Acta Psychologica 141 (2012) 243–249

Contents lists available at SciVerse ScienceDirect

Acta Psychologica journal homepage: www.elsevier.com/ locate/actpsy

Effects of intrinsic and extrinsic motivation on attention and memory
Lucy J. Robinson, Lucy H. Stevens, Christopher J.D. Threapleton, Jurgita Vainiute,
R. Hamish McAllister-Williams, Peter Gallagher ⁎
Institute of Neuroscience, Newcastle University, UK

a r t i c l e

i n f o

Article history:
Received 21 February 2012
Received in revised form 22 May 2012
Accepted 31 May 2012
Available online 26 June 2012
PsycINFO classification:
2300 Human Experimental Psychology
2360 Motivation and Emotion
2346 Attention

a b s t r a c t
It is well recognised that motivational factors can influence neuropsychological performance. The aim of this study was to explore individual differences in intrinsic motivation and reward-seeking and the effect of these on attentional and mnemonic processes, in the presence or absence of financial incentives. Forty participants
(18–35 years) completed two testing sessions where the Attentional Network Test (ANT) and the Newcastle
Spatial Memory Test (NSMT) were administered. After a baseline assessment, participants were re-tested after randomisation to a non-motivated (control) group or to a motivated group, where payment was contingent upon performance. Performance in the motivated group was significantly improved compared to the control group on the NSMT (condition by session; F(1,33) = 4.52, p = 0.041) and the ANT, with participants increasing performance to cued presentations within the alerting network (F(1,36) = 5.48, p = 0.025) and being less distracted by incongruent stimuli in the executive control network (F(1,36) = 6.74, p = 0.014).
There were significant negative correlations between the ‘Interest/ Enjoyment’ Intrinsic Motivation Inventory subscale and both NSMT between-search errors and ANTalerting. In the motivated group, those who had higher self-reported internal motivation were less susceptible to- or affected by- the external motivation of financial incentive. The effects of motivational factors should not be overlooked when interpreting absolute levels of performance in neuropsychological processes.
© 2012 Elsevier B.V. All rights reserved.

1. Introduction
It is well-known in cognitive research that participants vary widely in their motivation to perform well in neurocognitive tests (Locke
& Braver, 2008). Some individuals may be highly driven to perform to their best, while others may be anxious about failing or fairly indifferent, perhaps taking part for other motives (e.g. for student coursecredit or for monetary incentive). It is only by understanding more about the interaction between motivation and measured levels of cognitive function that we can develop a clearer idea of which processes are affected and how testing conditions can be optimised so as to minimise any discrepancy between an individual's potential and actual level of performance. This is not just of scientific interest, but is especially important given the extensive use of tests of cognitive function for clinical and diagnostic purposes.
A distinction has been drawn between internal (or intrinsic) and external (or extrinsic) motivation. The former indicates that behaviour is driven by factors internal to the person and has inherent value or meaning to them irrespective of the outcome. This is in

⁎ Corresponding author at: Institute of Neuroscience, Newcastle University, Academic
Psychiatry, Campus for Ageing and Vitality, Wolfson Research Centre, Newcastle General
Hospital, Newcastle upon Tyne, NE4 5PL, UK. Tel.: +44 191 208 1370; fax: +44 191 208
1387.
E-mail address: peter.gallagher@ncl.ac.uk (P. Gallagher).
0001-6918/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.actpsy.2012.05.012 contrast to extrinsic motivation, where behaviour is driven by outcome or external factors. As such, the concept of motivation can be viewed as a group of different types of driving forces, rather than a singular construct (Deci & Ryan, 1985), and one that can have state and trait elements (Tremblay, Goldberg, & Gardner, 1995). Classifications of types of extrinsic motivation have also been proposed (e.g.
Self-Determination Theory Deci & Ryan, 1985), although for the present study the distinction between extrinsic motivation (i.e. payment) and intrinsic motivation is the most relevant. Extrinsic and intrinsic motivations influence each other. Extrinsic influences are known to have a negative impact on enjoyment and motivation.
Rewards (Deci, Koestner, & Ryan, 1999), threats (Deci & Cascio,
1972), deadlines (Amabile, DeJong, & Lepper, 1976, cited in Ryan &
Deci, 2000) and competitive pressure (Reeve & Deci, 1996) can diminish one's interest and enjoyment and thus internal motivation in a task (Ryan & Deci, 2000). However, the relationship between intrinsic motivation and sensitivity to rewards or punishments needs further exploration.
Studies investigating the effects of payment on neuropsychological test performance in healthy adults have generally shown that performance improves (in terms of reaction time or accuracy) when participants are offered rewards or threatened with (financial) punishments.
Nielson and Bryant (2005) found an effect of motivation on long-term memory on an immediate and delayed recall task in healthy volunteers.
Subjects who were rewarded with a gift after immediate recall of a list

