Intelligence 64 (2017) 45–51

Contents lists available at ScienceDirect

Intelligence
journal homepage: www.elsevier.com/locate/intell

The relation between intelligence and reaction time in tasks with increasing MARK
cognitive load among athletes with intellectual impairment
Debbie Van Biesena,⁎, Katina McCullocha, Luc Janssensa,b, Yves C. Vanlandewijcka
a
KU Leuven, Faculty of Kinesiology and Rehabilitation Sciences, Department of Rehabilitation Sciences, Belgium
b
KU Leuven, Campus Groep T Leuven, Electrical Engineering (ESAT) TC, Belgium

A R T I C L E I N F O A B S T R A C T

Keywords: Intelligence has long been associated with chronometric performance measures, such as reaction time (RT);
Intellectual disability however, few studies have investigated this relation in humans at the lower end of the intelligence range. The
Sport purpose of this study is to examine the inter-relation between RT, and intelligence in the population of athletes
Flanker with mild to moderate (IQ 40–75) intellectual impairment (II), according to the DSM-5 diagnostic criteria. IQ-
Athletes
scores were retrieved from the athlete-database, assessed prior to the study by means of the most recently
standardized versions of the Stanford-Binet, Raven Progressive Matrices or Wechsler Intelligence Scales, i.e.,
WISC-V or WAIS-IV. Three RT tasks with increasing cognitive load were administered in this study. The sample
consisted of 103 young individuals (32% females, mean age = 24.4 years, SD = 5.8) with II (mean IQ = 60.6,
SD = 9.2), performing the tests prior to competing in the Global Games; the highest level of sport competition,
organized by the International Federation for Intellectual Disability Sport. We tested whether IQ and RT were
correlated in this sample by means of Pearson's correlations. Afterwards, a comparison sample of tertiary edu-
cation students was recruited (n = 103; mean age = 21.1, SD = 2.4), whereby for each individual participant
with II, a peer was selected on the basis of equal gender, practicing the same sport, and equal accumulated sport
expertise. We focused on the possibility that RT tasks requiring higher cognitive load differentiate more between
the samples. The 2 × 3 ANOVA demonstrated a significant group (athletes with II versus comparison group) by
task (simple RT, choice RT, complex RT) interaction effect, indicating that the progression of RTs increase across
the three tasks with increasing complexity, was more pronounced in the athletes with II than in the comparison
sample. This is the first study investigating RT in a large sample of well-trained active individuals with II. It
provides a benchmark for other studies and suggests that the impact of lower levels of intelligence on RT is most
apparent in RT tasks with the highest cognitive load.

1. Introduction
Reaction times (RT) are a widespread, important and informative
tool in the study of intelligence (Der & Deary, 2003; Lee & Chabris,
2013; Nissan, Liewald, & Deary, 2013). The relationship between RT
and measures of intelligence has been studied since the second half of
the 20th century (Jensen, 2006). Reviews of this research have in-
dicated that there is a significant inverse correlation between measures
of RT, inspection time and other measures of information processing
time on the one hand; and general intelligence (or g) on the other hand
(Vernon, 1983; Nettelbeck, Edwards, & Vreugdenhil, 1986; Deary,
Der, & Ford, 2001; Sheppard & Vernon, 2008; McGrew, 2009a,b).
Many types of RT tasks exist (Cinaz, Vogt, Arnrich, & Tröster, 2012).
Simple Reaction Time (SRT) refers to the time taken to respond to a
single stimulus, and only one response option is available (Nissan,
Liewald, & Deary, 2013). A task of this kind typically has a low g-
loading (Schneider & McGrew, 2012). SRT does not require substantial
cognitive effort; it is enough for the person being tested to simply in-
dicate that the stimulus is perceived (Johnson & Deary, 2011). Never-
theless, there is a long history of researchers having interest in the study
of the relation between SRT and intelligence; and faster RTs (i.e.,
shorter processing speed) in these very simple tasks are associated with
complex reasoning (Schneider & McGrew, 2012). Choice RT tasks in-
volve tasks with a higher g-loading, in which some minimal processing
of information content is required; for example, in a four-choice RT-
task, arrows may appear facing upwards, downwards, left or right; and
the corresponding keyboard arrow must be chosen (Johnson & Deary,
2011).
Processing speed is a key area of intelligence, and the processing
speed provides an indication of the rate at which tasks of a specified

⁎
Corresponding author at: KU Leuven, Faculty of Kinesiology and Rehabilitation Sciences, Department of Rehabilitation Sciences, Tervuursevest 101, 3001 Leuven, Belgium.
E-mail address: debbie.vanbiesen@kuleuven.be (D. Van Biesen).

http://dx.doi.org/10.1016/j.intell.2017.06.005
Received 19 August 2016; Received in revised form 28 June 2017; Accepted 28 June 2017
0160-2896/ © 2017 Elsevier Inc. All rights reserved.
D. Van Biesen et al.
kind and difficulty are performed (McGrew, 2009a,b;
Schneider & McGrew, 2012) which can be evaluated via RTs (Carroll,
1993). Nissan et al. (2013) demonstrated that people who score high on
intelligence tests also tend to have faster RTs and less variance in RT
performance. Deary, Der, & Ford (2001) showed that SRT indices were
significantly inversely associated with cognitive test scores. The corre-
lation between a four-choice RT and general intelligence in their study
was − 0.49 in a sample of 900 participants. Overall, the inverse re-
lationship between intelligence and RT is clear (for reviews, see Jensen,
1998; Deary, 2000; Jensen, 2006); however, the strength of the re-
lationship is debatable (Deary, Der, & Ford, 2001; Fry & Hale, 2000).
