Differences in working memory between gifted or talented students and community samples: A meta-analysis.
When detecting highly capable students, we must consider the theoretical model the measurement is based on, as it will guide the identification and subsequent intervention (Pfeiffer, 2012). Even though the first studies about high abilities by Terman (1925) who considered gifted those with an IQ equal or above 130, there has been an evolution of the concept including talent, creativity, innovation and excellence (Gagne, 2004; Hernandez-Torrano & Gutierrez-Sanchez, 2014; Touron & Touron, 2011). The identification may be carried out by measuring intelligence based on standardized tests, or including more variables, such as general and specific capabilities, personal variables, and the valuation of the environment (Harder, Vialle, & Ziegler, 2014).
The characteristics of high-ability students include the development of cognitive and motivational strategies, which makes their learning style different from that of normative students. This is defined by different learning rhythms, precocity and depth, abstract style, and a greater understanding that differentiates them from their other classmates (Van Tassel-Baska, 2013). The cognitive differential functioning of this group can be explained by greater plasticity and efficiency, which contributes to having extensive attentional processes facilitating the high level of cognitive skills, complexity, and--sometimes--precocity of manifestation (Geake, 2009). Working memory (WM), flexibility, and inhibition contribute to better complex cognitive functioning in these students, emphasizing the high performance in WM as an executive process for convergent and divergent intellectual functioning (Sastre-Riba & Viana-Sanz, 2016).
WM is a system that temporarily maintains and manipulates information (Tirapu-Ustarroz & Munoz-Cespedes, 2005). The concept of WM has evolved from a simple memory store system to a multicomponent memory system, which shows an evolution to a more systematic and dynamic understanding of what WM is (Yuan, Steedle, Shavelson, Alonzo, & Oppezzo, 2006).
The most important conceptualization of WM was developed by Baddeley & Hitch (1974). They introduced the multicomponent model with a central executive system and two slave storage systems: the visuospatial sketchpad and the phonological loop. This model has been in continuous development and reflects the main theoretical framework in WM based on a multiprocess activity that relies on a variety of systems (Baddeley, 2003). However, other models, called state-based, also described WM mechanisms o from a cognitive neuroscience perspective (D'Esposito & Postle, 2015).
The relationship between intelligence and WM has been widely studied, although it is still a matter of debate. Recent studies have clarified underlying processes that explain this relationship (Rey-Mermet, Gade, Souza, von Bastian, & Oberauer, 2019; Waunguparaja, 2018) and the mechanisms of WM (Chekaf, Gauvrit, Guida, & Mathy, 2018). However, the discussion is focused around the amount of that relation and in what way both constructs are the same.
Nevertheless, this relationship seems to be more complex. A meta-analysis showed that WM and intelligence are related, but they are not the same (Ackerman, Beier, & Boyle, 2005). These authors found a moderate positive correlation across studies (p = 0.397), concluding that WM could be classified as a lower cognitive ability in hierarchical models of cognitive functioning. In a later study, Alloway and Alloway (2010) considered that WM is not a proxy for IQ, but it represents a cognitive ability dissociable with a greater importance to predict school performance in young children than IQ. On similar lines, Rey-Mermet, Souza, Gade and von Bastian (2019) related executive control with fluid intelligence and WM and showed that they are different factors but correlated. This relation is also showed by Redick et al. (2016) but not in an isomorphic way.
Other studies established a strong link between fluid intelligence and WM stating that WM and the g factor of intelligence are (almost) isomorphic constructs (Barbey, Colom, Paul, & Grafman, 2014; Colom, Abad, Rebollo, & Shih, 2005; Engle, 2002; Jastrzebskia, Ciechanowskab, & Chuderskib, 2018).
Given the existence of primary studies--in which children who are gifted and talented are compared with their normative peers--in this research, a meta-analysis will be carried out to determine the role that WM plays in the cognitive evolution of this student area.
The aim of this research is to evaluate WM in gifted children across different studies. Specifically, the main goals of this review are (a) to compare differences between gifted and talented children in WM, (b) to compare differences in verbal and visual WM between gifted and community children, (c) to analyze the age effect on WM in gifted children, and (d) to analyze the methodological issues affecting research on WM and gifted children. We hypothesized that talented students have a higher score in WM, which supports the idea of a high correlation between intelligence and WM.
The 33 studies analyzed included a total of 609 talented students and 969 community samples. The mean and standard deviation of the age in gifted/talented are 11.08 (range 7.44 - 17.05) and 3.16, respectively, and community samples are 10.11 (range 6.83-175) and 3.31, respectively. The percentage of males in the experimental group was 66.60%, but in the community samples, it was 54.08%. In table 1 and 2, sample sizes of each study are showed, as well as the average age of the participants of each study.
