ARTICLE IN PRESS

Can the Clobber

game become a
classroom-based tool
for screening students’
executive functions?
Telma Paráa,*, Luı́s Alfredo Vidal de Carvalhoa, Paulo Mattosa,
Simone Dantasa, Sylvain Gravierb, and Sue Johnston-Wildera
a
Fundação de Apoio à Escola T

ecnica—FAETEC-RJ, Rio de Janeiro, Brazil
b
CNRS, Universit

e Grenoble Alpes, Grenoble, France
*Corresponding author: e-mail address: telma.para@faetec.rj.gov.br

Abstract
The use of games for cognitive screening is not new and involves employing simple tasks as
well as virtual reality. In this work, we introduce the use of the combinatorial game Clobber,
created by the mathematicians Albert, Grossman, Nowakowski and Wolfe in 2001 in a
classroom-based experiment and analyzed how it can assess cognitive functions. Specifically,
this study tries to address how the use of the Clobber game can target executive functions (EFs)
and why it may be a valuable game to assess EFs. Executive functions have an extremely com-
plex nature and combine abilities which involve planning, decision-making, productive action,
and self-regulation, among others. We performed a cross-sectional study with a sample of 111
participants aged 9–30 from three educational levels in which Clobber was applied in four dif-
ferent configurations varying in complexity. The findings identify two variables that can guide
future experiments with Clobber: the game configuration and the time spent solving the game.

Keywords
Classroom-based intervention, Executive functions, Cognitive screening, Clobber game,
Mathematics

1 Introduction
Executive functions (EFs) are described as cognitive and behavioral processes that
enable self-control and planning (Lezak et al., 2004). Planning is defined as the pro-
cess of formulating a sequence of operations intended to achieve a goal in an
Progress in Brain Research, ISSN 0079-6123, https://doi.org/10.1016/bs.pbr.2023.07.001
Copyright © 2023 Elsevier B.V. All rights reserved.
1
 ARTICLE IN PRESS

2 Can the clobber game become a classroom-based tool?

organized, strategic, and efficient manner (Hayesroth and Hayesroth, 1979). Some
“hands-on” neuropsychological tests suggest that positive motivation affects plan-
ning skills, and that motivation can be enhanced by adding structure to a specific
task. For example, dividing large tasks into smaller steps to keep the adolescent/child
actively involved may help resolve daily life planning problems (Boyer et al., 2018).
A successful attempt to organize this suit of abilities described as EFs is the
model proposed by Miyake et al. (2000). This model operationalizes EFs as being
composed of three main domains: Inhibition (ability to prevent pre-potent responses
or suppression of certain types of activation); Updating (ability to manipulate infor-
mation in real-time); and Shifting (ability to switch between tasks and processes).
These EF domains help us to perform activities of daily living (Cahn-Weiner
et al., 2002) and are age-related, increasing from childhood to adulthood and declin-
ing during aging (Hasher et al., 1991; Hasher and Zacks, 1988). One of the features of
EFs is their association with a variety of categories of responses such as temporary
suppression of reflex action and dealing with new and unexpected situations
(Shallice, 1988). A challenge for the neuropsychological assessment of EFs, in
particular, is that tests are often well-structured. The EFs assessments typically pro-
vide instructions that can, non-intentionally, give participants tips on how to solve
the tasks, leading to overestimation of their abilities. That can mask difficulties in
planning, decision-making, productive action, and self-regulation, making the task
unable to capture existing dysfunctions in these skills.
Another challenge in EF assessments concerns tasks’ validity across sociodemo-
graphic groups, such as different gender and age-groups (Parkin, 1996). For instance,
the literature review by Grissom and Reyes (2019) supports the findings that there are
small and subtle gender differences in EFs although individual factors may show a
tendency toward a sex bias (e.g., increased impulsive action in males, reduced reac-
tion time in males, avoidance of frequent punishment in females, improved working
memory in females). According to these authors: “The fact that many studies draw
different conclusions on whether there is a gender/sex difference in executive func-
tions based on small modifications to task design suggest that differences in strategy
and outcome preference drive apparent effects on executive function, rather than a
difference in ability between genders/sexes.” (Grissom and Reyes, 2019, p. 7). The
authors observe that future studies should adopt gender differences in strategy sug-
gesting that males and females utilize diverse brain circuits and/or molecular mech-
anisms to solve the same cognitive problems, that is, even though the ability may be
the same, the strategies employed are unlikely to be supported by the same neuro-
biological mechanisms.
Regarding aging, EFs begin to emerge early in infancy, with basic skills needed
for EFs emerging before 3 years of age, and more specific skills developing into early
childhood (Garon et al., 2008). It has been suggested that each component of EF
grows at its own rate across childhood and adolescence, reaching maturity at differ-
ent ages (Diamond, 2013). For example, Zelazo et al. (2014) found that cognitive
flexibility abilities continue to improve at the ages of 20–29 years old and reinforce
the importance of using different approaches and tasks to assess EFs abilities.
 ARTICLE IN PRESS