Furthermore, most of the studies in this field of research comprised
samples of young people at the higher end of the intelligence range
(e.g., college and university students) (Der & Deary, 2003). The present
study extends on the current knowledge by including a large sample of
people with intellectual impairment, situated at the lower end of the
intelligence range (IQ ≤ 75).
Intellectual impairment (II) is used through the remainder of this
paper to denote deficits in intellectual functioning, which refers to
general mental capacities, such as learning, problem solving, thinking
abstractly etc., according to the American Association of Intellectual
and Developmental Disabilities (AAIDD, 2010). According to the A-
AIDD's definition of intellectual disability, the individual should not
only have II (IQ ≤ 75) to be clinically diagnosed, but two other criteria
should also be met: (1) limitations in adaptive behavior, and (2) both
manifested before the age of 18. More specifically, the sample included
in the present investigation was diagnosed as having a mild to moderate
II (IQ between 40 and 75). Various researchers included participants
with II in their RT studies, and demonstrated that these individuals
have slower RTs compared to peers without II, as well as more inter-
and intra-individual variance in RT performance (Klotz, Johnson, Wu,
Isaacs, & Gilbert, 2011; Koichi, Hideyuki, & Mitsuru, 2011; Kosinski,
2013;).
RT is also considered as a key factor determining performance in
many sports. Sport expertise has been associated with improved RTs
(Williams & Ericsson, 2005). However, it is not clear whether the RT of
more experienced athletes is a product of their sport expertise or an
artifact of individuals with naturally high processing speed being at-
tracted to, and excelling in sport. To investigate this relationship be-
tween sport expertise and RT, a research design can be applied in which
athletes/active individuals versus non-athletes/non-active individuals
are compared (e.g., Khezri, Shahbazi, Kashan, & Pashabadi, 2011; Nuri,
Shademehr, Moghaddam, & Ghotbi, 2012; Atan & Akyol, 2014). For
example, the study by Nuri, Shadmehr, Moghaddam, and Ghotbi (2012)
focused on comparing RTs between volleyball players and non-athletes,
with the athlete-group displaying faster visual choice RTs and visual
complex choice RTs than the non-athletes. An earlier study by Abourezk
and Toole (1995) revealed that older active individuals reacted sig-
nificantly faster than non-active controls. However, they only found
differences for complex choice RT paradigms, but not on simple RT
tests.
There is a gap within the literature to what extent sport expertise or
specific interventions could improve the RT in athletes with II. The
World Health Organization has highlighted being involved in sport and
physical activity as a public health priority focus since 2004 (WHO,
2016). Specifically for the population of individuals with II, it is vital to
be physically active, as Bartlo and Klein (2011) found moderate to
strong evidence on the effectiveness of physical activity interventions
on their balance, muscle strength, and overall health and quality of life.
We are not aware of any studies investigating the effect of exercise on
RT or on cognitive factors other than RT in athletes with II. Alter-
natively, a new body of literature is emerging investigating the impact
of II on sport performance (Burns, 2015; Van Biesen, Mactavish,
McCulloch, Lenaerts, & Vanlandewijck, 2016a). Van Biesen, Kerremans,
Mactavish, and Vanlandewijck (2016b) investigated the relationship
between cognition and tactical proficiency in well-trained table tennis
46
Intelligence 64 (2017) 45–51
players with II and they found that 18% of the variance in tactical
proficiency was attributed to spatial visualization and simple RT. In
another recent investigation with well-trained athletes with II from
various types of sports (Van Biesen et al., 2016a) the cognitive skills
component approach was used, i.e., focusing on the assessment of
generic cognitive skills (i.e., outside the sport-specific context) that tap
into the intrinsic cognitive demands of the sport. From that study it was
concluded that athletes with II scored significantly below athletes
without II on cognitive ability measures relevant to sport (i.e., Fluid
Reasoning, Short-term Memory, Reaction and Decision Speed). How-
ever, there were large inter-individual differences, and about 2% of the
II-athletes' sample obtained higher scores than the average score in the
comparison group, most apparent in the purely speed-based tests, such
as simple RT and complex RT.
In a study involving table tennis players with and without mild to
moderate II, slower RTs and slower speed of upper limb were found for
the table tennis players with II, and these measurements correlated
positively with their table tennis proficiency levels (Van Biesen,
Verellen, Meyer, Mactavish, Van de Vliet, & Vanlandewijck, 2010). In a
more recent study, Van Biesen et al. (2016a) assessed RT in a large
sample of well-trained athletes with and without mild to moderate II
from various sports: track and field, swimming, table tennis and bas-
ketball. The RT measures included four pure speed-based measures, at
the most basic level that did not require substantial cognitive effort.
