Impaired or Not-impaired: The non-verbal Hebb Effect in Poor Readers in Grade 4
Ceyrine M. Pellikaan 10274960 18 EC from February 2012 – November 2013 MSc. In Brain and Cognitive Sciences, Cognitive Science track Supervisor: Dr. T. van Zuijen Co-Supervisor: Dr. P.F. de Jong Graduate School of Child Development and Education University of Amsterdam
Abstract
Dyslexia has been prominently attributed to phonological deficits. Recent research, however, suggests a deficit in long-term learning of serial information, including visuo-spatial
information, in dyslexic adults. We tried to replicate this finding in poor reading children. We examined differences in the Hebb Repetition Effect (i.e. a transition of recalled serial
information from short term to long term memory) between 25 poor readers and 25 normal readers from 4th grade of elementary school. The tasks were: 1) a visuo-spatial recall task (presented with series of fixed length and increasing length) in which repeated trials were interspersed with random trials and 2) a repetitive reading task in which 3 columns of increasing difficulty were read 6 times. The results of the visuo-spatial task showed that poor reading children performed lower on both Hebb and random trials than normal children for series of fixed length, but equally in series of increasing length. The results of the repetitive reading task showed that poor reading children had a smaller improvement in reading fluency than normal reading children. We did not find support for an impaired Hebb Repetition Effect
in poor reading children.
Keywords: dyslexia, Hebb repetition effect, serial information, repetitive reading task,
Introduction
Dyslexia is a disorder manifested by difficulty in learning to read and spell despite conventional instruction, adequate intelligence, sensory abilities, and sociocultural
opportunity (Schumacher, Hoffmann, Schmal, Schulte-Korne, & Nothen, 2007). It affects 5-10% of the population (Shaywitz, 1996) and although the symptoms are rapidly recognized, the causes are still not fully identified.
Ample research has been done to understand the possible causes and underlying deficits of dyslexia. The most investigated theory is the phonological deficit theory (Hulme & Snowling, 2007) which identifies poor phonological processing as the core deficit. Examples are dyslexics’ inability or deficiency in using speech codes to represent information in the form of words and word parts (Vellutino, Fletcher, Snowling, & Scanlon, 2004) and in hearing fine distinctions between sounds or phonemes. Even though the phonological deficit is the most prominent underlying deficit of dyslexia, a different approach has been suggested by Szmalec, Loncke, and Duyk (2011): They suggest that the cause of dyslexia is a deficit in the long-term learning of serial order information, and investigated this by measuring the Hebb Repetition effect.
Donald Hebb (1961) introduced the notion of the transition of recalled serial
information from short term to long term memory. He demonstrated that serial information was better recalled when presented repeatedly within a block of trials (e.g., every third trial) than when presented only once. This effect is since then known as the Hebb Repetition Effect. The serial information can consist of sequences of digits, letters, words, sounds, or can be visually presented through a Corsi Block task. The Corsi Block task is composed of
nine identical blocks with an irregular position on a board in which the subject is requested to
sequences are increased in length until the subject can no longer recall them correctly
(Vandierendonck, Kemps, Fastame, & Szmalec, 2004).
Szmalec et al. (2011) measured Hebb learning in dyslexic adults through three
different conditions: The first one was the verbal-visual condition and consisted of a list of
nine non-sense syllables presented visually to the participants to assess their immediate
recall. The list was presented 20 times in random order (i.e., the syllables changed position
within the list) and 10 times in the same order, i.e., the sequence (the Hebb sequence) was
repeated 10 times. During recall, the stimuli appeared in a noisy circle around a question
mark. Participants had to tap on the syllables in the same order as presented. The second was
the verbal-auditory condition which also consisted of a sequence of nine non-syllables,
constructed in the same way as in the verbal-visual condition, except that these were
presented aurally. During recall the participants had to name out loud each syllable in the
correct order. The third condition was visuo-spatial in which nine dots were presented on
different locations on a computer screen. The participants were requested to recall the order
in which the sequences occurred. In all three conditions they found that the dyslexic group
benefited less from repetition than the control group. For this reason, Szmalec and colleagues suggest that dyslexics have a deficit in long-term learning of serial order information across different sensory modalities, and view dyslexia as a modality independent learning problem instead of a problem of phonological processing.
