Intern. J.

Neuroscience, 117:869882, 2007

C 2007 Informa Healthcare
Copyright

ISSN: 0020-7454 / 1543-5245 online
DOI: 10.1080/00207450600910077

RELATIONSHIP OF WORKING MEMORY AND EEG

TO ACADEMIC PERFORMANCE: A STUDY
AMONG HIGH SCHOOL STUDENTS

DALIA M. AGUIRRE-PEREZ
Centro de Investigacion
en Ciencias Medicas

Universidad Autonoma
del Estado de Mexico

Toluca, Mexico

GLORIA A. OTERO-OJEDA
F. BERNARDO PLIEGO-RIVERO
Facultad de Medicina
Universidad Autonoma
del Estado de Mexico

Toluca, Mexico

ALONSO A. FERREIRA-MARTINEZ
Centro de Investigacion
en Ciencias Medicas

Universidad Autonoma
del Estado de Mexico

Toluca, Mexico

Some biological and behavioral elements which could explain differences between
high and low academic attainment (HA/LA) students were identified. The qEEG
of subjects under the 1020 derivation system was recorded at rest and while
completing a 3-back working memory (WM) task. While completing the task LA
students showed more theta and total absolute potency at rest, and HA individuals
showed more energy in delta and theta frequencies in frontal regions; LA students
Received 6 April 2006.
This work was done with support from the Secretara de Investigacion y Estudios Avandos,
UAEM, Project No. 1803/2003.
Address correspondence to F. Bernardo Pliego-Rivero, Facultad de Medicina-Universidad
Autonoma del Estado de Mexico, Paseo Tollocan esq. Jesus Carranza (s/n), Toluca, Mex., CP
50180, Mexico. E-mail: bpliego@uaemex.mx
869


D. M. AGUIRRE-PEREZ
ET AL.

870

made a higher number of mistakes while executing the WM task with no differences
in reaction time between groups. We conclude that a diminished WM capacity is
present in LA students.
Keywords working memory, 3-back task, high school, students, qEEG, academic
achievement, teenage, adolescent

INTRODUCTION
In this new century, it is assumed that knowledge will constitute a fundamental
added value in all aspects of goods manufacture and service supply. Mastering
knowledge thus has become the main factor of self-sustained development in
our societies (Castaneda, 2004). The process of learning must be understood
as a multicausal and multidimensional progression in which biological and
psychosocial aspects are related and affect each other. These give place to
multiple and very complex interactions which, in the case of students, reveal
their academic performance throughout the different school stages (LopezOliva, 2000).
The main objective of this work was to identify biological and behavioural
components which could let us determine differences between teenage high
school students who show a high school performance level from those whose
academic achievement is poor. In this last group were included students
whose academic notes were not failing ones but just above the minimum
national approval standard (60%). This is the beginning of a research study
which will help to appraise biological and behavioral parameters allowing us
partly to explain if there are variables of this type pointing out differences in
academic performance between students. This kind of study will then allow
us to propose alternative solutions to reduce the critical problem of academic
underachievement, which in turn has a considerable economic and social impact
in our society. It must be stated that there are very few works dealing with this
aspect of teenage academic performance.
Cognition requires memory. All intelligent beings have different types of
memory playing important roles in their cognitive functions. One of these
is the system of working memory (WM), which has been considered a
temporary deposit of information related to the completion of other more
complex cognitive functions like comprehension, learning and reasoning
(Baddeley, 1992). Typically, WM has a limited information storage capacity.
WM constitutes a system of stored information which can easily be updated,
coded and recovered. It is labile, however, and cannot be kept for a long time