1 Introduction 3

In contrast, cognitive performance peaks in young adulthood with declines emerging

as early as 20 or 30 years old (Hartshorne and Germine, 2015).
Concerning the neural basis of individual differences in Executive Functioning,
most neuroimaging studies have focused on the frontoparietal and cingulo-opercular
areas that tend to activate across individuals during EF tasks (Collette et al., 2005;
Dosenbach et al., 2006; Duncan, 2010; Fedorenko et al., 2013; Nee et al., 2013;
Niendam et al., 2012). According to Yarkoni and Braver (2010), the brain regions
that activate consistently across individuals for any given task may not necessarily
be sensitive to individual differences.
Reineberg et al. (2022) highlight two issues in studies that investigate the neural
basis of individual differences in EFs. First, most studies have focused on single EF
tasks and EFs domain-general processes that control lower-level processes in diverse
contexts. However, to get a full picture of an EF, it is necessary to measure perfor-
mance in multiple contexts or tasks and look at what is common across these contexts
(Miyake et al., 2000). This multitask approach is popular in behavioral literature but
is rarely adopted within the neuroimaging literature. Second, most fMRI studies of
EF individual differences to date are modest in sample size. This is limiting for fMRI
studies of individual differences, which require larger sample sizes to detect reliable
and reproducible associations, compared to studies that focus on group mean effects
(Yarkoni, 2009).
Addressing these gaps, Rodrı́guez-Nieto et al. (2022) conducted a systematic
review and meta-analysis of neuroimaging studies with large samples to identify
functional activation that is shared across all commonly used executive functioning
paradigms (global executive network). The study revealed a widespread activation
across bilateral frontoparietal regions, insula, basal ganglia, and occipital regions,
extending to the cerebellum. The authors also performed a conjunction analysis to
detect brain areas that were consistently recruited across the three key domains of
executive functioning (core executive network) that derived from the conjunction
among Inhibition, Shifting and Updating. This analysis showed a substantial overlap
across cortical and subcortical regions, less extensive activation in the bilateral
middle frontal gyrus and did not engage the cerebellum (Rodrı́guez-Nieto et al., 2022).
EFs have been linked to various life outcomes (Diamond, 2013), including
school success (Duncan et al., 2007; St Clair-Thompson and Gathercole, 2006).
In particular, better EFs are linked to better educational outcomes such as school
achievement, school engagement, and self-regulated classroom behaviors (Diamond,
2013; Obradovi
c et al., 2012). By contrast, EF deficits—which are common in children
with Attention-Deficit/Hyperactivity Disorder (ADHD)—have an important negative
impact on student’s classroom experience (Cocchi et al., 2012; Figueiredo et al.,
2023; Hinshaw et al., 2011; Nejati, 2021).
The conventional approach in clinical or research settings is to measure students’
EFs skills using standardized EFs tasks developed in artificial settings, such as
laboratories. This conventional approach, although effective in most cases, lacks
ecological validity regarding students’ use of EF in school settings or in daily life
(Obradovi
c et al., 2018). Kiili and Ketamo (2017) claim that, as EFs are assessed
 ARTICLE IN PRESS

4 Can the clobber game become a classroom-based tool?

outside the classroom environment, and the examiner works closely with partici-
pants, external factors may artificially inflate EFs’ performance compared with
the natural setting of the classroom. One possible solution to improve the ecological
validity of EF tasks is to use game-based assessments.

2 Using games for EF screening

In this section, we discuss the use of games in EF screening and introduce the
Clobber game as a potential tool to assess executive functions in classroom-based
activities.
The use of games to measure executive functioning is not new. Several games
have been used as neuropsychological measures, such as the Tower of Hanoi Test,
Tower of London Test, and Stroop Test. The Tower Test assesses several key EFs,
including planning. The objective is to move disks, varying in size from small to
large, across three pegs to build a designated tower in the fewest number of moves
possible. In constructing target towers, the examinee must follow two rules: (a) move
only one disk at a time and (b) never place a larger disk over a smaller disk (Boyer
et al., 2018). There are also games that are based on traditional psychological tasks.
For example, the Stroop task, which measures selective attention, cognitive flexibil-
ity, processing speed, and interference control (Doyler et al., 2000; Lansbergen et al.,
2007; Lezak et al., 2004; MacLeod, 1991; Strauss et al., 2006). The Stroop effect can
be observed in paper and pencil and through a game-based adaptation, participants
usually show slower speed in naming the ink color of a word if the color is incon-
gruent with the meaning of the word, e.g., say blue when the ink is red vs say red
when the word ink is also red. Many authors found a significant correlation between
the amount of time spent on the parts of the game that required EFs and the
interference scores on a Stroop task.
“Serious” games, designed to have a primary purpose other than entertainment
(Charsky, 2010), have been used for cognitive assessment by many authors
(Abd-alrazaq et al., 2022; Giglioli et al., 2021; Tong et al., 2021; Wang et al.,
2022). Games have been used to evaluate EFs using virtual reality (Parsons and
Reinebold, 2012) or using simple tasks such as cooking (Manera et al., 2015). Other
researchers have used games to distinguish adults with or without cognitive decline
(Jimison et al., 2004; Tarnanas et al., 2015). In this study, we propose the use of a
mathematical game, Clobber, more accurately classified as a combinatorial game,
and seek to identify the variables that make Clobber a possible test to assess and even
train EFs.