Szmalec and colleagues observed a Hebb learning impairment in dyslexics through the use of different tasks. However, they did not use a task which involved the reading process on itself in spite of the reading difficulty being the core deficit of dyslexia. This raises the question whether Hebb learning is associated with the reading process on itself. There is evidence for the Hebb effect to be related to language learning. Page and Norris (2008) for example, compare the Hebb Repetition Effect with naturalistic word learning.
They claim repeated hearings of new sounds, which are composed by sequences of phonemes, can lead to the long-term representation of that sequence. In this way, Hebb learning is associated with language learning.
Children are in the process of learning to read and are therefore an ideal sample for research on learning and reading. The experiment of Szmalec et al. (2011) only included adults. Nonetheless, if the theory of the impaired Hebb effect holds, it should also be present in children. Following this line, we wanted to investigate whether the deficit in long-term learning of serial order information was also evident in children with poor reading skills, and if the Hebb Repetition Effect is related to learning to read. To test this we designed a visuo-spatial task and a repetitive reading task. The visuo-visuo-spatial task, which assesses recall of serial information presented visually, includes two variants; one with a fixed number of serial information in each sequence, i.e., fixed span; the other one with an adaptive span, in which the length of serial information increases. The reason to include both variants is due to the possibility of different maximum span lengths across the groups; i.e., it could be that poor reading children have a shorter span length than normal reading children. The fixed task has a supraspan which provides room for learning. The adaptive span task allows us to measure the participants’ span length. The repetitive reading task, which assesses reading fluency,
includes three columns of words of increasing difficulty to avoid ceiling effects. We
hypothesize children of 4th grade of elementary school with poor reading skills to benefit less from repetition than normal reading children, i.e., an impaired Hebb Repetition Effect on both visuo-spatial tasks and a smaller improvement in reading fluency in the repetitive reading tasks. If the performance on the visuo-spatial task is related to reading fluency in the repetitive reading task, it would confirm Hebb learning to be associated to the reading
process. Since dyslexia and dyscalculia are frequently co-morbid (Landerl & Moll, 2010), we assessed arithmetic ability in our study in order to control for arithmetic skills avoiding poor
arithmetic skills to cause differences on Hebb learing in our visuo-spatial task. To our knowledge, this is the first research that investigates the Hebb Repetition Effect in children with poor reading skills.
Method Participants
Our participants were selected by screening 203 children, from 4th grade of regular elementary schools in Hilversum (110 boys, 93 girls). The group obtained for the experiment consisted of 50 children, 25 with poor reading skills and 25 with normal reading skills. The poor reading children represent 16% of the Dutch population obtaining a standard score ≤7 on these tests. Normal reading children represent average reading children. All children were native Dutch speakers with adequate intelligence and normal or corrected to normal vision. The groups were matched on their reading fluency, verbal and non-verbal intelligence (see section Screening tests), age, and gender. The descriptive statistics of the groups are shown in table 1.
Materials
Screening tests.
The Een Minuut Test (EMT). (Brus & Voeten, 1973) assesses word reading fluency.
From a list of 116 words with increasing difficulty, children were requested to read as many words as possible in one minute time. The score was the amount of correctly read words. The score was converted to a norm score (mean 10, SD 3).
The Klepel (Van den Bos, Lutje Spelberg, Scheepstra & de Vries, 1994) assesses
technical reading of pseudo-words. From a list of 116 words with increasing difficulty children are requested to read as many words as possible in two minutes. The purpose of using pseudo-words is to prevent word recognition, and forces the reader to read all phonemes and combine them in proper speech sounds as according to the Dutch
pronunciation rules. The score is the amount of correctly read words converted to a norm score (mean 10, SD 3).
