Cognitive Development 43 (2017) 224–234

Contents lists available at ScienceDirect

Cognitive Development
journal homepage: www.elsevier.com/locate/cogdev

The effects of working memory training on improving fluid MARK

intelligence of children during early childhood
⁎,1
Jun Peng1, Lei Mo , Ping Huang, Ying Zhou
Center for Studies of Psychological Application & School of Psychology, South China Normal University, Guangzhou, China

AR TI CLE I NF O AB S T R A CT

Keywords: Early childhood is a phase of rapid development in many aspects, and early interventions can be
Working memory more effective than interventions provided later on. In prior research, working memory training
Intervention has demonstrated the possibility to improve the fluid intelligence of both adults and school-aged
Fluid intelligence children. This study examined the training effects of enhancing the development of fluid in-
Early childhood
telligence in the long-term for preschoolers. Seventy-four preschoolers participated in the ex-
periment. Participants in the experimental group were trained for 14 days using n-back program.
Fluid intelligence tests were administered four times: pretest, post-test, 6-month follow-up, and
12-month follow-up. Results indicated that, the experimental group significantly enhanced their
working memory performance. The experimental group also significantly outperformed the two
control groups on the fluid intelligence post test and maintained their superior performance for
up to 12 months. Implications are discussed within the context of providing effective early
childhood interventions to improve fluid intelligence.

1. Introduction

From birth to 6 years old, children experience a critical period of development in multiple domains (Piaget, 1981). Important
cognitive functions such as language, perceptual processing, emotional control, and social skills yield rapid development during this
period. For example, Fenson et al. (1994) reported a period of rapid language acquisition between 2 and 4 years. Besides, children
rapidly improve their verbal memory during this period: between the age of 3 and 5 years, children are able to verbally report events
using their developing memory ability (Simcock & Hayne, 2003). Development of communication skills and self-discipline is also
significant during early childhood (Man, 2013). Many evidences from neuroscience researches illustrate that early childhood is the
important period of brain development. The brain develops rapidly in early childhood and reaches approximately 95% of the size of
an adult brain by age 6 (Lenroot & Giedd, 2006). The developmental trajectory of the brain is influenced by both genetic and
environmental factors (Lenroot & Giedd, 2006). If children suffer from severe sensory deprivation, their brains may be significantly
smaller than average and demonstrate abnormal development of the cortex (Perry & Pollard, 1997).
The development of intelligence during early childhood is rapid. Intelligence is defined as a general mental ability that mainly
involves comprehending, reasoning, problem solving, and efficient learning. Individual intelligence is associated with the lateral
prefrontal cortex and parietal association cortex (Deary, Penke, & Johnson, 2010). These brain areas are highly malleable during
early childhood (Lenroot et al., 2009). For children with normal intelligence, the thickness of their cortex reaches its maximum when
they are about 5.6 years old (Shaw et al., 2006), suggesting the rapid brain growth during this critical stage of development.

⁎
Corresponding author at: South China Normal University, Guangzhou 510631, China.
E-mail address: molei@scnu.edu.cn (L. Mo).
1
These authors contributed equally to the work.

http://dx.doi.org/10.1016/j.cogdev.2017.05.006
Received 3 June 2015; Received in revised form 27 March 2017; Accepted 17 May 2017
0885-2014/ © 2017 Published by Elsevier Inc.
J. Peng et al. Cognitive Development 43 (2017) 224–234

Special attention should be given to children's development of fluid intelligence at this stage. According to Cattell, fluid in-
telligence is an important dimension of general intelligence. It refers to the ability to solve abstract relational problems and has been
considered to be free of cultural influences (Carpenter, Just, & Shell, 1990). Unlike crystallized intelligence, which is shaped by past
experiences, fluid intelligence is considered partly hereditary and independent of acquired knowledge (Cattell, 1971). Fluid in-
telligence improves steadily during childhood, typically peaks during adolescence, and exhibits an early and regular aging loss
(Baltes & Kliegl, 1986; Horn & Cattell, 1967), which indicates that fluid intelligence is primarily associated with biological func-
tioning and brain status (Aizpurua & Koutstaal, 2010; Manard, Carabin, Jaspar, & Collette, 2014; Schretlen et al., 2000). Fluid in-
telligence is highly predictive for academic achievement and future accomplishments (Deary, Strand, Smith, & Fernandes, 2007;
Neisser et al., 1996; Rohde & Thompson, 2007; te Nijenhuis, van Vianen, & van der Flier, 2007).
Experience and environmental factors play essential roles in shaping intelligence during early childhood (Bruer, 1997; Viadero,
1996). A substantial body of research (Barnett, 1995; Batty, Deary, & Gottfredson, 2007) has established the importance of providing
intervention programs during early childhood to improve intelligence over the long term. Barnett (1995) reported that programs
designed for disadvantaged children can yield immediate boosts of about eight IQ points. Nisbett et al. (2012) suggested that early
childhood interventions aimed at improving IQ can lead to significant effects for later academic and life achievements. Therefore, it is
possible that effective early interventions can produce long-term effects in individuals’ intelligence.

