g factor (also known as general intelligence , general mental ability or general intelligence factor ) is a construct developed in psychometric investigation of cognitive abilities and human intelligence. These are variables that encapsulate positive correlations among different cognitive tasks, reflecting the fact that individual performance on one type of cognitive task tends to be proportional to that person's performance on other types of cognitive tasks. The g factor typically accounts for 40 to 50 percent of the difference in performance between individuals on given cognitive tests, and composite scores ("IQ scores") based on many tests often perceived as individual estimates' Standing above g factor. The term IQ, general intelligence, general cognitive abilities, general mental abilities, or simply intelligence are often used interchangeably to refer to these common nuclei that are shared by cognitive tests. g factors target custom sizes general intelligence .
The existence of the g factor was originally proposed by the English psychologist Charles Spearman in the early years of the 20th century. He observed that children's performance appraisals, across seemingly unrelated school subjects, were positively correlated, and reasoned that these correlations reflected the underlying general mental effect that went into performance on all types of mental tests. Spearman suggests that all mental performance can be conceptualized in terms of a single common ability factor, which he labels g , and a large number of narrow specific task-ability factors. Soon after Spearman proposed the existence of g , his existence was challenged by Godfrey Thomson, which provided evidence that Spearman's intercorrelation findings among the test results were inconsistent with the existence of g or in his absence. The current intelligence factor model usually represents cognitive ability as a three-level hierarchy, where there are a large number of narrow factors at the bottom of the hierarchy, some broad factors, more common at the middle level, and at the peak of a single factor, called the g factor , which represents the common variance for all cognitive tasks.
Traditionally, research on g has concentrated on psychometric investigation of test data, with particular emphasis on factor analytic approaches. However, empirical research on the properties of g has also been drawn on experimental cognitive psychology and mental chronometry, brain anatomy and physiology, quantitative and molecular genetics, and primate evolution. While the existence of g as statistical regularity is well established and not controversial, there is no consensus as to what causes a positive correlation between tests.
Research in the field of behavioral genetics has determined that the construct g is strongly inherited. It has a number of other biological correlations, including brain size. It is also a significant predictor of individual differences in many social outcomes, particularly in education and employment. The most widely accepted contemporary theories include the g factor. However, critics of g have stated that the emphasis on g is misplaced and requires another important devaluation of ability, and supports a revised realistic view of human intelligence. Some critics have gone so far as to state that g "... is for what psychometric Huygens ether is for the early physicist: nonentity taken as an article of faith, not one that requires verification with real data."
Video G factor (psychometrics)
Cognitive proficiency test
Cognitive ability tests are designed to measure different aspects of cognition. The specific domains judged by the tests include math skills, verbal fluency, spatial visualization, and memory, among others. However, individuals who excel in one type of test tend to excel in other types of tests, too, while those who do one test tend to do it on all tests, regardless of the test content. The English psychologist Charles Spearman was the first to describe this phenomenon. In a famous research paper published in 1904, he observed that the performance measures of children across seemingly unrelated school subjects were positively correlated. These findings have been replicated many times. The consistent findings of the universal positive correlation matrix of mental test results (or "positive manifolds"), despite the large differences in test content, have been described as "arguably the most replicated results in all psychology". Zero or negative correlations between tests indicate a sampling error or a limitation of the range of capability in the sample under study.
Using factor analysis or related statistical methods, it is possible to compute a common factor that can be considered as a summary variable that characterizes the correlation between all the different tests in the test battery. Spearman refers to this general factor as a common factor , or just g . (By convention, g is always printed as italics.) Mathematically, the g factor is the source of discrepancies between individuals , which requires that one can not meaningfully speak of a person's mental abilities consisting of g or other factors of a certain degree. A person can only talk about a person's status on g (or other factors) compared to other individuals in the relevant population.
Different tests in the test battery can correlate with (or "load to") the battery factor to a different degree. This correlation is known as g loading. An individual test taker g factor score, representing the relative standing in the g factor in the total group of individuals, can be estimated using g loading. A full-scale IQ score of the test battery will typically be highly correlated with the g factor score, and they are often considered to be estimated g . For example, the correlation between g factor scores and full-scale IQ scores from David Wechsler tests has been found to be greater than.95. The term IQ, general intelligence, general cognitive abilities, general mental abilities, or simple intelligence are often used interchangeably to refer to the common core possessed by cognitive tests.
The g load of mental tests is always positive and typically ranges between 0.10 and 0.90, with an average of about 0.60 and a standard deviation of about 0.15. Raven's Progressive Matrices is one of the highest-loaded g tests, about.80. Vocabulary tests and general information are also usually found to have high g loading. However, the g loading of the same test may be somewhat different depending on the composition of the test battery.
The complexity of the tests and demands they place on mental manipulation is related to the ' g loading test. For example, in a future digit range test, the subject is asked to repeat the sequence of numbers in the order of presentation after hearing them once at a single digit rate per second. The test of the rear digit range is the exact opposite except that the subject is required to repeat the digits in the reverse order to which they are presented. The rear digit range test is more complex than the front-digit range test, and has a significantly higher g load. Similarly, g load arithmetic calculations, spelling, and word reading tests are lower than arithmetic problem solving, text composition, and reading comprehension tests, respectively.