WORKING MEMORY, EEG AND ACADEMIC PERFORMANCE

871

unless a process of continuous attention is involved. WM is a system which

has evolved adapting itself to serve higher mental processes (Baddeley, 1992)
and it is proposed to be a system of transient memory storage subject to further
processing (Colom & Flores-Mendoza, 2001).
The speed of information processing and WM are two distinguishable
dimensions, though not independent of each other. That is, if a WM task
is of low complexity, a positive relationship between speed of processing
and precision is observed. As the complexity of the task increases, the
relationship between speed of processing and precision is reduced, and
when the WM task is highly demanding, the relationship between speed of
processing and precision turns into a negative one. In this respect, observations
in the scientific literature allow us to presume and hypothesize about the
possibility of finding differences between WM capacity and performance
ability while executing mental tasks involving WM (Colom & Flores-Mendoza,
2001).
To identify the relationship between brain structure and cognition, it is
necessary to analyze those biological processes in cortical and subcortical
areas undergoing transitory changes in the scale of milliseconds (Mesulam,
1990). The technique of electroencephalography allows an analysis of such
relationship, i.e., recording cerebral activity simultaneously to the completion
of a mental task. Diverse changes are observed in the electroencephalogram
(EEG) frequency bands while a mental task is performed and these relate to the
type of undergoing cognitive processes. Delta activity increases during verbal
and sustained attention tasks (Dolce & Waldeier, 1974: Alcaraz et al., 1991;
Fernandez et al., 1993; Valentino et al., 1993). Other authors report changes in
theta (Gevins et al., 1997; Klimesch, 1996; Harmony, 1999), in alpha and beta
activity (Gevins et al., 1979a,b,c; Gundel & Wilson, 1992) while diverse mental
WM tasks are executed. These findings show that brain activation associated
to the completion of WM tasks parallels the electrical activity recorded in the
EEG.
Considering the relevant role of WM in attention, learning and memory
processes and, according to Gevins and Smith (2000), it is very likely that
differences between subjects in terms of their general cognitive abilities could
be related to noticeable differences in the activity of those brain systems
involved in WM. In this study work brain activity was investigated during
the performance of a high load WM task (3-back) by two academically distinct
groups of high school teenage students: low (non-failing) and high academic
attainment (LA, HA) groups. In this paper are presented the behavioural results
and the EEG characteristics of each group.


D. M. AGUIRRE-PEREZ
ET AL.

872

MATERIALS AND METHODS

Sample
The selected sample was formed by 21 students, 1517 years old, belonging
to the public high school Preparatoria Lic. Adolfo Lopez Mateo Universidad
Autonoma del Estado de Mexico (UAEM), Toluca City, State of Mexico.
Participating students and their parents signed an informed agreement in which
all procedures, benefits and possible risks of the study were indicated. Only
subjects with a clinical diagnosis of normal health were accepted. The general
health condition was assessed by qualified medical and clinical personnel at
the UAEMs Centro de Investigaciones en Ciencias Medicas after a complete
haematological study, determination of iron status and EEG analysis at
rest.
The sample was divided into two groups taking into account the results
obtained by the students in the National Examination to Access Pre-University
Stage Education (Examen Nacional de Ingreso a la Educacion Media
SuperiorEXANI-1). All the students in the sample approved the EXANI-1
and were selected from the above mentioned high school when already assisting
to regular lessons:
1. Low Academic Attainment (LA) Group: 10 individuals with academic
scores one standard deviation below the mean in the subjects of mathematics
and Spanish. This group was formed by students not failing their
subjects, with approval notes above the minimum national standard (60%)
requirement.
2. High Academic Attainment (HA) Group: 11 individuals with academic
marks one standard deviation above the mean in the subjects of mathematics
and Spanish.
Procedures:
The next studies were performed in all participating students:
1. Quantitative EEG (qEEG) at rest using the derivations of the 1020 system
under monopolar recording, with linked ear lobules as reference. Out of line
15 to 20 artifact-free, 3.5 ms windows were edited.
2. EEG recording while performing a WM verbal task under the same
conditions as in qEEG recording at rest.

WORKING MEMORY, EEG AND ACADEMIC PERFORMANCE

873

WM Task
The task designed for WM evaluation consisted of a Stenberg-type protocol
(3-back task) with the next features: The student was placed seating in front
a computers monitor in which upper and small case letters were presented
one by one. The individuals task consisted in determining if the letter on the
screen matched the letter which had appeared three presentations back. The
examined individual was asked to press the left mouse button if the letter
was the same presented three places back, regardless of it as an upper or a
lower case (Figure 1a). A total of 200 stimuli were presented with a 50%:50%

Figure 1. a) 3-back task, b) Working memory study design.