1.1. Enhancing development of fluid intelligence through working memory training

Although it is commonly agreed upon that academic interventions can improve academic achievements, it is still controversial
whether cognitive interventions can improve the fluid intelligence. Recent studies have found that through cognitive training,
especially working memory training, it is possible to enhance individuals’ performances in tasks measuring fluid intelligence (Jaeggi,
Buschkuehl, Jonides, & Perrig, 2008; Jaeggi, Buschkuehl, Jonides, & Shah, 2011; Jausovec & Jausovec, 2012; Klingberg et al., 2005;
Klingberg, Forssberg, & Westerberg, 2002; Stephenson & Halpern, 2013). Working memory is a memory system with limited capacity.
It temporarily maintains and manipulates important information while simultaneously inhibiting irrelevant information (Baddeley,
2003; Baddeley & Hitch, 1974; Baddeley & Hitch, 1994). Working memory capacity is predictive of multiple cognitive abilities such as
reading (Baddeley, 1992; Daneman & Carpenter, 1980), writing (McCutchen, 1996; Swanson & Berninger, 1996), learning English as
a second language (Abu-Rabia, 2003), and arithmetic (De Smedt et al., 2009; Destefano & Lefevre, 2004; Gathercole et al., 2004).
Children with learning disabilities in reading or mathematics were found to have small working memory capacity (Chiappe,
Hasher, & Siegel, 2000; Swanson, 1993, 1994; Swanson & Beebefrankenberger, 2004).
A strong correlation between working memory and fluid intelligence has been documented in prior research (Colom, Abad,
Quiroga, Shih, & Flores-Mendoza, 2008; Colom, Rebollo, Palacios, Juan-Espinosa, & Kyllonen, 2004; Conway, Cowan, Bunting,
Therriault, & Minkoff, 2002; Conway, Kane, & Engle, 2003; Engel de Abreu, Conway, & Gathercole, 2010; Kane, Hambrick, & Conway,
2005). For instance, a confirmatory factor analysis by Kyllonen and Christal (1990) reported high correlations between working
memory capacity and factors contributing to reasoning ability. Neuropsychological studies have indicated that working memory and
fluid intelligence share similar neural networks, which are mainly located in the dorsolateral prefrontal and parietal cortex (Gray,
2002; Gray, Chabris, & Braver, 2003; Gray & Thompson, 2004; Kane & Engle, 2002), which may explain the mechanism underlying
the relation between working memory and fluid intelligence.
An increasing number of studies have emerged during the last decade reporting that working memory training can enhance
cognitive abilities, including fluid intelligence (Klingberg, 2010). A recent meta-analysis (Weicker, Villringer, & Th & ne-Otto, 2016)
analyzed 103 studies of working memory training and found that training lead to long-lasting improvements on reasoning/in-
telligence and cognitive control functions. They suggested that the working memory training had a long-lasting beneficial effect on
cognitive function of brain injured patients. Besides, Au et al. (2015) put forward that there was a small but significant positive effect
of a specific training program (n-back training) on improving fluid intelligence among healthy adults in their meta-analysis. Simi-
larly, another meta-analysis (Karbach & Verhaeghen, 2014) suggested that the working memory trainings were effective on near- and
far-transfer (i.e., measuring the same and different task constructs, respectively). Another meta-analysis (Melby-Lervåg & Hulme,
2013) and two systematic reviews (Morrison & Chein, 2011; Shipstead, Redick, & Engle, 2012) synthesized empirical studies ex-
amining the effects of working memory training on related cognitive skills and reported consistent and positive immediate im-
provements but mixed transfer effects. There are multiple types of training programs, including Cogmed training (RoboMemo© from
Cogmed Cognitive Medical Systems AB, Stockholm, Sweden; Beck, Hanson, Puffenberger, Benninger, & Benninger, 2010; Bergman
Nutley et al., 2011; Dahlin, 2011), Jungle memory (Alloway & Alloway, 2009), n-back training (Jaeggi et al., 2008, 2010; Li et al.,
2008; Seidler et al., 2010), and running span training (Dahlin, Neely, Larsson, Bäckman, & Nyberg, 2008; Dahlin, Nyberg,
Bäckman, & Neely, 2008; Zhao, Wang, Zhou, Wang, & Tan, 2011). Among the various training methods, the n-back training was
found to be more effective than training using simple/complex tasks (Shipstead et al., 2012).
While positive effects of working memory training were reported, researchers also expressed concerns about the effects of the
training (Melby-Lervåg & Hulme, 2013; Shipstead et al., 2012). Firstly, some studies reported that participants were unable to
transfer the improved working memory abilities to enhance their performance on other tasks, including academic achievement tests
(Holmes, Gathercole, & Dunning, 2009; Horowitz-Kraus & Breznitz, 2009; Van Der Molen et al., 2010), intelligence tests (Bergman
Nutley et al., 2011; Redick et al., 2013; Westerberg et al., 2008), and attention tasks (Dahlin, Nyberg et al., 2008; Van Der Molen
et al., 2010; Westerberg et al., 2007). Secondly, the sources of these changes are still unclear; further research is necessary to show
that the improvement represents a true growth of cognitive abilities instead of a test-retest effect (Colom et al., 2010).
Some researchers suspected that the improvement is only temporary and that the maintenance effects are questionable. In

225
J. Peng et al. Cognitive Development 43 (2017) 224–234

addition, mixed maintenance effects for both verbal and visual working memory training were reported in a meta-analysis by Melby-
Lervåg and Hulme (2013). Only a very small number of existing studies have reported the maintenance effects and it may causes
biases in meta-analyses that summarize and evaluate the WM training effects. More empirical research is warranted to examine the
maintenance effects of working memory training with different programs and for different groups of individuals.

1.2. Working memory training for preschool-aged children

Age was found to significantly moderate the effectiveness of working memory training (Melby-Lervåg & Hulme, 2013). Specifi-
cally, young children (under the age of 10) showed the greatest improvement after receiving working memory training compared
with older children (11–18 years old) and adults. A review on working memory training effectiveness (Shipstead et al., 2012)
suggested that working memory training programs targeting children yielded positive results whereas training programs for adults
reported mixed effects.
Despite the significant training effects found in young children, preschoolers and infants were rarely included in the reported
studies. In the review by Shipstead et al. (2012), only two of the 11 reviewed studies included 4- to 6-year old children. In the Thorell
et al. (2009) study, 35 4- to 5-year old children participated in the experiment. They were randomly assigned to receive either
inhibition training or working memory training using a software product (RoboMemo© from Cogmed Cognitive Medical Systems AB,
Stockholm, Sweden) and were tested on tasks measuring working memory, executive function, problem solving, and attention.
Results showed that children who received working memory training significantly improved their working memory performance and
that the improvement was successfully transferred to both visual and auditory attention. Given the encouraging results of working
memory transfer effects in young children, it is plausible to hypothesize that working memory training during early childhood can
produce relatively long-lasting benefits on children's development of fluid intelligence.

1.3. The present study

This study aimed to examine the immediate and long-term training effects of working memory training on improving pre-
schoolers’ fluid intelligence development. For this purpose, we trained preschoolers with an N-back task, and measured their fluid
intelligence before, immediately after, and 12 months after training. We hypothesized that (a) working memory training would
significantly enhance fluid intelligence in preschool-aged children, and (b) the fluid intelligence gains from working memory training
could be maintained in preschool-aged children for up to 12 months.

2. Material and methods

2.1. Participants

Ninety-six children from a preschool in the urban area of Guangzhou (China) initially participated in the experiment. They were
randomly assigned to an experimental group, an active control group, and a passive control group; thus, there were 32 participants in
each group initially. However, only 90 children (32 in experimental group, 27 in active control group, and 31 in passive control
group) completed their pretest and only 26 participants of each group completed the posttest. In addition, three male children in the
experimental group and one in the active control group were excluded from the data analysis because they missed a number of
training sessions or had difficulties understanding the training tasks. Therefore, the data of 74 participants (23 in the experimental
group, 25 in the active control group, and 26 in the passive control group) who completed all testing and intervention sessions from
pretest to posttest were used to examine the immediate effects of working memory training. Likewise, we only included the parti-
cipants with complete data (i.e., 22 participants of the experimental group, 21 participants of the active control group, and 26 of the
passive control group completing the 12-month maintenance test) for analyzing the maintenance effects. Demographic characteristics
of each group are shown in Table 1. Twelve participants (6 in experimental group and 6 in active control group) missed pretest or
posttest but all of them had the chances to participate in the intervention sessions. Even though their data were not included in the
analysis regarding fluid intelligence improvement, the data were included in evaluating the attitudes towards the intervention
program if they attended more than 10 intervention sessions and completed the questionnaire assessing attitude and motivation. As a
result, 7 participants (1 in experimental group and 6 in active control group) who missed the pretest or posttest were included in the
evaluating data. By doing so, we intended to collect the evaluating data of the two inventions as much as we can.
To ensure that the participants who did not complete the experiment were not systematically different from the participants who
completed the whole experiment, we conducted a 2 (intervention completion: participants who did not complete the experiment vs.
participants who completed the experiment) × 3 (group type: experimental group vs. active control group vs. passive control group)
ANOVA with the participants’ fluid intelligence pretest scores as the dependent variable. No significant effect was observed
(Fintervention (1, 84) < 1; Fintervention*group (2, 84) < 1, Fgroup (2, 84) < 1), suggesting that participants who did not complete the
training did not differ systematically from those who stayed in the training. Therefore, it is reasonable to assume that the missing data
were randomly distributed.

2.2. Measures

Two research assistants, who were naive to the participants’ assignments, performed the fluid intelligence tests and collected the

226
J. Peng et al. Cognitive Development 43 (2017) 224–234

Table 1
Demographic Characteristics of Each Group.

