Psy 5330

Moral Issues-Test takers have the right to:

1. Know who will have access to test data. 2. Confidentiality of test results.

Influential Approaches to Intelligence- Biological Approaches (which to some degree overlaps with psychometric and multiple-forms approaches). part 3

3. Reciprocal Causation Between Brain Morphology/Function and Intellectual Function. Folks with certain brain morphology (that is, shape and structure) tend to be smarter in specific ways, and folks who are attempt to be smarter in specific ways tend to systematically alter their brain morphology. For example, the hippocampus helps you ind your way around in the world (that is, not get lost). Cabbies in London who have been on the job for a long time have bigger hippocampus than London cabbies who are less experienced. Also, folks who played Tetris showed increases in cortical thickness in certain brain regions.

Responsibilities of Test Users and Constructors

A. Make sure tests are reliable and valid for client's group. B. Make sure tests are appropriate to client's language. C. Take into account situational, cultural, or personal factors that might influence test interpretation. D. Must know: 1. Why test is being used 2. How to use it most effectively 3. How unfairness is to be minimized 4. What psychometric properties of the measures are 5. Literature relevant to test 6. Whether interpretations are justified.

Invasion of Privacy

APA code: Personal info. communicated only with person's consent. However, test data can be subpoenaed and clinician can communicate info. if it is felt that person might be a danger. In many settings, clinicians' loyalties divided between institution and client. For example, psychologists working in business settings use tests both for the benefit of their employers and the individuals who are tested. Clinicians are required to let the clients know exactly to whom they are most beholden/loyal. They must also inform the clients of how the test data are to be used, and the limits of confidentiality.

Adaptations to Stereotype Threat part 2

Bright and motivated folks who are constantly operating under conditions conducive to stereotype threat may cope with these extra pressures on performance by: (1) Disindentifying with the threatened domain. If the domain is math, for example, they may come to view math as being less important to their self-concepts (perhaps by changing majors from engineering to psychology), And/Or (2) Dropping out of academics altogether. (For example, competent and capable young woman who dropped out college after she had completed three years of school work and accumulated a 4.0 GPA. She was a chemistry major and was raised in a culture in which scientific pursuits among women were somewhat discouraged).

Social Issues-Dehumanization

E.G., Computerized test interpretations of, say, the Minnesota Multiphasic Inventory-2, increase the chances that we will one day become the slaves of humanoid robots who can diagnose and label us with some psychopathology and subsequently deem that we are fit for no other work than the salt mine.

Adaptations to Stereotype Threat.

For Steele (1997), stereotype threat is likely to primarily affect the best and brightest among the stereotyped. That is, individuals who are both confident in and care about their performance should become anxious and distracted when they fear that their failure will tend to confirm a negative group stereotype. This should occur when (a) they're aware that the stereotype exists, and (b) the task is challenging for them.

Sex Differences

Generally speaking, there are no large IQ score differences between males and females on the most commonly used standard tests of intelligence. However, there are striking differences in a few specific cognitive abilities.

Socioeconomic factors

Groups that perform poorly on IQ tests tend to live in more impoverished conditions than do groups that perform well. Poor nutrition, fewer intellectually stimulating resources, and inadequate prenatal care may systematically lower IQ scores. However, when we, for example, match African and European Americans on socioeconomic status (SES), European Americans still have a higher average IQ score Moreover, the direction of the relationship between SES and IQ is not clear (that is low IQ may lead to low SES, rather than the other way around).

Nutrition

Guatemalan preschool kids who more frequently availed themselves of a protein supplement scored higher on school achievement tests ten years later. Other studies have shown beneficial effects of protein and vitamins on kids' IQ.

Intelligence Tests and Their Correlates

Here, we are interested in those things that IQ scores predict. For example, do they predict themselves at a later time? (are they reliable). And do they predict school grades, job success, social status, other cognitive measures, and the like?

To know

Heritability does not imply immutability! For example, although height is highly heritable, it can also be substantially influenced by nutrition. Despite recent relevant technical advances, little progress has been made in identifying the specific genes involved in intelligent behavior. As Nisbett et al. (2012) put it,"It may simply be that the number of genes involved in . . . intelligence is very large, and therefore the contribution of any individual locus is just as small as the number of genes is large . . . ."

Predictor of School Performance

IQ scores have been found to predict school performance and years of education attained (both rs about .50). IQ scores also have been found to predict social status, income, and job performance. And IQ is a negative predictor of criminality (i.e., higher IQ scores are associated with less criminal behavior).

Predictors of "Processing Speed."

IQ test scores are modestly correlated with the speed of both paired-associates learning and "same-different" judgments. It has also been found that more intelligent people need less exposure time to make accurate perceptual judgments.