244

L.J. Robinson et al. / Acta Psychologica 141 (2012) 243–249

of words subsequently recalled more words at delayed recall (1 week later) compared to subjects who received no gift. It was suggested by the authors that monetary reward induced a positive emotional state, which influenced memory consolidation. A similar positive effect was found for working memory. Taylor et al. (2004) used an object working memory task and found not only that financial incentives improved reaction times (RTs) overall, but also that there was a dose-dependent relationship in that subjects responded more quickly in high reward tasks compared to low reward tasks. This provides evidence for a direct positive, and incremental, effect of extrinsic motivation on working memory task performance.
The mnemonic effects described may be a consequence of increased attentional focus to the task demands. A number of research studies have investigated the effects of motivation on aspects of attention. Small et al. (2005) found that monetary incentives enhanced performance on a visual–spatial attentional task and underlying brain regions consistent with top–down control mechanisms. Begleiter,
Porjesz, Chou, and Aunon (1983) found that subjects RTs were faster when given a financial incentive to perform quickly (within
350 msec) compared to subjects that received no incentive. However, this was addressing attention as a singular system while evidence has shown attentional tasks to activate differing networks of cortical areas dependent on the nature of the task (e.g. Posner & Petersen,
1990). Posner and Petersen (1990) and Fan, McCandliss, Sommer,
Raz, and Posner (2002) proposed a model of attention fractionated into three unique, but functionally related components: alerting, orienting and conflict (executive control). Alerting is referred to as the individual's capacity to establish and maintain an alert state, orienting involves prioritising attention to particular stimuli, and executive control is said to be responsible for resolving conflict among responses. To date, there has been no published investigation of the effect of motivation on all components of this model of attention.
Engelmann and Pessoa (2007) investigated orienting and reorienting elements of attention and found that perceptual sensitivity increased linearly as a function of the value of the monetary incentive. Furthermore,
Locke and Braver (2008) investigated the effects of motivation on executive control and found that financial incentive lead to an increase in executive control (indicated by increased activity in related brain areas) which significantly reduced RTs. It is uncertain how the alerting element of attention would be affected by motivation due to a paucity of research, though there have been suggestions that motivation can increase sustained attention (as proposed by Sohlberg & Mateer, 1989) in young adults (Tomporowski & Tinsley, 1996), which shares similarities with the alerting component of Fan et al.'s (2002) model.
There remain questions about the scope for external regulation to improve neuropsychological performance—which ‘higher-order’ attentional or neuropsychological control processes are susceptible to improvement using financial rewards? For the present study, we sought to focus on two tasks: (i) the Attentional Network Test
(ANT), which permits the fractionation of alerting, orienting and conflict processes/networks described above, to establish which components of this model of attention are susceptible to change; (ii) a self-ordered visuo-spatial working memory task which places demands on executive control of working memory. An additional rationale for the selection of these specific tasks was in order to focus on the neuropsychological processes (i.e. attention and memory) that are often impaired in clinical populations, especially those associated with anhedonia and loss of motivation, such as psychiatric illnesses
(Glahn et al., 2003; Porter, Gallagher, Thompson, & Young, 2003;
Robinson et al., 2006; Thompson et al., 2005). The tasks here were designed or selected to tap the processes behind commonly utilised tasks, whilst allowing an increase in level of difficulty which is acceptable for use in younger, healthy participants, thereby avoiding potential ceiling effects. A second question is what is the interaction between intrinsic motivation and external regulation—are people who are highly intrinsically motivated more or less susceptible to

the effects of offering financial incentive compared to those with lower levels of intrinsic motivation? The present study aims to investigate this question by measuring intrinsic motivation and rewardseeking then exploring the relationship between these variables and improvement in attentional and cognitive performance following monetary incentive.
2. Methods
2.1. Subjects
The sample comprised 40 healthy male subjects aged between 18 and 35 (mean 21.5 years, SD 2.5) recruited from Newcastle University and surrounding area via advert. Exclusion criteria were a serious medical condition, personal history of psychiatric illness or history of mood or psychotic disorder in a first-degree relative (ascertained through self-report), score on the Beck Depression Inventory (Beck,
Ward, Mendelson, Mock, & Erbaugh, 1961) of greater than 12, and premorbid IQ, as measured by the National Adult Reading Test
(Nelson, 1982), less than 90. The project was approved by the
Psychology Ethics Committee at Newcastle University and all subjects provided written informed consent.
2.2. Materials
The Attentional Network Test (ANT) (Fan et al., 2002) is an attentional cueing paradigm that was developed to differentiate three components of attention; alerting, orienting and executive control.
The task alters two key features of the stimulus display to probe the different components of attention, 1) the nature of the cue given to the participant to indicate when or where the target array will appear, and 2) whether the target appears in the context of congruent or incongruent stimuli. The task comprises a central fixation cross followed by an asterisk cue and then the target, which is a central arrow pointing either to the right or the left and flanked by either four other arrows or four hyphens. The participant's task is to indicate the direction that the central arrow is pointing (right or left) by clicking the right or left mouse button. There are four possible cueing conditions (no cue, a centre cue, a double cue, or an orienting cue) and three possible target conditions (congruent, incongruent, or neutral). 1 The task begins with 24 practise trials where feedback is given.
This is followed with three blocks of 48 experimental trials (n = 144 trials in total) where no feedback is given. The task is described in full elsewhere (Fan et al., 2002).
Primary outcome measures are the number of correct responses, and alerting, orienting and executive control indices.
- The alerting effect was calculated by subtracting the mean RT of the double-cue conditions from the mean RT of the no-cue conditions (therefore higher values represent the increased efficiency or speed that participants can respond when primed that an event is about to occur).
- The orienting effect was calculated by subtracting the mean RT of the spatial cue conditions from the mean RT of the centre cue
(therefore higher values represent the increased efficiency or speed that participants can respond when cued to allocate spatial attention to the location of interest).
- The conflict (executive control) effect was calculated by subtracting the mean RT of all congruent flanking conditions, summed across cue types, from the mean RT of incongruent
1
The cue provides the indication to the participant that the target ‘event’ is about to occur. In the case of the centre cue this simply primes the participant, as does the double cue condition although it does indicate that the target will be in one of the two locations. In the case of the orienting (or ‘spatial’) cue, this specifies precisely where the target will appear.