Group Age Gender Group Age Gender Group Age Gender

Experimental group(N = 23) 5.21 M Active control group 4.72 F Passive control group(N = 26) 4.69 M
5.06 M (N = 25) 4.52 M 4.38 M
5.44 F 4.90 M 4.60 M
5.24 F 4.93 M 5.06 F
5.15 M 4.65 F 5.00 F
5.00 M 5.02 F 4.44 M
5.20 F 4.93 M 4.45 M
4.59 F 4.85 F 5.32 F
4.80 M 5.00 M 5.28 M
5.08 F 5.04 F 5.14 M
4.81 M 4.70 F 4.58 M
5.13 M 4.65 F 4.46 F
5.16 M 4.99 F 4.43 M
5.25 M 4.99 F 4.36 M
4.79 M 4.65 F 4.61 F
4.69 M 2.59 M 5.17 F
4.70 F 4.90 M 4.78 F
5.09 F 4.93 M 4.73 M
4.94 F 4.94 M 5.26 M
4.84 M 5.00 M 4.89 F
5.14 F 5.21 M 4.99 M
4.71 M 5.21 M 5.09 F
4.98 M 4.74 M 4.65 F
4.92 M 5.25 F
4.90 M 4.72 M
5.03 M

Note: M = male, F = female.

self-report questionnaires for attitude and motivation. Raven's Standard Progressive Matrices (SPM) (Raven, Raven, & Court, 1990)
were used as measures of fluid intelligence in the pretest and the immediate posttest. The parallel versions of Test of Nonverbal
Intelligence 4th edition (TONI-4) (Brown, Sherbenou, & Johnsen, 2010) were used during the two follow-up tests (6 and 12 months
after training) as measures of fluid intelligence. We replaced the SPM with the TONI-4 as a measure of fluid intelligence to rule out
practice effects that may occur when repeatedly using the SPM. All three groups participated in the SPM tests during pretest and
posttest to examine the immediate effects of the working memory training. The pretest was arranged 2 days before the training and
the immediate posttest was arranged the next day of the last training session. All three groups completed the TONI-4 test 12 months
after training. We did not measure the fluid intelligence of passive control group in the 6-month follow-up test because of school
scheduling issues.

2.2.1. Raven's Standard Progressive Matrices (SPM)

The language-free SPM test was an untimed assessment administered individually. SPM included five sets (A to E) and each of the
five sets comprised 12 items with increasing difficulty. The first three sets were used during the pretest and posttest. The items in the
pretest and posttest were identical and their order counterbalanced. The number of correctly solved problems was recorded as the
participant's scores on the test.

2.2.2. Test of Nonverbal Intelligence (TONI)

Parallel versions (A and B) of the TONI-4 were administered individually at the first and second follow-up tests. Participants were
encouraged to correctly solve as many problems as possible without time limits. The tests were stopped when three or more errors
occurred within five consecutive items. The number of correctly solved problems was recorded as the participants’ scores.

2.2.3. Self-report questionnaire for attitude and motivation

Participants in the experimental and active control group were assessed on their attitude and motivation toward their assigned
tasks at the end of the training. The self-report questionnaire included five questions (e.g., “Did you like your game?” “How well did
you perform on your game?” etc.). Participants answered on a 5-point scale with cartoon-style faces whose emotion valence re-
presented the extent of their positive or negative attitude and motivation (e.g., a face with a big smile indicated “very much,” whereas
an angry face indicated “not at all”). Participants completed the questionnaire with the instructions given by their teachers to avoid
any misunderstandings.

2.3. Training materials and procedures

During the training sessions, participants in the experimental group were trained with our computerized training game, whereas

227
J. Peng et al. Cognitive Development 43 (2017) 224–234

Table 2
Difficulty of Each Level.

Level Numer of Locations N-back

Level 1 3 1-back
Level 2 4 1-back
Level 3 4 2-back
Level 4 5 2-back
Level 5 5 3-back
Level 6 6 3-back

participants in the active control group were assigned to play the game Fruit Ninja. Participants in the passive control group received
no interventions. The active control group and the experimental group received game time training for 15 min per session and 14
sessions in total. The whole training lasted for 3 weeks, including 14 training sessions (10 training sessions arranged in 10 weekdays
in the first two weeks and another 4 training sessions arranged from Monday to Thursday of the third week).
Games were provided to the experimental and active control group using tablet PCs (screen resolution 1024 by 768 pixels) with
identical brightness and sound settings.

2.3.1. Experimental group with the computerized n-back training game

This game was assigned to the experimental group for working memory training and was adapted from the n-back training
program used by Jaeggi et al. (2011). Participants were trained on the video-like game involving six levels of difficulty. The difficulty
of each level is illustrated in Table 2. Participants were presented with a random sequence of stimuli that appeared in different
locations; the presentation of each stimulus lasted for 500 ms. The participants had to determine whether the present stimulus was in
the same location as the stimulus n items back (an adaptive n-back paradigm) by pressing the “yes” key within 2500 ms. Another key
for “no” had to be pressed if the present stimulus was not in the same location as the one n items back. There were 15 + n trials in a
block, and five of them were programmed to come from the same location as the stimulus n items back. All participants were asked to
start from level 1 (i.e., n = 1). If they made three or fewer errors, they moved to the next level (i.e., n = 2, n = 3, etc.); in contrast,
they had to move back to the lower level if they made four or more errors within two successive blocks of the same level. Participants
were required to complete 10 blocks of the game in each training session. The achieved level of the 10th block was recorded as the
score of the session and was defined as the starting level of the next training session.

2.3.2. Active control group with the Fruit Ninja game

Participants in the active control group were asked to play the Fruit Ninja game on tablet PCs. The duration of this game was
identical as the duration of the training. In this game, participants were required to slice fruits thrown into the air to earn scores; at
the same time, they had to watch out for bombs to avoid explosions. We recorded participants’ highest score of the day as the measure
of their performance on this game.

3. Results

3.1. Descriptive statistics

Table 3 presents the mean scores and standard deviations on the fluid intelligence test by group and testing time. The results of
the pretest and posttests were obtained from analyzing the data from the 74 students who completed the pretest and posttest,
including 23 of the experimental group, 25 of the active control group, and 26 of the passive control group. The results of the 6-month
maintenance test were from 22 participants of the experimental and 21 of the active control group. The results of the 12-month
maintenance test were from 22 participants of the experimental, 21 of the active control, and 26 of the passive control group.
Table 4 presents the mean scores and standard deviations of the self-reported attitude and motivation questionnaire. The results
suggested that at the end of the game intervention, both the experimental and active control group have comparable attitudes toward
the games they played during the training. There were no significant differences between the two groups on the five questions

Table 3
Scores on Fluid Intelligence Tests.

Pretest Posttest 6-Month Follow-up 12-Month Follow-up

Mean (SD) N Mean (SD) N Mean (SD) N Mean (SD) N

Experimental 21.61(5.10) 23 26.96(3.14) 23 25.64(4.86) 22 26.41(5.58) 22

Active Control 23.04(6.85) 25 21.32(6.15) 25 17.33(6.00) 21 19.76(5.28) 21
Passive Control 20.46(5.46) 26 21.81(3.81) 26 23.22(5.92) 26

Note: A, B and C sessions of SPM were administered during the pre- and posttest. Session A of TONI-4 was administered during 6-month follow-up test and session B
was administered during 6-month follow-up test.