874

D. M. AGUIRRE-PEREZ
ET AL.

correspondence of matching to nonmatching letters. This variant of the Stenberg

letters task is a WM verbal task due to the fact that all the stimuli presented
are linguistically significant units (Baddeley, 1984). Each stimulus, i.e, a single
letter, was presented for 500 ms with an interval between stimuli of 2500 ms.
For EEG analysis during the execution of the task, 15 to 20 artifact-free windows
of 1280 ms, time taken up preceding the presentation of the stimuli (Figure
1b). Before each test, a short trial assay was run to guarantee every student
understood the task to complete.
EEG recordings were done with a digital electroencephalography recorder
(Neuronic MD) and the task was applied through the Mind Tracer software
coupled to the EEG recorder. This allowed relating corresponding events to
EEG recordings, i.e., presentation of the stimulus letter and the individuals
correct or wrong answers, differentiated these by a colour code. The software
also calculated reaction times and quantified correct and incorrect answers.
Statistical Analysis
1. Behavioral study. The U Mann-Whitney test was applied to establish mean
differences between groups for reaction times and task execution (number
of correct answers and number of errors).
2. Resting EEG: To the edited EEG a broad band frequency analysis (delta,
1.53.5 Hz; theta, 3.57 Hz; alpha, 7.512 Hz; beta, 12.519 Hz and; total,
1.519 Hz) was applied. A Loge transformation was performed on the
absolute power (AP) value for each frequency band to guarantee the normal
distribution of the data (John et al., 1980; Gasser et al., 1982; Pollock et al.,
1990). To establish EEG AP differences of means between groups a U
Mann-Whitney test was applied.
3. EEG throughout the WM task: Out of line, 1520 artifact-free windows of
1280 ms each were edited (Figure 1b). To all edited EEG a narrow band
frequency analysis was performed every 0.78 Hz, from 0.78 to 18.78 Hz.
For the same frequencies we obtained cross correlation matrices.
For the detection of differences between groups, a non-parametric, multivariate
analysis of permutations (Galan et al., 1997) was applied to the EEG during
the completion of the task. This analysis tests one global and several marginal
hypothesis. The first one with the intention of proving if two topographical
distributions are equal between them. The marginal tests consisted of
determining if two amplitudes, for a frequency (f) and for a derivation (d) are the
same. For these, two stadigraphs were used. In the global one, the maximum

WORKING MEMORY, EEG AND ACADEMIC PERFORMANCE

875

Students t for any f,d combination is used. In the marginal tests a multiple
comparison is performed: in a set f for all d or for a single topographical
localization (d) for all f. The empirical distributions of both stadigraphs are
calculated through the technique of permutations. The original Students ts,
i.e. before permutation, for the global or marginal hypothesis are compared to
the empirical distribution of the maximum Students ts of their respective f,d
combination or multiple comparison. The technique of permutations is used
assuming that if the samples to be compared are the same, in the permutation
of observations or individuals belonging to the same sample equality must be
preserved. This is advantageous in EEG studies due to the fact that when there
are too many variables and few individuals the ANOVAs main assumption
does not hold up. Additionally, a normal distribution of the data is not required
because the test itself constructs its own empirical distribution.
Only answers to stimuli in which letters matched were analyzed. This was
because it was observed that both groups of students showed more difficulty
in recognizing letters when they matched. With matching letters, compared to
non matching letters, reaction times were longer and there was also a higher
number of mistakes.

RESULTS
Behavioral Study
No differences were found between groups in relation to reaction times.
Nevertheless, the LA group showed worst performance in the number of
incorrect answers compared to the HA group (Table 1).

Table 1. Number of correct answers, errors and reaction times

while completing the WM task. Differences between groups
HA
Answers
Correct
Errors
Reaction time

LA

Mean

SD

Mean

SD

88
12
1106.9

2.6
4.6
354.2

74
26
1053.5

3.5
9.0
358.1

P<
0.002
0.015
0.44

HA/LA, high and low academic attainment groups. SD =

standard deviation; Errors = Incorrect responses + Not answered.
Reaction time, milliseconds.