L.J. Robinson et al. / Acta Psychologica 141 (2012) 243–249

flanking conditions (therefore higher values represent the increased efficiency to overcome conflicting information).
The Newcastle 2D/3D Spatial Memory Test (NSMT) is a computerised measure of spatial memory based on the principles of several widely used paradigms, including the Executive Golf Task (Feigenbaum,
Morris, & Polkey, 1996; Morris, Pickering, Abrahams, & Feigenbaum,
1996), CANTAB Spatial Working Memory (Robbins et al., 1998), and the Box Task (van Asselen, Kessels, Wester, & Postma, 2005). The task can run in both 2D and 3D modes, although only the former was used in the present study. For the first level, participants are presented with an array of three coloured circles on the computer screen and advised to think of them as if they are looking down on three up-turned
‘cups’ on a table top. They are told that the computer has hidden marbles under the cups and it is their job to find the marbles. To look whether there is a marble under a cup, the participant clicks on the circle and either the cup is empty (solid black circle) or it contains a marble
(a white circle). Participants are told that the computer has hidden only the same number of marbles as there are cups on the screen and it will never hide a marble under the same cup twice. Participants are specifically instructed not to look under cups where they have found marbles already, as this will score an error. Two trials with three cups are used for practise, and then the experimental trials follow with two trials at each of 5 levels (4 cups, 6 cups, 8 cups, 10 cups and 12 cups). The outcome measures for this task are the number of between search errors
(the number of times a cup is searched where a marble has been found previously) and the number of within search errors (the number of times a cup is searched more than once between finding two marbles). Questionnaires: self-report rating scales were administered to assess aspects of motivational style and personality.
- The Behavioural Inhibition System and Behavioural Activation
System (BIS/BAS) questionnaire (Carver & White, 1994) assesses dimensions of aversive motivation and appetitive motivation respectively. The BIS regulates sensitivity to punishment/non-reward and novelty, and inhibits behaviours that may lead to negative consequences. The BAS scales reflect sensitivity to reward and goal-directed behaviour; as such they are separated into dimensions of drive, fun-seeking and reward responsiveness.
- The Intrinsic Motivation Inventory (IMI) was designed to assess subjective experience of an activity specifically in an experimental setting (Ryan, 1982). A shorter-form was used, comprising of four sub-scales: Interest/Enjoyment, Effort/Importance, Competence, and Pressure/Tension. Although a composite score can be derived the Interest/Enjoyment subscale is considered the self-report measure of intrinsic motivation and is therefore the primary measure of interest in the present study.

245

to one of two conditions: motivated or control. Allocation was arranged in blocks of 4 to ensure balanced allocation. Both participants and the examiner were blind to allocation for the baseline session, but in order to reveal the reward schedule, neither party were blind to allocation for the second session. Participants completed each of the tasks twice on the same day—once at 1:40 pm (baseline session) and again at 3:45 pm (experimental session). Testing was undertaken individually and there was a break between the two sessions where participants were free to do as they pleased. Participants were paid for their participation. At the start of the second testing session group allocation was revealed and the payment schedule was made known to participants. The non-motivated (control) group received a fixed honorarium of £25. The motivated group payment varied between £20 and £30 depending on performance (mean honorarium:
£25.53, SD 3.2). It was possible to earn up to £10 on each test and the payment schedule was explained before each one. For the ANT, participants were rewarded with £0.09 for each correct response that was quicker than their median reaction time from the previous testing session. For the NSMT, as the aim is to conduct an efficient search and avoid returning to boxes already searched or having contained a target, participants began with a reward of £12 and lost £0.07 for every error they made (the maximum pay-out was capped at
£10). The total amount earned was revealed and paid only at the end of the session. No running total was provided throughout.

2.6. Statistical analysis
Data were analysed using SPSS version 17. Performance on the neuropsychological tests was analysed using ANOVA with one between subjects factor (group) with two levels (motivated or control), and one within subjects factor (session) with two levels (baseline and experimental). Main effects or interaction effects were followed up by post hoc t-test. In the case of the ANT, because each of the three network components are derived by subtracting two mean reaction times under different cueing/flanker conditions (see methods section), significant interaction effects were explored further using the change scores (baseline minus the experimental session) for these additional measures to ascertain why precisely any performance change occurred (for example, an increase in the alerting measure could either be a consequence of faster responding to the doublecue or a slowing in response to the no-cue condition). To examine the correlation between questionnaires and task performance
(accounting for any imbalance between the groups at baseline), the percentage improvement from baseline to session 2 was calculated for each participant. The significance level was set to p b 0.05 (twotailed), with p b 0.1 taken to indicate a trend.

2.3. Apparatus

3. Results

The neuropsychological tests were all presented on computers with a Windows operating system. The ANT was presented using
E-prime and the NSMT was a custom-made script written and developed by Newcastle University Informatics Research Institute.

Three subjects had some missing screening data; one subject on the BDI (in the motivation condition) and two subjects did not respond to the question regarding number of years in education (one in each condition). Scores obtained from the NART for two subjects
(one from each condition) did not meet the criteria for the study and their data was not entered into the analysis. Data from a total of 38 subjects were therefore available to take forward for analysis.
Three of these subjects had missing data for the NSMT task (2 from motivated and 1 from control group).

2.4. Design
The study has a mixed factorial design with one within subjects factor (testing session: baseline or experimental) and one between subjects factor (group: motivated or non-motivated ‘control’).
2.5. Procedure

3.1. Demographic measures

Participants took part in a screening session to evaluate whether any exclusion criteria were met before the first testing session took place. On entry into the study, participants were randomly allocated

Independent samples t-tests revealed no significant differences between motivated and control subjects in age, years of education or premorbid IQ (all p > 0.1; see Table 1).