Testing difficulties and g loading are different concepts that may or may not be related empirically in certain situations. Tests that have the same level of difficulty, as indexed by the proportion of test items failed by the test participants, may show various g charges. For example, memory memorization tests have been shown to have the same level of difficulty but much lower g loading than many tests involving reasoning.
Maps G factor (psychometrics)
Theory
While the existence of g as statistical regularity is well established and not controversial among experts, there is no consensus as to what causes positive intercorrelation. Several explanations have been put forward.
Energy or mental efficiency
Charles Spearman reasoned that the correlation between tests reflects the influence of common causal factors, general mental skills that go into performance on all types of mental tasks. However, he thinks that the best indicator of g is a test that reflects what he calls relationship-related and correlated education, which includes skills such as deduction, induction, problem-solving. , grasping relationships, summing up rules, and finding differences and similarities. Spearman hypothesizes that g is equivalent to "mental energy". However, this is more of a metaphorical explanation, and he remains agnostic about the physical basis of this energy, hoping that future research will reveal the exact physiological properties of g .
Following Spearman, Arthur Jensen stated that all mental tasks enter g to some extent. According to Jensen, the factor g represents a "distillate" score on a different test than the sum or average score, with factor analysis acting as a distillation procedure. He argues that g can not be explained in terms of item characteristics or information content of tests, indicating that very different mental tasks may have almost the same g loading. Wechsler also argues that g is not a capability at all but rather a public property of the brain. Jensen hypothesizes that g corresponds to individual differences in the speed or efficiency of neural processes associated with mental abilities. He also suggested that given the relationship between g and basic cognitive tasks, it should be possible to construct a scale test g which uses time as a unit of measurement.
Sampling theory
The so-called sampling theory g , was originally developed by E.L. Thorndike and Godfrey Thomson, suggest that the existence of a positive manifold can be explained without reference to the underlying underlying capacity. According to this theory, there are a number of uncorrelated mental processes, and all tests use different samples of these processes. The intercorrelation between tests is caused by the overlap between processes that are intercepted by the test. Thus, positive manifolds arise because of measurement problems, the inability to measure finer mental processes, which may not be correlated.
It has been proven that it is not possible to distinguish statistically between the Spearman model from g and the sampling model; both are equally able to account for the intercorrelation between tests. Sampling theory is also consistent with the observation that more complex mental tasks have higher loading, because more complex tasks are expected to involve larger nerve element samples and therefore have more in common with others. task.
Some researchers have argued that the sampling model cancels g as a psychological concept, since the model indicates that the g factor derived from different test batteries only reflects the shared elements of a particular test. contained in every battery rather than g that is common to all tests. Similarly, the high correlation between different batteries can be due to their measuring the same set of capabilities rather than the same capability.
Critics argue that the sampling theory is not aligned with certain empirical findings. Based on sampling theory, one might expect that related cognitive tests share many elements and thus are highly correlated. However, some closely related tests, such as forward and backward digit ranges, are just simple correlates, while some seemingly completely unequal tests, such as the Vocabulary and Raven matrix tests, are consistently highly correlated. Another problematic finding is brain damage that often causes certain cognitive impairments rather than common disorders that may occur based on sampling theory.
mutualism
The "mutualism" model of g proposes that cognitive processes are initially uncorrelated, but that positive manifolds arise during individual development because of the mutually beneficial relationship between cognitive processes. So there is no single process or capacity underlying a positive correlation between tests. During the development process, this theory applies, every highly efficient process will benefit other processes, with the result that those processes will end up correlating with each other. Thus, the same high IQ in different people may come from the very different initial gains they have. Critics argue that the observed correlations between the loading and the heritability coefficients of the subtes are problematic for the theory of mutualism.
Structure of cognitive ability factor
Factor analysis is a family of mathematical techniques that can be used to represent correlations between tests of intelligence in terms of a small number of variables known as factors. The goal is to simplify the correlation matrix by using hypothetical underlying factors to explain the patterns within it. When all the correlations in the matrix are positive, as they are in the case of IQ, factor analysis will produce general factors that are common to all tests. The general factor of the IQ test is called the g factor, and this typically accounts for 40 to 50 percent of the variants in the IQ test battery. The presence of correlations between many varied cognitive tests has often been taken as evidence of the existence of g , but McFarland (2012) suggests that the correlation does not provide more or less support for the existence of g than existence several factors of intelligence.
Charles Spearman developed a factor analysis to study the correlations between tests. Initially, he developed a model of intelligence in which variations in all intelligence test scores are explained by only two types of variables: first, the specific factors for each test (denoted s ); and secondly, the factor g contributed a positive correlation across the test. This is known as Spearman's two-factor theory. Then research based on test batteries more diverse than those used by Spearman shows that g alone can not explain all the correlations between tests. In particular, it was found that even after controlling g , some tests were still correlated with each other. This leads to the postulation of group factors representing the variance that the test group with the same task demands (eg, verbal, spatial, or numerical) have similarities in addition to the g variance.
Through factor rotation, in principle, it is possible to produce an infinite number of factor solutions that are mathematically equivalent in their ability to account for intercorrelation between cognitive tests. This includes solutions that do not contain the g factor. So factor analysis alone can not determine the basic structure of intelligence. In choosing between different factor solutions, the researcher should examine the results of factor analysis along with other information about the cognitive ability structure.
