducted by Soetens

et al. (1993) who found that Dex improved word list

retention during a free recall task compared to a

placebo for up to 3 d post-learning in a group of

healthy males. A subsequent study found that Dex

improved word list recognition up to 1 wk post-

learning (Soetens et al., 1995). As Dex did not improve

performance immediately after learning, Soetens et al.

(1995) concluded that the drug acted by enhancing

memory consolidation, rather than arousal or short-

term memory processes. However, even though both

study tasks required short-term memory, participants

were required to learn lists of real words, which al-

lowed the use of existing phonological representations

of the words within long-term memory (Gathercole,

1999). Thus, the results may have reflected the effects

of Dex on the activation of words in long-term

memory, rather than short-term memory processes.

New word learning may provide a means of circum-

venting this issue, as during new word learning, there

is no existing phonological representation of the word.

Instead, the learner is required to hold the new word

form temporarily in short-term memory until a more

permanent representation is established in long-term

Address for correspondence : Dr D. Copland, Division of Speech

Pathology, The University of Queensland, St Lucia, QLD 4072,

Australia.

Tel. :+61 7 3365 2817 Fax :+61 7 3365 1877

E-mail : d.copland@uq.edu.au

International Journal of Neuropsychopharmacology (2007), 10, 805–816. Copyright f 2007 CINPdoi:10.1017/S1461145706007516

ARTICLE

CINP

memory (Baddeley et al., 1988, 1998 ; Gupta and

MacWhinney, 1997 ; Martin and Gupta, 2004 ; Service

and Craik, 1993).

Indeed, Breitenstein et al. (2004) demonstrated that

the facilitatory effects of Dex can extend to new word

learning. Participants received Dex (0.25 mg/kg) or a

placebo before learning new names for familiar objects

over five consecutive days. The new names were

learnt by implicit associative learning mechanisms,

a process that involved subconsciously comparing

the relative occurrences of ‘correct ’ and ‘incorrect ’

picture-to-new word pairings. At the end of the fifth

training session, participants were required to trans-

late the words into their native language (German).

The Dex group showed enhanced new word learning

compared to the placebo group, with the difference

maintained 1 wk and 1 month later.

Unlike most word learning studies, Breitenstein

et al. (2004) investigated new word learning within

an implicit associative learning paradigm. Broadly

defined, implicit learning involves the acquisition

of information without conscious awareness, while

explicit learning involves the conscious acquisition of

information (Church and Schacter, 1994; Cohen et al.,

1997 ; Squire et al., 1993 ; Ullman, 2004 ; Zacks et al.,

2000). Thus far, the role of each memory system in

lexical learning remains contentious. The study by

Breitenstein et al. (2004) involved implicit learning,

however, the words were retrieved in an explicit

memory-based task, which suggested that participants

may have had some explicit knowledge about the

words. Consequently, Breitenstein et al. (2004) sug-

gested that both the explicit and implicit memory

systemsmay be involved in word learning. In contrast,

Ullman (2004) proposed that lexical learning may be

linked exclusively to the explicit (declarative) memory

system. Therefore, given our current level of knowl-

edge about memory systems, it is unknown whether

the effects of Dex on new word learning observed by

Breitenstein et al. (2004) are exclusive to implicit word

learning or extend to other types of learning. As the

study by Breitenstein et al. (2004) involved learning

new names for familiar objects, similar results may

not be observed when learning names for unfamiliar

objects, given evidence from bilingual speakers (e.g.

Kroll and Stewart, 1994).

In summary, previous research has found that

Dex can facilitate word learning in list-based and

implicit associative learning contexts. However, the

effect of Dex on the learning of names for unfamiliar

objects within an explicit learning paradigm remains

unknown. Additionally, no research has examined the

effects of Dex on new word learning with semantic

information. It might be expected that providing

semantic information in conjunction with a phono-

logical form would facilitate word learning compared

to when only a phonological form is provided, given

that deeper processing of information has been found

to improve learning (Craik and Tulving, 1975 ; Maestu

et al., 2003). As a consequence, the aims of the present

study were to (a) determine whether the effects of

Dex on word learning extend to explicit new word

learning, and (b) examine whether Dex boosts new

word learning in the presence of additional semantic

information. By including mood, cognitive, and

cardiovascular assessments, we were able to investi-

gate whether learning was related to mood or other

cognitive functions, or cardiovascular arousal. Over-

all, it was predicted that Dex would improve both the

rate of newword learning and the retention of the new

words. In addition, it was predicted that the inclusion

of extra semantic information would further improve

new word learning, compared to when only a phono-

logical form was provided. Finally, it was predicted

that new word learning would be related to baseline

learning and memory skills, and changes in mood

and cardiovascular arousal, reflecting dopaminergic

or adrenergic mechanisms (Breitenstein et al., 2004;

Knecht et al., 2001).

Methods

Participants

The participants were 37 right-handed native English-

speaking adults (11 males, 26 females) aged between

18 and 34 yr. Participants were randomly assigned to

the Dex (M 5, F 14) or placebo (M 6, F 12) groups in a

between-subjects, double-blind design (see Table 1 for

the demographic characteristics of each group). There

were no significant differences between the Dex and

placebo groups in terms of age, body weight, cigarette

consumption, or baseline neuropsychological scores

(all p>0.10) (see Table 1 for a summary of the pre-test

neuropsychological scores). Participants possessed

hearing and visual abilities (with or without corrective

devices) within normal limits. All participants com-

pleted a medical review with a doctor, including

an electrocardiogram, prior to participation in the

study. Exclusion criteria included pregnancy or any

psychiatric, neurological, cardiac, metabolic, degener-

ative condition, or motor disorders or diseases. Any

individuals with a history of substance abuse (any

substance), as determined by a government registry of

individuals with a known history of substance abuse,

were excluded.

806 E. Whiting et al.

Participants were requested to abstain from alcohol,

caffeine, and food for at least 6 h before each session.

Written informed consent for participation in the

study was obtained from each participant. All partici-

pants received monetary reimbursement for partici-

pation. In order to encourage participants to attend to

the task, there was an additional monetary bonus for

the participant with the highest accuracy score at the

end of the five learning sessions.