228
J. Peng et al. Cognitive Development 43 (2017) 224–234

Table 4
Scores of Self-reported Attitude and Motivation Questionnaire.

Experimental group Active Control group

Mean(SD) N Mean(SD) N

Q1: Did you like your game? 4.63(1.13) 24 4.97(0.18) 31

Q2: Did you play well? 4.33(1.52) 24 4.71(0.46) 31
Q3:How well did you perform on your game? 4.83(0.82) 24 4.52(0.68) 31
Q4:Did you want the awards? 4.83(0.82) 24 4.87(0.43) 31
Q5:Did you make the effort in your game? 4.64(0.66) 24 4.74(0.44) 31

Note: N refers to the number of participants who participated in the intervention and completed the Questionnaire, including some students who did not take the
pretest or posttest.

(pQ1 = 0.156, pQ2 = 0.253, pQ3 = 0.131, pQ4 = 0.839, and pQ5 = 0.517).

3.2. Pretreatment comparability

An ANOVA on the pretest scores showed no significant main effect of group on fluid intelligence (F (2, 71) = 1.236, p = 0.297,
ƞ2partial = 0.034), indicating group equivalency before the working memory training was implemented.

3.3. Differential fluid intelligence performance as a function of treatment

3.3.1. Training effects on working memory

Fig. 1 illustrates the working memory performance in the n-back working memory training program in the experimental group. A
paired-sample t-test was conducted for the experimental group to examine if there were any gains in working memory throughout the
training. We compared mean training performance (defined as the achieved level in the last block) during the first five and last five
sessions of the n-back training game. A significant improvement in the training task was observed, suggesting that our training
effectively improved performance in a working memory task (Mfirst5 = 2.12, Mlast5 = 3.69, t (22) = 6.971, p < 0.001). In contrast,
we did not find a significant improvement of participants’ scores on the Fruit Nijia game (Mfirst5 = 156.21, Mlast5 = 158.62, t (24)
= 0.376, p = 0.710).

3.3.2. Immediate training effects on fluid intelligence

To illustrate the immediate training effect of fluid intelligence, we ran a 3 (group: experimental vs. active control vs. passive
control) × 2 (session: pretest vs. posttest) mixed-design ANOVA with the fluid intelligence scores as the dependent variable, the
group as the between subject factor and the test session as the within subject factor. The main effect of test session (F (1, 71) = 8.37,
p < 0.01, ƞ2partial = 0.105), main effect of group (F (2, 71) = 11.06, p < 0.001, ƞ2partial = 0.238) and the interaction (F (2, 71)
= 12.37, p < 0.001, ƞ2partial = 0.258) of group and test session were significant. The post hoc analysis indicated that during the
posttest, the experimental group significantly outperformed the active control group (mean difference = 5.64, SE = 1.32, p < .001)
and passive control group (mean difference = 5.15, SE = 1.31, p < .001), whereas the difference between the active and passive
control group was not significant (mean difference = 0.49, SE = 1.28, p = 0.705).
To better illustrate the immediate effects of training, we conducted two separate ANOVAs. A 2 (group: experimental vs. active
control) × 2 (session: pretest vs. posttest) ANOVA was conducted with the group as the between subject factor and the session as the
within subject factor. The main effect of session was significant (F (1.46) = 6.827, p = 0.012, ƞ2partial = 0.129), the main effect of

Fig. 1. Working memory performance of the experimental group.

229
J. Peng et al. Cognitive Development 43 (2017) 224–234

Fig. 2. (1) Pretest scores of fluid intelligence (SPM) in three groups. (2) Posttest scores of fluid intelligence (SPM) in three groups.

group was not significant (F (1.46) = 2.128, p = 0.151, ƞ2partial = 0.044), and there was a significant interaction (F (1.46)
= 25.914, p < 0.001, ƞ2partial = 0.360). Then we conducted a simple effect analysis. According to the results, there was no sig-
nificant difference between experimental group and active control group (mean difference = −1.43, SE = 1.76, p = 0.419) in
pretest. Experimental group got significant higher scores in posttest (mean difference = 5.64, SE = 1.43, p < 0 .001) than active
control group. In addition, the posttest scores of experimental group were significantly higher than pretest (mean difference = 1.43,
SE = 1.76, p = 0.419) when there was no significant difference between pretest and posttest scores of active control group (mean
difference = 1.72, SE = 0.96, p = 0.080).
Another ANOVA was conducted in 2 (group: experimental vs. passive control) × 2 (session: pretest vs. posttest). The main effects
of session (F (1, 47) = 23.907, p < 0.001, ƞ2partial = 0.337) and group (F (1.47) = 8.375, p = 0.006, ƞ2partial = 0.151) were both
significant. Also the interaction was significant (F (1.47) = 8.544, p = 0.005, ƞ2partial = 0.154). The results of simple effect analysis
showed that there was no significant difference between two groups in pretest (mean difference = 1.15, SE = 1.52, p = 0.453), but
experimental group had significant higher scores in posttest than passive group (mean difference = 5.15, SE = 1.01, p < 0.001).
Besides, there was significant difference between two sessions of test in experimental group (mean difference = 5.35, SE = 1.00,
p < 0.001), but no difference in passive control group (mean difference = 1.35, SE = 0.94, p = 0.158). The comparison between
the three groups on the pretest and posttest are illustrated in Fig. 2. The effect sizes of the working memory training on the posttest
and 12-month follow-up test are listed in Table 5.

3.3.3. The maintenance effects on fluid intelligence

To examine the maintenance effect of training, we first standardized the raw scores of the two comparable measures of fluid
intelligence (i.e., SPM and TONI-4) and then performed statistical analysis with these standardized scores. We converted the raw
scores into Z scores (z = (x − μ)/σ) and then transformed the Z scores into T scores (T = z*10 + 50). Because the passive control
group did not participate in the 6-month maintenance test, we only included the experimental group and the active control group in
this analysis. Table 6 presents the means and standard deviations of the T scores for the two groups. A 4 (session: pretest vs. posttest
vs. 6-month maintenance test vs. 12-month maintenance test) × 2 (group: experimental vs. active control) repeated measures
ANOVA was performed. The results showed that the main effect of session (F (3, 123) = 0.39, p = 0.760, ƞ2partial = 0.009) was not
significant, whereas the main effect of group (F (1, 41) = 14.52, p < 0.001, ƞ2partial = 0.261) and the interaction between group
and session (F (3, 123) = 12.00, p < 0.001, ƞ2partial = 0.226) were significant. According to the simple effect analysis, there was no
significant difference between the two groups in the pretest (mean difference = −2.59, SE = 2.62, p = 0.328) while the experi-
mental group significantly outperform the active control group in posttest (mean difference = 8.99, SE = 2.47, p = 0.001), 6-month
maintenance test (mean difference = 12.17, SE = 2.43, p < 0.001) and 12-month maintenance test (mean difference = 10.49,
SE = 2.62, p < 0.001). The pretest T scores of experimental group were significantly lower than posttest (mean difference = -6.25,
SE = 1.62, p = 0.002), 6-month maintenance test (mean difference = −6.60, SE = 2.28, p = 0.036) and 12-month maintenance
test (mean difference = -5.78, SE = 2.08, p = 0.048) while there were no significant differences among the other three test sessions.
And the pretest T scores of active control group were significantly higher than posttest (mean difference = 5.34, SE = 1.66,
p = 0.015), 6-month maintenance test (mean difference = 8.16, SE = 2.34, p = 0.007) and 12-month maintenance test (mean

Table 5
Effect Sizes on Post Test and Maintenance Test.