There are many relevant psychological reasons for choosing factor solutions that contain the g factor. These include the existence of a positive manifold, the fact that some types of tests (generally more complex) have larger g loads, substantial invariant than g factors across different battery tests, the impossibility of building test batteries that do not produce the g factor, and the wide practical validity of g as predictors of individual results. The factor g , together with the group factor, best represents an empirically established fact that, on average, the overall difference in capabilities between individuals is greater than the difference between the capabilities > within individuals, while factor solutions with orthogonal factors without g obscure this fact. In addition, g seems to be the most heritable component of intelligence. Research using confirmatory factor analysis techniques has also provided support for the existence of g .
The g factor can be calculated from the correlation matrix of test results using several different methods. These include exploratory factor analysis, principal component analysis (PCA), and confirmatory factor analysis. Different factor extraction methods produce very consistent results, although PCA has sometimes been found to produce an increased estimate of the effect of g on the test scores.
There is broad contemporary consensus that cognitive variation between people can be conceptualized at three hierarchical levels, distinguished by their general level. At the lowest level, at least there are a large number of first-order factors that are narrow; at a higher level, there is a relatively small amount - somewhere between five and ten - of the second order factors (ie, group factors) wider (ie, more general); and at its peak, there is a third-order factor, g , a common factor common to all tests. The g factor typically accounts for the majority of the total variants of the general factors of the IQ test battery. Models of contemporary hierarchical intelligence include three stratum theories and the Cattell-Horn-Carroll theory.
"Indicator ignorance"
Spearman proposes the principle of indifference from the indicator, which he says the exact content of the intelligence test is not important for the purpose of identifying g , since g goes into performance at all type of test. Each test can be used as an indicator of g . Following Spearman, Arthur Jensen recently argued that the extracted factor of one test battery will always be the same, within the measurement error limits, as extracted from other batteries, provided the battery is large. and diverse. According to this view, every mental test, no matter how typical, calls g to some extent. Thus the combined score of a number of different tests will be loaded into g stronger than any individual test score, since the components g converge into the composite score, while the non- g uncorrelated will cancel each other. Theoretically, the composite score of a very large and diverse test battery would be a perfect size g .
In contrast, LL Thurstone believes that the extracted factor of the test battery reflects the average of all capabilities that are called by a particular battery, and therefore varies from one battery. batteries to the other and "has no fundamental psychological significance." In line with the same line, John Horn argues that the g factor is meaningless because they are not invariant in the battery test, maintaining that the correlation between different capability measures arises because it is difficult to determine human actions that depend on just one ability.
To show that different batteries reflect the same g , one has to administer multiple test batteries to the same individual, extract g the factor of each battery, and show that the factor is highly correlated.. This can be done within the framework of confirmatory factor analysis. Wendy Johnson and colleagues have published two such studies. The first found that the correlations between the g factors extracted from three different batteries were 0.99, 0.99, and 1.00, supporting the hypothesis that the g factor of the battery is different is the same. and that the identification of g is not dependent on the speci fi c capabilities assessed. The second study found that the g factor derived from four of the five test batteries correlated between 0.95 to 1.00, while the correlation ranged from 0.79 to 0.96 for the fifth battery, Cattell Culture Fair Intelligence Test. (CFIT). They correlate somewhat lower correlations with CFIT batteries with a lack of diversity of content because they contain only matrix types, and interpret the findings to support the assumption that the g factors derived from different test batteries are the same as those that battery is quite diverse. The results show that the same g can be consistently identified from different test batteries.
Population distribution
The population distribution form g is unknown, since g can not be measured on a ratio scale. (The distribution of scores on IQ tests is generally more or less normal, but this is achieved by construction, that is by normalizing the raw score.) It has been argued that there is good reason to assume that g is usually distributed in the general population, range  ± 2 standard deviation from the mean. Specifically, g can be considered as a composite variable reflecting the additive effect of a large number of independent genetic and environmental influences, and that variable must, in accordance with the central limit theorem, follow a normal distribution.
Spearman's law of diminishing returns
Some researchers have suggested that the proportion of variations calculated by g may not be uniform across all subgroups within a population. The decreasing Spearman spider law (SLODR), also called the cognitive capability hypothesis of differentiation , predicts that the positive correlation between different cognitive abilities is weaker among more subgroups of individuals intelligent. More specifically, (SLODR) predicts that the g factor will account for the smaller proportion of individual differences in cognitive test scores at higher scores on the g factor.
(SLODR) was originally proposed by Charles Spearman, who reported that the average correlation between 12 cognitive ability tests was 0.466 in 78 normal children, and 0.782 in 22 "disabled" children. Detterman and Daniel rediscovered this phenomenon in 1989. They reported that for subtests of both WAIS and WISC, the subtestation of the subtest decreased monotonically with the ability group, ranging from about 0.7 average intercorrelation among individuals with an IQ of less than 78 to 0.5 among individuals with an IQ greater than 122.