Stimuli

The same stimulus items were viewed by participants

in the Dex and placebo groups. Each of the stimulus

items consisted of a black-and-white line drawing of

an object and a randomly assigned non-word name

(e.g. barve, dreint, peam) with/without a fictitious

description of the object’s function (e.g. a tool used for

digging trenches) written below. The stimulus items

consisted of a combination of critical and filler items.

In total, participants viewed 240 stimulus items during

the sessions. There were 50 critical items with a

drawing of an unfamiliar object, a non-word name and

a description of the object’s function, and 50 critical

items with a drawing of an unfamiliar object, a non-

word name but no description of the object’s function.

There were 10 matched fillers for each of the two types

of unfamiliar critical items. There were 120 additional

items consisting of drawings of familiar objects paired

with non-words for a companion study.

The drawings of the unfamiliar objects had been

used in an earlier experiment by Laine et al. (2003)

and were used with permission from the author.

The non-word names for all items were obtained from

the Australian Research Council (ARC) non-word

naming database (Rastle et al., 2002), and consisted of

combinations of consonants and vowels that adhered

to the rules of written English. Homophones and

words that closely resembled real words in English

were excluded. In addition, only non-words that had

Table 1. Participant demographic and neuropsychological characteristics

Measure

Group

Dexamphetamine (n=19) Placebo (n=18)

Mean S.D. Mean S.D.

Age (yr) 23.00 4.65 23.44 4.79

Years of education 15.42 3.37 15.00 3.55

TOMAL Digits backwards (maximum score : 88) 48.21 17.17 49.50 21.40

RBANSa

Immediate memory index score 103.21 15.74 96.94 10.96

Visuospatial/constructional index score 94.84 15.47 94.78 15.53

Language index score 107.68 9.43 101.06 13.42

Attention index score 106.74 15.71 107.44 14.58

Delayed memory index score (average range : 85–115) 97.16 11.43 95.11 6.54

CTOPP Non-word repetition (maximum score : 18) 13.68 2.83 13.53 3.83

Woodcock Word Attack (maximum score : 45) 41.11 3.78 40.67 3.27

Hayling overall scaled score (maximum score : 7) 5.95 0.71 5.94 0.42

NART-R Predicted full scale IQ (average range : 85–115) 108.63 4.80 108.22 5.40

CANTAB

Big/little circle 100.00 0.00 100.00 0.00

ID/ED shift stage 8.79 0.63 8.56 1.89

ID/ED shift total errors adjusted 17.32 14.92 23.28 44.47

PAL total errors adjusted 0.89 1.49 1.00 1.24

TOMAL, Test of Memory and Learning (Reynolds and Bigler, 1994) ; RBANS, Repeatable Battery for the Assessment of

Neuropsychological Status (Randolph, 1998) ; CTOPP, Comprehensive Test of Phonological Processing (Wagner et al., 1999) ;

NART-R, National Adult Reading Test – Revised (Nelson and Willison, 1991) ; CANTAB, Cambridge Neuropsychological Test

Automated Battery (Cambridge Cognition, 2004) ; ID/ED shift, intra-extra dimensional set shift ; PAL, paired associates learning.a As all participants were aged<35 yr, we used the lowest range of standardized scores when calculating Index scores (20–39 yr)

for every participant. Standard scores for 20-yr-olds were used for participants aged <20 yr.

Dexamphetamine and word learning 807

a one-to-one phoneme–grapheme correspondence

were selected. The length of the non-words ranged

from four to six letters.

Experimental design

The study was conducted in accordance with the

Declaration of Helsinki. Ethical clearance was ob-

tained from the University of Queensland Medical

Research Ethics Committee. The study was divided

into one pre-test session, five learning sessions, and

two post-test sessions. Participants did not receive a

tablet during the pre-test or post-test sessions. During

the pre-test session, participants completed a series of

formal and informal tests designed to assess learning,

attention, memory (immediate, delayed, visuospatial,

auditory, numeric, and linguistic), word suppression,

reading, vocabulary, and executive function (Burgess

and Shallice, 1997; Cambridge Cognition, 2004;

Nelson and Willison, 1991 ; Randolph, 1998 ; Reynolds

and Bigler, 1994 ; Wagner et al., 1999 ; Woodcock, 1998)

(see Table 1 for a summary of the tests and results).

During the learning sessions, participants attended

five 3–3.5 h learning sessions over five consecutive

mornings. Each learning session consisted of a tablet

ingestion phase, a waiting phase, an initial learning

phase, and then a short test phase.

At the start of each learning session, the participants

were given either two 5 mg Dex tablets or two placebo

tablets. There was a 2-2.5 h wait between tablet inges-

tion and commencement of the learning tasks each

session. This time-period was selected so that optimal

Dex plasma levels were reached during the learning

tasks (Angrist et al., 1987 ; Asghar et al., 2003).

During the learning phase, each object was seen

with its corresponding non-word name once for 5 s.

The order of the items was randomized. A break was

provided after each block of 60 items. Participants

were not required to respond to the stimuli when

viewing them at the start of each learning session.

At the end of each learning session, participants

completed recall and recognition tasks, which are

described below. The post-test sessions involved recall

and recognition tasks and were conducted 1 wk and

1 month after participants had completed the learning

sessions. E-prime software (Psychology Software

Tools, Pittsburgh, PA, USA) was used for all of the

learning tasks.

Recall task

During the recall task, participants viewed each of the

120 pictures of the critical unfamiliar items one-by-one

on the computer screen without non-word names or

descriptions. The objects were presented in random

order, with the order of the items differing between

each session. Participants were instructed to type the

non-word name of each object. For a response to be

recorded as correct, participants had to type all of the

letters of the non-word name in the correct order.

Recognition tasks

After completing the recall task each session, partici-

pants then completed two recognition tasks : a name

recognition task and a description recognition task.

The order of the two recognition tasks was random-

ized each session.