246

L.J. Robinson et al. / Acta Psychologica 141 (2012) 243–249

25

Table 1
Means and standard deviations of the demographic variables and the questionnaire measures for the subjects in the two different conditions.

Mean

t

SD

Mean

SD

21.3
16.0
111
3.8

2.5
1.7
7.9
2.3

21.7
16.5
107
5.1

2.9
1.4
10.3
4.4

20

p

15
− 0.51
− 0.91
1.61
− 1.11

0.615
0.369
0.116
0.276

10.6
17.1
12.7
20.6

2.3
1.8
2.1
3.8

11.6
17.5
12.9
19.3

2.2
1.7
2.2
3.5

− 1.31
− 0.22
− 0.64
1.06

10.8
18.6
19.1
18.2
20.7

4.1
4.9
3.7
2.7
3.7

10.5
20.3
20.4
18.9
20.3

4.4
3.4
5.5
3.1
4.0

0.23
− 1.23
− 0.86
− 0.78
0.34

0.821
0.229
0.394
0.441
0.739

19.4
30.5
28.0
29.7
30.4

8.0
6.2
5.3
5.0
4.9

20.9
31.9
24.6
28.8
31.4

8.7
5.1
6.7
5.0
7.4

− 0.51
− 0.72
1.62
0.55
− 0.46

0.612
0.477
0.115
0.585
0.646

10

5

0.199
0.824
0.528
0.295

-5

NART: National Adult Reading Test; BDI: Beck Depression Inventory.

3.2. Attentional Network Test
3.2.1. Mean accuracy
There was no statistically significant main effect of condition
(F(1,36) = 2.51, p = 0.122), session (F(1,36) = 0.03, p = 0.874), or condition by session interaction (F(1,36) = 0.23, p = 0.633).
3.2.2. Alerting
There was a significant main effect of session (F(1,36) = 7.18, p = 0.011) and a significant condition by session interaction
(F(1,36) = 5.48, p = 0.025). There was no significant main effect of condition (F(1,36) = 2.08, p = 0.158). Post hoc t-tests on the interaction revealed that there was no significant change in alerting in the control condition from session one to session two (t(18) = − 0.25, p = 0.803). However, alerting was increased significantly at session
2 in the motivated group (t(18) = −3.38, p = 0.003). To fully interpret this effect, change scores (from session 1 to session 2) in the two components of the alerting measure (no-cue and double-cue
RTs) were analysed further by ANOVA and a significant cue by condition interaction was observed (F(1,36) = 5.64, p = 0.023). As can be seen in Fig. 1, the increased alerting effect in the motivated group was a consequence of an improvement in reaction time to the double-cue trials.
3.2.3. Orienting
There was no significant main effect of session (F(1,36) =
0.01, p = 0.922) and no significant condition by session interaction
(F(1,36)= 0.33, p = 0.570). There was a trend towards a main effect of condition, with orienting being non-significantly higher in the motivated group (F(1,36)= 3.55, p =0.068).
3.2.4. Executive control (conflict)
There were no significant main effects of session (F(1,36)=0.86, p=0.361) or condition (F(1,36)=0.27, p=0.609) although there was a significant condition by session interaction (F(1,36)=6.74, p=0.014).
Post hoc t-tests on the interaction revealed that there was no significant change in the conflict measure in the control condition from session 1 to session 2 (t(18)=−1.03, p=0.316). However, the conflict measure

-10
Control

Motivated
No Cue

Control

Motivated

Double Cue

Fig. 1. Change in RT in ANT alerting components (baseline to session 2) Positive values indicate improved performance (faster RTs) at session 2.

decreased significantly at session 2 in the motivated group (t(18)=
3.00, p=0.008). To interpret this effect, the individual components of the conflict effect (RTs to congruent and incongruent conditions) were analysed further. ANOVA conducted on the change scores revealed a significant cue by condition interaction (F(1,36)=6.66, p=0.014). As can be seen in Fig. 2, the decreased conflict effect in the motivated group was a consequence of an improvement in reaction time to incongruent
‘flankers’ at session 2.
3.3. NSMT
3.3.1. Between search errors
The repeated measures ANOVA revealed no significant main effect of condition (F(1,33) = 0.25, p = 0.621), but there was a significant effect of session (F(1,33) = 5.67, p = 0.023) and a significant condition by session interaction (F(1,33) = 4.52, p = 0.041). Further exploring the interaction indicated that performance improved significantly between sessions in the motivated group (t(16) = 2.58,
40

30

20

RT (ms)

Demographics
Age (years)
Full-time education (years)
NART
BDI
BIS/BAS scales
Behavioural Activation System
Drive
Reward Responsiveness
Fun-seeking
Behavioural Inhibition System
Intrinsic Motivation Inventory
Pressure and tension
Interest/enjoyment
Value/usefulness
Perceived competence
Effort/importance
Personality Inventory
Neuroticism
Extraversion
Openness
Agreeableness
Conscientiousness

Control

RT (ms)

Motivated

10

-10

-20

Control

Motivated

Incongruent

Control

Motivated

Congruent

Fig. 2. Change in RT in ANT conflict components (baseline to session 2). Positive values indicate improved performance (faster RTs) at session 2.