Significant differences in complex and simple RT were found between
athletes with and without II. It is important to note that the athletes
with II had significantly larger variability in RT performance. Schweizer
(2001) provides a possible explanation, proposing that the speed ad-
vantage of people with higher intelligence test scores is more dominant
when using more sensitive alternative RT tests requiring complex re-
sponses. This finding was confirmed in the large scale review performed
by Lee and Chabris (2013). Combined data from 172 studies, with a
total of 53,542 participants indicated that measures of intelligence were
significantly positively correlated with RT and that for some measures
this relationship became stronger as the complexity of the tasks in-
creased. For the current project, a series of three RT paradigms will be
used with increasing cognitive load, i.e., one pure speed-based test
(simple reaction time), a choice RT test requiring decision speed and
processing visual information and an adapted version of the Flanker test
(Eriksen, 1995). The Flanker test has a higher g-loading compared to
the two other RT tasks, as it is an example of a visual search task, in
which participants need to indicate as rapidly as possible whether a
given target item is present in a display, in which other items (called
distracters) are present (Trick & Enns, 1998). During the Flanker task,
stimuli present targets and distracters simultaneously, and the task is to
ignore the distracters (response inhibition) and only respond to the
targets. Since its introduction, the Flanker task has been widely used for
studying attention, detection, distraction, executive control, and other
aspects of cognition (McClean et al., 2013). The Flanker task also
provides the opportunity to examine response time and response ac-
curacy, as an inability to inhibit distracting stimuli leads to slower RTs
and greater susceptibility to errors. Previous research has shown that
individuals with II are characterized by medium to large inhibition
deficits (Eriksen, 1995), and groups with higher scores on trait anxiety
showed greater interference effects, a greater number of errors, and
reduced efficiency than low trait anxiety groups (Gomez-Iniguez et al.,
2014).
The purpose of this study is to investigate the inter-relation between
intelligence, RT and sport expertise. The study aimed to (1) explore the
relation between intelligence and RT in healthy, active individuals at
the lower ends of the intelligence range; (2) compare the differences in
RT performance between individuals with II and the comparison group;
and (3) investigate differences in processing speed on three RT tasks
with increasing cognitive load. The hypotheses are that (1) RT and
intelligence are negatively correlated at the lower end of the in-
telligence range; (2) that the athletes in the comparison group score
D. Van Biesen et al. Intelligence 64 (2017) 45–51

Table 1 questions of interest for this study. It should be noted that participants'
Demographics of both participant groups (athletes with II versus comparison group). nationality was not a matching criterion. All participants in the com-
parison group had Belgian nationality, which we should take into
Athletes with II Comparison group
consideration in the interpretation of our findings, as the athletes with
N Mean SD N Mean SD II represented various nationalities. Furthermore, IQ scores were not
available for the comparison group; however, all participants in the
Gender (male/female) 70/33 70/33
comparison group were college or university students, who, at
Age 103 24.4 5.8 103 21.1 2.4
IQ 103 60.6 9.2 Data not available minimum, had graduated from regular secondary education. Data from
Accumulated training the standardization of the Wechsler-Adult Intelligence Scale Revised
experience across (WAIS-R) was used by Matarazzo and Herman (1984) to investigate the
the lifespan/sport relation between IQ and education. Based on their data, years of
(hours)
schooling may be used as an estimation of Full Scale IQ. Individuals,
Athletics 4 6474.0 7539.0 4 5915.0 2569.5
Basketball 34 3688.6 3111.5 34 3042.0 1428.4 like those in our comparison group, who completed 12 years of regular
Cycling 5 9188.4 7631.7 5 9305.4 8486.6 education, have an average IQ of 100.1 (SD 15.3). Hence, it can be
Soccer 26 2426.1 1375.6 26 3505.0 1865.0 assumed (although not guaranteed) that all the participants in the
Swimming 25 7763.0 5467.5 25 4857.8 2789.3
comparison group are athletes without II. A written informed consent
Tennis 9 3068.0 2483.9 9 2241.8 1649.9
was obtained from all participants before the start of the experiment.
Note. II = intellectual impairment. The Medical Ethics Committee of KU Leuven approved the study.
Whereas all participants in the comparison group completed all
better than athletes with II on all measures of RT; and (3) that the ef- tests, n = 102 athletes with II had complete data on all RT measures.
fects are more pronounced for RT tasks requiring higher cognitive load. One athlete refused to perform the final subtest (Flanker) after per-
forming the SRT and ChRT.

2. Methods 2.2. Procedure & instruments

2.1. Participants
The primary sample for this study consisted of 103 participants
(32% female) who participated at the Global Games in Ecuador,
September 2015, organized by The International Federation for
Intellectual Disability Sport (better known by its acronym INAS), M
age = 24.4, SD = 5.8 years, M IQ = 60.6, SD = 9.2. The athletes
participated in six different sports: basketball (n = 34), soccer
(n = 26), cycling (n = 5), swimming (n = 25), tennis (n = 9) and
athletics (n = 4). Seventeen athletes (16.5%) were left-handed, and 86
athletes (83.5%) were right-handed. The training experience data (see
Table 1) was self- reported by the athletes and their coaches, including
the total years of previous involvement in the sport, and the average
time spent on sport practice/training per week. An estimation of ac-
cumulated training hours across the lifespan was calculated per athlete
by means of a simple formula multiplying the previous years of in-
volvement by the average hours of deliberate practice per week
(Ericsson, Krampe, & Tesch-Römer, 1993).
As stated earlier, all athletes included in this study met the criteria
for diagnosis of intellectual disability (AAIDD, 2010). The participants
represented nine different countries (Australia, USA, Mexico, The
Netherlands, Ecuador, France, Portugal, South-Africa, and Japan). None
of the participants had severe or profound II, or a known chromosomal
disorder (e.g., Down Syndrome). The detailed IQ and adaptive behavior
data of the participants were retrieved from the INAS database. INAS
recognizes the most recently standardized versions of the Stanford-
Binet (Thorndike, Hagen, & Sattler, 1986), Raven Progressive Matrices
(Raven, Raven, & Court, 1998) and the Wechsler Intelligence Scales,
i.e., WISC-V (Wechsler, 2014), and WAIS-IV (Wechsler, 2008).