3 The Clobber game

Combinatorial games satisfy the following conditions (Berlekamp et al., 2018): there
are two players, player A and player B; player A is the first to play and B is the sec-
ond; there are several usually finitely many positions where tokens (or stones) can be
 ARTICLE IN PRESS

3 The clobber game 5

placed, and often a particular starting position; there are clearly defined rules that
specify the two sets of moves that the players can make from a given position;
the players move alternately. In the game as a whole, there are no chance moves;
no dealing of cards; no rolling of dice; a player who is unable to move loses; the rules
are such that a play will always come to an end because some player will be unable to
move (ending condition). Some popular examples of combinatorial games are Chess;
Checkers; Hex; Go; and Geography.
Clobber is a two-player combinatorial game introduced by Albert et al. (2005) at
the Seminar on Algorithmic and Combinatorial Game Theory in Dagstuhl, Germany
(Albert et al., 2005). It is played with black and white stones occupying some subset
of the squares of an n by m checkerboard (see Fig. 1). The two players, White and
Black, move alternately by picking up one of their own stones and they clobber
(or “eat”) an opponent’s stone on a horizontally or vertically adjacent square. The clob-
bered stone is removed from the board, and it is replaced by the picked stone. The game
ends when one player, on their turn, is unable to move, and then this player loses.
In this paper, we study a Clobber game modeled using graph theory. A graph
G ¼(V, E) is a mathematical structure defined by a non-empty collection V of points,
called vertices, and a set E of segments that connects a pair of vertices of V, called
edges. Two vertices u and v are adjacent if they share a common edge e ¼ (u,v).
In this case, e is said to be incident with u (and v).
In 2004, Demaine et al. (2004) introduced a new version of the game, called the
Impartial Solitaire Clobber (ISC), with the following rules: given a graph G ¼ (V, E),

FIG. 1
Clobber 5 by 6 checkerboard with black and white stones.
 ARTICLE IN PRESS

6 Can the clobber game become a classroom-based tool?

J1

u v w
1 2 3

FIG. 2
An example of Clobber game played on Configuration J1.

place a black or white stone in each vertex v in V. A graph with a stone placed in
every vertex of V is called configuration. A move consists of picking up a stone
of any color located in a vertex v and clobbering another one of the opposite colors,
located in a vertex adjacent to v, taking its place (eating it). The vertex v is then
removed from the graph with all its incident edges. The player is not forced to alter-
nate moves with white and black stones, and the game ends when no moves are
allowed, that is, it is not possible to eat any more stones. The game’s main goal is
to minimize the number of stones left in the graph.
In Fig. 2, we present an example Clobber game played on Configuration J1,
where the numbers on the red arrows represent a sequence of moves. Playing this
configuration, in the player’s first move, the black stone on vertex v clobbers the
white stone on vertex w. We observe that another possible move is the black stone
on vertex v clobbers the white stone on vertex u.
Note that, in Fig. 2, the edges are displayed in the form of either arcs or line
segments. Considering only line segments, we say that an edge is of reach k, if
the number of line segments between its endpoints is k. For example, in Fig. 2,
the edge (u,v) has reach 2 and the edge (v,w) has reach 1.

3.1 The present study

We carried out a classroom-based assessment to measure the performance of
students of different levels of education in the game. Participants’ age, gender,
and school level were analyzed, relating them to the time spent in each con-
figuration of the game, success in the game and the configuration of the game
itself.
Specifically, this study focuses on the following research questions:

1. How could Clobber measure executive functions?

2. Why may Clobber be a useful game to measure EFs?
 ARTICLE IN PRESS

4 Methods 7

In the Clobber game, the player must analyze the available set of moves available and
choose moves that reduce the graph aiming at a minimum number of stones at the end
of the game. This involves the ability to update one’s working memory with infor-
mation (Updating) as well as the ability to switch between tasks and processes
(Shifting). Participants may also use an important EF skill, planning, to anticipate
movements. The game presents different levels of difficulty, for example, the distri-
bution of the coloring of the stones, and the number of edges incident with a vertex v
in a given turn.

4 Methods
This was a cross-sectional study approved by the Ethics Committee of the Institute of
Psychiatry, at the Federal University of Rio de Janeiro (Book 07, protocol number
14, 11/2010). Informed Consent forms were obtained from all parents or guardians,
in the case of minors. The parents and adult participants also filled out a sociodemo-
graphic questionnaire (i.e., the Brazilian Institute of Geography and Statistics form).
The research carried out forms part of one doctoral thesis.

4.1 Participants
A convenience sample of 126 participants recruited from three educational
institutions in Rio de Janeiro took part in the study. Data from participants with
a previous history of learning, neurological, or psychiatric disorders were ex-
cluded from the analyses (n ¼ 15). The final sample included in the data analysis
consisted of 111 participants aged from 9 to 30 years (65 males and 46 females;
M ¼ 16.2; SD ¼ 4.2). Regarding schooling, participants were distributed as fol-
lows: 20 students from Elementary School I (ES-I); 28 students from Elementary
School II (ES-II), 25 students from High School (HS), and 38 students from Un-
dergraduate Education (UE). Participant characteristics by gender are included in
Table 1.

Table 1 Sample characteristics separated by gender.

Male Female

Participants, N (%) 65 (58.6) 46 (41.4)

Mean age in years, (SD) 16.9 (4.3) 15.3 (3.9)
Schooling, N(%)
UE 29 (76.3) 9 (23.7)
HS 13 (52.0) 12 (48.0)
ES-II 13 (46.4) 15 (53.6)
ES-I 10 (50.0) 10 (50.0)
 ARTICLE IN PRESS