Name recognition task

During the name recognition task, participants saw

the 120 critical unfamiliar objects with a single non-

word written below. The non-word was either the

‘correct ’ non-word name of the object or an ‘incorrect ’

non-word name. The ‘ incorrect ’ names consisted of

(a) non-words that differed from the ‘correct ’ names

by 2–3 letters, (b) the names of other objects seen

during the session, or (c) non-words that had not

been seen by the participants previously. The correct/

incorrect pairings of the non-words and objects were

randomized in order to reduce expectancy effects,

however, half of the object/non-word pairings were

correct each session.

Participants were required to indicate whether the

non-word was the ‘correct ’ or ‘ incorrect ’ name of the

object by pressing either a ‘yes’ or ‘no’ button on a

SRB200 response box (Psychology Software Tools).

Participants were encouraged to respond to the stimuli

as quickly as possible. The accuracy (correct/incorrect)

and response latency (in milliseconds) of participants ’

responses were automatically recorded by E-prime.

Description recognition task

The description recognition task was included so that

participants would attend to the descriptions of the

items’ functions. The task was only conducted with

the 50 items that included descriptions of the objects ’

functions. The description recognition task was the

same as the name recognition task described above,

however participants were required to indicate which

description matched an item rather than a non-word

name. Data from the task was not included in the

analyses.

Physiological measures

To provide an indication of the effects of Dex on

physiological functioning, each participant’s blood

808 E. Whiting et al.

pressure (diastolic and systolic) and heart rate were

measured at the start of each learning session before

tablet ingestion and then at half-hourly intervals

throughout the session.

Rapid Visual Information Processing (RVIP) task

To examine the effects of Dex on sustained attention,

participants completed a computerized RVIP task

during the middle learning session. The task was

based on similar experiments used by researchers to

investigate the effects of pharmacological agents on

information processing (Wesnes and Warburton, 1983,

1984). During the RVIP task, participants viewed a

series of digits (1–9) presented one-by-one at a rate of

100 digits per minute for a total of 10 min. The digits

appeared in black font against a white background

and were pseudo-randomized so the same digit never

appeared successively. Participants were instructed to

press a response button as soon as they saw three

consecutive even or odd digits in a row. The time limit

for responding to an item was 1500 ms. The number

of correct responses, misses, and false alarms, and

response times were automatically recorded.

Mood rating scales

To measure the effects of Dex on mood, participants

completed the Mood Rating Scale (Bond and Lader,

1974) twice during each learning session: once im-

mediately before tablet ingestion and then y2 h after

tablet ingestion. The timing of the secondmeasure was

selected to correspond with peak plasma Dex levels, as

identified in previous studies (Angrist et al., 1987 ;

Asghar et al., 2003 ; Mattay et al., 2000). The Mood

Rating Scale consists of a 16-item visual analogue scale

designed to measure drug effects. Each scale rep-

resented a continuum between positive and negative

feelings (e.g. alert–drowsy, troubled–tranquil).

Data analysis

Analyses were conducted with the independent

variable ‘group’ (dexamphetamine, placebo) and the

dependent variables ‘accuracy’ (for the recall and

recognition tasks) and ‘response time’ (for the recog-

nition tasks only). In order to equalize variance, an arc

sine transformation was applied to data that involved

proportions (i.e. the accuracy data from the recall and

recognition tasks). A log transformation was applied

to the mood rating scale data to reduce skewness.

Data from the learning sessions were analysed

using a series of repeated-measures ANOVAs with

‘session’ (1–5) and ‘description’ (description, no

description) as within- subjects factors, and ‘group’

(dexamphetamine, placebo) as a between-subjects

factor. For the data pertaining to blood pressure and

heart rate, ‘ time’ (baseline, 2 h post-tablet ingestion)

was added as an additional within-subjects factor.

The difference in self-reported feelings between base-

line and 2 h post-tablet ingestion each session, as

measured by the mood rating scales, was analysed

using linear mixed models with ‘group’ (dexam-

phetamine, placebo), ‘subject ’, and ‘session’ (1–5) as

factors. Data from the 1 wk and 1 month post-tests

were analysed using independent-samples t tests.

Spearman’s correlation coefficients were used

to analyse the effect of different variables (e.g.

mood, baseline neuropsychological characteristics,

and changes in blood pressure and heart rate) on

training success and changes in response latency over

the five learning sessions. As in Breitenstein et al.

(2004), training success was measured by calculating

the difference in accuracy scores between sessions 1

and 5 on the recall and recognition tasks. In order to

minimize the risk of Type I errors due to multiple

comparisons, an a-level of 0.01 was employed for

analyses involving correlations. An a-level of 0.05 was

employed for all other analyses.

Preliminary analyses indicated that there were no

significant correlations between body weight and

training success or response time, or between gender

and training success or response time (all p>0.10).

As a consequence, body weight and gender were not

included as factors in subsequent analyses.

Results

Learning data

Recall

Both groups improved in terms of accuracy on

the recall task over the five learning sessions

[main effect – session, linear trend: F(4, 32)=60.559,

p<0.001)], however, the Dex group was consistently

more accurate than the placebo group during each

session [interaction – sessionrgroup: F(4, 32)=6.858,

p<0.001] (see Figure 1). The difference in accuracy

scores between the two groups was still apparent 1 wk

and 1 month after the completion of the learning

sessions, with the Dex group still performing signifi-

cantly more accurately on the task than the placebo

group [t(35)=2.997, p=0.005 and t(35)=3.400, p=0.002 respectively].

Participants responded more accurately to items

that did not include a description of the object’s func-

tion compared to items that included a description

Dexamphetamine and word learning 809

[main effect – description : F(1, 35)=19.048, p<0.001]

(see Figure 2), but this pattern was not affected by

drug. Participants ’ recall of both sets of items

increased each session, however, items without a de-

scription were recalled more accurately each session

than items that included a description [interaction –

sessionrdescription : F(4, 32)=2.755, p<0.05].

Correlations with pre-test neuropsychological measures

There were no significant correlations between train-

ing success on the recall task and any of the pre-test

neuropsychological measures for either the Dex or

placebo groups (all p>0.01).