Affirming Global Sense of Self-Worth (Cohen et al. 2006):

In a couple of experiments that examined the effects of self-affirmation on academic performance among African American (AA) and European American (EA) 7th graders, these folks found that: AA students who wrote about important personal values showed a remarkably lower failure rate compared with both historical norms and AA students who wrote about a neutral-control topic. The positive effect of self-affirmation was not found among EA kids (who perhaps managed to spontaneously affirm themselves and operated under conditions of less academic threat). So, affirming a general sense of self-worth goes a long way to improving adaptive coping and productive behavior.

Stability of IQ scores

Jones & Bayley (1941): The correlation between IQ at 6 years old and IQ at 18 is about 86. "Habituation" measures of infant IQ (e.g., how does it take for an infant to get bored of one stimulus and prefer another) show modest (about .36) correlations with IQ at 2 to 6 years of age. Important to remember that these data reflect relative-standing, not absolute, stability.

Perinatal factors

Low-birth-weight infants tend to have lower IQs later in childhood Controlling for mothers' SES and IQ, breastfeeding has been shown to boost IQ by as much as 3 points. This positive effect of breastfeeding appears to occur most reliably for kids having a particular genetic allele at a site that regulates fatty acids.

Social Issues-Usefulness of Tests

One must always weigh the risks of testing (e.g., labeling, misdiagnosis, discrimination) against the benefits (e.g., accurate diagnosis and subsequent treatment, guidance into fulfilling careers).

Moral Issues-Human Rights

People who don't wanna be tested shouldn't hafta. However, APA guidelines assert that "informed consent is implied because testing is a routine part of [clinical/job-related] activity . . . [or when] the purpose of test is to evaluate decisional capacity." Also if the government/law says that you must be tested, you don't have the right to refuse.

Lead and Alcohol

Postnatal exposure to lead and prenatal exposure to alcohol have been shown to have a detrimental effect on IQ.

Social Issues-Access to Psychological Testing Service.

Some clinicians charge thousands of dollars to administer a battery of tests, interpret the scores, and write a report.

A threat in the air everywhere: The ubiquity of stereotype-threat effects.

Stereotype threat has been found among: Older adults on an alleged test of memory. Low-socioeconomic-status students on a test when it was purportedly diagnostic of intelligence. Latinas but not Latinos on a math test when ethnicity had been made contextually salient. Women drivers for whom the women-are-bad-drivers stereotype had been made salient.

Interpreting Group Differences

The distinction between Native and many "Hispanic" Americans is often arbitrary - both descend predominantly from peoples who came to the Americas from Asia long before the Europeans arrived in the 15th and 16th centuries. we often group people into different categories based on an amalgam of cultural, linguistic, and genetic differences.

Stereotype Threat Definition

The event of a negative stereotype about a group to which one belongs becoming self-relevant, usually as a plausible interpretation for something one is doing, for an experience one is having, or for a situation one is in, that has relevance for self-definition. Ex. A female may feel that her performance on a math test might confirm a negative stereotype about female math ability.

The genetic hypothesis

There is no reliable evidence that between-groups genetic differences are responsible for between-groups IQ score differences.

Influential Approaches to Intelligence-The Psychometric Approach

This is the most widely used and known approach. The most commonly administered tests developed in this tradition are the various Wechsler and Stanford-Binet tests As a reminder, however, both of these tests employ scales with means of 100 and standard deviations of 15. Approximately 95% of folks will achieve scores between 70 and 130.

Interventions (cont)

b. Interventions to raise luid intelligence [g(f)].(Remember that g(f) is the ability to solve problems and learn new information) A great deal of research demonstrates that training on working memory tasks (tasks that force you to keep various types of visual and auditory information in mind) can result in substantial gains on various measures of g(f) These types of working-memory training programs have been shown to increase g(f) in kids with ADHD, elderly folks, and young adults But this type of training has not been shown to affect crystallized IQ and verbal reasoning.

Interventions (cont 2)

c. Cognitive-Enhancing Pharmaceuticals. It has been found that about 1 in 4 college students use stimulants for "cognitive enhancement." Although there is some evidence that they can be somewhat useful for this purpose, the long-term risks associated with them are not well understood. d. Physical Exercise. Aerobic exercise has been found to be very important for the maintenance of IQ as folks grow older.

The Genes and Intelligence

comparing (a) monozygotic twins with dizygotic twins and (b) children who live together who are either genetically related or adopted. The statistical indices that are computed from these studies are h2 and c2. h2 gives an estimate of the proportion of variance in IQ that is due to common genes, or genetics. c2 gives an estimate of the proportion of variance that is due to common environments (e.g., similarities in intelligence that are due to being reared in the same home). However, it must be noted that due to the operation of the individual multiplier, a process by which folks with better genes tend to be selectively exposed to better gene-relevant environments, the effects of c2 may be systematically underestimated by twin studies.