8 Can the clobber game become a classroom-based tool?

4.2 Materials and procedures

The study was conducted in classrooms with different groups of primary, second-
ary, and undergraduate students. The task was explained to each participant at the
start of the session. The entire process took place at a pre-agreed time, agreed with
the school board and the class teacher. The workshop did not interfere with the
scheduled activities and students were advised that there was no obligation to take
part in the experiment and that they could withdraw at any time without
consequence.
The experiment was carried out in the whole class for Elementary School II
(ESII), Highschool (HS) and Undergraduate (UE) participants; and in groups of 4
children with Elementary School I students (ESI). The workshop lasted for a
maximum of 2 h to allow time to introduce the game and 50 min of gameplay.
In this paper, we apply the Clobber game in one experiment with four config-
urations for each participant, from Configurations J1 through to J4, as we add
more stones, changed the color of the stones, and add more edges. The game’s
main goal is to obtain the minimum number of stones at the end of each given
configuration (J1–J4). Participants were allowed to redo their solution as many
times as they wished, within a total time limit of 50 min for all four configura-
tions. A correct response means that the solution was achieved with the minimum
number of stones.
The proposed configurations of the Clobber game were given as a paper-based
activity. Each configuration was presented on a sheet of paper where the participant
individually had to draw their moves (step-by-step) and exhibit the resultant config-
uration after each step. At the end of the experiment, we showed them the optimum
results for each configuration with the possible moves.
In this study, the game configurations were planned to assess participants’ per-
formance and identify variables that can be useful in future applications of the game
aiming at executive functions. A participant who randomly plays this game will
probably be left with many stones at the end of the task. The set of configurations
was designed in such a way that the participant needed a strategy that involves
working memory (since when performing the second move in the game, the partic-
ipant will need to retain and operate information for the next move) and attention
since the game involves restrictions (it is only possible to make a move with stones
of different colors that can be reached). It is expected that the participant will not
have difficulties when performing J1 and J2; due to the small number of existing
stones the responses may be related to innate number sense (Wynn, 1998). In
Configuration J3, however, participants are expected to have more difficulty and take
longer to solve, since this configuration, which is more complex, has a greater num-
ber of stones maintaining edges of reach 1 and 2. In Configuration J4, the coloring is
maintained, and more paths are introduced to carry out the moves (edges of reach 1, 2
and 3). It is expected that the participant will take less time with this configuration
than the previous one. In the set of configurations applied in the experiment, the
smallest number of stones (chance of success) is 1 stone for configurations J1, J2
and J4, and 2 stones for Configuration J3.
 ARTICLE IN PRESS

5 Results and analysis 9

J1 J2

J3 J4

FIG. 3
Clobber configurations used in the intervention.

Participants were not allowed to change their moves on the paper sheet once it
was delivered to the researcher. We present in Fig. 3 the game configurations
(J1–J4) used in this experiment.

5 Results and analysis

Data analyses were performed using R Software, specifically the LME4 package
(Faraway, 2016; R Foundation, 2023). The evaluated models considered the vari-
ables age, gender, school level and game configuration, as well as interactions
between them, the gameplay success and the time spent to solve the configura-
tions. As we had data from four levels of education (EI, EII, HS, UE), it seems
reasonable that performance in Clobber depends on these levels. Therefore, we
worked with multilevel models by placing a hierarchical level at the school level.
This is reflected in the inclusion of a random effect in the models that we
considered.

5.1 Chance of success

First, we examine associating the probability of the participant being able to
achieve the lowest number of stones in the game configuration (named chance
of success) with the covariates considered previously. As the variable of interest
is a probability (value between zero and one), it is appropriate to use the logistic
regression model with a random effect on the intercept (Myers et al., 2012). We
observed that the game configuration variable is introduced in the model by in-
cluding three variables (i ¼ 2, 3, 4) in which the i-th variable indicates whether
the observation was taken using the i-th configuration. A variable associated with
the first configuration was not introduced as this would generate an identifiability
 ARTICLE IN PRESS

10 Can the clobber game become a classroom-based tool?

model problem (Hoffmann, 2004). Thus, the coefficients associated with the
game configurations should be interpreted as the effect of replacing the first con-
figuration with the i-th configuration.
Eight models with different combinations of covariates were then considered.
These models were:
Model 1: intercept
Model 2: intercept + gender
Model 3: intercept + age
Model 4: intercept + configuration
Model 5: intercept + gender + age
Model 6: intercept + gender + configuration
Model 7: intercept + age + configuration
Model 8: intercept + gender + age + configuration
To identify the model to be chosen, the AIC (Akaike Information Criterion) was used
(Agresti, 2012) and the model that had the smallest AIC was chosen. We obtained the
following AIC values for the eight models listed above.
According to Table 2, the model chosen was model 6 (configuration and gender).
For this model, we obtained the following estimates for the coefficients (beta-values)
and the respective p-values (Table 3).
The p-values associated with all coefficients were <0.05 except for the p-value
associated with the coefficient of Configuration J2, indicating that they were

Table 2 AIC values associated with each of the

models considered in the study of the probability of
success in the game.
Model Name AIC

Model 1 Empty model (intercept) 549.3

Model 2 Gender 544.5
Model 3 Age 550.6
Model 4 Configuration 463.2
Model 5 Gender + age 546
Model 6 Gender + configuration 456.7
Model 7 Age + configuration 463.9
Model 8 Gender + age + configuration 457.8

Bold values highlight the chosen model (the smallest AIC).

Table 3 Estimates of coefficients (beta-values) and p-values in the study of the

probability of success in the game.
Intercept Gender J2 J3 J4

p-values 0.023 0.003 0.259 3.97*10(11) 3.46*10^(12)

Beta-values 1.349 0.715 0.365 2.374 2.555
 ARTICLE IN PRESS

5 Results and analysis 11

statistically significant at the 5% level. This means that there is statistical evidence
that both gender and changing the game configuration from J1 to J3 or J4 affect the
probability of success in the game. All beta values were negative, except for the in-
tercept, which allows us to see that the estimated probability of success in the game is
higher for males on average and that this probability decreases when we replace Con-
figuration J1 with configurations J3 or J4.
As a multilevel logistic regression with a random effect on the intercept was per-
formed, it is also possible to obtain estimates of the intercepts conditioning to a spe-
cific school level.
The estimates obtained are described in Table 4, which shows a difference in the
intercepts when there is a change in the level of education. ELI-level participants are
less likely to succeed, while UE-level participants are more likely. At levels ELII and
HS, the difference in the chance of success is minor.