To do the meta-analysis, a coding book was prepared (available contacting the reference autor). The elaboration process of the coding book is detailed below.
Once the final articles were selected, a coding book was designed, in which the modulating variables of interest were recorded and divided into substantive and methodological variables (Botella & Sanchez-Meca, 2015).
The substantive variables are those related to and that allow the characterization of the topic studied in the meta-analysis. These were (a) the mean and SD of the age in gifted/ and community samples and (b) the procedure in which talent was diagnosed.
The methodological variables, which are related to the research design and the instruments, were (a) the total size of the study sample, (b) the total size of the experimental group, (c) the total size of the control group, (d) the type of the experimental group (gifted [IQ > 130] or talented [IQ < 130 or not specified]), (e) the instrument for measuring talent, and (f) how to measure WM.
Two of the authors participated in the analysis of the coding process's reliability with a random sample of 6 articles (Botella-Ausina & Sanchez-Meca, 2015). Cohen's Kappa index (Cohen, 1960) was calculated using SPSS v21, without reaching an excellence value of .75 according to the criteria of Fleiss (1981). The discrepancies were solved by consensus, doing again the codification. After obtaining an adequate value (.93), we proceeded to code all the articles to identify variables that influence the variability of the study results. Those that involved visual and verbal WM measures were coded as different studies.
To search for relevant investigations, different procedures were used. The main one consisted of the electronic search of articles in the Academic Search Premier databases, Educational Resources Information Center (ERIC), MeDline, PsychArticle and Psychlnfo using the terms "working memory" AND (gifted (*)) OR "talented student" OR "high ability students." The search was restricted to material published in English, Spanish, and French, and no temporal limitation in the publication of the articles was applied. The first search, conducted on November 23, 2016, yielded a total of 1,173 publications. The duplications were automatically deleted on January 1, 2017, leaving a total of 973 documents. A screening was carried out through by reading the titles and abstracts to select the articles that met the inclusion criteria, which were previously designed. In addition, other search strategies included (a) a review of titles and abstracts of articles suggested by the databases, (b) contact with the authors, (c) Google Scholar, and (d) the ability to track the bibliographic citations in the articles. The search procedure lasted 2 months, from November 2016 to January 5, 2017
Selection and exclusion criteria
The following inclusion criteria were used on the articles sampled:
1. Studies measuring WM in populations diagnosed as "gifted," "talented," or "high ability student" were included.
2. Measures of WM should include the mean and variances in chosen studies.
3. Studies with full text available were included.
4. Studies with a community comparison group were included.
5. The WM measurement procedures must be identified in included studies.
6. Peer-reviewed publications chosen must be written in English, Spanish, or French.
The exclusion criteria follow:
1. Studies including children with double exceptionality or savant were excluded.
2. Studies with a control group with any neurological or psychiatric pathology were excluded.
3. Studies for which the article's full text was not available were excluded.
4. Studies without a community comparison group were excluded.
The means and standard deviations of WM measurements were recorded in each group and in each study. Subsequently, the calculation of the effect size for each of the studies was made from the standardized mean difference using Hedges' g. The effect size between the talented group and the control group of normative intelligence was found. For each value of g, the 95% confidence interval was calculated to determine its significance. Positive g values indicated a better performance in WM for the gifted group than for the control group. Following Cohen's (1988) guidelines, effect sizes around 0.2, 0.5, and 0.8 were interpreted as reflecting low, moderate, and large practical relevance.
The calculation of effect sizes and confidence intervals in the study set was done using a random-effects model. This model considers a within-study variability, depending on the sampling error, and between-studies variability that reflects the heterogeneity in methods and sample characteristics among studies. Once the total effect size was found, separate effect sizes were also calculated for verbal WM and visual WM. A forest plot was constructed, and heterogeneity among the effect sizes was assessed with the homogeneity Q statistic and the P index. Publication bias was assessed by constructing funnel plots with the trim and fill method. Analysis of the moderating variables was accomplished by applying metaregression models for those that were continuous and ANOVAs for categorical ones. These analyses were carried out through Comprehensive Meta-Analysis v.3 (Borenstein, Hedges, Higgins, & Rothstein, 2014).
Selection of studies and characteristics
A total of 17 articles met the inclusion criteria and were selected (Figure 1); they allowed us to obtain 33 independent studies or comparisons between an experimental and a control group. Tables 1 (Visual WM) and 2 (Verbal WM) show the characteristics of the studies analyzed:
- 23 studies are composed of samples of more than 30 gifted children, and 10 are made up of samples of up to 80 gifted children.
- 25 studies are made up of samples from the control group of more than 80 children and adolescents, and 8 studies are made up of samples of up to 80 children and adolescents.