(SLODR) has been replicated in various samples of children and adults who have been measured using a wide range of cognitive tests. The most common approach is to divide individuals into groups of abilities to use observable proxies for their general intellectual abilities, and then compare between averages among subtests among different groups, or to compare the proportion of variations calculated by a single common factor, in different groups. However, both as Deary et al. (1996). and Tucker-Drob (2009) have demonstrated, dividing the continuous distribution of intelligence into a number of groups of arbitrary discrete abilities less than ideal for checking (SLODR). Tucker-Drob (2009) extensively reviews the literature on (SLODR) and various methods previously tested, and suggests that (SLODR) can be appropriately captured by installing a general factor model that allows the relationship between factors and indicators to be nonlinear. He applied the factor model to the national representative data of children and adults in the United States and found consistent evidence for (SLODR). For example, Tucker-Drob (2009) found that common factors accounted for about 75% of variation in seven different cognitive abilities among very low IQ adults, but only accounted for about 30% variation in ability among very high IQs. adults.
A recent meta-analytic study by Blum and Holling also provided support for the differentiation hypothesis. Unlike most studies on this topic, this work makes it possible to study capacity and age variables as a sustainable predictor of saturation g , and not only to compare the lower vs. the higher or the younger. the older group of test takers. The results show that the mean correlation and g imposition of cognitive ability tests decreases with increasing ability, but increases with age of respondents. (SLODR), as described by Charles Spearman, can be confirmed by g -saturation decrease as a function of IQ as well as an increase in g from middle age to aging. Specifically speaking, for samples with an average intelligence of two standard deviations (ie, 30 points IQ) higher, the expected average correlation decreased by about 0.15 points. The question remains whether this large difference can result in clearer factorial complexity when cognitive data are taken into account for higher-capacity samples, compared with low-ability samples. It seems likely that larger factor dimensions should tend to be observed for higher ability cases, but the magnitude of these effects (ie, how likely and how many factors) is uncertain.
Practical validity
The practical validity of g as a predictor of educational, economic, and social outcomes is broader and more universal than any other known psychological variables. The validity of g is greater the greater the complexity of the task.
The practical validity of a test is measured by its correlation with performance on some criteria beyond the test, such as the average college grade, or job performance rating. The correlation between test scores and the size of some criteria is called the coefficient of validity . One way to interpret the validity coefficients is to squarce them to obtain the variance calculated by the test. For example, the validity coefficient of 0.30 corresponds to the 9 percent variant described. This approach, however, has been criticized as misleading and uninformative, and several alternatives have been proposed. One approach that can be interpreted is to look at the percentage of test participants in each quintile test score that meets some agreed standard of success. For example, if the correlation between test and performance values ​​is 0.30, the expectation is that 67 percent of those in the top quintile will perform above average, compared with 33 percent in the bottom quintile.
Academic achievement
Predictive validity g is the most striking in the domain of scholastic performance. This is apparently because g is closely related to the ability to learn new material and understand concepts and meanings.
In primary school, the correlation between IQ and grade and achievement score is between 0.60 and 0.70. At a more advanced educational level, more students from the lower end of the IQ distribution drop out, which limits the IQ range and results in lower validity coefficients. In secondary schools, colleges, and graduate schools, the validity coefficients are.50-.60,.40-.50, and.30-.40, respectively. The high IQ score implies, but it may be that some of the IQ's validity in predicting scholastic attainment is due to factors measured by IQ independent of g . According to a study by Robert L. Thorndike, 80 to 90 percent of the variance predicted in scholastic performance is due to g , with the remainder associated with non- g factor measured by IQ and other tests.
The achievement test score is more correlated with IQ than the school score. This may be because the value is more influenced by the students' idiosyncratic perceptions. In a longitudinal English study, the g score measured at age 11 correlated with all 25 national GCSE exam test subjects taken at age 16. The correlation ranged from 0.77 for the math test to 0.42 for the test art. The correlation between g and the general education factor calculated from the GCSE test is 0.81.
Research shows that SAT, which is widely used in college admissions, is basically a measure of g . A correlation of 0.82 has been found between g scores calculated from IQ test batteries and SAT scores. In a study of 165,000 students in 41 US colleges, the SAT score was found to correlate at 0.47 with the first year of college-average grade after correcting range restrictions in SAT scores (correlation rose to 0.55 when course difficulties were held constant, ie, if all students attend the same class).
Work achievement
There is a high correlation of 0.90-0.95 between job prestige ratings, which are rated by the general population, and the average common intelligence score of people employed in each job. At the level of individual employees, the relationship between job prestige and g is lower - one large US study reported a correlation of 0.65 (0.72 corrected for attenuation). The average rate is g so it increases with the prestige of perceived work. It has also been found that general intelligence scatter deployment is smaller in more prestigious jobs than in lower level work, indicating that higher employment levels have minimum requirements g .
Job performance
Research shows that the g test is the best single predictor of job performance, with an average validity coefficient of 0.55 in some meta-analysis studies based on the supervisor's ratings and job samples. The average meta-analytic validity coefficient for performance in the training job is 0.63. The validity of g in work with the highest complexity (professional, scientific, and top management jobs) has been found to be greater than in jobs with lowest complexity, but g has predictive validity even for jobs the simplest. The study also showed that a customized intelligence test adjusted for each job provides little or no increased predictive validity for general intelligence tests. It is believed that g affects job performance mainly by facilitating the acquisition of work-related knowledge. Predictive validity g is greater than work experience, and increased experience on the job does not reduce the validity of g .