Name recognition

Accuracy

The accuracy of participants in both groups on the

name recognition task improved over the five learning

sessions [main effect – session, linear trend: F(4, 32)=

80.505, p<0.001] (see Figure 3). However, the Dex

group consistently performed more accurately during

each session than the placebo group [interaction –

sessionrgroup: F(4, 32)=7.424, p<0.001], with the

difference remaining significant 1 wk and 1 month

later [t(35)=4.109, p<0.001 and t(35)=3.395, p=0.002

respectively] (see Figure 3). Overall, participants

performed more accurately on items that did not

include a description of the object’s function [main

effect – description : F(1, 35)=5.436, p=0.026]. Whilst

participants ’ accuracy scores increased each session,

participants were consistently more accurate each

session on items that did not include a description

[interaction – sessionrdescription: F(4, 32)=3.005, p=0.033] (see Figure 4).

Correlations with pre-test neuropsychological measures

For the Dex group, there was a significant positive

correlation between training success on the name rec-

ognition task and the RBANS attention index score

1.0

0.8

0.6P

rop

ort

ion

of

corr

ect

resp

on

ses

0.4

0.2

01 2 3 4 5

Session

1 weekpost-test

1 monthpost-test

Figure 1. Overall proportion of non-word names correctly

produced during the recall task $, Dexamphetamine ; m,

placebo.

1.0

0.8

0.6

0.4

Pro

po

rtio

n o

fco

rrec

t re

spo

nse

s

0.2

01 2 3 4 5 1 week

post-test

Session

1 monthpost-test

Dex + DescDex − DescPla + DescPla − Desc

Figure 2. Proportion of correct responses for items with and

without descriptions during the recall task. Dex,

Dexamphetamine ; Pla, placebo; +Desc, item included a

description ; xDesc, item did not include a description.

1.0

0.8

0.6

Pro

po

rtio

n o

fco

rrec

t re

spo

nse

s

0.4

0.2

01 2 3 4 5

Session

1 weekpost-test

1 monthpost-test

Figure 3. Overall proportion of non-word names correctly

identified during the name recognition task. $,

Dexamphetamine ; m, placebo.

Dex + DescDex − DescPla + DescPla − Desc

1.0

0.8

0.6

0.4

Pro

po

rtio

n o

fco

rrec

t re

spo

nse

s

0.2

01 2 3 4 5 1 week

post-test

Session

1 monthpost-test

Figure 4. Proportion of correct responses for items with and

without descriptions during the name recognition task. Dex,

Dexamphetamine ; Pla, placebo; +Desc, item included a

description ; xDesc, item did not include a description.

810 E. Whiting et al.

(r=0.652, p=0.002, all other p>0.01). There were no

significant correlations between training success on

the name recognition task and the pre-test neuro-

psychological measures for the placebo group (all

p>0.01).

Response time

Statistical analyses were conducted only on correct

‘yes’ responses. Participants responded increasingly

faster on the name recognition task over the five

learning sessions [main effect – session, linear trend:

F(4, 32)=36.014, p<0.001], however, there was no

significant difference between the two groups (p>0.10) (see Table 2). At the 1 wk post-test, participants

in the placebo group responded to the stimuli in the

name recognition task significantly faster than the Dex

group [t(35)=2.229, p<0.05], although the difference

was no longer significant at the 1 month post-test

(p>0.10). There was no significant difference between

items that included a description of the object’s func-

tion and items without a description for participants in

either group (p>0.10).

Correlations with pre-test neuropsychological measures

For the Dex group, the change in response times be-

tween session 1 and 5 was negatively correlated with

the Repeatable Battery for the Assessment of Neuro-

psychological Status (RBANS) immediate memory

index score (r=x0.583, p=0.009) and RBANS delayed

memory index score (r=x0.598, p=0.007, all other

p>0.01). In contrast, there were no significant corre-

lations between the change in response times between

sessions 1 and 5 and the pre-test neuropsychological

measures for the placebo group (all p>0.01).

Physiological measures

Systolic blood pressure

Baseline measurements indicated that there were

no differences in systolic blood pressure between

the Dex or placebo groups (Dex group: mean=102.41, S.D.=1.93 ; placebo group: mean=103.37, S.D.=2.32) (p>0.05). The change in systolic blood pressure

between the initial and final measurements each

session did not differ significantly between the groups

(final measurement – Dex group: mean=104.00,

S.D.=2.48 ; placebo group: mean=102.97, S.D.=0.85)

(p>0.05).

Diastolic blood pressure

There were no differences in diastolic blood pressure

between the Dex or placebo groups at baseline

(Dex group: mean=68.78, S.D.=2.50 ; placebo group:

mean=70.04, S.D.=3.34) (p>0.05). The change in

diastolic blood pressure between the initial and final

measurements each session did not differ significantly

between the groups (Final measurement – Dex group:

mean=69.37, S.D.=1.75 ; placebo group: mean=68.79,

S.D.=2.25) (p>0.05).

Heart rate

There was no difference in baseline heart rate

(Dex group: mean=72.04, S.D.=3.36 ; placebo group:

mean=70.59, S.D.=2.51) (p>0.05), or in the change

in heart rate between initial and final measure-

ments each session (final measurement – Dex group:

mean=74.76, S.D.=4.21 ; placebo group: mean=70.97,

S.D.=2.89) (p>0.05) for either the Dex or placebo

groups.

RVIP task

There were no differences between the Dex or placebo

groups on the RVIP task with respect to the proportion

of correct positive hits, misses, false alarms, or the

response latency of correct positive hits (all p>0.10)

(see Table 3). In addition, there were no significant

correlations between the proportion of correct positive

hits or response latency on the RVIP task and training

success or response latency on the recall or recognition

tasks for either group (all p>0.01).

Mood rating scales

Four mood ratings were not available for four

different participants (i.e. one record sheet for each

of the four participants was unavailable on four

separate days).

Table 2. Mean reaction times on the name recognition task in

milliseconds

Session

Group

Dexamphetamine Placebo

Mean S.D. Mean S.D.