- 14 of the studies mediated verbal work memory compared to 19 that mediated visual work memory through standardized tests in 15 studies and experimental tasks in 18 studies.
- In 26 studies, psychometric tests are used to determine the category of gifted children, and in 7 of the studies, this group presents a prior identification.
- To measure intelligence, 15 studies used Wechsler scales (Wechsler, 1991), 12 studies used Raven (Raven, Raven, & Court, 2003) and four other procedures, and two studies did not specify how to measure intelligence.
- With respect to the type of talent, 12 of the studies presented a sample with an IQ higher than 130 with or without mathematical talent. In the remaining 21 studies, talent diagnoses or undifferentiated talents were included, with unspecified IQ or mathematical talent or IQ below 130.
Average effect size and heterogeneity
The calculation of the effect sizes, using Hedges' g, was carried out by separately analyzing the data related to verbal and visual language, as well as the joint effect. Figure 3 shows the forest plot of the meta-analysis performed for all studies, differentiating between visual and verbal work memory. In addition, a total effect is calculated without this differentiation.
The analysis of all the 33 studies (Figure 2) as a whole showed an average effect of [g.sub.+] = 0.80 (95% CI: 0.621, 0.976), of large and significant magnitude, and large heterogeneity (<2(32) = 196.966; P
< .001; P = 83.754%). Analyzing the two subgroups separately, we found the following results. There were 14 studies measuring verbal WM, with an average effect of [g.sub.+] = 0.969 (95% CI: 0.697, 1.241) of large magnitude and large heterogeneity (<2(13) = 78.388; p < .001; P = 83.416%). A total of 19 studies included visual memory, with an average effect size of [g.sub.+] = 0.674 (95% CI: 0.443, 0.906) of moderate magnitude and large heterogeneity (<2(18) = 108.078; p
< .001; P = 83.345%).
Figures 3 and 4 present funnel plots for visual and verbal WM meta-analyses, respectively. When the trim and fill method was applied to both funnel plots, no effect sizes were added to symmetrize the aspect of these funnel plots. Therefore, publication bias is not considered to be a threat in regard to the results of these meta-analyses.
The variables that were analyzed as moderators were the total sample size, the sample size of the gifted group, the sample size of the control group, the nomination, the type of talent, the method of measuring intelligence, the method of measuring WM, and the age. Table 3 shows the results of the ANOVAs and meta-regressions applied for each moderator variable. The only variable that reached a statistically significant relationship with the effect sizes was the method of measuring the WM (p = .019), with a larger average effect size for studies that measured the WM with standardized tests ([g.sub.+] = 1.044) than for experimental tasks ([g.sub.+] = 0.661).
In light of the results obtained, some important conclusions can be reached. First, there is a clear difference between the students with high abilities and the control group in visual WM and verbal WM, the size of the effect being greater in the case of verbal WM, which is considered large, but the visual is moderate (Cohen, 1988). This result support partially our hypothesis that states that there is a strong relation between WM and intelligence. However, a greater clarification in how is this relation established is needed (Chekaf et al., 2018; Jastrzebskia et al., 2018; Redick et al., 2016; Rey-Mermet et al., 2019; Wongupparaja et al., 2018).
The analysis carried out by moderating variables also yields relevant information. On the one hand, the only moderator that presents significant results is the procedure by which the WM concept is measured, which has a greater effect when using standardized tests against experimental tasks. This result highlights the idea that, throughout any measurement process, it is convenient to use a standard measure, as it allows greater comparability. The greater development at the psychometric level assumes criteria of goodness in the instruments that enable them to obtain more reliable data. This allows us to work through interchangeable measures while, when using laboratory measures, we must consider the lower applicability and the greater relationship with circumstantial aspects subject to the time and environment in which they are extracted. This result enhances the importance of psychometrics and highlights the need to use measurement instruments with good psychometric qualities.
Some important aspects should be highlighted within the absence of significant results: When the moderator is the type of talent diagnostic test, regardless of the test that has been measured, the type of talent is what sets differences, not the form of diagnosis made. On the other hand, there are no differences between the two groups depending on the participants' ages. The differences that are given are the equivalents by age, but this variable does not produce a differential effect depending on the groups.
In the face of the controversy raised about the type of talent and the great conceptual differentiations, the data obtained in this work do not show differences depending on the type of talent. The fundamental difference lies in the level of intellectual capacity. Although this result must be considered as one obtained when analyzing a sample of studies, it may reflect a problem present in the study of intelligence: The immense breadth of knowledge expands in different and multiple ways and may harm or postpone the search of common elements and may be mostly accepted in the scientific world. This makes reconsidering the extent to which so much variability in how the conception of talent allows real progress in the field of high intellectual capacities important.