5.2 Time needed

Second, we associate the time needed to solve a given game configuration consid-
ering the observations in which the participant obtained the smallest number of
stones in the game configurations (chance of success) with the covariates considered
previously. As the variable of interest is a strictly positive continuous value, it is
adequate to use the general linear model with a gamma distribution for the response
variable and with a random effect on the intercept (Dobson and Barnett, 2018). Eight
models with different combinations of covariates were then considered. We obtained
the following AIC values for them (see Table 5).

Table 4 Estimates of the intercepts associated with each level of

education in the study of the probability of success in the game.
Schooling level ELI ELII HS UE

Beta-values 0.381 1.078 0.856 3.089

Table 5 AIC values associated with each of the models

considered in the study of the time required to solve the game.
Model Name AIC

Model 1 Empty model (intercept) 900.1

Model 2 Gender 902.1
Model 3 Age 902.0
Model 4 Configuration 670.8
Model 5 Gender + age 904
Model 6 Gender + configuration 672.5
Model 7 Age + configuration 672.8
Model 8 Gender + age + configuration 674.5

Bold values highlight the chosen model (the smallest AIC).

 ARTICLE IN PRESS

12 Can the clobber game become a classroom-based tool?

Table 6 Estimates of coefficients (beta-values) and p-values in the study of the

time needed to solve the game.
Intercept J2 J3 J4

p-values <2*10^(16) 6.73*10^(7) <2*10^(16) <2*10^(16)

Beta-values 0.681 0.456 0.501 0.481

Table 7 Estimates of the intercepts associated with each level of education in

the study of the time spent playing the game.
Schooling level ELI ELII HS UE

Beta-values 0.701 0.647 0.715 0.625

Again, we chose the model that has shown the smallest AIC, the model chosen
was model 4 (configuration).
For this model, we obtained the following estimates for the coefficients
(beta-values) and the respective p-values (Table 6). The p-values associated with
all coefficients were <0.05 indicating that they were statistically significant at the
5% level. This means that there is statistical evidence that changing the game con-
figuration from J1 to any other considered configuration affected the time taken to
solve the game. From the beta values, the estimated time needed to complete the
game decreases when replacing Configuration J1 with J2. On the other hand, this
estimate increases when replacing Configuration J1 with J3 or J4.
It is also possible to obtain estimates of the intercepts conditioning to a specific
school level.
The estimates obtained are described in Table 7.
There is little difference in the intercepts when there is a change in the level of
education. This indicates that changing the level of education does not lead to sig-
nificant changes in the time spent solving the game. This interpretation is confirmed
when we observe that only 0.461% of the variability in time spent is explained by the
level of education.

6 Discussion
This cross-sectional study investigated the performance of students of different
levels of education in the Clobber game, seeking to identify variables that the
Clobber game generates that may be useful for cognitive screening.
Regarding the study associating the participant’s probability of reaching the
lowest number of stones when changing the game configuration from J1 to J3 or
J4, the probability of success decreases. We can infer that there is an increase in the
difficulty of performing the game, requiring greater cognitive ability. Configurations
J3 and J4 seem to require more working memory effort (the capacity to hold and
 ARTICLE IN PRESS

7 Limitations and future prospects 13

manipulate information “on-line” in real time) due to the increased amount of choice
of moves. Another significant result is the impact of school level on the probability of
success in the game as it shows differences in the performance of students with a
greater chance of success for UE. We suggest that the Clobber game has the potential
to be used for training at initial levels of education (ELI) to enhance executive func-
tions as according to Dias et al. (2022) playing the game regularly could help chil-
dren to hold and work with information in their mind, focus their attention and inhibit
disruptive behavior, becoming more flexible, adapting to new situations and
demands in learning and day-to-day functioning.
Regarding the time needed to solve a given game configuration, changing the
game configuration increases the time spent to run it, reinforcing the above-
mentioned increase in game complexity. However, it was found that there was no
considerable difference in the execution time of the game in relation to the school
level. This shows the different patterns of participants related to cognitive abilities
which suggests that student performance possibly is associated with the environ-
ment, family income and parents’ education, particularly in a developing country
such as Brazil. Therefore, there is a need, at least locally, for greater investments
in public education environments that promote stimulation and consistent studies
on the benefits of such stimulation.

7 Limitations and future prospects

This is the first time that the Clobber game has been studied as a tool to examine
executive functions. The present study was focused on describing in-game features
and key demographic variables interfering with performance. However, partici-
pants’ baseline of EFs skills was not obtained, which could explain individual dif-
ferences in performance. Moreover, no validity assessments were carried out as a
proof of concept. The findings indicate that this might be a novel venue for future
studies to examine EFs through the Clobber game. This opens a new research area
involving the comparison of Clobber with other neuropsychological tests. After
further validity studies, the Clobber game could be adapted for online research
and implemented in more interactive interfaces. It would allow better control of
reaction time and in-game progress, for instance, with indicators of focused atten-
tion, number of trials, etc. A digital approach could also lead to the use of more
configurations for screening or training in less time. Hopefully, after more
advanced designed studies, the Clobber game may be adapted as a digital interven-
tion tool to stimulate EFs. However, many schools are still having scarce resources
such as gadgets and Internet connection, as it depends on the social and economic
context of a country. Future studies with Clobber are desirable. For example, getting
larger samples with more age bands and preferentially including clinical samples for
each of the configurations (J1–J4); performing more configurations; and making mea-
sures over time (response time) to help determine whether the participant is exhibiting
cognitive increase or decline.
 ARTICLE IN PRESS

14 Can the clobber game become a classroom-based tool?

8 Conclusions
Many steps are still required before this game can be used as a screening tool in
schools. The study identified two variables that could guide future research con-
cerning the interface between the Clobber game and EFs: the game configuration
and the time spent to solve the game. However, we suggest that new studies
should collect quantitative and qualitative data in order to elaborate parameters
for measuring EFs.