Cohen's d Effect-size r

Posttest Experimental Group vs. Active Control Group 1.15 .50

Experimental Group vs. Passive Control Group 1.47 .59

12-month Follow-up Test Experimental Group vs. Active Control Group 1.22 0.52
Experimental Group vs. Passive Control Group 0.57 0.27

230
J. Peng et al. Cognitive Development 43 (2017) 224–234

Table 6
T Scores of Experimental Group and Active Control Group on Fluid Intelligence Tests.

Pretest Post test 6-Month Follow-up 12-Month Follow-up

Mean(SD) N Mean(SD) N Mean(SD) N Mean(SD) N

Experimental Group 49.34(8.15) 22 55.59(5.33) 22 55.94(7.11) 22 55.12(8.80) 22

Active Control Group 51.93(9.03) 21 46.60(10.22) 21 43.78(8.80) 21 44.63(8.33) 21

Note: Data of passive control group was not included because they did not participate in the 6-month follow-up test.

difference = 7.30, SE = 2.13, p = 0.008) while there were no significant differences among the other three test sessions. In active
control group, the significant differences between pretest and other three test sessions did not caused by the performance decline, but
the outperformance of the experimental group in the three test sessions. The results suggesting that the experimental group made
greater improvement from the training than the active control group and sustained this superiority during the maintenance tests.
To test whether the experimental group sustained the superiority over the other two groups on the 12-month maintenance test, we
conducted a one-way ANOVA with the raw scores on the 12-month maintenance test as the dependent variable and group as the
independent variable. The analysis yielded a significant main effect of group type (F (2, 68) = 8.20, p = 0.001, ƞ2partial = 0.199). The
post hoc analysis indicated that the experimental group significantly outperformed the active control group (mean difference = 6.65,
SE = 1.64, p < 0.001) and it also outperformed the passive control group in marginal trend toward significance (mean differ-
ence = 3.10, SE = 1.56, p = 0.051). These results indicated that the experimental group not only improved more than the two other
groups on the posttest but also maintained the improvement for 12 months.

4. Discussion

Here, we examined whether working memory training with the n-back computerized game can effectively improve preschool
children's fluid intelligence. Among the three groups of preschool participants, the experimental group, who received the compu-
terized n-back training program, obtained the greatest improvements in fluid intelligence tests immediately, 6 months, and 12
months after training. The results support our hypothesis that working memory training using the n-back task lead to improvements
in fluid intelligence tests, and the improvements can maintain as long as 12 months.
Our results are in line with previous research, such as the studies by Jaeggi et al. (2011) and Klingberg et al. (2005). Furthermore,
our study contributes to the literature by providing evidence for the maintenance effects of the working memory training for im-
proving preschoolers’ fluid intelligence. Few studies have examined whether the working memory training effect (i.e., improved
performance in fluid intelligence assessment) can be maintained over the long term (Melby-Lervåg & Hulme, 2013). Advocates for
long-term effectiveness argued that working memory training effects were attributed to substantial development of mental function
(Rueda, Checa, & Cómbita, 2012), whereas critics claimed that the effects were short-term and the result of simple skill development
or even practice effects (Owen et al., 2010). In the present study, participants who received systematic cognitive training out-
performed the two control groups during posttests of fluid intelligence; moreover, they maintained their superior performance for up
to 12 months after training. Therefore, such stable training effects on fluid intelligence could not be interpreted simply as a warm-up
or practice effect.
Considering the sparse reports about effective long-term effects on adult participants (Melby-Lervåg & Hulme, 2013), we propose
that the maintenance effect in preschoolers is unique and consistent with the literature showing that interventions provided during
early childhood can generate more durable effects than interventions provided during later childhood or adulthood. The above
findings about the greater training effects in early childhood than in adulthood are congruent with the literature that early childhood
interventions usually produce more influential impact than interventions provided later on (Burger, 2010; Ramey & Ramey, 1998).
For example, reading interventions occurring before second grade were more effective than reading interventions provided after
second grade (Shaywitz, Morris, & Shaywitz, 2008). Likewise, early interventions for children with speech and language disorders
were significantly more effective than interventions provided later on (Hallahan, Kauffman, & Pullen, 2012). The superior effects of
early intervention can be associated with the rapid brain development and sensitive developmental periods for multiple cognitive
skills.
This study further contributes to the literature by validating the intervention effects by eliminating a possible threat to its internal
validity. Some researchers (Boot, Kramer, Simons, Fabiani, & Gratton, 2008) argued that an expanded attention capacity acquired
from playing video-like games may contribute to the observed improvements of fluid intelligence, that is, the improvement of fluid
intelligence was not caused by the working memory training but by an expanded attention span obtained from the games
(Green & Bavelier, 2003). Therefore, to better illustrate our training effects, the Fruit Ninja Game was assigned to participants in the
active control group. The Fruit Ninja Game presumably required equal involvement as the n-back training game, which can be
illustrated by the participants’ results of self-reported attitude and motivation questionnaire in Table 4. This comparison demon-
strated that the training effects were not merely due to video-game playing, because only participants who received the adaptive
working memory training improved in test performance, whereas participants who played Fruit Ninja showed no improvements.
However, a test-retest effect or a maturation effect can be ruled out from threats to the internal validity by including the passive
control group, which did not show any improvement throughout the intervention. Therefore, the improvement in fluid intelligence