In this way, this result is considered to be an aspect of great interest. How the different nomenclature or criterion for cataloging the condition of "high capacity" does not mean a difference in its superior performance in WM is observed as well. These data may be relevant to highlight how, in literature with different categories and labels for a population, a tendency of disaggregation can be implemented. The superior performance in WM without differences, depending on the label, raises the need to look for more common elements and gives way to an approach that seeks greater convergence.
As far as the limitations of this study are concerned, few primary studies have been used, which requires a greater number of them to strengthen and ensure these results. The small number of studies analyzed could explain the absence of significant effects of the moderating variables. Therefore, more primary studies should be done in this topic so the sample could be increased and test whether there are any effects due to these moderating variables.
The results obtained here corroborate the existence of differences at a cognitive level among the most capable students, which results in strengthening the need for a different educational approach based on this population's defining abilities. The clarification of the differentiating characteristics of talented students is fundamental, especially when there is a clear maintenance of biased ideas and myths (Perez-Tejera, Borges, & Rodriguez-Naveiras, 2017), both in the cognitive and socio-affective fields (Borges, Hernandez-Jorge, & Rodriguez-Naveiras, 2011), so, fundamentally in order to receive the appropriate educational response, it is essential to know the real characteristics of these students.
We would like to thank Dr. Julio Sanchez Meca for his helpful comments on previous versions of the paper.
"References marked with an asterisk indicate studies included in the meta-analysis" (American Psychological Association [APA] Publication Manual, 5th ed., section 4.05, p. 222)
Ackerman, P. L., Beier, M. E., & Boyle, M. O. (2005). Working memory and intelligence: The same or different constructs? Psychological Bulletin, 131(1), 30-60. https://doi.org/10.1037/0033-2909.131.130
(*)Alloway, T. P., & Elsworth, M. (2012). An investigation of cognitive skills and behavior in high ability students. Learning and Individual Differences, 22, 891-895.
Alloway, T. P., & Alloway, R.G. (2010). Investigating the predictive roles of working memory and IQ in academic attainment. Journal of Experimental Child Psychology, 106, 20-29. https://doi.org/10.1016/j.jecp.200911.003
Baddeley, A. (2003). Working memory: Looking back and looking forward. Nature Reviews Neuroscience, 4(10), 829-839. https://doi.org/10.1038/nrnl201
Baddeley, A. D., & Hitch, G. (1974). Working memory. In H. Bower (Ed), The psychology of learning and motivation: Vol. 8. Advances in research and theory (pp. 47-89). New York: Academic Press.
Barbey, A. K., Colom, R., Paul, E. J., & Grafman, J. (2014). Architecture of fluid intelligence and working memory revealed by lesion mapping. Brain Structure and Function, 219(2), 485-494 https://doi.org/10.1007/s00429-013-0512-z
Borenstein, M., Hedges, L., Higgins, J., & Rothstein, H. (2005). Comprehensive Meta Analysis (Version 3) [Computer software]. Englewood, NJ: BioStat.
Borges, A., Hernandez Jorge, C, & Rodriguez-Naveiras, E. (2011). Evidence against the myth of adjustment problems of people with high intellectual abilities. Psicothema, 23, 362-367
Botella-Ausina, J., & Sanchez-Meca, J. (2015). Meta-analisis en Ciencias Sociales y de la Salud [Meta-analysis in social and health sciences]. Madrid: Sintesis.
(*)Calero, M. D., Garcia-Martin, M. B., Jimenez, M. I., Kazen, M., & Araque, A. (2007). Self-regulation advantage for high-IQ children: Findings from a research study. Learning and Individual Differences, 17 328-343.
Chekaf, M., Gauvrit, N, Guida, A., & Mathy, F. (2018) Compression in working memory and its relationship with fluid intelligence. Cognitive Science, 42, 904-922. https://hal.univ-rennes2.fr/hal-01873321
Cohen, J. (1960). A coefficient of agreement for nominal scales. Educational and Psychological Measurement, 20, 37-46. doi http://doi.org/10.1023A
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2 (nd) ed.). New York: Academic Press.
Colom, R., Abad, F J, Rebollo, I., & Shih, P. C. (2005). Memory span and general intelligence: A latent-variable approach. Intelligence, 33, 623 -642. https:// doi.org/10.1016/j.intell.2005.05.006
Dai, Y., & Chen, F (2014). Paradigms of gifted education. A guide to theory-based practice-focused research. Texas, Usa: Prufrock Press. (*)Desco, M., Navas-Sanchez, F J., Sanchez-Gonzalez, J., Reig, S., Robles.