Acknowledgments
This study was partially funded by the Coordenação de Aperfeiçoamento de Pessoal de Nı́vel
Superior—Brazil (CAPES)—Finance Code 001, by the Fundação de Amparo à Pesquisa do
Estado do Rio de Janeiro—Brazil (FAPERJ), and by Conselho Nacional de Desenvolvimento
Cientı́fico e Tecnológico—Brazil (CNPq). We would also like to thank Rafael Santos from the
Federal University of Rio de Janeiro for his support with the Statistics. The authors acknowl-
edge the reviewers for the critical comments that improved our article.

Disclosure
The author(s) declared no potential conflicts of interest with respect to the research, author-
ship, and/or publication of this article.

References
Abd-alrazaq, A., Alhuwail, D., Ahmed, A., Househ, M., 2022. Effectiveness of serious games for
improving executive functions among older adults with cognitive impairment: systematic re-
view and meta-analysis. JMIR Serious Games 10 (3), e36123. https://doi.org/10.2196/36123.
Agresti, A., 2012. Categorical Data Analysis. vol. 792 John Wiley & Sons.
Albert, M.H., Grossman, J.P., Nowakowski, R.J., Wolfe, D., 2005. An introduction to Clobber.
Integers: J. Comb. Number Theory 5 (2), 1–12. A1.
Berlekamp, E.R., Conway, J.H., Guy, R.K., 2018. Winning Ways for Your Mathematical
Plays, Volume 3. CRC Press.
Boyer, B.E., Geurts, H.M., Van der Oord, S., 2018. Planning skills of adolescents with ADHD.
J. Atten. Disord. 22 (1), 46–57. https://doi.org/10.1177/1087054714538658.
Cahn-Weiner, D.A., Boyle, P.A., Malloy, P.F., 2002. Tests of executive function predict in-
strumental activities of daily living in community-dwelling older individuals. Appl. Neu-
ropsychol. 9 (3), 187–191. https://doi.org/10.1207/S15324826AN0903_8.
Charsky, D., 2010. From edutainment to serious games: a change in the use of game charac-
teristics. Games Cult. 5 (2), 177–198. https://doi.org/10.1177/1555412009354727.
Cocchi, L., Bramati, I.E., Zalesky, A., Furukawa, E., Fontenelle, L.F., Moll, J., Mattos, P.,
2012. Altered functional brain connectivity in a non-clinical sample of young adults with
attention-deficit/hyperactivity disorder. J. Neurosci. 32 (49), 17753–17761. https://doi.
org/10.1523/JNEUROSCI.3272-12.2012.
 ARTICLE IN PRESS