The RAKIT (Bleichrodt, Drenth, Zaal & Resing, 1987) measures the receptive
vocabulary as part of the verbal intelligence. It consists of 60 words of increasing difficulty which the experimenter reads out loud one by one. The participants are requested to match the word that is read with one of four pictures presented for this word. The score is the number of correct answers.
The Raven Standard Progressive Matrices (Raven, Court & Raven, 1988) measures
the non-verbal intelligence. It consists of 60 items containing patterns of increasing difficulty. The participants are requested to complete the patterns by the correct option out of six
possible answers. The children should complete the test within 45 minutes. The score is the number of correct answers.
Table 1
Characteristics of the participants
Reading group Poor (n=25) Normal (n=25)
M SD Range M SD Range t-value
Age 10.20 0.49 9.5-11 10.02 0.41 9.5-10.6 0.13 Gender
1.48 0.51 1.40 0.51 0.00
EMT (standard score)
5.76 1.51 1-7 10.04 0.80 9-11 -12.57*** Klepel (standard score) 7.00 1.71 3-9 10.84 1.62 8-14 -8.14***
Rakit 50.12 2.60 44-55 49.80 2.90 43-54 0.41
Raven 42.68 4.75 32-50 43.00 3.64 36-48 -0.27
*** p < .001.
Experimental tasks.
The Visuo-Spatial Fixed Length task measures the recall of serial information of serial
information presented visually when series are fixed in number. This task was presented on a laptop and displayed six squares on a screen and a frog jumping from square to square. The
frog made six movements for each trial (16 trials in total) which the child had to imitate with the mouse pointer. The task consisted of two types of trials, Hebb trials and random (or filler) trials. All the odd trials were Hebb trials, thus the same as the first trial (Fixed Length Hebb Condition). All even trials were random, thus not repeated (Fixed Length Random
Condition). On average the path length and number of path crossings was the same for the random and the Hebb trials. The score for each trial was the number of correct serial movements that the participants recalled and reproduced from the movements the frog had made.
The Visuo-Spatial Increasing Length task measures the recall of serial information
presented visually when series are increasing in length. The information presented consisted of a frog making movements on a screen which the child had to imitate with the mouse pointer. The task consisted of two conditions presented separately, the Increasing Length Hebb Condition and the Increasing Length Random Condition. The Increasing Length Hebb Condition, presented in twofold, started with two movements of the frog and repeated the same movements building up by one during every consecutive trial until a maximum of 9 movements was reached after 8 trials. The first run ended when two errors were made or when completing 8 trials after which the second run commenced with alternative build-up movements. After two errors or completion the test ended. In the Increasing Length Random Condition, the child was requested to operate similarly, only this time the movements were presented randomly, i.e. there was no repetition after each consecutive trial. The first run ended after 2 errors or completion and the second run commenced with alternative build-up movements. The score per condition was the maximum correctly reproduced sequence (maximum span length) out of the two runs.
The Repetitive Reading task consisted of a list of 30 pseudo-words presented in 3 columns of 10 pseudo-words each. The columns were of increasing difficulty. Since all
words were pseudo-words they were equally unfamiliar to all participants. Children were assessed by their time and number of words read correctly per column. The scores were converted to fluency scores, i.e. the time it took to read the correctly read words was converted to the number of words that could have been read in one minute.
The Math Timed Test (Tempo Toets Rekenen) (De Vos, 1992), consisted of 5 columns, one of addition, one of subtractions, one of multiplications, one of divisions, and one of all operations mixed. The participant had one minute per column to solve as many operations as possible. Raw scores were the amount of correct answers.
Procedure
Screening.