O., Franco, C, ... & Arango, C. (2011). Mathematically gifted adolescents use more extensive and more bilateral areas of the fronto-parietal network tan controls during executive functioning and fluid reasoning tasks. Neurolmage, 57, 281-292.
D'Esposito, M., & Postle, B. R. (2015). The Cognitive neuroscience of working memory. Annual Review of Psychology, 66, 115-142. https://doi.org/10.1146/annurev-psych-010814-015031
Duval, S. (2005). The trim and fill method. In H. Rothstein, A. J. Sutton & M. Borenstein (Eds.), Publication bias in meta analysis: Prevention, assessment and adjustments (pp. 127-144). Chichester. England: Wiley.
Engle, R. W. (2002). Working memory capacity as executive attention. Current Directions in Psychological Science, 11, 19-23.
Fleiss, J. L. (1981). Statistical methods for rates and proportions (2 (nd) ed.) New York: Wiley.
Gagne, F (2004). Transforming gifts into talents: The DMGT as a developmental theory. High Ability Studies, 15, 119-147
Geake J. G. (2009). Neuropsychological characteristics of academic and creative giftedness. In L.V. Shavinina (Ed), International handbook of giftedness (pp. 261-273). Dordrecht: Springer.
Harder, B., Vialle, W., & Ziegler, A. (2014). Conceptions of giftedness and expertise put to the empirical test. High Ability Studies, 25(2), 83-120.
(*) Haring, I. L. (2016). Thepredictive value of working memory and creativity in average performing and gifted children. Universiteit Utrecht, Utrecht. Retrieved from https://dspace.library.uu.nl/handle/1874/335028
Heller, K. A. (2004). Identification of gifted and talented students. Psychology Science, 46(3), 302-323.
Hernandez-Torrano, D., & Gutierrez-Sanchez, M. (2014). The study of high intellectual ability in Spain: Analysis of the current situation. Revista de Educacion, 364, 251-272. https://10.4438/1988-592X-RE-2014-364-261
(*)Hoard, M. K. (2005). Mathematical cognition in gifted children: Relationships between working memory, strategy use, and fluid intelligence. Dissertation abstracts international. B. The sciences and engineering, 67(9-B), 5443.
(*)Howard, S. J, Johnson, J., & Pascual-Leone, J. (2013). Measurement of mental attention: Assessing a cognitive component underlying performance on standardized intelligence tests. Psychological Test and Assessment Modeling, 55(3), 250-273.
Jastrzebskia, J., Ciechanowskab, I., & Chuderskib, A. (2018). The strong link between fluid intelligence and working memory cannot be explained away by strategy use. Intelligence, 66, 44-53. https://doi.org/10.1016/j.intell.201711.002
(*)Johnson, J., Im-Bolter, N., & Pascual-Leone, J. (2003). Development of mental attention in gifted and mainstream children: The role of mental capacity, inhibition, and speed of processing. Child Development, 74(6), 1594-1614. https://doi.org/10.1046/j.1467-8624.2003.00626.x
(*)Khosravi-Fard, E., Keelor, J. L., Akbarzadeh-Bagheban, A. R., & Keith, R.W. (2016). Comparison of the Rey Auditory Verbal Learning Test (RAVLT) and digit test among typically achieving and gifted students. Iranian Journal of Child Neurology, 10(2), 26-37.
(*)Kormmann, J., Zettler, I., Kammerer, Y., Gerjets, P., & Trautwein, U. (2015). What characterizes children nominated as gifted by teachers? A closer consideration of working memory and intelligence. High Ability Studies, 26(1), 75-92. http://dx.doi.org/10.1080/13598139.2015.1033513
(*)Leikin, M., Paz-Baruch, N., & Leikin, R. (2013). Memory abilities in generally gifted and excelling-in-mathematics adolescents. Intelligence, 41, 566-578.
(*)Leikin, R., Paz-Baruch, N., & Leikin, M. (2014). Cognitive characteristics of students with superior performance in mathematics. Journal of Individual Differences, 35(3), 119-129 https://10.1027/1614-0001/a000140
Matthews, D. I, & Dai, D Y. (2014). Gifted education: changing conceptions, emphases and practice. International Studies in Sociology of Education, 24(4), 335-353. https://doi.org/10.1080/096202142014979578
(*)Navarro, J. I., Ramiro, P., Lopez, J. M., & Aguilar, M. (2006). Mental attention in gifted and nongifted children. European Journal of Psychology of Education, XXI(4), 401-411.
(*)Paz-Baruch, N., Leikin, R., & Leikin, M. (2016). Visual processing in generally gifted and mathematically excelling adolescents. Journal for the Education of the Gifted, 39(3), 237-258. https: //doi org/10.1177/0162353216657184
Perez, J., Borges, A., & Rodriguez-Naveiras, E. (2017) Knowledge and myths about high abilities. Talincrea, 4(1), 40-51.