L.J. Robinson et al. / Acta Psychologica 141 (2012) 243–249

p = 0.020) while there was no significant change in the control group
(t(17) = 0.25, p = 0.808). There was no significant difference between the groups at baseline (t(33) = 0.64, p = 0.530), although there was a trend towards a significant difference at the second session (t(33) = − 1.84, p = 0.074) with errors being lower in the motivated group (see Fig. 3 and Table 2).
3.3.2. Within search errors
The repeated measures ANOVA revealed no significant main effect of condition (F(1,33) = 0.39, p = 0.536), but a significant effect of session
(F(1,33) = 6.46, p =0.016) and a significant group by session interaction
(F(1,33) = 4.83, p = 0.035). Further exploring the interaction indicated that performance improved significantly between sessions in the motivated group (t(16) = 3.73, p = 0.002) while there was no significant change in performance in the control group (t(17)= 0.23, p = 0.824).
There were no significant differences between the groups at either baseline (t(33) = 1.53, p = 0.136) or the second session (t(33) = −0.81, p = 0.421) (See Table 2).
3.4. Practise effects
In order to exclude the possibility of simple practise effects accounting for changes, paired t-tests were used to examine performance in the control condition over the two sessions. There was no significant change in performance on any of the primary outcome measures (p > 0.3 for all).
3.5. Relationship between performance improvement and motivation measures Correlations between self-report motivation measures and change in test scores between sessions in the motivated group are shown in
Table 3. The main measures of interest are ‘reward responsiveness’ from the BIS/BAS and the principal intrinsic motivation measure from the IMI (the Interest/Enjoyment subscale). As can be seen, there were no significant correlations with reward responsiveness although there was a trend towards a greater improvement in NSMT within search errors being associated with higher reward responsiveness. Similarly there was a positive relationship between BAS ‘fun seeking’ and an increased orienting response when externally

247

Table 2
Means and standard deviations (SD) for motivated and control subjects on session one
(baseline) and session two (experimental) of testing.
Task Scores

NSMT; within search errors
ANT; accuracy (%)
ANT; alerting
ANT; orienting
ANT; executive control

1
2
1
2
1
2
1
2
1
2
1
2

Motivated

Control

Mean
NSMT; between search errors

Session

SD

Mean

SD

64.1
39.4
14.6
5.5
97.5
97.3
32.5
45.0
47.3
48.8
103.5
90.2

(35.7)
(22.4)
(11.9)
(5.2)
(1.9)
(2.1)
(20.6)
(26.5)
(18.6)
(18.2)
(19.9)
(22.1)

56.7
55.3
8.6
7.9
94.5
95.0
47.7
48.6
39.8
37.6
89.7
96.0

(32.6)
(28.2)
(11.7)
(11.1)
(9.2)
(5.4)
(19.7)
(18.3)
(19.3)
(16.4)
(30.4)
(32.4)

NSMT: Newcastle Spatial Memory Test; ANT: Attentional Network Task

motivated. Interestingly, there were negative relationships between the ‘Interest/Enjoyment’ Intrinsic Motivation subscale and both
NSMT between search errors and ANT alerting. The direction of the relationship indicated that higher scores on this intrinsic motivation subscale (which indicate a higher level of interest/enjoyment) were associated with less improvement in memory and a reduced ‘cueing’ advantage in attention, following external motivation. No significant correlations were observed between these measures in the nonmotivated control group.
3.6. Relationship between baseline task performance and intrinsic motivation One question is whether individuals who are highly intrinsically motivated approach the tasks differently to those with lower motivation and consequently perform better. Correlations between baseline
ANT and NSMT performance in the whole sample, and IMI ‘Interest/
Enjoyment’ were examined. A significant negative correlation was observed between the intrinsic motivation measure and NSMT between search errors (rs = − 0.431, p = 0.01) and a trend for within search errors (rs = − 0.330, p = 0.053). No significant correlations were found on ANT measures.

40

4. Discussion
35

Errors (change from baseline)

Control
30

Motivated

The current study investigated the effects of monetary incentive on neuropsychological test performance assessing attention and visuo-spatial memory. Monetary incentive improved performance

25
20

Table 3
Spearman's correlations between intrinsic motivation, BIS/BAS and change in performance from baseline to session two in the motivated group.

15

NSMT

10

BSE

5

ANT
WSE

a

Alerting

Orienting Conflict

Intrinsic Motivation Inventory
Interest/Enjoyment
− 0.615b − 0.289 − 0.520c − 0.018
BIS/BAS
BIS
0.006 − 0.014 − 0.181 −0.075
0.106
0.154
(BAS) reward responsiveness
0.052
0.414d
(BAS) drive
− 0.074
0.359 − 0.141
0.224
(BAS) fun seeking
− 0.204
0.023
0.288
0.472c

0
-5
-10
BSE (change)

WSE (change)

Fig. 3. Change in Spatial Working Memory between and within search error rate (baseline to session 2). Higher values indicate greater improvement at session 2.

− 0.020
0.168
− 0.147
− 0.257
− 0.179

a
The signs of the coefficients are reversed so as to be congruent and comparable with the NSMT values. b p b 0.01. c p b 0.05. d p b 0.1.