References 15

Collette, F., Van der Linden, M., Laureys, S., Delfiore, G., Degueldre, C., Luxen, A.,
Salmon, E., 2005. Exploring the unity and diversity of the neural substrates of executive
functioning. Hum. Brain Mapp. 25 (4), 409–423. https://doi.org/10.1002/hbm.20118.
Demaine, E.D., Demaine, M.L., Fleischer, R., 2004. Solitaire Clobber. Theor. Comput. Sci.
313, 325–338. https://doi.org/10.1016/j.tcs.2003.02.001.
Diamond, A., 2013. Executive functions. Annu. Rev. Psychol. 64, 135–168. https://doi.org/10.
1146/annurev-psych-113011-143750.
Dias, N.M., Pereira, A.P.P., Seabra, A.G., 2022. Executive functions in the prediction of ac-
ademic performance in elementary education. Psicologia: Teoria e Pesquisa 38.
Dobson, A.J., Barnett, A.G., 2018. An Introduction to Generalized Linear Models. CRC press.
Dosenbach, N.U.F., Visscher, K.M., Palmer, E.D., Miezin, F.M., Wenger, K.K., Kang, H.C.,
Burgund, E.D., Grimes, A.L., Schlaggar, B.L., Petersen, S.E., 2006. A core system for the
implementation of task sets. Neuron (Cambridge, Mass.) 50 (5), 799–812. https://doi.org/
10.1016/j.neuron.2006.04.031.
Doyler, A.E., Biederman, J., Seidman, L.J., Weber, W., Faraone, S.V., 2000. Diagnostic ef-
ficiency of neuropsychological test scores for discriminating boys with and without atten-
tion deficit-hyperactivity disorder. J. Consult. Clin. Psychol. 68, 477–488. https://doi.org/
10.1037/0022-006X.68.3.477.
Duncan, J., 2010. The multiple-demand (MD) system of the primate brain: mental programs
for intelligent behaviour. Trends Cogn. Sci. 14 (4), 172–179. https://doi.org/10.1016/j.tics.
2010.01.004.
Duncan, G.J., Dowsett, C.J., Claessens, A., Magnuson, K., Huston, A.C., Klebanov, P.,
Pagani, L.S., Feinstein, L., Engel, M., Brooks-Gunn, J., Sexton, H., Duckworth, K.,
Japel, C., 2007. School readiness and later achievement. Dev. Psychol. 43 (6),
1428–1446. https://0-doi-org.pugwash.lib.warwick.ac.uk/10.1037/0012-1649.43.6.1428.
Faraway, J.J., 2016. Extending the Linear Model With R: Generalized Linear, Mixed Effects
and Nonparametric Regression Models. CRC press.
Fedorenko, E., Duncan, J., Kanwisher, N., 2013. Broad domain generality in focal regions of
frontal and parietal cortex. Proc. Natl. Acad. Sci. U. S. A. 110 (41), 16616–16621. https://
doi.org/10.1073/pnas.1315235110.
Figueiredo, T., Sudo, F.K., Serra-Pinheiro, M.A., Mattos, P., 2023. Interpersonal negotiation
impairment in ADHD: the critical role of comprehension processing. Curr. Psychol. (New
Brunswick, N.J.) 42 (6), 4842–4850. https://doi.org/10.1007/s12144-021-01767-8.
Garon, N., Bryson, S.E., Smith, I.M., 2008. Executive function in preschoolers: a review using
an integrative framework. Psychol. Bull. 134 (1), 31. https://doi.org/10.1037/0033-
2909.134.1.31.
Giglioli, I.A.C., Carrasco-Ribelles, L.A., Parra, E., Marı́n-Morales, J., Alcañiz Raya, M.,
2021. An immersive serious game for the behavioral assessment of psychological needs.
Appl. Sci. 11 (4), 1971. https://doi.org/10.3390/app11041971.
Grissom, N.M., Reyes, T.M., 2019. Let’s call the whole thing off: evaluating gender and sex
differences in executive function. Neuropsychopharmacology 44 (1), 1–11. https://doi.
org/10.1038/s41386-018-0179-5.
Hartshorne, J.K., Germine, L.T., 2015. When does cognitive functioning peak? The asynchro-
nous rise and fall of different cognitive abilities across the life span. Psychol. Sci. 26 (4),
433–443. https://doi.org/10.1177/0956797614567339.
Hasher, L., Zacks, R.T., 1988. Working memory, comprehension, and aging: a review and a
new view. In: Bower, G.H. (Ed.), Psychology of Learning and Motivation. Academic
Press, San Diego, pp. 193–225.
 ARTICLE IN PRESS

16 Can the clobber game become a classroom-based tool?

Hasher, L., Stoltzfus, E.R., Zacks, R.T., Rypma, B., 1991. Age and inhibition. J. Exp. Psychol.
17 (1), 163–169. https://doi.org/10.1037/0278-7393.17.1.163.
Hayesroth, B., Hayesroth, F., 1979. A cognitive model of planning. Cognit. Sci. 3 (4),
275–310. https://doi.org/10.1016/S0364-0213(79)80010-5.
Hinshaw, S.P., Scheffler, R.M., Fulton, B.D., Aase, H., Banaschewski, T., Cheng, W.,
Mattos, P., Holte, A., Levy, F., Sadeh, A., Sergeant, J.A., Taylor, E., Weiss, M.D.,
2011. International variation in treatment procedures for ADHD: social context and recent
trends. Psychiatr. Serv. 62 (5), 459–464. https://doi.org/10.1176/ps.62.5.pss6205_0459.
Hoffmann, J.P., 2004. Generalized Linear Models: An Applied Approach. Pearson College
Division.
Jimison, H., Pavel, M., McKanna, J., Pavel, J., 2004. Unobtrusive monitoring of computer in-
teractions to detect cognitive status in elders. IEEE Trans. Inf. Technol. Biomed. 8 (3),
248–252. https://doi.org/10.1109/TITB.2004.835539.
Kiili, K., Ketamo, H., 2017. Evaluating cognitive and affective outcomes of a digital game-
based math test. IEEE Trans. Learn. Technol. 11 (2), 255–263. https://doi.org/10.1109/
TLT.2017.2687458.
Lansbergen, M.M., Kenemans, J.L., van Engeland, H., 2007. Stroop interference and attention
deficit/hyperactivity disorder: a review and meta-analysis. Neuropsychology 21, 251–262.
https://doi.org/10.1037/0894-4105.21.2.251.
Lezak, M.D., Howieson, D.B., Loring, D.W., Fischer, J.S., 2004. Neuropsychological Assess-
ment. Oxford University Press, USA.
MacLeod, C.M., 1991. Half a century of research on the Stroop effect: an integrative review.
Psychol. Bull. 109, 163–203. https://doi.org/10.1037/0033-2909.109.2.163.
Manera, V., Petit, P.-D., Derreumaux, A., Orvieto, I., Romagnoli, M., Lyttle, G., Robert, P.H.,
2015. Kitchen and cooking,’ a serious game for mild cognitive impairment and Alzhei-
mer’s disease: a pilot study. Front. Aging Neurosci. 7, 24. https://doi.org/10.3389/
fnagi.2015.00024.
Miyake, A., Friedman, N.P., Emerson, M.J., Witzki, A.H., Howerter, A., Wager, T.D., 2000.
The unity and diversity of executive functions and their contributions to complex “frontal
lobe” tasks: a latent variable analysis. Cogn. Psychol. 41 (1), 49–100. https://doi.org/
10.1006/cogp.1999.0734.
Myers, R.H., Montgomery, D.C., Vining, G.G., Robinson, T.J., 2012. Generalized Linear
Models: With Applications in Engineering and the Sciences. John Wiley & Sons.
Nee, D.E., Brown, J.W., Askren, M.K., Berman, M.G., Demiralp, E., Krawitz, A., Jonides, J.,
2013. A meta-analysis of executive components of working memory. Cereb. Cortex
(New York, N.Y. 1991) 23 (2), 264–282. https://doi.org/10.1093/cercor/bhs007.
Nejati, V., 2021. Program for attention rehabilitation and strengthening (PARS) improves
executive functions in children with attention deficit-hyperactivity disorder (ADHD).
Res. Dev. Disabil. 113, 103937. https://doi.org/10.1016/j.ridd.2021.103937.
Niendam, T.A., Laird, A.R., Ray, K.L., Dean, Y.M., Glahn, D.C., Carter, C.S., 2012. Meta-
analytic evidence for a superordinate cognitive control network subserving diverse exec-
utive functions. Cogn. Affect. Behav. Neurosci. 12 (2), 241–268. https://doi.org/10.3758/
s13415-011-0083-5.
Obradovi
c, J., Portilla, X.A., Boyce, W.T., 2012. Executive functioning and developmental
neuroscience: current Progress and implications for early childhood education. In:
Pianta, R.C., Justice, L., Barnett, W.S., Sheridan, S. (Eds.), Handbook of Early Childhood
Education. Guilford, New York, pp. 324–351.
Obradovi
c, J., Sulik, M.J., Finch, J.E., Tirado-Strayer, N., 2018. Assessing students’ executive
functions in the classroom: validating a scalable group-based procedure. J. Appl. Dev. Psy-
chol. 55, 4–13. https://doi.org/10.1016/j.appdev.2017.03.003.
 ARTICLE IN PRESS