In the 2011 meta-analysis, researchers found that general cognitive abilities (GCAs) predict job performance better than personality (Five-factor model) and three streams of emotional intelligence. They tested the relative importance of these constructs in predicting job performance and found that cognitive ability explained most of the variance in job performance. Other studies have shown that GCA and emotional intelligence have independent and linear complementary contributions to job performance. CÃÆ'Â'tÃÆ' Â © and Miners (2015) found that these constructs are interrelated when assessing their relationship to two aspects of job performance: organizational citizenship behavior (OCB) and task performance. Emotional intelligence is a better task performance predictor and OCB when GCA is low and vice versa. For example, an employee with a low GCA will offset the performance of his duties and OCB, if his emotional intelligence is high.
Although this compensatory effect supports emotional intelligence, GCA still remains the best job performance predictor. Some researchers have studied the correlation between GCA and job performance among different job positions. For example, Ghiselli (1973) found that the seller had a higher correlation than the sales clerk. The first obtained a correlation of 0.61 for GCA, 0.40 for perceptual ability and 0.29 for psychomotor abilities; whereas the sales clerk obtained a correlation of 0.27 for GCA, 0.22 for perceptual ability and 0.17 for psychomotor ability. Another study compares GCA - the correlation of job performance between jobs with different complexities. Hunter and Hunter (1984) developed a meta-analysis with over 400 studies and found that this correlation was higher for high complexity work (0.57). Followed by work with moderate complexity (0,51) and low complexity (0,38).
Job performance is measured by objective ranking performance and subjective ratings. Although the former is better than the subjective ranking, most studies in job performance and GCA have been based on performance ratings of supervisors. This ranking criterion is considered problematic and unreliable, mainly because of its difficulty in determining good and bad performance. Monitoring ratings tend to be subjective and inconsistent among employees. In addition, job performance supervisor ratings are influenced by various factors, such as halo effects, facial attractiveness, racial or ethnic bias, and employee height. However, Vinchur, Schippmann, Switzer and Roth (1998) found in their study with sales employees that the objective sales performance had a 0.04 correlation with GCA, while the supervisor's performance appraisal got a correlation of 0.40. This finding is surprising, given that the main criterion for assessing these employees is an objective sale.
In understanding how GCA is associated with job performance, some researchers conclude that GCA affects the acquisition of job knowledge, which in turn improves job performance. In other words, people with high GCAs can learn faster and gain more work knowledge easily, allowing them to work better. Conversely, the lack of ability to acquire knowledge of work directly affects job performance. This is because GCA levels are low. In addition, GCA has a direct effect on job performance. On a daily basis, employees continue to face challenges and problem-solving tasks, whose success depends entirely on their GCA. These findings are discouraging to the government entity responsible for protecting workers' rights. Due to GCA's high correlation on job performance, the company employs employees based on GCA test scores. Undeniably, this practice denies the opportunity to work for many people with low GCA. Previous researchers have found significant differences in GCA between race/ethnic groups. For example, there is a debate on whether the study biased against Afro-Americans, who scored significantly lower than white Americans in the GCA test. However, findings on the correlation of GCA performance should be done with caution. Some researchers have warned of the existence of statistical artifacts associated with performance measurement and GCA test scores. For example, Viswesvaran, Ones and Schmidt (1996) argue that it is highly unlikely to get a perfect measure of job performance without causing a methodological error. In addition, studies on GCA and job performance are always vulnerable to range restrictions, since data is collected mostly from current employees, ignoring those not employed. Therefore, the sample comes from employees who successfully pass through the recruitment process, including the GCA steps.
Earnings
The correlation between income and g , as measured by an IQ score, averaged about 0.40 across the study. The correlation is higher at higher levels of education and it increases with age, stable when people reach their highest career potential in middle age. Even when education, employment and socioeconomic backgrounds remain constant, correlations are not lost.
More correlations
The g factor is reflected in many social outcomes. Many social behavior problems, such as school dropouts, chronic wellbeing, accident sterility, and crime, are negatively correlated with g not dependent on the social class of origin. Health outcomes and deaths are also associated with g , with higher childhood test scores predicting better health and mortality results in adulthood (see Cognitive epidemiology).
Genetic determinants and environment
Heritability is the proportion of phenotypic variants in traits in the population that can be attributed to genetic factors. Heritability g has been estimated to fall between 40 and 80 percent using twin design, adoption, and other family designs as well as molecular genetic methods. Estimates based on the totality of evidence place the heritability of g around 50%. It has been found to increase linearly with age. For example, a large study involving more than 11,000 pair of twins from four countries reported heritability g to 41 percent at age nine, 55 percent at age twelve, and 66 percent at age seventeen. Other studies have estimated that heritability is as high as 80 percent in adulthood, although it may decline in old age. Much of the research on heritability has been done in the United States and Western Europe, but studies in Russia (Moscow), former East Germany, Japan, and rural India have produced similar heritability estimates as Western studies.
Behavioral genetic research has also established that the shared (or interpersonal) environmental effects on g are strong in childhood, but decrease thereafter and may be neglected in adulthood. This shows that the important environmental effects for the development of g are unique and are not shared among the same family members.