A comparison group of 103 Belgian tertiary education students (M
age = 21.1, SD = 2.4 years) was recruited afterwards. In the compar-
ison group, 23 participants (22.3%) were left-handed and 80 partici-
pants (77.7%) were right-handed. The demographics of both partici-
pant groups are provided in Table 1. In any analysis of this type, the
characteristics of the particular comparison group have some impact on
the results. In this study, each individual with II was matched with a
student in the same age category, same gender and practicing the same
principal sport. Every subsample (e.g., male soccer players) was re-
cruited taking into account the training volume reported, making sure
that the total training experience of both subsamples was equal (see
Table 1). The comparison sample was selected to address the main
Data collection consisted of one test session with a total duration of
approximately 15 min. Prior to starting the experiment, participants
(and their guardian if necessary) signed an informed consent, as well as
a questionnaire concerning age, sport activity, training history and
training volume. Parental consent was also attained if the participant
was younger than 18 years. After consent was given, three computer-
ized RT tests were performed in fixed order with gradually increasing
cognitive load; simple reaction time (SRT), choice reaction time (ChRT)
and complex reaction time or Flanker test (Flanker). We used this ap-
proach in order to gradually build up the task difficulty, but also the
complexity of the test instruction, to accommodate the specific needs of
individuals with II. Standardized instructions, demonstration, and
practice trials were provided for each of the three tests, with completion
time and number of practice trials determined by how quickly each
participant understood the desired task. A 16-inch screen (resolution of
1280 × 720 pixels, blue background, white stimuli) and a keyboard,
attached to the laptop, was used. The screen was adjusted with the
center at participants' eye-height and a standardized distance (65 cm,
i.e., two times the screen width) between the screen and the partici-
pants' forehead. The test took place in a quiet room, free from any
distractions. No feedback was provided during test performance, to
avoid participants being distracted. After completion of the subtests,
the test instructor did not communicate information about the test re-
sult, and only general positive encouragement (i.e., “you are doing a
good job”, “well done”, “continue to perform as fast as you can”).
2.3. Instruments
The SRT (see Fig. 1a) included one stimulus (i.e., an arrow, pointing
to the right), appearing in the middle of screen, with time intervals
between 0 ms and 2000 ms, randomly generated in LabVIEW™, Na-
tional Instruments. The practice trials were followed by 12 test trials.
The participant rested the index finger of the dominant hand on the
right arrow of the keyboard with the instruction to respond as fast as
possible, every time as soon as the stimulus appeared on the screen. The
result registered by the computer was the mean RT over 12 trials, ex-
pressed in milliseconds.
The ChRT (see Fig. 1b) included one arrow, pointing randomly to
any of the four directions (up, down, left, right), appearing in random
positions all over the screen. There were no time intervals between each
trial; rather, a new arrow appeared immediately after the participants'

47
D. Van Biesen et al.
Fig. 1. Example of stimulus representation during the three versions of the computerized
reaction time tasks (a) simple reaction time, (b) choice reaction time, and (c) Flanker test.
response. The stimulus continued for 30 s. The participant rested the
index finger of the dominant hand close to the arrow keys on the
keyboard with the instruction to respond by pressing the corresponding
arrow key (up, down, left, right) as fast as possible, as soon as the sti-
mulus appeared on the screen. The total number of responses in 30 s,
and the number of correct and wrong taps were automatically regis-
tered.
The adapted Flanker task (see Fig. 1c) included one stimulus target
(i.e., arrow pointing left, right, up or down) and four distractors
(flanker arrows surrounding the target), with no time intervals between
each trial. The participant rested the index finger of the dominant hand
close to the arrow keys on the keyboard, with the instruction to ignore
the flanker arrows, and to focus on the central arrow only, responding
as fast as possible by pressing the corresponding arrow key (up, down,
left, right), as soon as the stimuli appeared on the screen. The Flanker
configurations appeared randomly at nine possible locations around the
screen to increase the cognitive load (add visual search capacity) and to
avoid that the participants would tap random buttons as fast as possible
continuously without information processing. The total number of re-
sponses in 30 s, and the number of correct and wrong taps were auto-
matically registered.
2.4. Data analysis
All the statistical analyses were performed in the IBM SPSS Statistics
program version 22, with level of significance set at p < 0.05 to test
the three hypotheses. Measures of RT were defined as the average RT to
respond per trial and expressed in milliseconds (ms) (respectively
SRTms, ChRTms, and Flankerms). As the RT measures' means and
48
Intelligence 64 (2017) 45–51
standard deviations were positively skewed, they were log-transformed,
using log base 10, to produce normally distributed variables, resulting
in three RT variables: SRTmslog, ChRTmslog, and Flankermslog.
(1) Bivariate descriptive statistics and Pearson's correlations were used
to model the bivariate relationship between each of the three RT
measures and IQ scores in the sample of athletes with II.
(2) F-tests were used to compare differences in RT measures between
the group of athletes with II and the comparison group of equally
well-trained peers.
(3) Finally, to assess differences in RT task performance with increasing
cognitive load, between athletes with II and the comparison group,
a 2 × 3 ANOVA repeated measures analysis was performed. The
normality assumptions to perform the ANOVA were met after per-
forming log transformations (i.e., independent and continuous
variables, no significant outliers, normal data distribution).