D. M. AGUIRRE-PEREZ
ET AL.

876

Table 2. Theta absolute power of EEG at rest. Differences by group

LA
Lead
O1
O2
T3
T4
Fz
Cz

HA

Mean

SD

Mean

SD

P<

598.06
618.12
753.87
803.98
1389.47
1526.55

159.5
175.4
622.4
348.9
450.8
514.2

1087.58
946.20
1223.12
1387.36
2007.13
2193.07

397.5
401.6
629.6
632.3
943.14
850.4

0.002
0.048
0.050
0.015
0.048
0.025

HA/LA, high and low academic attainment groups. SD =

standard deviation.

EEG at Rest
Significant differences were obtained for theta AP in occipital and temporal
regions, in the middle line in frontal and central regions and for total AP
in diverse frontal, central, occipital and temporal regions. In all cases, LA
individuals showed higher values (Tables 2 and 3).

EEG During Completion of the Task

Comparing HA and LA individuals, the nonparametric permutations analysis
revealed significant differences between both groups (global hypothesis). The
significant differences for frequency and derivation were found in the delta and
Table 3. Total absolute power of EEG at rest. Differences by group
HA
Lead
F3
C3
O1
T3
T5
T6
Fz

LA

Mean

SD

Mean

SD

P<

4895.90
4998.48
4361.47
3122.40
3772.99
5383.31
5207.85

2642.09
2938.26
2744.01
3383.52
2590.86
3410.41
2838.94

6830.61
8932.52
9106.36
3711..22
9133.59
10495.68
7827.82

1957.58
4479.93
4668.13
1049.07
5245.61
4675.48
2365.51

0.052
0.041
0.005
0.041
0.006
0.012
0.035

HA/LA, high and low academic attainment groups. SD = standard

deviation.

WORKING MEMORY, EEG AND ACADEMIC PERFORMANCE

877

Table 4. EEG while completing the working memory task. Differences by group. Global significance:
p < 0.049
Leads
Fp1
Fp2
F3
P3
T5

Frequencies

p<

2.24
3.90
2.34
4.68
1.56
6.24
11.70

0.049
0.049
0.049
0.049
0.010
0.019
0.049

HA group energy values were higher than LA

group energy values in these derivations and
frequencies. HA/LA, high and low academic attainment groups.

theta range of frequencies in frontal and parietal regions. In the alpha range,
the differences were found in the temporal posterior region (Table 4).
Comparing the LA group to the HA group there were no significant
differences (global signification, p < .81). The LA group individuals did not
show higher AP values in any frequency or derivation.

DISCUSSION
Countless factors participate in the process of learning; some of social
nature, others psychological and importantly some others due to biological
characteristics. All these factors are interrelated and influence each other.
The main objective of this work was to pin point some of the possible
biological factors that could eventually influence academic achievement. The
study presented here forms part of a broader research project which, in addition
to dealing with the biological approach it includes the study of psychosocial
factors. In this work are presented only the biological results obtained after
EEG recording in a group of high school students comparing between high and
low academic performance individuals. This last group was not conformed by
failing students but by individuals whose average grades are merely above the
minimum approval condition (60%). Here WM was analysed as a key element
implicated in a determinant way in the process of learning.
It has been established that for a successful performance of higher order
cognitive functions, a process of intentional self control of attention is an