The children were assessed per school in their classroom in an examination setup during the screening phase. The RAKIT (20 min), RAVEN (45 min) and TTR (5 min) were provided in a groupwise fashion and the EMT (1 min) and Klepel (2min) individually. Since this project is part of a larger project, there were different tasks involved in the experiment which deviate from the research topic developed in this paper, yet they deserve mentioning as they influenced the order in which the tests were taken. These tasks were: Rapid Automatized Naming (RAN) (1 min) provided after the EMT and Klepel. The TTR is technically part of the research phase but time-wisely it was presented during screening phase, as it could be provided in a classical fashion instead of individually.
Research Phase.
The experiment of the present study consisted of two assessments: A visuo- spatial assessment and a repetitive reading assessment. With the visuo-spatial assessment we measured the Hebb Repetition Effect and with the reading assessment, the improvement in reading ability after repetition. As part of a larger project, a verbal task which consisted of a paper based task and a computer task, not to be discussed further in present paper, was also
included. This influenced the order in which the tasks were presented and might have
influenced the participants’ motivation (e.g., tiredness). The tasks were administered in semi-random order under the condition that the reading task had to be presented in two parts: one part was presented after two other tasks and the other part after the third task. The sessions were scheduled individually and in such a way that interruptions such as recess or other activities, were reduced. In the cases in which this was not possible we assured the interruptions took place between (and not during) the tasks. Each session lasted approximately one hour per child.
The Fixed Length (Hebb and Random) Conditions and the Increasing Length Hebb condition had ten different versions appointed counterbalanced to the participant previous to participation to avoid stimulus specific effects. The Increasing Length Random condition did not have different versions because all series of movements in the task were random
themselves. Each task was followed by a pause of approximately three minutes before starting the next one. The subjects were not informed of the possible presence of Hebb trials, and when spotted it was not confirmed to them to minimize a possible expectation factor.
In the reading assessment participants were requested to read out loud each column of words 6 times in two blocks of 3 consecutive trials, with a verbal or visuo-spatial task between both blocks. The reading assessment was divided in two blocks to avoid tiredness. The session ended after this task.
Results Visuo-spatial tasks
Table 2 shows the descriptive statistics of the performance of poor and normal readers on the visuo-spatial tasks. The analysis for the Visuo-Spatial Fixed Length task was
conducted with a Mixed ANOVA to compare conditions (Hebb and random trials) with trial (8) and group (poor and normal readers).
Figure 1 illustrates the average score (mean span length) obtained for the Visuo-Spatial Fixed Length task. We found a significant main effect of condition (Hebb vs. random condition), F(1, 48) = 12.84, p < .05. This means the overall performance in the condition with Hebb trials was better than in the condition with random trials. We also found a significant main effect of group, F(1, 48) = 5.076, p < .05. This points out that the normal readers had a better performance on the task than poor readers. However, we did not find an interaction between trial and condition, F(7, 336) = .720, p >.05, which shows there was not an improvement on Hebb trials over the trials compared to random trials. Neither did we find an interaction between trial x condition x group, F(7, 336) = 1.54, p >.05. Thus the learning curve on the Hebb trails (relative to the random trials) was not steeper for normal readers than for poor readers.
The analysis for the Visuo-Spatial Increasing Length task was conducted with a Repeated Measures ANOVA to compare condition (Hebb and random trials) with group (poor and normal readers). Figure 2 illustrates the average score obtained per condition for the Visuo-Spatial Increasing Length Task. We found a significant main effect of the condition, F(1, 48) = 80.38, p < .05. This means that the Hebb trials were better performed than the random trials by both groups. We did not find an interaction between condition and Table 2
Descriptive Statistics visuo – spatial tasks
Task Poor (n=25) Normal (n=25)
M SD M SD
FLH 25.92 9.39 32.56 11.63 FLR 22.68 8.08 27.64 10.36 ILH 7.08 1.71 7.10 1.90 ILR 4.54 0.83 4.66 1.12
Note. FLH = Fixed Length Hebb; FLR= Fixed Length Random; ILH
group, i.e. the better performance on the Hebb trials relative to the filler trials was not
different for the groups. In addition, we did not find a main effect of the group, meaning there was no significant difference between the performances of the poor readers compared to the normal readers.