Homogeneity of variance was not guaranteed; however, the large
and equally distributed sample size justified the choice for para-
metric statistics.
3. Results
The results of the study are described in three sections. First, we
provide the evaluation of the relationship between IQ scores and RT
measures in athletes with II (see Table 2). Next, we examine the com-
parison of performance on the three RT tasks in both samples, to de-
termine whether differences in RT exist between both samples (see
Table 3). In the third and final section, we provide the results of the
repeated measures ANOVA to evaluate the task by group interactions.
The relationship between RT and IQ was investigated using Pearson
product-moment correlation coefficient. There was a moderate negative
correlation between IQ and RT, ranging from r = − 0.34, p < 0.001
for the SRT task, r = − 0.46, p < 0.001 for the ChRT task, and
r = − 0.49, p < 0.001 for the Flanker task.
As shown in Table 3, significantly slower RTs were observed for
athletes with II compared to the comparison group on all three RT tasks.
The 2 × 3 ANOVA was conducted between group (athletes with II
versus comparison) and task with increasing cognitive load (SRTlog,
ChRTlog and Flankerlog). The results of the analysis are displayed in
Fig. 2. This analysis revealed a significant main effect for task, F(1,
204) = 1610.7, p < 0.001, η2 = 0.94. The RT significantly increased
with increasing cognitive load. A significant main effect was also found
for the between factor group, F(1, 204) = 163.59, p < 0.001,
η2 = 0.44, with athletes in the comparison group performing better on
all three RT tasks than II-athletes. Interestingly, there was an interac-
tion effect between the cognitive load of the task and the group, F(1,
204) = 27.13, p < 0.001, η2 = 0.21). The difference between the
groups was larger in the ChRT test compared to the SRT, and the dif-
ference between groups was larger in the Flanker test compared to the
ChRT and SRT test.
Table 2
Pearson's r correlations among the measures of the three reaction time tasks, IQ, age and
training volume in athletes with intellectual impairment.
Variable SRTmslog ChRTmslog Flankermslog IQ age Training
volume
SRTmslog 1 0.54⁎⁎ 0.48⁎⁎ − 0.34⁎⁎ 0.09 − 0.03
ChRTmslog / 1 0.85⁎⁎ − 0.46⁎⁎ 0.44⁎⁎ 0.26⁎
Flankermslog / / 1 − 0.49⁎⁎ 0.37⁎⁎ 0.21⁎
Note. SRTmslog = simple reaction time, ChRT = choice reaction time, Flanker = adapted
Flanker test, ms = milliseconds, log = log 10-transformed, training volume = accumu-
lated sport experience over the lifespan.
⁎
p < 0.05.
⁎⁎
p < 0.001.
D. Van Biesen et al.
Table 3
Descriptive statistics for the RT tasks by participant group, and repeated measures ANOVA results for between-group means.
Athletes with II
Median Range M (SD)
SRT (ms) 349.0 218.0–1489.0 409.8 (190.8)
ChRT (ms) 857.1 555.6–1666.7 892.7 (244.6)
Flanker (ms) 1111.1 666.7–3750.0 1256.7 (500.6)
Note. SRT = simple reaction time, ChRT = choice reaction time, Flanker = adapted Flanker test, ms = milliseconds.
Fig. 2. Comparison of response time for each RT task between both groups (athletes with
II and comparison group). Note. II = intellectual impairment, SRT = simple reaction
time, ChRT = choice reaction time. Error bars represent standard deviations. *p < 0.05.
4. Discussion
The purpose of this study was to investigate how RT and intelligence
were related in two samples of well-trained athletes. The relationship
between intelligence and RT was first investigated in a sample of in-
dividuals at the lower end of the intelligence range. IQ was moderately
negatively correlated with all measures of RT in the athletes with II and
the correlation varied between r = −0.34 for the simple RT, and
r = − 0.49 for the most complex RT task. As also expected, the dif-
ferences in RT performance between two samples of healthy, active
athletes with II and the comparison group was more pronounced when
the RT tasks required a higher cognitive load.
In general, slower RTs in individuals with II compared to individuals
without II have been found in numerous previous studies (Klotz,
Johnson, Wu, Isaacs, & Gilbert, 2011; Koichi, Hideyuki, & Mitsuru,
2011; Kosinski, 2013; Nissan et al., 2013). Not only the response time,
but also intra-individual variability in RT (which is measured as the SD
of the participant's RT over the various trials) has a large inverse cor-
relation with IQ (Jensen, 2006).