231
J. Peng et al. Cognitive Development 43 (2017) 224–234

measures in the experimental group was not resulted from the possible enhanced attention by playing video games, nor from retest or
natural maturation. The improvement found only in experimental group suggested that our working memory training program
includes certain steps that are critical components of visual working memory, such as requiring the participants to maintain the
briefly presented stimuli in mind and to determine if the upcoming stimulus was at the same position as the previous N-back one. This
is congruent with previous explanations that the cognitive and neural processes underlying fluid intelligence and the n-back program
overlap (Au et al., 2015). Although in this study we were unable to explicitly identify the underlying mechanism of working memory
improvement in the experimental group, it is plausible to assume that the intensive practice on maintaining visual information in the
training program may help the young participants biologically improved their brain function, given the plasticity of the neural system
at the early childhood stage (e.g., Hempel et al., 2004; Dahlin et al., 2008). It is also possible that the participants successfully
developed certain strategies (Morrison & Chein, 2011), such as rehearsal (Ford, Pelham, & Ross, 1984; Turley-Ames & Whitfield,
2003) or elaborative encoding (Carretti, Borella, & De Beni, 2007; Cavallini, Pagnin, & Vecchi, 2003), to help them with encoding,
maintenance, and/or retrieval from working memory, and they may transfer these strategies to the items in the intelligence test
where working memory is required for the problem solving. Future research is needed to further explore the mechanism of the
training effects and accordingly design working memory training programs with greater effectiveness.
Unlike some existing research (Liu, Zhu, Ziegler, & Shi, 2015) that reported both significant training effects on the Fruit Ninja
game and on the fluid intelligence in preschoolers, the present study did not find significant improvement on the game scores in the
active control group. The Fruit Ninja game is usually considered a training for inhibitory control (Liu et al., 2015), which is a central
component of executive functions and involves the ability to inhibit automatic but incorrect responses or to resist interference from
distracting stimuli (Diamond, 2013). Although previous research shows a close relationship between inhibitory control and fluid
intelligence (Dempster, 1991; Witth & ft, Sander, Süss, & Wittmann, 2009), Liu et al. (2015) demonstrates that inhibitory control
training can not be significantly transferred to fluid intelligence, and other existing research (Enge et al., 2014) also reported no
transfer effects of inhibitory training on fluid intelligence. Likewise, results of the present study showed no transfer effects on fluid
intelligence in the participants who played Fruit Ninja game. Considering the significant improvement and maintenance effects of the
computerized n-back training game, it seems more plausible that the intelligence improvement was closely related to the capacity of
maintaining information in working memory rather than merely the inhibitory control.
This study has implications for early childhood educators, as it reveals the potential to apply working memory training to early
childhood educational practices to promote the cognitive development of preschoolers. Early intervention programs for children with
intellectual disabilities might consider to include the working memory training program as evidence-based intervention in their
Individual Family Service Plan (IFSP, for children younger than 3 years) or Individualized Educational Program (IEP, for children
older than 3 years), which are individualized special educational services required by the Individuals with Disabilities Education Act
(IDEA, 2004). For many children with other types of disabilities associated with working memory deficits, for instance, children who
are at risk of or suffering from learning disabilities (Geary, 2004) or attention deficit hyperactivity disorder (Zentall, 2006), the
working memory training programs could be beneficial for enhancing their working memory capacity, consequently improving their
academic performance and social skills (Mezzacappa & Buckner, 2010).
Lacking the parallel versions of SPM is a limitation in current study and it may be better to measure the participants’ performance
by using parallel versions of tests. Moreover, future research should further explore the changes at the neuropsychological level
throughout the working memory training. While some researchers ascribed training effects to newly developed skills needed to solve
test problems or simply to test-retest effects, our findings, from a behavioral perspective, support the argument that substantive
improvement occurred in mental function during working memory training. However, the underlying neural mechanisms of training
effects and its transfer are still unclear. Further research should explore the possible development of neural functions associated with
working memory training. A previous study showed that after 2 months of working memory training with n-back tasks, increased
structural integrity of the white matter in the parietal regions and neighboring corpus callosum was detected (Takeuchi et al., 2010).
Thus, it is plausible to conclude that our training effects could lead to a substantive brain change in preschool children that directly
improved fluid intelligence. A follow-up study examining the mediation effects of brain changes between working memory training
and development of fluid intelligence is underway.
This research was funded by grants from the Project of Key Institute of Humanities and Social Sciences, MOE, China (No.
16JJD190001).

References

Abu-Rabia, S. (2003). The influence of working memory on reading and creative writing processes in a second language. Educational Psychology, 23(2), 209–222.
Aizpurua, A., & Koutstaal, W. (2010). Aging and flexible remembering: Contributions of conceptual span, fluid intelligence, and frontal functioning. Psychology and
Aging, 25(1), 193–207.
Alloway, P. T., & Alloway, R. G. (2009). The efficacy of working memory training in improving crystallized intelligence. Nature Proceedings.
Au, J., Sheehan, E., Tsai, N., Duncan, G. J., Buschkuehl, M., & Jaeggi, S. M. (2015). Improving fluid intelligence with training on working memory: a meta-analysis.
Psychonomic Bulletin & Review, 22(2), 366–377.
Baddeley, A. D., & Hitch, G. J. (1974). Working memory. In G. A. Bower (Ed.), The psychology of learning and motivation (pp. 47–89). New York: Academic Press.
Baddeley, A. (1992). Working memory: The interface between memory and cognition. MIT Press.
Baddeley, A. D., & Hitch, G. J. (1994). Developments in the concept of working memory. Neuropsychology, 8(4), 485.
Baddeley, A. (2003). Working memory: looking back and looking forward? Nature Reviews Neuroscience, 4(10), 829–839.
Baltes, P. B., & Kliegl, R. (1986). On the dynamics between growth and decline in the aging of intelligence and memory. Neurology, . Berlin Heidelberg: Springer, 1–17.
Barnett, S. W. (1995). Long-term effects of early childhood programs on cognitive and school outcomes. The Future of Children, 25–50.
Batty, G. D., Deary, I. J., & Gottfredson, L. S. (2007). Premorbid (early life) IQ and later mortality risk: systematic review. Annals of Epidemiology, 17(4), 278–288.
Beck, S. J., Hanson, C. A., Puffenberger, S. S., Benninger, K. L., & Benninger, W. B. (2010). A controlled trial of working memory training for children and adolescents