Genetic correlation is a statistic that shows the extent to which the same genetic effect affects two different traits. If the genetic correlation between the two properties is zero, the genetic effect on them is independent, whereas the correlation of 1.0 means that the same set of genes explains the heritability of both traits (regardless of how high or low their heritability is). Genetic correlations between certain mental abilities (such as verbal ability and spatial ability) have been found to be consistently very high, close to 1.0. This suggests that genetic variation in cognitive abilities is almost entirely due to genetic variation in anything g . It also shows that what is common among cognitive abilities is largely due to genes, and that independence among abilities is largely due to environmental effects. It has thus been argued that when genes for intelligence are identified, they will become "generalist genes", each affecting many different cognitive abilities.
The g loads of mental tests have been found to correlate with their heritability, with correlations ranging from moderate to perfect in various studies. So the heritability of mental tests is usually higher the greater the g loading.
Many research points for g to be highly polygenic properties that are influenced by a large number of common genetic variants, each having only a small effect. Another possibility is that the inherited distinction in g is because individuals have different "loads" of rare and destructive mutations, with genetic variation among individuals who survive because of the selection-mutation balance.
A number of candidate genes have been reported to be associated with differences in intelligence, but the effect size is small and almost no findings are replicated. There is no individual genetic variant that is conclusively related to intelligence within the normal range so far. Many researchers believe that very large samples will be needed to reliably detect individual genetic polymorphisms associated with g . However, while genes that affect variations within g in the normal range have proved difficult to find, a large number of single gene disorders with mental retardation among their symptoms have been found.
Several studies have shown that tests with greater loading are influenced by inbreeding depression that decreases test scores. There is also evidence that tests with greater g loading are associated with greater positive heterotic effects on test scores. Inbreeding depression and heterosis show the effect of genetic dominance for g .
Neuroscientific findings
g has a number of correlations in the brain. Studies using magnetic resonance imaging (MRI) have determined that g and total brain volume are moderately correlated (r ~.3-.4). The size of the external head has a ~.2 correlation with g . MRI studies in the brain region show that the volume of the frontal, parietal and temporal cortices, and the hippocampus also correlate with g , generally at 0.25 or more, whereas the correlation, the average over many studies, gray and white matter overall have been found.31 and.27, respectively. Some but not all studies also found a positive correlation between g and cortical thickness. However, the reasons underlying this association between the quantity of brain tissue and differences in cognitive abilities are largely unknown.
Most researchers believe that intelligence can not be localized to a single region of the brain, such as the frontal lobes. Studies of brain lesions found a small but consistent association that showed that people with white matter lesions were more likely to have lower cognitive abilities. Research using NMR spectroscopy has found a rather inconsistent but generally positive correlation between intelligence and white matter integrity, supporting the idea that white matter is important for intelligence.
Some studies show that apart from the integrity of the white matter, also the organizational efficiency associated with intelligence. The hypothesis that brain efficiency has a role in intelligence is supported by functional MRI research that shows that smarter people generally process information more efficiently, that is, they use fewer brain resources for the same task than less intelligent people.
A small but relatively consistent relationship with intelligence test scores includes brain activity, as measured by EEG or event-related potential, and nerve conduction velocity.
g on non-human
Evidence of a common factor of intelligence has also been observed in non-human animals. Studies have shown that g is responsible for 47% of individual variance in primates and between 55% and 60% in mice. Although it can not be assessed using the same measure of intelligence used in humans, cognitive ability can be measured by interactive and observational tools focused on innovation, habitual reversal, social learning, and responsiveness to new things.
Non-human models such as mice are used to study genetic influences in intelligence and neurological developmental studies into the back mechanisms and biological correlations g .
g (or c ) in the human group
Similar to g for individuals, the new research pathway aims to extract common collective intelligence factors c for groups that display the group's general ability to perform various tasks. The definition, operationalization and statistical approach to this c factor are from and similar to g . Causes, predictive validity as well as additional parallel for g are investigated.
Other biological associations
Elevation correlates with intelligence (r ~.2), but this correlation is not generally found in the family (ie, among siblings), indicating that the result of assortative cross-breeding is high and intelligence. Myopia is known to be associated with intelligence, with a correlation of about 0.2 to 0.25, and this relationship has been found in the family as well.
Group similarities and differences
Cross-cultural studies show that the g factor can be observed whenever the battery varies, complex cognitive tests are given to human samples. The factor structure of the IQ test has also been found to be consistent across sexes and ethnic groups in the US and elsewhere. The g factor has been found to be the most distinct factor of all factors in cross-cultural comparison. For example, when the g factor was calculated from the American standardized sample of the Wechsler IQ battery and from a large sample that completed the Japanese translation of the same battery compared, the congruence coefficient was 0.99, indicating the virtual identity. Similarly, the coefficient of conformity between the g factors obtained from the white and black standardization sample of the WISC battery in the US is 0.995, and the variance in the test scores recorded by g is very similar for both groups.
Most studies show that there is a negligible difference in mean levels between sexes, and that gender differences in cognitive abilities can be found in more narrow domains. For example, men generally outperform women in spatial tasks, whereas women generally outperform men in verbal tasks. Another difference that has been found in many studies is that males show more variability in both general and specialized abilities than females, with men proportionately better on the low end and high end of the test scores distribution.