In the present study, three RT tasks were included with increasing
cognitive load, ranging from SRT (one visual stimulus – one stimulus
location – one response), over ChRT (four types of stimuli at various
locations - four corresponding response options), to the Flanker Task
(four types of stimuli – various locations – visual search - distracters/
response inhibition added). As expected, information-processing load
increased across the three RT tasks and the response time increased
accordingly in both populations. According to Schweizer (2001), RT
tasks that require a higher cognitive load (e.g., when inspection time is
involved) yield higher correlations between IQ scores and RT measures
than RT tasks with less cognitive demands (e.g., SRT). The results of our
study support this assumption, whereby there was a greater increase in
RT with increasing cognitive load in II-athletes compared to their
equally well-trained counterparts in the comparison group. In the
Intelligence 64 (2017) 45–51
Comparison group F η2 p
Median Range M (SD)
292.0 218.0–454.0 296.3 (41.3) 35.8 0.21 < 0.001
612.2 500.0–810.8 620.2 (59.8) 123.9 0.43 < 0.001
750.0 588.2–1000.0 745.5 (77.5) 107.8 0.47 < 0.001
adapted Flanker task, the response demands were comparable with
those in the ChRT task (i.e., four choices); however, additional cogni-
tive processes were involved in response selection, because distracters
were added and response inhibition was required on top of the visual
search task. As hypothesized, differences between athletes with II and
the comparison group became more pronounced with increasing cog-
nitive load. The challenge in this type of research is to find RT tasks
requiring sufficient cognitive load to distinguish between athletes with
II and the comparison group without II; however, higher-order cogni-
tive engagement should not include cognitive abilities such as working
memory or fluid reasoning, to make sure the RT task still measures
what it should be measuring, i.e., speed of response (Miller and Vernon,
1996; Nissan, Liewald, & Deary, 2013).
In a number of previous studies regarding the relation between RT
and measures of intelligence; not only the central tendency, but also the
different aspects of the shape of RT distributions have been examined
(Coyle, 2003; Schmiedek, Oberauer, Wilhelm, Süb, & Wittmann, 2007).
It has been claimed that aspects of RT distributions beyond means or
medians, for example RT variability, might carry important information
regarding the relationship between SRT tasks and intelligence (Jensen,
1998). For example, the worst performance rule states that the slowest
RTs on a cognitive task are more strongly related to intelligence than
are the faster RTs (Schmiedek, Oberauer, Wilhelm, Süb, & Wittmann,
2007). One possible explanation for this theory is that individuals at the
lower end of the intelligence range are more susceptible to attentional
lapses while performing RT tasks, which can also explain the RT versus
intelligence distributions observed within the present study for all three
RT tasks. Attentional lapses can occur when the participant completely
disengages from the task during one or more trials (Adam, Mance,
Fukuda, & Vogel, 2015). Attentional lapses could also manifest as de-
graded attentional control rather than as a complete lapse. Individuals
with II may have attentional lapses, mostly while performing the RT
tasks with the highest cognitive load. As a suggestion for further re-
search, it could be investigated whether the characteristics of the intra-
individual RT distribution are associated with the individual differences
in intelligence of well-trained athletes with and without II.
The specific target population for the study of RT in the present
study involved well-trained athletes with II, competing in international
events. Previous studies demonstrated that experienced athletes have a
better RT than novices (Abourezk & Toole, 1995; Khezri et al., 2011;
Nuri et al., 2012; Atan & Akyol, 2014). Nuri et al. (2012) showed that
volleyball players had better RT compared to non-athletes. The authors
attributed the superior perceptual abilities of the volleyball players to
the acquired knowledge while performing sport specific patterns and to
the extraction of this knowledge in similar conditions. Atan and Akyol
(2014) found that athletes from various sports (i.e., football, basketball,
judo, athletics and taekwondo) had faster RT parameters compared to
non-athlete peers. However, the question remains; what is the causal
relation for their finding: are people with faster RTs more inclined to
excel in sports, or did high-level athletes acquire faster RTs as a product
of their sport expertise? Exercise can affect RT; however, inconclusive
results have been reported regarding the precise nature of the effect of
exercise on RT and other measures of cognitive function (Kosinski,
2013). Some studies showed that vigorous exercise has the potential to
improve RTs; however, the strongest effects were found immediately
49
D. Van Biesen et al.
after exercise, and less in the long term (Tomporowski, 2003;
Kashihara & Nakahara, 2005; Davranche, Audiffren, & Denjean, 2006).
This might contribute to the explanation of the present findings that
being well-trained seems to have only limited positive effects on RT in
individuals with II; however, if there is a positive effect, it is more
apparent in the simple RT tasks compared to the higher order cognitive
tasks.
4.1. Strengths and limitations
Our study consisted of a unique sample of 103 international well-
trained athletes with II, and an equally well-trained comparison group
matched for gender, sport discipline, and sport experience. The results
provided insight in the relation between RT, sport expertise and cog-
nition. However, due to the field-based nature of this study, some
limitations need to be addressed. We were not able to systematically
eliminate some of the variables that might account for a proportion of
the observed variance in RT. The participants in the comparison group
of college and university students were younger than the II-athletes.
The average age difference was small (2 years), yet significant, there-
fore, it needs to be considered when interpreting the results. As already
stated previously, a more important limitation of this study was the
heterogeneity of samples regarding nationality. Whereas the compar-
ison group included all Belgian citizens, the international athletes with
II represented nine different countries. However, even the best per-
forming II-athletes were performing far below the average scores of the
comparison group, independent of what country they came from.
Finally, the order of testing was not randomized; to ensure that all tasks
were well understood by the participants with II, the tests were pre-
sented in increasing order of difficulty. Although mental fatigue and/or
attentional lapses may possibly influence performance on RT tasks in-
cluding multiple trials (Adam, Mance, Fukuda, & Vogel, 2015), it was
not expected to impact the results of the present study, as the total
duration of the three tests (including instruction) was maximum
15 min, and the total number of trials per test was limited.