878

D. M. AGUIRRE-PEREZ
ET AL.

important requisite. This process will allow the individual to manipulate WM

stored information. WM provides the ability to control and sustain attention
over particular cerebral representations in front of distracting influences, a
process which positively correlates with results obtained in psychometric tests
of cognitive ability and also with school abilities indicators (Carpenter et al.,
1990; Kyllonen & Cristal, 1990). Taking this into account, it could well be that
the differences between individuals in terms of general individual cognitive
abilities could be due to ostensible activity differences in cerebral processes
underlying WM (Gevins & Smith, 2000).
Studying brain electric activity in humans is useful for exploration at two
fundamental levels: a. it allows an assessment of the higher CNS functional and
anatomical integrity and; b. it provides information about cognitive activity. In
this work, EEG activity at rest was initially investigated in order to determine
the basal activity features of each group (HA and LA) and observe if differences
between groups existed throughout the whole range of EEG frequencies. The
traditional EEG trace was normal for both groups, observation which at the
same time fulfilled the pre-requisite of normality for inclusion in the sample.
As shown in table 3, the LA group showed more AP in theta frequency in
frontal and central regions (Fz, Cz) as well as in posterior derivations (O1, O2,
T5 and T6). Diverse degrees of learning aptitudes, from best to worst, appear to
be related to a slowing of basal EEG activity. It is a known fact that individuals
with learning difficulties show a slow basal EEG. For instance, Fernandez et al.
(2002) found an increase in theta activity in frontal regions of children with
learning disabilities. Harmony et al. (1990) also found an increase in delta
activity in children with severe learning difficulties, while those with moderate
or light learning disabilities showed an increase in theta activity compared to
a control group with competent learning abilities. Studying the EEG at rest in
young people (1721 years old) to whom eight cognitive tests were applied, an
increase in theta AP was observed in central and occipital regions in individuals
with lower spatial abilities compared to those with higher spatial abilities (Arce
et al., 1995).
Our results correlate with those found by Harmony et al. (1995). Although
this last work deals with children with learning disabilities and our task
consisted of studying young people of low academic attainment, i.e. students
not failing their subjects, it is noticeable the fact that in both studies the increase
of theta activity in the EEG at rest parallels with poor academic achievement.
The work of Arce et al. (1995) is in agreement with our work confirming that an
excess of theta activity in the EEG at rest of young people correlates with a lower
cognitive performance. Considering the fact that individuals of LA showed a

WORKING MEMORY, EEG AND ACADEMIC PERFORMANCE

879

range of frequencies with more theta AP in the EEG at rest compared to those
of HA, these could in part explain the differences in academic performance
between groups.
In this study the students performance in the high load WM task (3-back
paradigm) and its correlation to the EEG trace was investigated with the aim
of finding out whether WM capacity could be one of the factors explaining
academic differences between both groups. From the behavioural point of view
there were no differences in terms of reaction time while completing the WM
task. Nevertheless, the LA group of students made a higher number of mistakes
than the HA group (Table 1).
In relation to the recorded EEG in parallel to the completion of the task,
narrow band analysis revealed higher energy in slow frequencies (delta and
theta, 1.564.68 Hz) among HA individuals in frontal and parietal leads as
well as in the 11.7 Hz frequency in the left temporal area. Vogel et al. (1968)
found a high correlation between the amount of slow waves and the efficiency
with which mental tasks were implemented, explaining such finding as a
manifestation of an inhibitory process selectively suppressing inappropriate
or irrelevant neuronal activity for the completion of the task. This observable
fact has been explained (Dolce & Waldeier, 1974) in terms of delta activity
revealing a process of elimination of external and internal influences, useful
for the execution of a given task. More recently (Fernandez et al., 1993) found
delta activity during the implementation of mental tasks, event explained as an
expression of self or inner concentration.
On the other hand, slow activity and in particular theta activity increases in
response to a considerable range of mental tasks, like the completion of verbal
and spatial tasks. Theta activity also increases in parallel to the difficulty of
the task while it is not sensitive to the type of cognitive information attended
or remembered (Gevins et al., 1997). In our study the HA group showed
more energy in frontal regions in frequencies corresponding to the delta range
(1.56, 2.24, 3.90 Hz), which could be construed as an electroencephalographic
expression of inner concentration. Precisely, the 3-back task requires an intense
and sustained concentration and attention for its execution. From these results,
it is obvious that LA students could not develop enough inner concentration
and sustained attention to perform adequately on the WM task.
It is specifically in the frontal lobule where the central executive WM
construct is located. According to Baddeley and Hitchs model (Baddeley, 1986;
Jonides & Smith, 1997), WM is formed by three fundamental components.
These are an attention control system (executive central) and two subordinated
subsystems responsible for the temporal storage and manipulation of visual