Figure 1. Scores per trial of the Fixed Length task for poor and normal readers. The lines represent the average score per trial, i.e., the mean span length
for the Fixed Length Hebb (blue) and the Fixed Length Random task (green).
Figure 2. Scores for the Hebb condition and Random condition of the Increasing Length Task. The poor readers (blue line) scored similarly
to normal readers (green line) on the Hebb and random condition showing poor readers do not benefit less from the Hebb effect. There is a significant main effect of condition; the Hebb condition was performed better by both groups than the Random condition. There is no interaction between the condition (ILH and ILR) and the group (poor and normal readers).
Repetitive Reading task
The analysis of the repetitive reading task was conducted with a Mixed ANOVA to compare condition (1-3 from easy to difficult words), repetition (1–6 trials) and group (poor
and normal readers). Figure 3 illustrates the average scores obtained for each trial for three conditions on the repetitive reading task. We found a significant main effect of condition,
F(1.23, 58.93) = 661.52, p < .05. This means that there was a difference in scores for each
condition; for the easiest condition the scores were higher than for the intermediate and difficult words respectively. We also found a significant main effect of trial, F(3.022, 145.04) = 55.12, p < .05 meaning that for each consecutive trial the scores improved. We found a significant interaction between trial and group, F(3.02,145.04) = 5.65, p < .05 which means that the normal readers had a steeper increment in their performance throughout the trials than poor readers. We also found a significant interaction between condition and group,
F(1.23, 58.93) = 22.03, p < .05, i.e. the degree of difficulty of each condition influenced the
performance of the groups, and the performance per condition differed between the groups. Normal readers always performed better than poor readers in all conditions, and both groups performed better on the easiest condition than on the intermediate and difficult condition consecutively. A significant interaction between trial and condition was also found, F(10, 480) = 13.36, p < .05 meaning a steeper increment in scores for each trial in the easy condition than in the difficult condition. There was, however, no interaction effect found between trial x condition x group.
Figure 3. Scores per trial of the Repetitive Reading task for poor and normal readers. The lines represent the progress in amount of words read per minute
Math Timed Test
The analysis of the Math Timed Test (MTT) was performed with an Independent-Samples T-Test. The difference between the scores of the poor readers and normal readers, means 81.04 (SD 3.58), 89.64 (SD 4.44) respectively, was not significant, t(48)= -1.50, p>.05, and had a low effect size, r = .22. The MMT was to be included to control for poor arithmetic skills to cause differences on Hebb learning in our visuo-spatial task. However, since we did not find a differential Hebb effect between the groups we did not include the MMT in the analyses.
Correlations
Table 3 shows the correlation between all variables. We made a distinction between reading for the first time (RRT1) and reading for the sixth time (RRT6) where actual learning has taken place. We took only the measures of the first condition (easy) since this condition showed the highest learning effect. The RRT1 is significantly correlated with the Fixed Length Hebb (FLH) condition. The FLH task involves both short-term memory and long-term memory, since it is based on repetition. The Fixed Length Random condition (FLR) only involves short-term memory, since it has no repetition. The RRT1contains no repetition and yet it is correlated with the task that has repetition, presumably because the FLH task has an association with the FLR task. To find the shared variance between RRT1 and FLH excluding the effect of FLR, we ran a partial correlation between the RRT1 task and FLH controlling for FLR. This time, we did not find a significant partial correlation for RRT1 indicating that there is no pure association between RRT1 and FLH. RRT6 is also
significantly correlated with FLH and not with FLR. RRT6 has a learning component and so does FLH, which might explain the correlation. To filter out the effect of FLR on FLH we ran a partial correlation between RRT6 and FLH controlling for FLR, r = .261, p = .070. This indicates that there is no pure association between RRT6 and FLH. There was no correlation
between RRT1 and RRT6 with the FLR task, neither when controlling for FLH. Neither was there a correlation with ILH or ILR. In summary, RRT1 and RRT6 are only associated with FLH.