Another factor that could possibly affect the results is the ability of
participants to work with computers, and more specifically the key-
board. Detailed information around the familiarity with IT devices was
not systematically collected; however, all participants had used com-
puters before, and sufficient time to get acquainted was foreseen prior
to each test. With respect to the three RT tasks used, these tasks were
selected on the basis of feasibility for an international population of
individuals with II, and increasing cognitive load. Some major differ-
ences between the test instruments need to be acknowledged. Time
intervals between subsequent trials were provided in the SRT task;
however, no time interval was provided in the ChRT and Flanker Tasks;
and an unequal number of trials was performed due to the constraint of
total time (30 s) rather than number of trials (SRT = 12 trails) being
utilized. These differences across the different paradigms need to be
taken into account when interpreting the results of this study.
5. Conclusion
The outcome of this study provided new insights regarding the re-
lation between RT and intelligence. Measures of intelligence (IQ) were
significantly related with all RT measures in individuals at the lower
end of the intelligence range, with people scoring higher on the in-
telligence tests also tend to have faster RTs. The present study also
confirmed that athletes with II (IQ ≤ 75) also have slower RTs com-
pared to equally well-trained athletes in the comparison group of col-
lege and university students. The group differences in RT among par-
ticipants differing in intelligence were amplified by increasing task
complexity, related to a task's g loading. The results of the present study
implied that the adapted version of the Flanker test is a speed-based test
with sufficiently high cognitive load to fully distinguish in RT perfor-
mance between samples of athletes with II versus athletes without II.
50
Intelligence 64 (2017) 45–51
Acknowledgements
The authors thank the International Federation for Intellectual
Disability Sport (INAS) and the International Paralympic Committee
(IPC) for their support to conduct this study.
References
AAIDD (2010). Intellectual disability, definition, classification and systems of support (11th
ed.). Washington, DC: AAIDD.
Abourezk, T., & Toole, T. (1995). Effect of task complexity on the relationship between
physical fitness and reaction time in older women. Journal of Aging and Physical
Activity, 3, 251–260.
Adam, K. C. S., Mance, I., Fukuda, K., & Vogel, E. K. (2015). The contribution of atten-
tional lapses to individual differences in visual working memory capacity. Journal of
Cognitive Neuroscience, 27(8), 1601–1616.
Atan, T., & Akyol, P. (2014). Reaction times of different branch athletes and correlation
between reaction time parameters. Procedia - Social and Behavioral Science, 116,
2886–2889.
Bartlo, P., & Klein, P. J. (2011). Physical activity benefits and needs in adults with in-
tellectual disabilities: systematic review of the literature. American Journal of
Intellectual and Developmental Disabilities, 116(3), 220–232.
Burns, J. (2015). The impact of intellectual disabilities on elite sports performance.
International Review of Sport and Exercise Psychology. http://dx.doi.org/10.1080/
1750984X.2015.1068830.
Carroll, J. B. (1993). Human cognitive abilities, a survey of factor-analytic studies.
Cambridge: Cambridge University Press (819pp.).
Cinaz, B., Vogt, C., Arnrich, B., & Tröster, G. (2012). Implementation and evaluation of
wearable reaction time tests. ElsevierPervasive and Mobile Computing, 8, 813–821.
Coyle, T. R. (2003). A review of the worst performance rule: Evidence, theory, and al-
ternative hypotheses. Intelligence, 31, 567–587.
Davranche, K., Audiffren, M., & Denjean, A. (2006). A distributional analysis of the effect
of physical exercise on a choice reaction time task. Journal of Sport Sciences, 24(3),
323–330.
Deary, I. J. (2000). Looking down on human intelligence. New York: Oxford University
Press.
Deary, I. J., Der, G., & Ford, G. (2001). Reaction times and intelligence differences: A
population-based cohort study. Intelligence, 29, 389–399.
Der, G., & Deary, I. J. (2003). IQ: Reaction time and the differentiation hypothesis.
Intelligence, 31, 491–503.
Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993). The role of deliberate practice
in the acquisition of expert performance. Psychological Review, 100(3), 363–406.
Eriksen, C. W. (1995). The flankers task and response competition: A useful tool for in-
vestigating a variety of cognitive problems. Visual Cognition, 2(2/3), 101–118.
Fry, A. F., & Hale, S. (2000). Relationships among processing speed, working memory,
and fluid intelligence in children. Biological Psychology, 54, 1–34.
Gómez-Íñiguez, C., Fuentes, L. J., Martinez-Sanchez, F., Campoy, G., Montoro, P. R., &
Palermo, F. (2014). Emotional cuing to test attentional network functioning in trait
anxiety. Psicológica, 35, 309–329.
Jensen, A. R. (1998). The g factor: The science of mental ability. New York: Praeger.
Jensen, A. R. (2006). Clocking the mind: Mental chronometry and individual differences.
Amsterdam: Elsevier Science.
Johnson, W., & Deary, I. (2011). Placing inspection time, reaction time, and perceptual
speed in the broader context of cognitive ability: The VPR model in the Lothian Birth
Cohort 1936. Intelligence, 39, 405–417.
Kashihara, K., & Nakahara, Y. (2005). Short-term effect of physical exercise at lactate
threshold on choice reaction time. Perceptual and Motor Skills, 100(2), 275–281.
Khezri, A., Shahbazi, M., Kashan, V., & Pashabadi, A. (2011). The comparison of simple
and choice reaction time between athletes and non-athletes at difference levels of
BMI. Electronic Physician, 3(3), 319.
Klotz, J. M., Johnson, M. D., Wu, S. W., Isaacs, K. M., & Gilbert, D. (2011). Relationship
between reaction time variability and motor skill development in ADHD. Child
Neuropsychology, 1–10.