248

L.J. Robinson et al. / Acta Psychologica 141 (2012) 243–249

on the spatial memory task and altered some aspects of the attentional task, with participants increasing performance to cued presentations within the alerting network and being less distracted by incongruent stimuli in the executive control network. Additionally, in the participants randomised to the motivated condition there were significant correlations between performance measures on both tasks and intrinsic motivation. Overall, individuals with higher intrinsic motivation made fewer errors on the NSMT at baseline.
Previous studies have found that overall sustained attention in younger adults is maintained longer when volunteers were paid for their participation compared to a decline when not paid (Tomporowski & Tinsley,
1996). However, to our knowledge this is the first study to directly examine the effects of motivation on the separable attentional networks from the ANT, although previous studies have utilised other paradigms to examine similar processes. Using a CPT-AX task (with high target frequency:
AX=70%), Locke and Braver (2008) examined motivational influences on cognitive/attentional control mechanisms, both behaviourally and with fMRI. Focussing on the former, in the reward blocks (where 25 cents was available for every response faster than an individual's overall median RT at baseline) participants were significantly faster in their responses in AX trials, without an increase in overall errors. When all trial conditions were analysed across baseline, reward blocks and penalty blocks (where $3 was taken away for every incorrect response) it was observed that when rewarded, average response times were faster than at baseline in all conditions with only ‘AY’ trials showing a parallel increase in errors. The overall pattern of results was described as being in line with the use of a ‘proactive control’ strategy in the reward condition i.e. the active maintenance of context information to prime the selection of responses, suggesting that extrinsic reward motivation elicited an increase in cognitive control to focus attention and limit interference (Locke & Braver, 2008). These findings appear to be in accord with the improved executive control (conflict) seen in the ANT in the present study. The lack of effect of extrinsic reward on the orienting component of the ANT contradicts the findings of Engelmann and Pessoa
(2007), who found a linear relationship between financial incentives and orienting efficiency in a spatial cueing task. It is not clear why orienting was not affected in the present study, although it may be down to differences in the specific demands of each task; this requires further examination. Of particular interest is the application and assessment of the framework proposed by Pessoa on the brain networks underpinning emotion and motivation, and their effects on higher cognitive control (Pessoa, 2008, 2009).
With regard to the visuo-spatial memory findings, although employing a different task to that used in the present study, Taylor and colleagues demonstrated that motivation, through the use of financial reward/loss, altered executive control of working memory.
Participants were more willing to experience higher false alarms to achieve more hits and fewer misses. There was also a trend towards better target detection in the higher motivation condition (Taylor et al., 2004).
In terms of the relationship observed between task performance and self-report internal motivation, the pattern of correlations suggests that those who had higher internal motivation – in terms of seeing the study as intrinsically valuable and enjoyable – were less susceptible to (or less affected by) the external motivation of financial intervention. Along with evidence of a negative impact of extrinsic motivation on intrinsic motivation (Deci et al., 1999) this suggests a bi-directional inhibition of one type of motivation on the other. This finding should be explored further to see if there are levels of external motivation which would overcome this effect and examine further the balance between level of internal vs. external motivation.
There are several limitations of the present study that need to be addressed in future studies. It was not established whether the different task outcome measures were equally sensitive for detecting change in performance. This may have accounted for the differences

and selectivity observed in response to financial incentive. Additionally, the different performance indices meant some tasks were motivated by increased reward, but others by threatened punishment i.e. for the ANT it was speed of response which was rewarded while for the NSMT it was efficient searching and therefore avoidance of error. While this was unavoidable given the specific demands of the tasks, participants may respond differently to these alternative external financial motivators, especially with regard to their go/no-go nature. The issue of how financial (monetary) incentives are applied in studies of this nature is particularly important given recent findings on the complexity and specificity of this effect. Dambacher, Hübner, and Schlösser (2011) have shown that within the context of a speeded categorization task (the Erikson flanker task), different effects were observed depending on how penalties were applied (i.e. to errors or to slow responses). When financial punishments were applied to slower responses, participants were strongly motivated to respond more quickly but without a concomitant effect on accuracy. However, with penalties applied to errors – but not slower responses – similar performance improvements did not occur. The authors argue that
“successful strategies to optimize profit are more easily exerted in settings that require control of speed rather than of accuracy. This notion appears to be compatible with the proposal of two systems that are involved in decision making: a fast pathway that makes rapid binary choices and a slow but more accurate cognitive system”. Interestingly,
Chiew and Braver (2011a) have highlighted the utility of applying diffusion model analysis as an alternative approach to examining speed–accuracy functions such as these. This approach permits the separation of multiple component features of decision making in
‘two-choice’ discrimination tasks (Ratcliff & McKoon, 2008). Using a similar flanker paradigm to Dambacher and colleagues, Chiew and
Braver reported that they also observed a speed–accuracy shift with reward, but only under conditions where a cue was displayed, relating to the occurrence or absence of conflict in the subsequent array, did the flanker-effect diminish with incentive. Following the application of a preliminary diffusion model analysis it was also noted that
“drift rate [the quality of accumulated information] improved under incentive only in the presence of these preparatory cues, while the speed– accuracy shift under reward in the absence of these cues was associated with a change in both response caution and non-decision time” (Chiew and Braver (2010); cited in Chiew and Braver (2011a).
Additional consideration should also be paid to the nature of the rewards. In the present study, both groups were paid a similar amount irrespective of improvement in performance from baseline to the second session. This was included to try and prevent an overly demotivating effect of allocation to the control group. Also, the payment schedule utilised was complex and may have been difficult to keep track of. Making the rewards and costs more transparent, for example by providing a running tally, may increase the saliency of the rewards/costs and may prove more motivating.
4.1. Implications and conclusions
The wider implications of these findings are also important to note.
There are a growing number of studies investigating ‘cognitive function’ in a range of clinical psychiatric conditions. It is increasingly accepted that people with major depression (Porter et al., 2003; Porter, Bourke,
& Gallagher, 2007; Zakzanis, Leach, & Kaplan, 1998), bipolar disorder
(Kurtz & Gerraty, 2009; Robinson et al., 2006; Thompson et al., 2005), obsessive compulsive disorder (Deckersbach et al., 2004; Purcell,
Maruff, Kyrios, & Pantelis, 1998), and eating disorders (CastroFornieles et al., 2010), to name but a few, show impairments when tested on neuropsychological measures. One of the most common interpretations of these deficits is that they reflect an illness-related disturbance in the structural or functional neural architecture called upon in completing the affected tests. By this view, test performance provides