232
J. Peng et al. Cognitive Development 43 (2017) 224–234

with ADHD. Journal of Clinical Child & Adolescent Psychology, 39(6), 825–836.
Bergman Nutley, S., Söderqvist, S., Bryde, S., Thorell, L. B., Humphreys, K., & Klingberg, T. (2011). Gains in fluid intelligence after training non-verbal reasoning in 4-
year-old children: a controlled randomized study. Developmental Science, 14(3), 591–601.
Boot, W. R., Kramer, A. F., Simons, D. J., Fabiani, M., & Gratton, G. (2008). The effects of video game playing on attention, memory, and executive control. Acta
Psychologica, 129(3), 387–398.
Brown, L., Sherbenou, R. J., & Johnsen, S. K. (2010). TONI-4, test of nonverbal intelligence. [Pro-Ed].
Bruer, J. T. (1997). Education and the brain: A bridge too far. Educational Researcher, 26(8), 4–16.
Burger, K. (2010). How does early childhood care and education affect cognitive development? An international review of the effects of early interventions for children
from different social backgrounds. Early Childhood Research Quarterly, 25(2), 140–165.
Carpenter, P. A., Just, M. A., & Shell, P. (1990). What one intelligence test measures: A theoretical account of the processing in the Raven Progressive Matrices Test.
Psychological Review, 97(3), 404.
Carretti, B., Borella, E., & De Beni, R. (2007). Does strategic memory training improve the working memory performance of younger and older adults? Experimental
Psychology, 54, 311–320.
Cattell, R. B. (1971). Abilities: their structure, growth and action. Boston, MA: Houghton Mifflin.
Cavallini, E., Pagnin, A., & Vecchi, T. (2003). Aging and everyday memory: The beneficial effect of memory training. Archives of Gerontology and Geriatrics, 37(3),
241–257.
Chiappe, P., Hasher, L., & Siegel, L. S. (2000). Working memory, inhibitory control, and reading disability. Memory & Cognition, 28(1), 8–17.
Colom, R., Rebollo, I., Palacios, A., Juan-Espinosa, M., & Kyllonen, P. C. (2004). Working memory is (almost) perfectly predicted by g. Intelligence, 32(3), 277–296.
Colom, R., Abad, F. J., Quiroga, M., Shih, P. C., & Flores-Mendoza, C. (2008). Working memory and intelligence are highly related constructs, but why? Intelligence,
36(6), 584–606.
Colom, R., Quiroga, M.Á., Pei, C. S., Martínez, K., Burgaleta, M., Román, F. J., et al. (2010). Improvement in working memory is not related to increased intelligence
scores? Intelligence, 38(5), 497–505.
Conway, A. R. A., Cowan, N., Bunting, M. F., Therriault, D. J., & Minkoff, S. R. B. (2002). A latent variable analysis of working memory capacity, short-term memory
capacity, processing speed, and general fluid intelligence. Intelligence, 30(2), 163–183.
Conway, A. R. A., Kane, M. J., & Engle, R. W. (2003). Working memory capacity and its relation to general intelligence. Trends in Cognitive Sciences, 7(12), 547–552.
Dahlin, K. E. (2011). Effects of working memory training on reading in children with special needs? Reading and Writing, 24(4), 479–491.
Dahlin, E., Neely, A. S., Larsson, A., Bäckman, L., & Nyberg, L. (2008). Transfer of learning after updating training mediated by the striatum? Science, 320(5882),
1510–1512.
Dahlin, E., Nyberg, L., Bäckman, L., & Neely, A. S. (2008). Plasticity of executive functioning in young and older adults: immediate training gains, transfer, and long-
term maintenance. Psychology and Aging, 23(4), 720.
Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory and reading. Journal of Verbal Learning and Verbal Behavior, 19(4), 450–466.
De Smedt, B., Janssen, R., Bouwens, K., Verschaffel, L., Boets, B., & Ghesquière, P. (2009). Working memory and individual differences in mathematics achievement: A
longitudinal study from first grade to second grade? Journal of Experimental Child Psychology, 103(2), 186–201.
Destefano, D., & Lefevre, J. A. (2004). The role of working memory in mental arithmetic. Journal of Cognitive Psychology, 16(3), 353–386.
Deary, I. J., Strand, S., Smith, P., & Fernandes, C. (2007). Intelligence and educational achievement. Intelligence, 35(1), 13–21.
Deary, I. J., Penke, L., & Johnson, W. (2010). The neuroscience of human intelligence differences. Nature Reviews Neuroscience, 11(3), 201–211.
Dempster, F. N. (1991). Inhibitory processes: a neglected dimension of intelligence. Intelligence, 15, 157–173.
Diamond, A. (2013). Executive functions. Annual Review of Psychology, 64, 135–168.
Enge, S., Behnke, A., Fleischhauer, M., Küttler, L., Kliegel, M., & Strobel, A. (2014). No evidence for true training and transfer effects after inhibitory control training in
young healthy adults. Journal of Experimental Psychology: Learning, Memory, and Cognition, 40(4), 987–1001.
Engel de Abreu, P. M. J., Conway, A. R. A., & Gathercole, S. E. (2010). Working memory and fluid intelligence in young children. Intelligence, 38(6), 552–561.
Fenson, L., Dale, P. S., Resnick, S. J., Bates, B., Thal, D. J., & Pethick, S. J. (1994). Variability in early communicative development. Monographs of the Society for
Research in Child Development, 59(5, Serial No. 242), 1–173.
Ford, C. E., Pelham, W. E., & Ross, A. O. (1984). Selective attention and rehearsal in the auditory short-term-memory task-performance of poor and normal readers.
Journal of Abnormal Child Psychology, 12(1), 127–141.
Geary, D. C. (2004). Mathematics and learning disabilities. Journal of Learning Disabilities, 37, 4–15.
Gray, J. R., & Thompson, P. M. (2004). Neurobiology of intelligence: science and ethics. Nature Reviews Neuroscience, 5(6), 471–482.
Gray, J. R., Chabris, C. F., & Braver, T. S. (2003). Neural mechanisms of general fluid intelligence. Nature Neuroscience, 6(3), 316–322.
Gray, J. R. (2002). Integration of emotion and cognition in the lateral prefrontal cortex? Proceedings of the National Academy of Sciences, 99(6), 4115–4120.
Green, C. S., & Bavelier, D. (2003). Action video game modifies visual selective attention. Nature, 423(6939), 534–537.
Hallahan, D. P., Kauffman, J. M., & Pullen, P. C. (2012). Exceptional learners: introduction to special education (12th ed.). Upper Saddle River, NJ: Pearson, Merrill
Prentice Hall.
Hempel, A., Giesel, F. L., Caraballo, N. M. G., Amann, M., Meyer, H., Wüstenberg, T., et al. (2004). Plasticity of cortical activation related to working memory during
training. American Journal of Psychiatry, 161(4), 745–747.
Holmes, J., Gathercole, S. E., & Dunning, D. L. (2009). Adaptive training leads to sustained enhancement of poor working memory in children. Developmental Science,
12(4), F9–F15.
Horn, J. L., & Cattell, R. B. (1967). Age differences in fluid and crystallized intelligence. Acta Psychologica, 26, 107–129.
Horowitz-Kraus, T., & Breznitz, Z. (2009). Can the error detection mechanism benefit from training the working memory? a comparison between dyslexics and
controls–an erp study. PLoS One, 4(9), e7141.
Individuals with Disabilities Education Act of 2004. 20 U. S. C.§1400 et seq. (2004).
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid intelligence with training on working memory. Proceedings of the National Academy of
Sciences, 105(19), 6829–6833.
Jaeggi, S. M., Studer-Luethi, B., Buschkuehl, M., Su, Y., Jonides, J., & Perrig, W. J. (2010). The relationship between n-back performance and matrix
reasoning—Implications for training and transfer. Intelligence, 38(6), 625–635.
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Shah, P. (2011). Short-and long-term benefits of cognitive training. Proceedings of the National Academy of Sciences,
108(25), 10081–10086.
Jausovec, N., & Jausovec, K. (2012). Working memory training: Improving intelligence-changing brain activity. Brain and Cognition, 79(2), 96–106.
Kane, M. J., & Engle, R. W. (2002). The role of prefrontal cortex in working-memory capacity, executive attention, and general fluid intelligence: An individual-
differences perspective. Psychonomic Bulletin & Review, 9(4), 637–671.
Kane, M. J., Hambrick, D. Z., & Conway, A. R. (2005). Working memory capacity and fluid intelligence are strongly related constructs: comment on Ackerman, Beier
and Boyle. The Psychological Bulletin, 131(1), 66–71.
Karbach, J., & Verhaeghen, P. (2014). Making working memory work: A meta-analysis of executive-control and working memory training in older adults. Psychological
Science, 25(11), 2027–2037.
Klingberg, T., Forssberg, H., & Westerberg, H. (2002). Training of working memory in children with ADHD. Journal of Clinical and Experimental Neuropsychology, 24(6),
781–791.
Klingberg, T., Fernell, E., Olesen, P. J., Johnson, M., Gustafsson, P., Dahlström, K., et al. (2005). Computerized training of working memory in children with ADHD: A
randomized: controlled trial. Journal of the American Academy of Child & Adolescent Psychiatry, 44(2), 177–186.
Klingberg, T. (2010). Training and plasticity of working memory. Trends in Cognitive Sciences, 14(7), 317–324.
Kyllonen, P. C., & Christal, R. E. (1990). Reasoning ability is (little more than) working-memory capacity? Intelligence, 14(4), 389–433.
Lenroot, R. K., & Giedd, J. N. (2006). Brain development in children and adolescents: Insights from anatomical magnetic resonance imaging.

233
J. Peng et al. Cognitive Development 43 (2017) 224–234

Neuroscience & Biobehavioral Reviews, 30(6), 718–729.