The EMT is significantly correlated with FLH and FLR. We ran a partial correlation between EMT and FLH controlling for FLR but did not find a significant correlation.
Inversely, we correlated EMT with FLR controlling for FLH, and did not find a significant correlation either. This means that EMT is correlated with the Fixed Length tasks but there is no difference in the contribution of the Hebb and Random trials in this correlation.
Table 3 – Correlation between measures
Measure MTT RRT1 RRT6 FLH FLR ILH ILR EMT
Bivariate correlations N=50 MTT - RRT1 .444** - RRT6 .274* .720** - FLH .113 .338* .373** - FLR .024 .271 .268 .704** - ILH .281** .055 .124 .293* .355* - ILR .161* .113 -.004 .429** 477** .227** - EMT .437** .771** .644** .309* .303* .057 .184* - Partial correlations FLHa .235 .261 .205 FLRb .035 .015 .078
Note. MTT= Math Timed Test; RRT1 = Repetitive Reading Task before repetition; RRT6 = Repetitive Reading Task after 6 repetitions; FLH = Fixed
Length Hebb; FLR = Fixed Length Random; ILH = Increasing in Length Hebb; ILR = Increasing in Length Random; EMT= Een Minuut Test. Spearman’s Rho was used for FLH, ILH, and EMT in Bivariate correlations.
a correlation between FLH and RRT1, 6, and EMT respectively, was controlled for FLR b correlation between FLR and RRT1, 6, and EMT respectively, was controlled for FLH
*p<.05, **p<.01
Discussion
The aim of current study was to examine whether poor reading children have a deficit in the
long-term learning of serial information, in the visuo-spatial domain. We investigated this by
of a frog jumping to different spaces on a screen with interspersed Hebb and Random trials.
We also wanted to find whether the Hebb Effect is associated with the reading process on
itself, and investigated this through a repetitive reading task. Our experiment was innovative
because it is the first experiment measuring the Hebb Repetition Effect in poor reading
children.
In contrast to our expectations, we found that children with a poor reading level did
not benefit less from repetition in the visuo-spatial task than children with a normal reading
level as found by Szmalec et al. (2011). Both groups differed in the condition with a fixed
span of 6 spatial positions: the poor readers obtained lower results than the normal readers on
the Hebb and random trials but neither of them seemed to benefit from repetition at all since
their performance did not improve over trials. The Hebb trials were performed better than the
random trials, but already since the first time presented; in other words, the trial that was to
be repeated was already better learnt even before repetition and the score remained higher in
every trial averaged over both groups. This might have happened due to a primacy effect; the
Hebb trial was always presented first.
All participants did much better on the Hebb condition than on the Random condition
in the Increasing Length task, suggesting a Hebb Repetition Effect, but there were no
significant differences between poor and normal readers, suggesting both groups learned
equally well on this task. Since the poor readers obtained lower scores than normal readers on
the Fixed Length Task but not on the Increasing in Length task, it is possible that a span of
six movements was too long for recall. This might point at a short term memory problem
which is not in line with Szmalec et al. (2011) who suggest the problem lays in the transition
from serial information from the short term to the long term memory. It might also be viewed
as a working memory problem. The working memory enables manipulations (or
1998). Our task demanded manipulation of the short term memory as children were requested to imitate the movements of the frog with the mouse pointer during recall. The Fixed Length task might have had a higher working memory demand than the Increasing in Length task that builds up instead of presenting the information all at once. Should this be the case, it would be analogue to the observations of Banai and Ahissar (2010); if the task had a higher working memory demand the performance of the adolescent dyslexics on that task became poorer compared to normal readers.