1 1720.28 458.89 1478.01 367.50

2 1596.52 361.22 1294.97 287.17

3 1525.19 290.31 1183.65 238.73

4 1281.37 300.26 1087.34 212.11

5 1275.81 229.61 1008.39 150.32

1 wk post-test 1313.86 336.01 1115.48 152.34

1 month post-test 1258.68 265.71 1155.28 248.55

Dexamphetamine and word learning 811

Alertness

There were no significant differences in subjective

alertness levels between the Dex or placebo groups

at the start of each learning session (Dex group:

mean=5.25, S.D.=0.08 ; placebo group: mean=5.59,

S.D.=0.05) (p>0.10). The difference in alertness

between participants in the two groups between

baseline and 2 h post-tablet ingestion was marginal,

with participants in the Dex group feeling more alert

than participants in the placebo group (Dex group:

mean=4.95, S.D.=0.07 ; placebo group: mean=5.49,

S.D.=0.03) [main effect : group, F(1, 30.391)=3.831,

p=0.06]. There were no correlations between the

change in alertness each session and training success

or response time on the recall or recognition tasks for

the Dex group (all p>0.01).

Contentedness

The Dex and placebo groups did not differ in terms

of subjective contentedness at baseline (Dex group:

mean=4.51, S.D.=0.10 ; placebo group: mean=4.76,

SD=0.02) or 2 h after tablet ingestion (Dex group:

mean=4.43, S.D.=0.07 ; placebo group: mean=4.74,

S.D.=0.06) (all p>0.10).

Calmness

There were no differences in subjective calmness

between the Dex or placebo groups at the start of each

learning session (Dex group: mean=3.56, S.D.=0.08 ;

placebo group: mean=3.89, S.D.=0.07) (p>0.10).

However, the difference in calmness levels between

baseline and 2 h post-tablet ingestion differed between

the groups. Participants in the Dex group reported

reduced feelings of calmness (i.e. participants felt

more tense and excited) y2 h after drug ingestion

compared to participants in the placebo group (Dex

group: mean=3.76, S.D.=0.15 ; placebo group: mean

=3.90, S.D.=0.10) [main effect : group, F(1, 17.607)=7.340, p<0.05]. However, the change in calmness each

session for the Dex group did not correlate with

training success or response times on the recall or

recognition tasks (all p>0.01). There was a significant

difference in the change in calmness levels from base-

line to 2 h later between the sessions [main effect :

session, F(4, 132.610)=4.769, p<0.05], with paired-

samples t tests indicating that there was less of a

change in participants ’ calmness levels from baseline

to 2 h later during session 2 compared to session 1

[t(33)=2.396, p<0.05, for all other paired-samples

t tests p>0.05].

Discussion

The present study found that within an explicit learn-

ing paradigm, Dex significantly improved (a) the rate

of word learning, and (b) the retention of the newly

learnt words at follow-up. Irrespective of drug group,

the inclusion of additional semantic information

hindered new word learning. Baseline attention and

memory skills were correlated with name recognition

accuracy and response time for participants in the Dex

group only. In contrast with Breitenstein et al. (2004),

changes in subjective mood ratings did not correlate

with new word learning success. The following dis-

cussion will centre on the mechanisms underlying the

effects of Dex on new word learning observed in the

present study.

Implicit learning mechanisms

As discussed previously, Breitenstein et al. (2004)

found that Dex enhanced new word learning within

an implicit associative learning paradigm. In contrast,

the present study found that Dex improved new

word learning within an explicit learning paradigm.

Consequently, the results of the present study suggest

that the facilitatory effects of Dex on new word learn-

ing are not solely due to implicit learning mechanisms.

Short-term memory and attention

Attention and memory, especially working memory

(a subcomponent of short-term memory linked to

the prefrontal cortex), are closely related to linguistic

learning (Arnsten, 1998 ; Baddeley et al., 1998;

Cornelissen et al., 2004 ; Duyck et al., 2003 ; Gupta,

1996 ; Gupta et al., 2004 ; Gupta and MacWhinney,

1997 ; Martin and Freedman, 2001 ; Schuchert, 2004;

Service and Craik, 1993). In the present study,

performance on the name recognition task correlated

Table 3. Mean proportion of positive hits and reaction

times in milliseconds on the rapid visual information

processing task

Measure

Group

Dexamphetamine Placebo

Mean S.D. Mean S.D.

Proportion of correct

positive hits

0.46 0.15 0.48 0.12

Mean reaction

time (ms)

580.00 70.00 550.00 80.00

812 E. Whiting et al.

significantly (in a positive direction) with baseline

attention levels for the Dex group only. Baseline at-

tention levels in the present study were measured

according to the R-BANS attention index (Randolph,

1998), a measure comprising of digit span and coding

tasks, which are designed to measure short-term

memory. Martin and Gupta (2004) argued that digit

recall and lexical learning are closely related and that

both are reliant on short-term memory. The results of

the present study suggest that Dex further enhanced

task performance in individuals who already had

higher baseline short-term memory capacity. Interest-

ingly, higher levels of dopamine have been associated

with higher performance on learning and memory

tests (Arnsten, 1998 ; Diamond et al., 2004 ; Knecht

et al., 2004). Further, Knecht et al. (2004) found that

there was a linear relationship between levodopa (a

precursor of dopamine) and implicit associative word

learning, with higher doses being linked to improved

word learning. Thus, one possible explanation for the

present data is that Dex acted through dopaminergic

mechanisms by enhancing short-term memory, a

cognitive skill that was critical for task success. It

should be noted that the effects of dopamine have also

been found to have an inverted U curve relationship

with prefrontal working memory, with too little or

too much dopamine leading to impaired functioning

(Arnsten, 1998 ; Mattay et al., 2000) ; however, for the

memory tests used in the present study this effect was

not apparent.