L.J. Robinson et al. / Acta Psychologica 141 (2012) 243–249

information about pathophysiology, either by indicating neural pathways affected by the disorder or highlighting processing biases that might have knock-on consequences for the sense that the sufferer is able to make of the world. This perspective assumes a particular view of cognitive function as a quantifiable and reliably measurable property of the brain. Accompanying this view is an implicit assumption that cognition, motivation and emotion are processes that can be separated and measured independently of one another (for a discussion of the distinction between emotion and motivation see Chiew & Braver, 2011b). Poor performance on a memory test is taken to reflect impaired memory processes, rather than deficiencies in any of the other processes involved in the task or poor motivation to perform well. The fact that patient groups often show selective impairments on only some neuropsychological tests in some way supports this view (although of course tests may differ in their sensitivity to detect impairment). Also, the fact that motivation itself has a neurobiology which may be core to the nature of some psychiatric illnesses has previously been noted (Austin, Mitchell, &
Goodwin, 2001). Nevertheless, the influence of these factors on performance should be considered and may offer unique insights into their role in mediating behaviour. As seen in the present study, intrinsic motivation may affect overall task performance and it is possible that this may be lowered in many clinical conditions. While it is possible that this negative internal state leads to poorer cognitive performance, it may be that – due to the overlap between the ‘emotional’ and ‘cognitive’ regions as localised within distributed networks within the brain – they actually represent highly overlapping processes (Pessoa, 2008). These represent important avenues for future research, especially within mood disorders.
The present study has provided further evidence that neuropsychological function is impacted by motivation. However, it has also indicated that not all cognitive processes may be influenced by external motivation and there may be a complex interaction with an individual's intrinsic motivation that modulates the impact of external rewards.
Acknowledgements
We thank Professor Patrick Olivier and Daniel Jackson (Culture Lab,
Informatics Research Institute, Newcastle University) for the design and programming of the Newcastle 2D/3D Spatial Memory Test (NSMT).
References
Amabile, T. M., DeJong, W., & Lepper, M. R. (1976). Effects of externally-imposed deadlines on subsequent intrinsic motivation. Journal of Personality and Social Psychology,
34, 92–98.
Austin, M. P., Mitchell, P., & Goodwin, G. M. (2001). Cognitive deficits in depression:
Possible implications for functional neuropathology. The British Journal of Psychiatry, 178, 200–206.
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561–571.
Begleiter, H., Porjesz, B., Chou, C. L., & Aunon, J. I. (1983). P3 and stimulus incentive value. Psychophysiology, 20(1), 95–101.
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation, and affective responses to impending reward and punishment: The BIS/BAS scales.
Journal of Personality and Social Psychology, 67, 319–333.
Castro-Fornieles, J., Caldú, X., Andrés-Perpiñá, S., Lázaro, L., Bargalló, N., Falcón, C., & Junqué,
C. (2010). A cross-sectional and follow-up functional MRI study with a working memory task in adolescent anorexia nervosa. Neuropsychologia, 48(14), 4111–4116.
Chiew, K. S., & Braver, T. S. (2010). Selective attention and conflict processing under primary motivational incentives: Evidence from the flanker task. Poster presented at the Annual Meeting of the Cognitive Neuroscience Society, Montreal, QC.
Chiew, K. S., & Braver, T. S. (2011a). Monetary incentives improve performance, sometimes: Speed and accuracy matter, and so might preparation. Frontiers in Psychology,
2(325), http://dx.doi.org/10.3389/fpsyg.2011.00325.
Chiew, K. S., & Braver, T. S. (2011b). Positive affect versus reward: Emotional and motivational influences on cognitive control. Frontiers in Psychology, 2(279), 1–10.
Dambacher, M., Hübner, R., & Schlösser, J. (2011). Monetary incentives in speeded perceptual decision: Effects of penalizing errors versus slow responses. Frontiers in
Psychology, 2(248), http://dx.doi.org/10.3389/fpsyg.2011.00248.
Deci, E. L., & Cascio, W. F. (1972). Changes in intrinsic motivation as a function of negative feedback and threats. Poster presented at the the meeting of the Eastern
Psychological Association, Boston.