880

D. M. AGUIRRE-PEREZ
ET AL.

content (visuospatial information storage subsystem) and another one of verbal

content (articulatory circuit). The executive central is considered to be formed
by a number of processes accounting for the selection of the WM task strategies
and planning at short term. The executive central also links information in WM
subsystems with long term memory and allows changing over attention from
one cognitive task to another (Jonides & Smith, 1997).
In the theta range (6.24 Hz) we observed higher energy in left parietal
leads. Harmony et al. (1999) observed an increase in theta energy in left
parietotemporal regions during the completion of mathematical calculation
tasks, suggesting that this energy frequency could be related to the production
of internal language, and storage and retrieval of verbal WM.
We also found an increase in alpha high energy frecuency (11.70 Hz)
in the HA group compared to the LA group. Contrary to our observations,
some authors point out that alpha activity is penchant to decrease when the
difficulty of the task increases (Gevins et al., 1979a,b,c; Gundel & Wilson,
1992). Presently, we are not able to explain this difference, however the
possibility exists this is a spurious result because the increase in alpha energy
was observed in a single frequency and a single lead.
Summing up, in this work we observed an EEG at rest with increased
theta activity in LA students which, as stated before, has been associated
to lower intellectual performance and diminished WM capacity (Carpenter
et al., 1990; Kyllonen & Cristal, 1990). This last one observed here as a
reduced AP in slow frequencies in frontal and parietal regions, suggestive of a
diminished inner concentration and helping partially to explain lower academic
performance. According to the observations made in this work and those
reported in the literature, students pertaining to the LA group appear to have
reduced WM abilities. We consider that although learning is a multifactorial
process and certainly there are many other issues involved in low academic
achievement, in this study is shown that at least one of them is a limitation
in WM. The results of this work directly lead us, and all those involved in
educational activities, to the challenge of working on the improvement of the
WM abilities of all students to attain better qualifications and educational
standards.

REFERENCES
Alcaraz, V. M., Daz de Leon, A. E., Garca, R., Guevara, M. A., & Ramrez, V. (1991).
La dinamica de la actividad electrica cerebral en el curso de la composicion de
frases. Revista Latina de Pensamiento y Lenguaje, 1, 104116.

WORKING MEMORY, EEG AND ACADEMIC PERFORMANCE

881

Arce, C., Ramos, J., Guevara, M. A., & Corsi-Cabrera, M. (1995). Effect of spatial
ability and sex on EEG power in high school students. International Journal of
Psychophysiology, 20, 1120.
Baddeley, A. D., Lewis, V., & Vallar, G. (1984). Exploring the articulatory loop. The
Quarterly Journal of Experimental Psychology, 36A, 233252.
Baddeley, A. D. (1986). Working Memory. Oxford: Oxford University Press.
Baddeley, A. D. (1992). Working Memory: The interface between memory and
cognition. Journal of Cognitive Neuroscience, 4, 281288.
Carpenter, P. A., Just, M. A., & Shell, P. (1990). What one intelligence test measures?
A theoretical account of the processing in the Raven Progressive Matrices Test.
Psychological Review, 97, 404431.
Castaneda-Figueiras, S. (2004). Evaluando y fomentando el desarrollo cognitivo y el
aprendizaje complejo. Psicologa desde El Caribe, 13, 109143.
Colom, R., & Flores-Mendoza, C. (2001). Inteligencia y Memoria de Trabajo: La
relacion entre Factor G, Complejidad Cognitiva y Capacidad de procesamiento.
Psicologa: Teora y Pesquisa, 17, 3747.
Dolce, G., & Waldeier, H. (1974). Spectral and Multivariate analysis of EEG during
mental activity in man. Electroencephalography and Clinical Neurophysiology,
36, 577584.
Fernandez, T., Harmony, T., Rodrguez, M., Reyes, A., Marosi, E., & Bernal, J.
(1993). Test-retest reliability of EEG spectral parameters during cognitive tasks. I.
Absolute and relative power. International Journal of Neurosciences, 68, 255261.
Fernandez, T., Harmony, T., Fernandez-Bouzas, A., Silva, J., Herrera, W., SantiagoRodrguez, E., & Sanchez, L. (2002). Sources of EEG activity in learning disabled
children. Clinical Electroencephalography, 33, 160164.
Galan, L., Biscay, R., Rodriguez, J. L., Perez-Abalo, M. C., & Rodriguez, R. (1997).
Testing topographic differences between event-related brain potentials by using
non-parametric combinations of permutation tests. Electroencephalography and
Clinical Neurophysiology, 102, 240207.
Gasser, T., Bacher, P., & Mocks, J. (1982). Transformation toward the normal
distribution of broad band several parameters of the EEG. Electroencephalography
and Clinical Neurophysiology, 53, 119124.
Gevins, A. S., Zeitlin, G. M., Doyle, J. C., Schaffer, R. E., & Callaway, E.
(1979a). EEG patterns during cognitive tasks. II. Analysis of controlled tasks.
Electroencephalography and Clinical Neurophysiology, 47, 704710.
Gevins, A. S., Zeitlin, G. M., Doyle, J. C., Yingling, C. D., Schaffer, R. E., Callaway,
E., & Yeager, C. L. (1979b). EEG correlates of higher cortical functions. Science,
203, 665668.
Gevins, A. S., Zeitlin, G. M., Yingling, C. D., Dedon, M. F., Schaffer, R. E.,
Roumasset, J. T., & Yeager, C. L. (1979c). EEG patterns during cognitive tasks.
Methodology and analysis of complex behaviors. Electroencephalography and
Clinical Neurophysiology, 47, 693703.