Altogether, the results of the visuo-spatial tasks show that poor reading children score
lower on larger serial trials (Fixed Length tasks) for both Hebb and random trials than normal
reading children, but that they experienced parallel Hebb learning compared to normal
readers in the Increasing Length task. We can therefore not conclude an impaired Hebb
Repetition Effect in poor reading children with the visuo-spatial assessment.
In the repetitive reading task we observed that the scores per trial were higher for
normal readers than for poor readers. We also observed that the differences in scores were
larger for the condition with the easy words, but smaller for the intermediate words and also
smaller for the difficult words. In accordance with our expectations, poor reading children
experienced smaller improvements in reading fluency than normal reading children.
Since we wanted to find out if there was a relation between Hebb learning and
repetitive reading, we ran a correlation analysis. For the visuo-spatial task with fixed span
length and a repetition component (Fixed Length Hebb Condition) we did not observe a Hebb
effect since the scores did not improve over the trials. Since the results were higher on the
Hebb condition than on the random condition, we still wanted to find whether there was an
association between the Repetitive Reading Task and the scores in the Fixed Length Hebb
Condition. We found significant associations between FLH and reading for the first time and
Length Hebb condition and repetitive reading controlling for the possible effect of the Fixed
Length Random Condition, no association was found. Neither did we find an association
between repetitive reading and the Fixed Length Random condition. A possible reason is that
the scores on the Hebb task were higher, already from the first trial presented, and therefore
may have contributed to a stronger association with repetitive reading than the random
condition. As for the Increasing in Length task, we did find a Hebb learning effect but we did
not find a correlation for neither condition (Hebb or random).This suggest that the
performance on the Repetitive reading task is more likely to be associated to the Fixed
Length Task per se, than on the Hebb effect specifically. Moreover, the reading level (EMT)
was correlated to both Fixed Length conditions but not to the Increasing Length conditions.
In our study we had several limitations. First of all, the difference in our expected
results, i.e. not finding an impaired Hebb effect as in Szmalec et al. (2011), is most likely due
to several factors; they did their research on adults, not on children. In addition, our study
group consisted of poor reading children, rather than dyslexic children. Another possible limitation is the possible influence of a ceiling effect on the performance on the Increase in Length Hebb task, being the maximum number of correct answers (nine) reached by 35% of the children, of which on 18 were normal readers and 17 poor readers.
Although the expected results were not obtained in the visuo-spatial task, the findings
suggest a Hebb Repetition Effect in the Increasing Length Task for both groups. This is an
important finding, since it might imply poor readers to benefit just as much from repetition of
visual serial information as normal readers for serial information that does not exceed a
certain span length. Possibly the deficit is manifested when learning large amounts of serial
information, even when presented repeatedly. To confirm this, it is important to expand this
research further in this direction by measuring the Hebb effect in poor reading children
Some recommendations for further research are to use a larger group of participants
and to include a wider range of age. This, aside from increasing the group size, makes the