Contrary to the findings of previous studies

(Fillmore et al., 2005 ; Fleming et al., 1995 ; Rapoport

et al., 1980), in the present study, there was no differ-

ence between the Dex or placebo groups in terms of

accuracy or response times on the RVIP task, a

measure of sustained attention. This finding suggested

that the effect of Dex on learning was not secondary to

any effects on sustained attention. However, the lack

of difference between the groups on the task may

have been due to the small dose (10 mg) of dexam-

phetamine. The dose may have been insufficient to

make a significant difference, given the substantial

cognitive demands of the RVIP task. The difference in

cognitive demands between the new word learning

and RVIP tasks may also explain why Dex improved

new word learning, but not RVIP task performance.

Long-term memory, long-term potentiation, and

memory consolidation

Both the present study and the study by Breitenstein

et al. (2004) found that Dex improved retention of the

newly learnt words when tested off drug several days

after the learning sessions ceased. Similarly, research

by Soetens and colleagues (Soetens et al., 1993, 1995)

found that Dex enhanced word list recall and recog-

nition up to 1 wk after initial learning. Soetens et al.

(1993) suggested that Dex acted by selectively en-

hancing the memory storage process (i.e. memory

consolidation), rather than memory encoding. Animal

studies have found that Dex can enhance memory

consolidation (Carr and White, 1984 ; Martinez et al.,

1980), with the improved memory retention being

linked to dexamphetamine-induced effects on dop-

aminergic function (Carr and White, 1984). Further-

more, both Jones (2004) and Knecht et al. (2004)

suggested that dopamine is involved in long-term

potentiation associated with memory consolidation.

Although the mechanisms underlying long-term

memory remain contentious, dexamphetamine-

induced neurophysiological changes may have con-

tributed to enhanced new word learning and retention

in the present study.

Mood

Recent research into the effects of emotion on memory

has produced equivocal results (Arntz et al., 2005 ;

Fenker et al., 2005). During the present study, Dex

increased subjective levels of alertness and tenseness/

excitability, however, the changes did not correlate

with new word learning success. This finding con-

trasts with Breitenstein et al. (2004) who found

that new word learning success was correlated with

dexamphetamine-induced changes in positive feel-

ings. One possible explanation for this difference is the

use of two different learning methods (i.e. explicit

vs. implicit associative learning), suggesting that

implicit new word learning may be more reliant on

mood-based mechanisms (but see Knecht et al., 2004)

and that increased positive emotions do not drive

improved new word learning under explicit learning

conditions.

Cardiovascular arousal

Previous studies have found that Dex can increase

blood pressure and heart rate, albeit not always

significantly (Asghar et al., 2003 ; Brauer et al., 1996 ;

Breitenstein et al., 2004 ; de Wit et al., 2002 ; Mattay

et al., 2000). At a neurophysiological level, nor-

adrenaline, a neurotransmitter that increases after Dex

ingestion (de Wit et al., 2002 ; Kuczenski and Segal,

1997 ; Martinsson and Eksborg, 2004), has been im-

plicated in the modulation of cardiovascular arousal

(Arnsten, 1998 ; Brauer et al., 1996 ; Breitenstein et al.,

2004 ; Knecht et al., 2004 ; Kuczenski and Segal, 1997 ;

Dexamphetamine and word learning 813

Soetens et al., 1995). However, changes in motor/

cardiovascular arousal have been found to be un-

correlated with the effects of Dex on word learning

(Breitenstein et al., 2004; Cooper et al., 2003). The

present study supported these findings, as during the

learning sessions, the Dex and placebo groups did not

differ on measures of cardiovascular arousal, despite

differences in the success of new word learning.

The finding that Dex did not modulate explicit new

word learning through cardiovascular mechanisms

suggests that noradrenergic mechanisms were not

chiefly responsible.

Effects of semantic information on new word

learning

The inclusion of additional semantic information

through descriptions of the objects’ functions did not

facilitate new name learning in the present study. This

finding resembles that of Gronholm and colleagues

(2005) who used some of the same stimuli as the

present study. Unlike Gronholm et al. (2005), the

present study included tasks designed specifically to

encourage participants to attend to both the phono-

logical form and the semantic information, however,

this did not result in superior learning for items with

additional semantic material. The most straight-

forward explanation for the poorer performance in

new word learning with vs. without a description in

the present study is that the participants did not have

sufficient time to read and subsequently learn the

name and description for each item compared to the

name alone. Participants may have found the semantic

information easier to learn, and thus concentrated on

learning the descriptions at the expense of the names.

Dex did not appear to influence this process.

As a final note, it must be mentioned that during

the present study all participants in the Dex group

received a 10 mg dose of the drug irrespective of body

weight. Despite the use of this uniform dose, beha-

vioural effects (i.e. improved word learning) were ob-

served with the drug and statistical analyses indicated

that body weight did not influence drug outcomes.

Conclusion

In conclusion, the results of the present study show

that Dex facilitates new word learning in an explicit

learning paradigm. This finding is important for the

development of relevant models of language rehabili-

tation and pharmacotherapy using healthy word

learning, as explicit learning more closely resembles

the type of relearning utilized during naming therapy

in aphasia than implicit associative learning, which

involves the repeated introduction of erroneous word

forms. This study further elucidates the role of Dex in

word learning by demonstrating enhanced learning

and development of word-object mappings when

competing pre-existing names are not present.

Importantly, the data suggest that the facilitatory

effects of Dex may be due to improved short-term

memory and/or long-term potentiation, rather than

alterations in general attention, mood, or cardio-

vascular arousal. These findings have important im-

plications for the future use of pharmacological agents

in the treatment of naming and learning disorders.

Acknowledgements

The authors thank Kerry Kennedy for assistance

with data entry, Anna Holmes and Dion Scott

for the RVIP task, and two anonymous reviewers

for comments on an earlier version of the manu-

script. This research was supported by a University

of Queensland Research and Development grant

awarded to D.C., J.C. and H.C.

Statement of Interest

None.

References

Angrist B, Corwin J, Bartlik B, Cooper T (1987).

Early pharmacokinetics and clinical effects of oral

d-amphetamine in normal subjects. Biological Psychiatry

22, 1357–1368.

Arnsten AFT (1998). Catecholamine modulation of prefrontal

cortical cognitive function. Trends in Cognitive Sciences 2,

436–447.