882

D. M. AGUIRRE-PEREZ
ET AL.

Gevins, A., Smith, M. E., McEvoy, L., & Yu, D. (1997). Highresolution EEG mapping
of cortical activation related to working memory: Effects of task difficulty, type of
processing, and practice. Cerebral Cortex, 7, 374385.
Gevins, A., & Smith, M. E. (2000). Neurophysiological measures of working memory
and individual differences in cognitive ability and cognitive style. Cerebral Cortex,
10, 829839.
Gundel, A., & Wilson, G. F. (1992). Topographical changes in the ongoing EEG related
to the difficulty of mental tasks. Brain Topography, 5, 1725.
Harmony, T., Hinojosa, G., Marosi, E., Becker, J., Rodrguez, M., Reyes, A., & Rocha,
C. (1990). Correlation between EEG spectral parameters and an educational
evaluation. International Journal of Neuroscience, 54, 147155.
Harmony, T., Marosi, E., Becker, J., Rodrguez, M., Reyes, A., Fernandez, T., Silva,
J., & Bernal, J. (1995). Longitudinal quantitative EEG study of children with
different performances on a reading-writing test. Electroencephalography and
Clinical Neurophysiology, 95, 426433.
Harmony, T., Fernandez, T., Silva, J., Bosch, J., Valdes, P., Fernandez-Bouzas, A.,
Galan, L., Aubert, E., & Rodrguez, D. (1999). Do specific EEG frequencies
indicate different processes during mental calculation? Neuroscience Letters, 266,
2528.
John, E. R., Ann, H., Prichep, L., Trepetin, M., Brown, D., & Kaye, H. (1980).
Developmental equations for the EEG. Science, 210, 12551258.
Jonides, J., & Smith, E. (1997). The architecture of working memory. In Cognitive
Neuroscience. (pp. 243276). London: The MIT Press.
Klimesch, W., Schimke, H., Doppelmayr, M., Ripper, B., Schwaiger, G., &
Pfurtscheller, G. (1996). Event-related desynchronization (ERD) and the Dmeffect: does alpha desynchronization during encoding predict later recall performance? International Journal of Psychophisiology, 24, 4760.
Kyllonen, P. C., & Cristal, R. E. (1990). Reasoning ability is little more than working
memory capacity? Intelligence, 14, 389433.
Lopez-Oliva, M. (2000). Material instruccional del curso de modelo, estrategias y
tecnicas de evaluacion e instruccion al servicio del orientador educativo, Ciudad
de Mexico, Prensa UNAM.
Mesulam, M. M. (1990). Large scale neurocognitive networks and distributed processing
for attention, language, and memory. Annals of Neurology, 28, 597613.
Pollock, V. E., Schneider, L. S., & Lyness, S. A. (1990). EEG amplitudes in healthy
late-middle age and erderly adults: normality distribution and correlations with
age. Electroencephalography and Clinical Neurophysiology, 76, 373375.
Valentino, D. A., Arruda, J. E., & Gols, M. S. (1993). Comparison of qEEG and response
accuracy in good vs. poorer performers during a vigilance task. International
Journal of Psychophysiology, 15, 123134.
Vogel, W., Broverman, D. M., & Klaiber, E. L. (1968). EEG and mental abilities.
Electroencephalography and Clinical Neurophysiology, 24, 166175.