References 17

Parkin, A.J., 1996. Explorations in Cognitive Neuropsychology. Psychology Press.

Parsons, T.D., Reinebold, J.L., 2012. Adaptive virtual environments for neuropsychological
assessment in serious games. IEEE Trans. Consum. Electron. 58 (2), 197–204. https://
doi.org/10.1109/TCE.2012.6227413.
R Foundation, 2023. The R Project for Statistical Computing. https://www.r-project.org/.
Reineberg, A.E., Banich, M.T., Wager, T.D., Friedman, N.P., 2022. Context-specific activa-
tions are a hallmark of the neural basis of individual differences in general executive func-
tion. NeuroImage (Orlando, Fla.) 249, 118845. https://doi.org/10.1016/j.neuroimage.
2021.118845.
Rodrı́guez-Nieto, G., Seer, C., Sidlauskaite, J., Vleugels, L., Van Roy, A., Hardwick, R.,
Swinnen, S., 2022. Inhibition, shifting and updating: inter and intra-domain commonalities
and differences from an executive function’s activation likelihood estimation meta-
analysis. NeuroImage (Orlando, Fla.) 264, 119665. https://doi.org/10.1016/j.neuroimage.
2022.119665.
Shallice, T., 1988. From Neuropsychology to Mental Structure. Cambridge University Press.
St Clair-Thompson, H.L., Gathercole, S.E., 2006. Executive functions and achievements in
school: shifting, updating, inhibition, and working memory. Q. J. Exp. Psychol. 59 (4),
745–759. https://0-doi-org.pugwash.lib.warwick.ac.uk/10.1080/17470210500162854.
Strauss, E., Sherman, E.M.S., Spreen, O., 2006. A Compendium of Neuropsychological Tests:
Administration, Norms, and Commentary, third ed. Oxford University Press, Oxford,
England.
Tarnanas, I., Laskaris, N., Tsolaki, M., Muri, R., Nef, T., Mosimann, U.P., 2015. On the com-
parison of a novel serious game and electroencephalography biomarkers for early demen-
tia screening. In: GeNeDis 2014. Springer, Cham, Switzerland, pp. 63–77. https://doi.org/
10.1007/978-3-319-08939-3_11.
Tong, T., Chignell, M., DeGuzman, C.A., 2021. Using a serious game to measure executive
functioning: response inhibition ability. Appl. Neuropsychol. Adult 28 (6), 673–684.
https://doi.org/10.1080/23279095.2019.1683561.
Wang, P., Fang, Y., Qi, J., Li, H., 2022. FISHERMAN: a serious game for executive function
assessment of older adults. Assessment. 10731911221105648 https://doi.org/10.1177/
10731911221105648.
Wynn, K., 1998. Psychological foundations of number: numerical competence in human in-
fants. Trends Cogn. Sci. 2, 296–303. https://doi.org/10.1016/S1364-6613(98)01203-0.
Yarkoni, T., 2009. Big correlations in little studies: Inflated fMRI correlations reflect low sta-
tistical power: commentary on Vul et al. Perspect. Psychol. Sci. 4 (3), 294–298. https://doi.
org/10.1111/j.1745-6924.2009.01127.x.
Yarkoni, T., Braver, T.S., 2010. Cognitive neuroscience approaches individual differences in
working memory and executive control: conceptual and methodological issues. In: Hand-
book of Individual Differences in Cognition: Attention, Memory, and Executive Control,
pp. 87–107, https://doi.org/10.1007/978-1-4419-1210-7_6.
Zelazo, P.D., Anderson, J.E., Richler, J., Wallner-Allen, K., Beaumont, J.L., Conway, K.P.,
Weintraub, S., 2014. NIH toolbox cognition battery (CB): validation of executive function
measures in adults. J. Int. Neuropsychol. Soc. 20 (6), 620–629. https://doi.org/10.1017/
S1355617714000472.