Arntz A, de Groot C, Kindt M (2005). Emotional memory is

perceptual. Journal of Behavior Therapy and Experimental

Psychiatry 36, 19–34.

Asghar SJ, Tanay VAMI, Baker GB, Greenshaw A,

Silverstone PH (2003). Relationship of plasma

amphetamine levels to physiological, subjective,

cognitive and biochemical measures in healthy

volunteers. Human Psychopharmacology 18, 291–299.

Baddeley A, Gathercole S, Papagno C (1998). The

phonological loop as a language learning device.

Psychological Review 105, 158–173.

Baddeley A, Papagno C, Vallar G (1988). When long-term

learning depends on short-term storage. Journal of

Memory and Language 27, 586–595.

Basso A, Marangolo P, Piras F, Galluzzi C (2001).

Acquisition of new ‘words’ in normal subjects : a

suggestion for the treatment of anomia. Brain and

Language 77, 45–59.

814 E. Whiting et al.

Bond A, Lader M (1974). The use of analogue scales in rating

subjective feelings. British Journal of Medical Psychology 47,

211–218.

Brauer LH, Ambre J, de Wit H (1996). Acute tolerance to

subjective but not cardiovascular effects of d-amphetamine

in normal, healthy men. Journal of Clinical

Psychopharmacology 16, 72–76.

Breitenstein C, Knecht S (2002). Development and validation

of a language learning model for behavioral and

functional-imaging studies. Journal of Neuroscience

Methods 114, 173–179.

Breitenstein C, Wailke S, Bushuven S, Kamping S,

Zwitserlood P, Ringelstein EB, Knecht S (2004).

D-amphetamine boosts language learning independent

of its cardiovascular and motor arousing effects.

Neuropsychopharmacology 29, 1704–1714.

Burgess PW, Shallice T (1997). The Hayling and Brixton

Tests – Manual. Bury St Edmunds, UK: Thames Valley

Test Company.

Cambridge Cognition (2004). CANTABeclipse Test

Administration Guide. Cambridge, UK: Cambridge

Cognition.

Carr GD, White NM (1984). The relationship between

stereotypy and memory improvement produced by

amphetamine. Psychopharmacology 82, 203–209.

Church BA, Schacter DL (1994). Perpetual specificity of

auditory priming : implicit memory for voice intonation

and fundamental frequency. Journal of Experimental

Psychology : Learning, Memory and Cognition 20, 521–533.

Cohen NJ, Poldrack RA, Eichenbaum H (1997). Memory

for items and memory for relations in the procedural/

declarative memory framework. Memory 5, 131–178.

Cooper JR, Bloom FE, Roth RH (2003). The Biochemical Basis

of Neuropharmacology. New York : Oxford University Press.

Cornelissen K, Laine M, Renvall K, Saarinen T, Martin N,

Salmelin R (2004). Learning new names for new objects :

Cortical effects as measured by magnetoencephalography.

Brain and Language 89, 617–622.

Craik FIM, Tulving E (1975). Depth of processing and the

retention of words in episodic memory. Journal of

Experimental Psychology : General 104, 268–294.

de Wit H, Enggasser JL, Richards JB (2002). Acute

administration of d-amphetamine decreases impulsivity

in healthy volunteers. Neuropsychopharmacology 27,

813–825.

Diamond A, Briand L, Fossella J, Gehlbach L (2004).

Genetic and neurochemical modulation of prefrontal

cognitive functions in children. American Journal of

Psychiatry 161, 125–132.

Duyck W, Szmalec A, Kemps E, Vandierendonck A (2003).

Verbal working memory is involved in associative word

learning unless visual codes are available. Journal of

Memory and Language 48, 527–541.

Fenker DB, Schott BH, Richardson-Klavehn A, Heinze H-J,

Duzel E (2005). Recapitulating emotional context : Activity

of amygdala, hippocampus, and fusiform cortex during

recollection and familiarity. European Journal of

Neuroscience 21, 1993–1999.

Fillmore MT, Kelly TH, Martin CA (2005). Effects of

d-amphetamine in human models of information

processing and inhibitory control. Drug and Alcohol

Dependence 77, 151–159.

Fleming K, Bigelow LB, Weinberger DR, Goldberg TE

(1995). Neuropsychological effects of amphetamine may

correlate with personality characteristics.

Psychopharmacology Bulletin 31, 357–362.

Gathercole S (1999). Cognitive approaches to the

development of short-term memory. Trends in Cognitive

Sciences 3, 410–419.

Gronholm P, Rinne JO, Vorobyev V, Laine M (2005).

Naming of newly learned objects : a PET activation study.

Cognitive Brain Research 25, 359–371.

Gupta P (1996). Verbal short-term memory and language

processing : a computational model. Brain and Language 55,

194–197.

Gupta P, Lipinski J, Abbs B, Lin P-H, Aktunc E, Ludden D,

Martin N, Newman R (2004). Space aliens and

nonwords : stimuli for investigating the learning of

novel word-meaning pairs. Behavior Research Methods,

Instruments, and Computers 36, 599–603.

Gupta P, MacWhinney B (1997). Vocabulary acquisition

and verbal short-term memory : computational and

neural bases. Brain and Language 59, 267–333.

Izquierdo I (1994). Pharmacology of memory : drugs acting

upon the neurotransmitter mechanisms involved in

memory consolidation. In : Delacour J (Ed.), The Memory

System of the Brain, vol. 4 (pp. 365–388). Singapore : World

Scientific.

Jones NE (2004). The neurobiology of memory consolidation.

In : Schumann JH, Crowell SE, Jones NE, Lee N,

Schuchert SA, Wood LA (Eds.), The Neurobiology of

Learning : Perspectives from Second Language Learning

(pp. 111–142). Mahwah, NJ : Lawrence Erlbaum.

Knecht S, Breitenstein C, Bushuven S, Wailke S,

Kamping S, Floel A, Zwitserlood P, Ringelstein EB

(2004). Levodopa: faster and better word learning in

normal humans. Annals of Neurology 56, 20–26.

Knecht S, Imai T, Kamping S, Breitenstein C,

Henningsen H, Lutkenhoner B, Ringelstein EB (2001).