Pfeiffer, S. I. (2012). Current perspectives on the identification and assessment of gifted students. Journal of Psychoeducational Assessment, 30, 3-9.
Raven, J., Raven, J. C, & Court, J. H. (2000). Standard Progressive Matrices. Oxford: Oxford Psychologists Press.
(*)Sacuzzo, D. P., Johnson, N. E., & Guertin, T. L. (1994). Information processing in gifted versus nongifted african american, latino, filipino, and white children: Speeded versus nonspeeded paradigms. Intelligence, 19, 219-243.
Sastre-Riba, S., & Viana-Sanz, L. (2016). Executive functions and high intellectual capacity. Revista de Neurologia, 62 (Supl 1), S65-S71.
(*)Segalowitz, S. J., Unsal, A., & Dywan, J. (1992). Cleverness and wisdom in 12-year-olds: Electriphysiological evidence for late maturation of the frontal lobe. Developmental Neuropsychology, 8(2 & 3), 279-298.
Sternberg, R. J. (1985). Beyond I. Q.: A triarchic theory of intelligence. Nueva York: Cambridge University Press.
Stroup, D. E, Berlin, J. A., Morton, S. C, Olkin, I., Williamson, G D., Rennie, D., & Thacker, S. B. (2000). Meta-analysis of observational studies in epidemiology. Journal of the American Medical Association, 283(15), 2008-2012. http://doi.org/10.1001/jama.283.15.2008
(*)Swanson, H. L. (2005). Cognitive processes that underlie mathematical precociousness in young children. Journal of Experimental Child Psychology, 93, 239-264
Redick, T. S., Shipstead, Z., Meier, M. E., Montroy, J. I, Hicks, K L., Unsworth, N.,... Randall W. (2016). Cognitive predictors of a common multitasking ability: Contributions from working memory, attention control, and fluid intelligence. Journal of Experimental Psychology: General, 145(11), 1473-1492. http://dx.doi.org.accedys2.bbtk.ull.es/10.1037/xge0000219
Rey-Mermet, A., Gade, M., Souza, A., von Bastian, C.C., & Oberauer, (2019). Is executive control related to working memory capacity and fluid intelligence? Journal of Experimental Psychology: General. Advance online publication, https://doi.org/10.1037/xge0000593
Terman, L. (1925). Mental and physical traits of a thousand grafted children. Stanford, CA: Stanford University Press.
Tirapu-Ustarroz, J., & Munoz-Cespedes, J. M. (2005). Memory and the executive functions. Revista de Neurologia, 41(8), 475-484. https://doi.org/10.33588/rn. 4108.2005240
Touron, J., & Touron, M. (2011). The center for Talented Youth Identification model: A review of the literature. Talent Development & Excellence, 3(2), 187-202.
Van Tassel-Baska, J. (2013). Curriculum for the gifted. A commitment to excellence. Gifted Child Today, 36, 213-240.
Wechsler, D. (1991). Wechsler Intelligence Scale for Children-Third Edition manual. San Antonio, TX: The Psychological Corporation.
Wongupparaja, P., Sumicha, A., Wickensd, M., Kumaria, V., & Morris, R. G. (2018). Individual differences in working memory and general intelligence indexed by P200 and P300: A latent variable model. Biological Psychology, 13, 96-105. https://doi.org/10.1016/j.biopsycho.2018.10.009
Yuan, K, Steedle, J., Shavelson, R., Alonzo, A., & Oppezzo, M. (2006). Working memory, fluid intelligence, and science learning. Educational Research Review, 1(2), 83-98. https://doi.org/10.1016/j.edurev.2006.08.005
Elena Rodriguez-Naveiras (1), Emilio Verche (2), Pablo Hernandez-Lastiri (3), Rubens Montero (3), and Africa Borges (3)
(1) Universidad Europea de Canarias, (2) Universidad Europea de Madrid, and (3) Universidad de La Laguna