249

Deci, E. L., Koestner, R., & Ryan, R. M. (1999). A meta-analytic review of experiments examining the effects of extrinsic rewards on intrinsic motivation. Psychological
Bulletin, 125(6), 627–668.
Deci, E. L., & Ryan, R. M. (1985). Intrinsic motivation and self determination in human behaviour. New York: Plenum Press.
Deckersbach, T., Savage, C. R., Reilly-Harrington, N., Clark, L., Sachs, G., & Rauch, S. L.
(2004). Episodic memory impairment in bipolar disorder and obsessive–compulsive disorder: The role of memory strategies. Bipolar Disorders, 6(3), 233–244.
Engelmann, J. B., & Pessoa, L. (2007). Motivation sharpens exogenous spatial attention.
Emotion, 7(3), 668–674.
Fan, J., McCandliss, B. D., Sommer, T., Raz, A., & Posner, M. I. (2002). Testing the efficiency and independence of attentional networks. Journal of Cognitive Neuroscience, 14(3),
340–347.
Feigenbaum, J. D., Morris, R. G., & Polkey, C. E. (1996). Deficits in spatial working memory after unilateral temporal lobectomy in man. Neuropsychologia, 34(3), 163–176.
Glahn, D. C., Therman, S., Manninen, M., Huttunen, M., Kaprio, J., Lonnqvist, J., &
Cannon, T. D. (2003). Spatial working memory as an endophenotype for schizophrenia. Biological Psychiatry, 53(7), 624–626.
Kurtz, M. M., & Gerraty, R. T. (2009). A meta-analytic investigation of neurocognitive deficits in bipolar illness: Profile and effects of clinical state. Neuropsychology
Review, 23(5), 551–562.
Locke, H. S., & Braver, T. S. (2008). Motivational influences on cognitive control: Behavior, brain activation, and individual differences. Cognitive, Affective, & Behavioral
Neuroscience, 8(1), 99–112.
Morris, R. G., Pickering, A., Abrahams, S., & Feigenbaum, J. D. (1996). Space and the hippocampal formation in humans. Brain Research Bulletin, 40(5–6), 487–490.
Nelson, H. E. (1982). National Adult Reading Test, NART. Windsor: Nelson Publishing
Company.
Nielson, K. A., & Bryant, T. (2005). The effects of non-contingent extrinsic and intrinsic rewards on memory consolidation. Neurobiology of Learning and Memory, 84(1), 42–48.
Pessoa, L. (2008). On the relationship between emotion and cognition. Nature Reviews
Neuroscience, 9(2), 148–158.
Pessoa, L. (2009). How do emotion and motivation direct executive control? Trends in
Cognitive Sciences, 13(4), 160–166.
Porter, R. J., Bourke, C., & Gallagher, P. (2007). Neuropsychological impairment in major depression—its nature, origin and clinical significance. The Australian and New
Zealand Journal of Psychiatry, 41, 115–128.
Porter, R. J., Gallagher, P., Thompson, J. M., & Young, A. H. (2003). Neurocognitive impairment in drug-free patients with major depressive disorder. The British Journal of Psychiatry, 182(3), 214–220.
Posner, M. I., & Petersen, S. E. (1990). The attention system of the human brain. Annual
Review of Neuroscience, 13, 25–42.
Purcell, R., Maruff, P., Kyrios, M., & Pantelis, C. (1998). Cognitive deficits in obsessive– compulsive disorder on tests of frontal–striatal function. Biological Psychiatry,
43(5), 348–357.
Ratcliff, R., & McKoon, G. (2008). The diffusion decision model: Theory and data for two-choice decision tasks. Neural Computation, 20(4), 873–922.
Reeve, J., & Deci, E. L. (1996). Elements of the competitive situation that affect intrinsic motivation. Personality and Social Psychology Bulletin, 22(1), 24–33.
Robbins, T. W., James, M., Owen, A. M., Sahakian, B. J., Lawrence, A. D., McInnes, L., &
Rabbitt, P. M. (1998). A study of performance on tests from the CANTAB battery sensitive to frontal lobe dysfunction in a large sample of normal volunteers: Implications for theories of executive functioning and cognitive aging. Journal of International Neuropsychological Society, 4(5), 474–490.
Robinson, L. J., Thompson, J. M., Gallagher, P., Goswami, U., Young, A. H., Ferrier, I. N., &
Moore, P. B. (2006). A meta-analysis of cognitive deficits in euthymic bipolar subjects. Journal of Affective Disorders, 93(1–3), 105–115.
Ryan, R. M. (1982). Control and information in the intrapersonal sphere: An extension of cognitive evaluation theory. Journal of Personality and Social Psychology, 43, 450–461.
Ryan, R. M., & Deci, E. L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25, 54–67.
Small, D. M., Gitelman, D., Simmons, K., Bloise, S. M., Parrish, T., & Mesulam, M. -M.
(2005). Monetary incentives enhance processing in brain regions mediating top– down control of attention. Cerebral Cortex, 15(12), 1855–1865.
Sohlberg, M. M., & Mateer, C. A. (1989). Introduction to cognitive rehabilitation: Theory and practice. New York: Guilford Press.
Taylor, S. F., Welsh, R. C., Wager, T. D., Luan Phan, K., Fitzgerald, K. D., & Gehring, W. J.
(2004). A functional neuroimaging study of motivation and executive function.
NeuroImage, 21(3), 1045–1054.
Thompson, J. M., Gallagher, P., Hughes, J. H., Watson, S., Gray, J. M., Ferrier, I. N., &
Young, A. H. (2005). Neurocognitive impairment in euthymic bipolar disorder.
The British Journal of Psychiatry, 186, 32–40.
Tomporowski, P. D., & Tinsley, V. F. (1996). Effects of memory demand and motivation on sustained attention in young and older adults. The American Journal of Psychology,
109(2), 187–204.
Tremblay, P. F., Goldberg, M. P., & Gardner, R. C. (1995). Trait and state motivation and the acquisition of Hebrew vocabulary. Canadian Journal of Behavioural Science, 27, 356–370. van Asselen, M., Kessels, R. P. C., Wester, A. J., & Postma, A. (2005). Spatial working memory and contextual cueing in patients with Korsakoff amnesia. Journal of Clinical and Experimental Neuropsychology, 27, 645–655.
Zakzanis, K. K., Leach, L., & Kaplan, E. (1998). On the nature and pattern of neurocognitive function in major depressive disorder. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 11(3), 111–119.