experiment applicable to all children, not only aged 9 and 10. Furthermore, since dyslexia is a
disorder that affects a vast amount of children, it is necessary to continue doing research but
References
Abadio, H.R & Turner, J. (2005). Reading Fluency: The Road to Developing Efficient and Effective Readers. The Nera Journal, 41(1), 50-6
Ahissar, M., Lubin, Y., Putter-Katz, H., & Banai, K. (2006). Dyslexia and the failure to form a perceptual anchor. Nature Neurosciences, 9(12), 1558–1564
Baddeley, A., Gathercole, S., & Papagno, C. (1998). The Phonological Loop as a Language Learning Device. Psychological Review, 105, 158–173
Baddeley AD. (2002) Is working memory still working? European Psychologist, 7 (2): 85-97 Banai, K. & Ahissar, M. (2010) On the Importance of Anchoring and the Consequences of Its
Impairments in Dyslexia. Wiley Interscience DYSLEXIA,16: 240–257 (2010)
Ben-Yehudah, G., Sackett, E., Malchi-Ginzberg, L., & Ahissar, M. (2001). Impaired contrast sensitivity in dyslexics is specific to retain-and-compare paradigms. Brain: A Journal
of Neurology, 124, 1381-1395
Ben Dror, I., Bentin, S., & Frost, R. (1995). Semantic, phonological, and morphological skills in reading disabled and normal children: Evidence from perception and production of spoken Hebrew. Reading Research Quarterly, 30, 876–893
Beneventi, H., Tonessen, F.E., Ersland, L., Hugdahl, K., (2010). Working Memory Deficit in Dyslexia: Behavioral and fMRI Evidence. International Journal of Neuroscience, 120, 51-59
Bleichrodt, N. D., Drenth, P. J. D., Zaal, J. N., & Resing, W. C. M. (1987). Revisie
Amsterdamse Kinder Intelligentie Test. Lisse: Swets & Zeitlinger
Bos, K. P. van den, Lutje Spelberg, H. C., Scheepstra, A. J. M., & Vries, J. R. de (1994).
De Klepel. Vorm A en B. Een test voor de leesvaardigheid van pseudowoorden.
Nijmegen: Berkhout
handleiding. Nijmegen: Berkhout
Cabeza, R., & Nyberg, L. (2000). Imaging Cognition II: An Empirical Review of 275 PET and fMRI Studies. Journal of Cognitive Neuroscience, 12(1), 1–47
Casey, B. J., Cohen, J.D., Jezzard, P., Turner, R., Noll,D. C., Trainor, R. J., et al. (1995). Activation of Prefrontal Cortex in Children During a Nonspatial Working Memory Task with Functional MRI. NeuroImage, 2(3), 221–229
Démonet, J.F., Taylor, M.J., Chaix, Y. (2004). Developmental Dyslexia. The Lancet, 363: 1451-60
Hulme, C. & Snowling, M. (1992): Deficits in Output Phonology: An Explanation of Reading Failure? Cognitive Neuropsychology, 9:1, 47-72
Landerl, K. & Moll, K. (2010). Comorbity of learning disorders: Prevalence and familial transmission. Journal of Child Psychology and Psychiatry, 51(3), 287-294
Mosse, E. K., & Jarrold, C. (2008). Hebb Learning, Verbal Short-term Memory, and the Acquisition of Phonological Forms in Children. Quarterly Journal of Experimental
Psychology, 61, 505–514
Raven, J. C., Court, J. H., & Raven, J. (1988). Standard Progressive Matrices. Londen, Engeland: H. K. Lewis & Co. ltd.
Repovs, G., & Baddeley, A.D. (2006) Multi-component model of working memory:
explorations in experimental cognitive psychology. Neurosc. Special Issue, 139, 5-21 Schumacher, J., Hoffmann, P., Schmal, C., Schulte-Korne, G., Nothen, M. M. (2007).
Genetics of dyslexia: the evolving landscape. Journal of Medical Genetics. 44: 289-297 Shaywitz, S.E. (1996). Dyslexia. Scie. Am., 275, 98-104
Szmalec A., & Loncke M., Page M.P.A., & Duyck W. (2011). Order or Disorder? Impaired Hebb Learning in Dyslexia. Journal of Experimental Psychology: Learning, Memory,
Vandierendonck, A., Kemps, E., Fastame, M.C., Szmalec, A. (2004). Working memory components of the Corsi blocks tasks. Br. J. Psychol, Feb; 95 (Pt 1): 57-79
Vellutino, F.R., Fletcher, J.M., Snowling, M.J., & Scanlon, D.M. (2004). Specific reading disability (dyslexia):What have we learned in the past for four decades? Journal of
Child Psychology and Psychiatry, and allied disciplines, 45(1), 2-40