Consistent differences between racial and ethnic groups in g have been found, especially in the US. A 2001 meta-analysis of millions of subjects showed that there was a 1.1 standard deviation gap in the average level between blacks and whites of America, supporting the former. The average American Hispanic score was found to be.72 standard deviations under non-Hispanic white. By contrast, East Asian Americans generally outperform white Americans. Some researchers have suggested that the magnitude of black-and-white gaps in cognitive ability tests depends on the magnitude of the test load g , with tests showing g greater production loads. gap (see Spearman's hypothesis). It has also been claimed that racial and ethnic differences similar to those found in the US can be observed globally.
Relationship with other psychological constructs
Basic cognitive tasks
Basic cognitive tasks (ECT) are also strongly correlated with g . ECT is, as the name implies, simple tasks that seem to require very little intelligence, but are still strongly correlated with more complete intelligence tests. Determining whether a light is red or blue and determining whether there are four or five boxes drawn on a computer screen are two examples of ECT. Answers to such questions are usually given by pressing the button quickly. Often, in addition to the buttons for the two options provided, the third button is pressed from the start of the test. When the stimulus is given to the subject, they release their hands from the start button to the correct answer button. This allows the examiner to determine how much time is spent thinking about the answer to the question (reaction time, usually measured in small fractions of a second), and how much time is spent on physical hand movements to the right button (movement time). Reaction time is strongly correlated with g , while the movement time is not strongly correlated. The ECT test has enabled quantitative testing of hypotheses about test bias, subject motivation, and group differences. Based on their simplicity, ECT provides a link between classic IQ tests and biological questions such as fMRI studies.
Working memory
One theory holds that g is identical or almost identical to the working memory capacity. Among other evidence for this view, several studies have found factors representing g and working memory to be perfectly correlated. However, in the meta-analysis, the correlation was found to be much lower. One criticism that has been made of a study that identifies g with working memory is that "we do not advance understanding by showing that one mysterious concept is related to another."
Piagetian task
Psychometric theory of intelligence aims to measure intellectual growth and identify differences in ability between individuals and groups. In contrast, Jean Piaget's theory of cognitive development seeks to understand qualitative change in the intellectual development of children. Piaget devised a number of tasks to verify hypotheses arising from his theory. The task is not meant to measure individual differences, and they are not equivalent in psychometric intelligence tests. For example, in one of Piagetian's most famous conservation tasks, a child is asked whether the amount of water in two identical glasses is the same. After the child agrees that the amount is the same, the investigator pours water from one glass into a glass with a different shape so the numbers look different although remain the same. The boy was then asked whether the amount of water in the two glasses was the same or different.
Regardless of the different research traditions in which psychometric tests and Piagetian tasks are developed, the correlation between the two types of actions has been found to be consistently positive and generally moderate in magnitude. Common common factors underlie them. It has been shown that it is possible to build a battery consisting of a Piagetian task which is a good measure of g as a standard IQ test.
Personality
The traditional view in psychology is that there is no meaningful relationship between personality and intelligence, and that both must be studied separately. Intelligence can be understood in terms of what a person can do, or what his/her maximum performance is, while personality can be thought of in terms of what an individual would normally > do, or what are the general trends of their behavior. Studies have shown that the correlation between the size of intelligence and personality is small, and thus it has been argued that g is a pure cognitive variable that does not depend on personality traits. In the 2007 meta-analysis, the correlations between g and the personality traits of "Big Five" were found as follows:
- conscience -.04
- hospitality.00
- extraversion.02
- .22 openness
- emotional stability.09
The same meta-analysis found a correlation of 0.20 between self-efficacy and g .
Some researchers argue that the relationship between intelligence and personality, though simple, is consistent. They have interpreted the correlation between measurement of intelligence and personality in two main ways. The first perspective is that personality traits affect performance on the intelligence test . For example, a person may fail to perform at the maximum level on an IQ test because of his anxiety and stress-prone. The second perspective considers intelligence and personality to relate conceptually, with personality traits determining how people apply and invest their cognitive abilities, leading to greater expansion of knowledge and cognitive differentiation.
Creativity
Some researchers believe that there is a threshold below g which is socially significant creativity rare, but otherwise there is no relationship between the two. It has been suggested that this threshold is at least one standard deviation above the population average. Above the threshold, personality differences are believed to be important determinants of individual variation in creativity.
Others have challenged the theory of thresholds. While not disputing such opportunities and personal attributes other than intelligence, such as energy and commitment, are important to creativity, they argue that g is positively related to creativity even at the end of a high-ability distribution. Longitudinal Youth Mathematics Study Today has provided evidence for this opinion. It has been shown that individuals identified by standardized tests as intellectually gifted in early adolescence achieve creative achievement (eg, securing patents or publishing literary or scientific works) at several times the general population level, and even in the top 1 percent of cognitive abilities, they who have a higher ability are more likely to make remarkable achievements. This study also shows that the level of g acts as a predictor of achievement levels, while certain patterns of cognitive ability predict the world of achievement.
Challenges
TheoryG f -G c
Raymond Cattell, a student of Charles Spearman, dismisses the factor model g and divides g into two relatively broad independent domains: fluid intelligence (G f ) and crystallized intelligence (G c ). G f is conceptualized as the capacity to find out new problems, and is best assessed by tests with little cultural or scholastic content, such as the Raven matrix. G c can be considered as consolidated knowledge, reflecting the skills and information that a person acquires and retains throughout his life. G c depends on education and other forms of acculturation, and is best assessed by tests that emphasize scholastic and cultural knowledge. G f can be thought of mainly consisting of the current reasoning and problem-solving capabilities , while G c reflects my previous results run the cognitive process.