Received: January 27, 2019 (*) Accepted: May 17, 2019
Corresponding author: Emilio Verche
Facultad de Ciencias Biomedicas y de la Salud
Universidad Europea de Madrid
28045 Madrid (Spain)
Table 1 Characteristics of the included studies by subgroup visual WM c Authors N Intelligence WM Nomination Talent test Task type Alloway & Elsworth 2012 82 Wechsler 2 1 2 Deseo et al 2011 27 Wechsler 2 2 1 Haring 2016 43 Other 2 1 2 Hoard 2005 217 Raven 2 1 2 Hoard 2005 211 Raven 2 1 2 Hoard et al 2008 211 Raven 2 1 2 Howard etal 2013 91 Unspecified 2 2 2 Johnson et al 2003 52 Wechsler 2 2 1 Johnson et al 2003 52 Wechsler 2 2 1 Johnson et al 2003 97 Wechsler 2 2 Johnson et al 2003 97 Wechsler 2 2 Khosravi et al 2016 148 Wechsler 2 Kormnann et al 2015 81 Other 2 Leikin etal 2013 157 Raven 1 Paz-Baracheta2016 96 Raven 1 Saccuzzo et al 1994 160 Raven 1 Saccuzzo et al 1994 160 Raven 1 Swanson 2005 127 Wechsler 2 Swanson 2005 127 Wechsler 2 Talented Commurity Authors n Age WM WMSD n Age Mean mean Mean Alloway & Elsworth 2012 44 10.04 128.0 12.5 38 9.80 Deseo et al 2011 13 13.40 75.8 12.4 14 13.80 Haring 2016 27 10.70 0.73 0.70 41 10.00 Hoard 2005 26 8.790 5.375 2.109 191 8.89 Hoard 2005 44 6.130 57.7 30.4 167 6.80 Hoard et al 2008 44 6.160 112.0 14.0 167 6.16 Howard etal 2013 47 13.72 0.77 0.15 44 13.72 Johnson et al 2003 17 8.05 4.41 1.37 35 8.27 Johnson et al 2003 17 8.05 4.24 1.15 35 8.27 Johnson et al 2003 40 10.42 5.60 0.74 57 10.64 Johnson et al 2003 40 10.42 4.95 1.20 57 10.64 Khosravi et al 2016 73 13.00 4.753 0.98 75 13.00 Kormnann et al 2015 42 9.87 9.02 2.29 39 9.60 Leikin etal 2013 70 17.50 11.2 2.40 87 16.70 Paz-Baracheta2016 40 17.00 6.00 0.78 56 17.00 Saccuzzo et al 1994 80 9.00 8.26 4.06 80 9.00 Saccuzzo et al 1994 80 9.00 11.09 4.62 80 9.00 Swanson 2005 50 7.44 2.14 4.51 77 7.30 Swanson 2005 50 7.44 4.00 5.74 77 7.30 samples Authors WM WMSD mean Alloway & Elsworth 2012 92.13 18.29 Deseo et al 2011 64.1 17.4 Haring 2016 0.73 0.11 Hoard 2005 3.98 1.41 Hoard 2005 28.6 28.8 Hoard et al 2008 97.0 14.0 Howard etal 2013 0.64 0.14 Johnson et al 2003 3.69 0.96 Johnson et al 2003 3.26 0.82 Johnson et al 2003 4.67 1.07 Johnson et al 2003 4.30 1.18 Khosravi et al 2016 3.65 0.797 Kormnann et al 2015 7.87 2.77 Leikin etal 2013 10.2 2.50 Paz-Baracheta2016 6.07 1.09 Saccuzzo et al 1994 7.46 3.81 Saccuzzo et al 1994 9.22 5.37 Swanson 2005 1.54 3.65 Swanson 2005 3.19 3.33 Note: Nomination = Psychometric test =1; Prior identification =2. Talent = Gifted/Talented: IQ<130; IQ not specified Mathematical talent (IQ not specified) =1; Gifted (IQ> 130). In scores and percentiles, they are coded with scores higher than the 98th percentile. O Gifted + Mathematical talent = 2 WM Task: Experimental tasks=l; Standardized tests =2 Table 2 Characteristics of the included studies by subgroup verbal WM Authors N Intelligence WM Nomination test Task Talent type Alloway & Elsworth 2012 82 Wechsler 2 1 2 Caero et al 2007 47 Other 1 1 1 Hoard 2005 211 Raven 2 1 2 Hoard et al 2008 211 Raven 2 1 2 Howard et al 2013 91 Unspecified 2 2 2 Kormnann et al 2015 81 Other 1 1 2 Leikin et al 2013 157 Raven 2 1 1 Leikin e tal 2014 49 Raven 2 1 1 Navarro et al 2006 110 Wechsler 2 1 1 Segaowitz et al 1992 48 Unspecified 2 2 1 Swanson 2005 127 Wechsler 1 1 2 Swanson 2005 127 Wechsler 1 1 2 Swanson 2005 127 Wechsler 1 1 2 Swanson 2005 127 Wechsler 1 1 2 Authors Talented n Age WM WM n Mean mean SD Alloway & Elsworth 2012 Caero et al 2007 44 10.04 125.73 16.31 38 Hoard 