D-amphetamine does not improve outcome of

somatosensory training. Neurology 57, 2248–2252.

Kroll JF, Stewart E (1994). Category interference in

translation and picture naming: evidence for

asymmetric connections between bilingual memory

representations. Journal of Memory and Language 33,

149–174.

Kuczenski R, Segal DS (1997). Effects of methylphenidate

on extracellular dopamine, serotonin, and norepinephrine :

comparison with amphetamine. Journal of Neurochemistry

68, 2032–2037.

Kumari V, Corr PJ, Mulligan OF, Cotter PA, Checkley SA,

Gray JA (1997). Effects of acute administration of

d-amphetamine and haloperidol on procedural learning

in man. Psychopharmacology 129, 271–276.

Laine M, Cornelissen K, Renvall K, Saarinen T, Martin N,

Salmelin R (2003). Learning new objects with new

Dexamphetamine and word learning 815

names : cognitive and neural correlates of language

acquisition. Brain and Language 87, 128.

Maestu F, Simos PG, Campo P, Fernandez A, Amo C,

Paul N, Gonzales-Marques J, Ortiz T (2003). Modulation

of brain magnetic activity by different verbal learning

strategies. Neuroimage 20, 1110–1121.

Martin N, Gupta P (2004). Exploring the relationship

between word processing and verbal short-term memory :

evidence from associations and dissociations. Cognitive

Neuropsychology 21, 213–228.

Martin RC, Freedman ML (2001). Relations between

language and memory deficits. In : Berndt RS (Ed.),

Language and Aphasia, 2nd edn, vol. 3 (pp. 239–256).

Amsterdam: Elsevier Science.

Martinez JL, Jensen RA, Messing RB, Vasquez BJ,

Soumireu B, Geddes D, Liang KC, McGaugh JL (1980).

Central and peripheral actions of amphetamine on

memory storage. Brain Research 182, 157–166.

Martinez JL, Schulteis G, Weinberger SB (1991). How to

increase and decrease the strength of memory traces : the

effects of drugs and hormones. In : Martinez JL, Kesner RP

(Eds.), Learning and Memory, 2nd edn. (pp. 149–198).

San Diego : Academic Press.

Martinsson L, Eksborg S (2004). Drugs for stroke

recovery : the example of amphetamines. Drugs and

Aging 21, 67–79.

Mattay VS, Callicott JH, Bertolino A, Heaton I, Frank JA,

Coppola R, Berman KF, Goldberg TE, Weinberger DR

(2000). Effects of dextroamphetamine on cognitive

performance and cortical activation. Neuroimage 12,

268–275.

Nelson HE, Willison JR (1991). The Revised National Adult

Reading Test – Test Manual. Windsor, UK: NFER Nelson.

Randolph C (1998). Repeatable Battery for the Assessment of

Neuropsychological Status (Manual). San Antonio, TX: The

Psychological Corporation.

Rapoport JL, Buchsbaum MS, Weingartner H, Zahn TP,

Ludlow C, Mikkelsen EJ (1980). Dextroamphetamine :

Its cognitive and behavioural effects in normal and

hyperactive boys and normal men. Archives of General

Psychiatry 37, 933–943.

Rastle K, Harrington J, Coltheart M (2002). 358,534

nonwords : the ARC nonword database. Quarterly

Journal of Experimental Psychology 55A, 1339–1362.

Reynolds CR, Bigler ED (1994). Test of Memory and

Learning, Manual. Austin, TX: Pro-Ed.

Riedel W, Hogervorst E, Lebroux R, Verhey F, Van Praag H,

Jolles J (1995). Caffeine attenuates scopolamine-induced

memory impairment in humans. Psychopharmacology 122,

158–168.

Schuchert SA (2004). The neurobiology of attention. In :

Schumann JH, Crowell SE, Jones NE, Lee N, Schuchert SA,

Wood LA (Eds.), The Neurobiology of Learning : Perspectives

From Second Language Learning (pp. 143–173). Mahwah,

NJ : Lawrence Erlbaum.

Service E, Craik FIM (1993). Differences between young

and older adults in learning a foreign vocabulary. Journal

of Memory and Language 32, 608–623.

Soetens E, Casaer S, D’Hooge R, Hueting JE (1995). Effect

of amphetamine on long-term retention of verbal

material. Psychopharmacology 119, 155–162.

Soetens E, D’Hooge R, Hueting JE (1993). Amphetamine

enhances human-memory consolidation. Neuroscience

Letters 161, 9–12.

Squire LR, Knowlton B, Musen G (1993). The structure

and organization of memory. Annual Review of

Psychology 44, 453–495.

Ullman MT (2004). Contributions of memory circuits to

language: the declarative/procedural model. Cognition

92, 231–270.

Wagner RK, Torgesen JK, Rashotte CA (1999). Comprehensive

Test of Phonological Processing, Manual. Austin, TX: Pro-Ed.

Walker-Batson D, Curtis S, Natarajan R, Ford J, Dronkers

N, Salmeron E, Lai J, Unwin H (2001). A double-blind,

placebo-controlled study of the use of amphetamine in

the treatment of aphasia. Stroke 32, 2093–2098.

Walker-Batson D, Smith P, Curtis S (1995). Amphetamine

paired with physical therapy accelerates motor recovery

after stroke. Stroke 26, 2254–2259.

Wesnes K, Warburton DM (1983). Effects of smoking on

rapid information processing performance.

Neuropsychobiology 9, 223–229.

Wesnes K, Warburton DM (1984). Effects of scopolamine

and nicotine on human rapid information processing

performance. Psychopharmacology 82, 147–150.

Woodcock RW (1998). Woodcock Reading Mastery

Tests – Revised. Forms G and H Examiner’s Manual.

Circle Pines, MN: American Guidance Service.

Zacks RT, Hasher L, Li KZH (2000). Human memory. In :

Craik FIM, Salthouse TA (Eds.), The Handbook of Aging

and Cognition, 2nd edn (pp. 293–357). Mahwah, NJ :

Lawrence Erlbaum Associates.

816 E. Whiting et al.