The rationale for the separation of G f and G c is to explain the individual's cognitive development over time. While G f and G c have been found to be highly correlated, they differ in the way they change for life. G f tends to peak around age 20, slowly declining thereafter. Conversely, G c is stable or increases throughout adulthood. One common factor has been criticized for obscuring this bifurcated development pattern. Cattell argues that G f reflects individual differences in the efficiency of the central nervous system. G c is, in Cattell's thinking, the result of someone "investing" it G f in a lifelong learning experience.
Cattell, along with John Horn, then extends the G f -G c model to include a number of other broad capabilities, such as G q (quantitative reasoning) and G v (visual-spatial reasoning). While all broad capability factors in the extended model f -G c are positively correlated and thus will allow extraction from the higher-order g factor, Cattell and Horn argues that it would be a mistake to assume that the general factor underlies this broad capability. They argue that the g factors calculated from different test batteries are not invariant and will give different values ​​of g , and that the correlation between the tests appears because it is difficult to test only one ability at a time.
However, some researchers have suggested that the G f -G c model is compatible with g -owners of cognitive ability comprehension. For example, John B. Carroll's three-stratum intelligence model includes G f and G c along with higher order factor g . Based on the factor analysis of many data sets, some researchers also argue that G f and g are one and the same factor and that the g factor of the battery substantially different tests as long as the battery is large and diverse.
Some theorists have proposed that there is an intellectual ability that does not correlate with one another. Among the earliest were L.L. Thurstone who created the model of primary mental ability that represents an intelligence domain that should be independent. However, Thurstone's tests of this ability were found to produce strong generalized factors. He argues that the lack of independence among his tests reflects the difficulty of constructing "factorially pure" tests that measure only one ability. Similarly, J.P. Guilford proposed a model of intelligence consisting of up to 180 different abilities, uncorrelated, and claimed to be able to test them all. Then the analysis shows that Guilford's factorial procedure is presented as evidence for his theory of not providing support for it, and that the test data he claims to provide evidence against g apparently shows the usual intercorrelation pattern. after correction for statistical artifacts.
Recently, Howard Gardner has developed a theory of multiple intelligences. He proposes the existence of nine different and independent intelligence domains, such as mathematical, linguistic, spatial, musical, kinesthetic-physical, meta-cognitive, and existential intelligence, and argues that individuals who fail in some may be superior in others. According to Gardner, tests and schools have traditionally emphasized only linguistic and logical abilities while ignoring other forms of intelligence. Although popular among educators, Gardner's theory has been criticized by psychologists and psychometricians. One criticism is that the theory violates the use of the word "intelligence" both scientifically and daily. Some researchers argue that not all Gardner's intelligence is in the cognitive sphere. For example, Gardner argues that successful careers in professional sports or popular music reflect the intelligence of physical intelligence and musical intelligence, respectively, although one may speak of athletics and musicals. skills , talent , or abilities instead. Another criticism of Gardner's theory is that many of his independently reported intelligence domains are in fact correlated with each other. In response to the empirical analysis that shows the correlation between the domains, Gardner argues that the correlation exists because of the general format of the test and because all tests require linguistic and logical skills. His criticism in turn suggests that not all IQ tests are given in paper and pencil formats, which in addition to linguistic and logical abilities, IQ test batteries also contain measures, for example, spatial abilities, and basic cognitive tasks. (Eg, inspection time and reaction time) that do not involve linguistic or logical reasoning correlate with conventional IQ batteries as well.
Robert Sternberg, who works with various colleagues, also states that intelligence has a dimension separate from g . He argues that there are three classes of intelligence: analytical, practical, and creative. According to Sternberg, traditional psychometric tests only measure analytic intelligence, and must be added to test creative and practical intelligence as well. He has compiled several tests for this effect. Sternberg equated analytic intelligence with academic intelligence, and compared it with practical intelligence, defined as the ability to deal with unclear real-life problems. Tacit intelligence is an important component of practical intelligence, which consists of knowledge that is not explicitly taught but required in many real-life situations. Assessing the independent creativity of the intelligence test has traditionally proved difficult, but Sternberg and colleagues claim to have made valid creativity tests as well. Validation of the Sternberg theory requires that the three capabilities tested are essentially uncorrelated and have independent predictive validity. Sternberg has conducted many experiments which he claims to confirm the validity of his theory, but some researchers have disputed this conclusion. For example, in the reanalysis of the STAT Sternberg validation test study, Nathan Brody showed that the STAT predictive validity, the test of three allegedly independent abilities, is almost solely due to a single common factor underlying the test, which equates Brody with g factors.
Model Flynn
James Flynn argues that intelligence should be conceptualized on three different levels: brain physiology, cognitive differences between individuals, and social trends in intelligence over time. According to this model, the g factor is a useful concept with respect to individual differences but its explanatory power is limited when the focus of investigation is brain physiology, or, in particular, the effects of social trends on intelligence. Flynn has criticized the idea that cognitive upgrades over time, or the Flynn effect, are "hollow" if they can not prove to be increasing in g . He argues that the Flynn effect reflects a shift in social priorities and individual adaptation to
Source of the article : Wikipedia
0 Comments