Lenroot, R. K., Schmitt, J. E., Ordaz, S. J., Wallace, G. L., Neale, M. C., Lerch, J. P., et al. (2009). Differences in genetic and environmental influences on the human
cerebral cortex associated with development during childhood and adolescence. Human Brain Mapping, 30(1), 163–174.
Li, S., Schmiedek, F., Huxhold, O., Röcke, C., Smith, J., & Lindenberger, U. (2008). Working memory plasticity in old age: Practice gain, transfer, and maintenance.
Psychology and Aging, 23(4), 731.
Liu, Q., Zhu, X., Ziegler, A., & Shi, J. (2015). The effects of inhibitory control training for preschoolers on reasoning ability and neural activity. Scientific Reports, 5.
Man, M. M. K. (2013). The early childhood experience in psychology development. International Journal of Management & Innovation, 5(2).
Manard, M., Carabin, D., Jaspar, M., & Collette, F. (2014). Age-related decline in cognitive control: the role of fluid intelligence and processing speed. BMC
Neuroscience, 15, 7.
McCutchen, D. (1996). A capacity theory of writing: Working memory in composition. Educational Psychology Review, 8(3), 299–325.
Melby-Lervåg, M., & Hulme, C. (2013). Is working memory training effective? A meta-analytic review. Developmental Psychology, 49(2), 270.
Mezzacappa, E., & Buckner, J. C. (2010). Working memory training for children with attention problems or hyperactivity: A school-based pilot study. School Mental
Health, 2(4), 202–208.
Van Der Molen, Mj, et al. (2010). Effectiveness of a computerised working memory training in adolescents with mild to borderline intellectual disabilities. Journal of
Intellectual Disability Research, 54(5), 33–447.
Morrison, A. B., & Chein, J. M. (2011). Does working memory training work? The promise and challenges of enhancing cognition by training working memory.
Psychonomic Bulletin & Review, 18(1), 46–60.
Neisser, U., Boodoo, G., Bouchard, T. J., Jr., Boykin, A. W., Brody, N., Ceci, S. J., et al. (1996). Intelligence: knowns and unknowns. American Psychologist, 51(2), 77.
Nisbett, R. E., Aronson, J., Blair, C., Dickens, W., Flynn, J., Halpern, D. F., et al. (2012). Intelligence: New findings and theoretical developments. American Psychologist,
67(2), 130.
Owen, A. M., Hampshire, A., Grahn, J. A., Stenton, R., Dajani, S., Burns, A. S., et al. (2010). Putting brain training to the test. Nature, 465(7299), 775–778.
Perry, B. D., & Pollard, R. (1997). Altered brain development following global neglect in early childhood. Proceedings from the society for neuroscience annual meeting.
(New orleans).
Piaget, J. (1981). Intelligence and affectivity: Their relationship during child development. (Trans & Ed TA Brown & CE Kaegi). Annual Reviews.
Ramey, C. T., & Ramey, S. L. (1998). Early intervention and early experience. American Psychologist, 53(2), 109.
Raven, J. C., Raven, J., & Court, J. H. (1990). Standard progressive matrices. Oxford, United Kingdom: Oxford Psychologists Press.
Rohde, T. E., & Thompson, L. A. (2007). Predicting academic achievement with cognitive ability? Intelligence, 35(1), 83–92.
Rueda, M. R., Checa, P., & Cómbita, L. M. (2012). Enhanced efficiency of the executive attention network after training in preschool children: Immediate changes and
effects after two months. Developmental Cognitive Neuroscience, 2, S192–S204.
Seidler, R. D., Bernard, J. A., Buschkuehl, M., Jaeggi, S., Jonides, J., & Humfleet, J. (2010). Cognitive training as an intervention to improve driving ability in the older adult.
No. M-CASTL 2010-01.
Shaw, P., Greenstein, D., Lerch, J., Clasen, L., Lenroot, R., Gogtay, N., et al. (2006). Intellectual ability and cortical development in children and adolescents. Nature,
440(7084), 676–679.
Shaywitz, S. E., Morris, R., & Shaywitz, B. A. (2008). The education of dyslexic children from childhood to young adulthood. Annual Review of Psychology, 59, 451–475.
Shipstead, Z., Redick, T. S., & Engle, R. W. (2012). Is working memory training effective? The Psychological Bulletin, 138(4), 628.
Simcock, G., & Hayne, H. (2003). Age-related changes in verbal and nonverbal memory during early childhood. Developmental Psychology, 39(5), 805.
Stephenson, C. L., & Halpern, D. F. (2013). Improved matrix reasoning is limited to training on tasks with a visuospatial component. Intelligence, 41(5), 341–357.
Swanson, H. L., & Berninger, V. W. (1996). Individual differences in children's working memory and writing skill. Journal of Experimental Child Psychology, 63(2),
358–385.
Swanson, H. (1993). Working memory in learning disability subgroups. Journal of Experimental Child Psychology, 56(1), 87–114.
Swanson, H. (1994). The role of working memory and dynamic assessment in the classification of children with learning disabilities? Learning Disabilities Research and
Practice, 9(4), 190–202.
Swanson, H. L., & Beebefrankenberger, M. (2004). The relationship between working memory and mathematical problem solving in children at risk and not at risk for
serious math difficulties. Cheminform, 100(2), 343–379.
Takeuchi, H., Sekiguchi, A., Taki, Y., Yokoyama, S., Yomogida, Y., Komuro, N., et al. (2010). Training of working memory impacts structural connectivity. The Journal
of Neuroscience, 30(9), 3297–3303.
te Nijenhuis, J., van Vianen, A. E. M., & van der Flier, H. (2007). Score gains on g-loaded tests: no g. Intelligence, 35(3), 283–300.
Thorell, L. B., Lindqvist, S., Bergman Nutley, S., Bohlin, G., & Klingberg, T. (2009). Training and transfer effects of executive functions in preschool children.
Developmental Science, 12(1), 106–113.
Turley-Ames, K. J., & Whitfield, M. M. (2003). Strategy training and working memory task performance. Journal of Memory and Language, 49(4), 446–468.
Viadero, D. (1996). Brain trust. Education Week, 16(3), 3.
Weicker, J., Villringer, A., & Thöne-Otto, A. (2016). Can impaired working memory functioning be improved by training? A meta-analysis with a special focus on brain
injured patients. Neuropsychology, 30(2), 190–212.
Witthöft, M., Sander, N., Süss, H. M., & Wittmann, W. W. (2009). Adult age differences in inhibitory processes and their predictive validity for fluid intelligence.
Neuropsychology Development & Cognition, 16(2), 133–163.
Zentall, S. S. (2006). ADHD and education: foundations, characteristics, methods, and collaboration. Upper Saddle River, N.J: Pearson Merrill Prentice Hall.
Zhao, X., Wang, Y. X., Zhou, R. L., Wang, L. J., & Tan, C. (2011). The influence on individual working memory during 15 days −6° head-down bed rest. Acta
Astronautica, 69(11–12), 969–974.

Peng Jun M.S. is a graduate student in the Center for Studies of Psychological Application & School of Psychology at South China Normal University, Guangzhou
510631, China. Her research focuses on working memory training for students with and without special needs.

Mo Lei Ph.D. is a full professor of the Center for Studies of Psychological Application & School of Psychology at South China Normal University. His research interests
include child development, the cognitive processing and cognitive-neurological deficits and interventions for students with reading disabilities.

Huang Ping Ph.D. is an experimentalist in the Center for Studies of Psychological Application & School of Psychology at South China Normal University, Guangzhou
510631, China.

Zhou Ying is an undergraduate student majoring in psychology at South China Normal University, Guangzhou 510631, China.

234