2005 24 8.19 4.46 0.45 23 Hoard et al 2008 44 6.13 48.5 28.9 167 Howard et al 2013 44 6.25 110.0 15.0 167 Kormnann et al 2015 47 9.81 0.59 0.12 44 Leikin et al 2013 42 9.87 8.97 0.61 39 Leikin e tal 2014 70 16.50 12.1 1.90 87 Navarro et al 2006 26 16.70 11.0 2.10 23 Segaowitz et al 1992 70 10.30 5.70 1.20 40 Swanson 2005 18 12.20 6.80 1.53 30 Swanson 2005 50 7.44 1.66 0.65 77 Swanson 2005 50 7.44 5.56 4.24 77 Swanson 2005 50 7.44 3.94 2.84 77 50 7.44 6.64 5.40 77 Authors Commurity samples Age WM WM Mean mean SD Alloway & Elsworth 2012 Caero et al 2007 9.80 93.42 14.76 Hoard 2005 7.81 2.81 0.63 Hoard et al 2008 6.80 23.9 23.7 Howard et al 2013 6.25 97.0 14.0 Kormnann et al 2015 9.81 0.47 0.12 Leikin et al 2013 9.60 8.52 0.63 Leikin e tal 2014 16.70 10.2 2.50 Navarro et al 2006 16.70 10.8 3.10 Segaowitz et al 1992 9.37 3.90 1.40 Swanson 2005 12.60 4.70 1.53 Swanson 2005 7.30 1.33 0.69 Swanson 2005 7.30 3.20 3.44 Swanson 2005 7.30 2.80 2.68 7.30 4.32 4.34 Note: Nomination = Psychometric test =1; Prior identification =2. Talent = Gifted/Talented: IQ<130; IQ not specified Mathematical talent (IQ not specified) =1; Gifted (IQ> 130). In scores and percentiles, they are coded with scores higher than the 98th percentile. O Gifted + Mathematical talent = 2 WM Task: Experimental tasks=l; Standardized tests =2 Table 3 Moderating variables Categorical moderators N total Average size: from 1 to 80 participants Large size: more than 80 participants N gifted Average size: from one to 30 participants Large size: from 31 to 80 participants Average size: from 1 to 80 participants N control Large size: more than 80 participants Psychometric test Nomination Prior identification Gifted >130 Talent Gifted<130 Raven Weschler type Intelligence test other tests No intelligence tests or unspecified Experimental tasks WM task Standardized tests Continuous moderators Intercept Sample size Average age gifted Intercept Average age control Llimit Ulimit Studies Hedges' g 95% 95% N total 9 0.917 0.549 1.285 24 0.762 0.557 0.968 N gifted 10 0.918 0.572 1.263 23 0.755 0.546 0.965 8 0.792 0.405 1.178 N control 25 0.801 0.598 1.005 25 0.769 0.567 0.971 Nomination 8 0.903 0.524 1.281 12 0.763 0.458 1.069 Talent 21 0.817 0.597 1.038 12 0.5642 0.347 0.936 15 0.867 0.595 1.140 Intelligence test 4 0.891 0.342 1.440 2 1.170 0.396 1.943 15 0.586 0.348 0.825 WM task 18 0.849 0.594 6.533 Continuous moderators Coefficient 1.695 0.4897 1.6493 Sample size -0.0270 -0.0820 0.0281 1.0748 0.4882 1.6614 -0.0276 -0.0836 0.0284 Z value p-value Qw P - value N total 4.883 0.001 0.516 (1) 0.473 7.271 0.001 N gifted 5.204 0.001 0.619 (1) 0.431 7.069 0.001 4.015 0.001 N control 0.002 (1) 0.965 7.727 0.005 7.446 0.001 Nomination 0.371 (1) 0.542 4.673 0.001 4.901 0.001 Talent 0.078 (1) 0.5780 7.267 0.001 4.266 0.001 6.236 0.001 Intelligence test 2.326 (1) 0.543 3.182 0.001 2.965 0.003 4.820 0.001 WM task 5.470 (1) 0.019 6.533 0.001 Continuous moderators 3.62 0.0001 0.92 (1) 0.337 Sample size -0.96 0.1683 3.59 0.002 0.94 (1) 0.336 -0.97 0.1667
|Printer friendly Cite/link Email Feedback|
|Author:||Rodriguez-Naveiras, Elena; Verche, Emilio; Hernandez-Lastiri, Pablo; Montero, Rubens; Borges, Africa|
|Date:||Jul 1, 2019|
|Previous Article:||Self-esteem and suicidal behaviour in youth: A meta-analysis of longitudinal studies.|
|Next Article:||Maladaptative personality traits in adolescence: Behavioural, emotional and motivational correlates of the PID-5-BF scales.|