Koichi, H. A., Hideyuki, O. B., & Mitsuru, K. B. (2011). Effects of age, intelligence and
executive control function on saccadic reaction time in persons with intellectual
disabilities. Research in Developmental Disabilities, 32, 2644–2650.
Kosinski, R. J. (2013). A Literature Review on Reaction Time. Clemson University. Retrieved
on 2 December 2015 from http://homepage.univie.ac.at/andreas.franz.reichelt/
intro2cogsci2/data/literature_review_reaction_time.pdf.
Lee, J. J., & Chabris, C. F. (2013). General cognitive ability and the psychological re-
fractory period: Individual differences in the mind's bottleneck. Psychological Science,
24, 1226–1233.
Matarazzo, J. D., & Herman, D. O. (1984). Relationship of education and IQ in the WAIS-R
standardization sample. Journal of Consulting and Clinical Psychology, 52(4), 631–634.
McGrew, K. S. (2009a). CHC theory and the human cognitive abilities project: Standing
on the shoulders of the giants of psychometric intelligence research. Intelligence, 37,
1–10.
McGrew, K. (2009b). CHC theory and the human cognitive abilities project: Standing on
the shoulders of the giants of psychometric research. Intelligence, 37, 1–10.
McLean, S. P., Garza, J. P., Wiebe, S. A., Dodd, M. D., Smith, K. B., Hibbing, J. R., &
Andrews Epsy, K. (2013). Applying the flanker task to political psychology: A re-
search note. Political Psychology. http://dx.doi.org/10.1111/pops.12056.
Miller, L. T., & Vernon, P. A. (1996). Intelligence, reaction time, and working memory in
D. Van Biesen et al.
4- to 6-year-old children. Intelligence, 22, 155–190.
Nettelbeck, T., Edwards, C., & Vreugdenhil, A. (1986). Inspection time and IQ: Evidence
for a mental speed-ability association. Personality and individual differences, 7(5),
633–641.
Nissan, J., Liewald, D., & Deary, I. J. (2013). Reaction time and intelligence: Comparing
associations based on two response modes. Journal of Intelligence, 41, 622–630.
Nuri, L., Shadmehr, A., Moghaddam, B. A., & Ghotbi, N. (2012). Comparison of reaction
time and anticipatory skill between female athletes and non-athletes. Journal of
modern Rehabilitation, 6(3), 37–43.
Raven, J. C., Raven, J., & Court, J. H. (1998). Manual for Raven's progressive matrices and
vocabulary scales. Section 4: Advanced progressive matrices. Pearson: San Antonio, TX.
Schmiedek, F., Oberauer, K., Wilhelm, O., Süb, H.-M., & Wittmann, W. W. (2007).
Individual differences in components of reaction time distributions and their relations
to working memory and intelligence. Journal of Experimental Psychology: General,
136(3), 414–429.
Schneider, W. J., & McGrew, K. S. (2012). The Cattell-Horn-Carroll Model of intelligence.
In D. P. Flanagan, & P. L. Harrison (Eds.), Contemporary Intellectual Assessment(3rd
Edition: Theories, Tests, and Issues). NY, US: The Guilford Press (926pp.).
Schweizer, K. (2001). Preattentive processing and cognitive ability. Intelligence, 29(2),
169–186.
Sheppard, L. D., & Vernon, P. A. (2008). Intelligence and speed of information-processing:
A review of 50 years research. Personality and Individual Differences, 44, 535–551.
Thorndike, R. L., Hagen, E. P., & Sattler, J. M. (1986). Guide for administering and scoring,
the Stanford-Binet intelligence scale (Fourth Edition). Chicago: Riverside.
51
Intelligence 64 (2017) 45–51
Tomporowski, P. D. (2003). Effects of acute bouts of exercise on cognition. Acta
Psychologica, 112, 297–324.
Trick, L. M., & Enns, J. T. (1998). Lifespan changes in attention: The visual search task.
Cognitive Development, 13, 369–386.
Van Biesen, D., Verellen, J., Meyer, C., Mactavish, J., Van de Vliet, P., & Vanlandewijck,
Y. (2010). The ability of elite table tennis players with intellectual disabilities to
adapt their service/return. Human Kinetics, 27, 242–257.
Van Biesen, D., Kerremans, J., Mactavish, J., & Vanlandewijck, Y. (2016a). Spatial vi-
sualisation and simple reaction time as predictors of performance among well-trained table
tennis players with intellectual disability. Accepted for publication Jan 2016.
Van Biesen, D., Mactavish, J., McCulloch, K., Lenaerts, L., & Vanlandewijck, Y. (2016b).
Cognitive profile of young well-trained athletes with intellectual disabilities. Research
in Developmental Disabilities, 53, 377–390.
Vernon, P. A. (1983). Speed of information-processing and general intelligence.
Intelligence, 7, 53–70.
Wechsler, D. (2008). Wechsler adult intelligence scale–fourth edition (WAIS-IV). San
Antonio, TX: Pearson.
Wechsler, D. (2014). Wechsler intelligence scale for children-fifth edition (WISC-V).
Bloomington, MN: Pearson.
Williams, A. M., & Ericsson, K. A. (2005). Perceptual-cognitive expertise in sport: Some
considerations when applying the expert performance approach. Human Movement
Science, 24, 283–307.
World Health Organization (2016). Global strategy on diet, physical activity and health.
Retrieved from http://www.who.int/dietphysicalactivity/background/en.