Ethnic Differences

i. Asian Americans Mean IQs for Japanese and Chinese Americans have consistently been about the same as those obtained for European Americans However, both Japanese and Chinese Americans consistently outperform European Americans in both school and the work place. Thus, Japanese and Chinese Americans seem to owe their relative success to factors other than intelligence as it is measured by most standard tests.

Ethnic Diferences (cont.)

ii. Hispanic Americans By "Hispanic Americans," we mean largely Mexican Americans, Puerto Ricans, Central and South Americans, and Cubans. Hispanic Americans typically score above African Americans and below European Americans. Partly because many Hispanics are not fluent in English, their verbal IQs scores tend to be relatively low However, their performance IQ scores tend not to be as low.

Ethnic Differences (cont.) 2

iii. Native Americans Like Hispanic Americans, Native Americans tend to score relatively low on verbal IQ and not as low on performance IQ. Although there are some subgroup differences. For example, the Inuit (who live in the extreme north) tend to have exceptional verbal-spatial skills It should also be noted that Native American children suffer disproportionately from hearing loss, which may adversely affect verbal IQ.

Environmental Effects on Intelligence

the effects of a wide range of social and environmental effects on intelligence.

Group Differences

the extent and nature of group differences in measured IQ (e.g., ethnic, racial, and sex differences).

Social Variables

"Westernized" folks tend to score higher on traditional measures of IQ (e.g., Wechsler and Binet tests). However, Liberian rice farmers are superior estimators of rice quantities Kids from Botswana tend to have excellent memories for stories Thus, culture and environment may substantially influence performance on specific intellectual tasks. And certainly social factors associated with social class (SES) appear to have a profound effect on IQ. Based on adoption studies, the most robust estimate is that relative to a low SES environment, a high SES environment confers about 12 IQ points on the individual.

Possible Reasons for the Flynn Effect?

1. Cultural Complexity/Modernity: folks today are simply living in a more complex, information-rich environment than were their ancestors.It is very likely that the Industrial Revolution created an environment in which superior intellectual skills were cultivated. And this new, intellectual-skills-selecting environment causes folks to compete with each other to increase cognitive abilities, further raising IQ. Dickens and Flynn refer to this competitive process as the social multiplier. On this point, it should be noted that the intergenerational improvements in IQ have been seen disproportionally on tests assumed to measure g(f). 2. Measurement Artifact: Although no doubt some substantial portion of the Flynn Effect reflects real gains in fluid IQ, our parents and grandparents simply could not have had been as lacking in intelligence as the data suggest! It is thus also plausible that these gains in measured IQ partially reflect something practically trivial about the nature of the measurement instruments themselves. Data are largely inconsistent with both an improved-nutrition hypothesis and a less-inbreeding hypothesis.

Influential Approaches to Intelligence-Multiple Forms of Intelligence

1. Gardner's Theory Argues that traditional approaches only measure linguistic, logical, and spatial intelligence but fails to measure other types such as musical, bodily-kinesthetic, and various forms of personal intelligence. 2. Sternberg's Triarchic Theory (1985). Argues that there are three distinct forms of intelligence: i. Analytical ii. Creative iii. Practical Argues that only analytical intelligence is measured by most IQ tests 3. Crystallized Versus Fluid Intelligence (which is actually derived from the psychometric approach). Crystallized:What folks know about the world and the operations that they have learned to perform on it. (geography, basic scientific principles, long division, how to bake a cake, etc.) Fluid: the ability to solve new problems and the ability to learn (similar to Sternberg's practical intelligence). ( solve puzzles, etc.)

Recommendations for Minimizing Stereotype Threat.

1. Wise Schooling (Steele, 1997): For the "domain identified" (those who are competent in and care about the subject matter), educators should: A. Affirm domain belongingness: (make it clear to the young woman that she indeed has what it takes to be a good mathematician.) B. Value multiple perspectives: publicly espouse the virtues of adopting multiple approaches to learning and scholarship in academic settings. This may reduce the perceived applicability of the negative stereotype in these situations. C. provide Role models: point out individuals who have been successful in the domain.

Recommendations for Minimizing Stereotype Threat. part 2

2. Affirming Global Sense of Self-Worth. "People are motivated to see themselves as adaptively and morally adequate, competent, good, coherent. I view these self-affirmation processes as being activated by information that threatens the perceived adequacy or integrity of the self and as running their course until this perception is restored." Claude Steele, 1988. Claude Steele's self-affirmation theory: folks are motivated to think well of themselves in an abstract and global sense. This generally positive sense of self serves to take the edge of various and sundry things that might otherwise impugn our self-concept.

Influential Approaches to Intelligence- Biological Approaches (which to some degree overlaps with psychometric and multiple-forms approaches). part 2

2. Cognitive Ability and Neural Eiciency. For relatively simple problems thought to involve fluid intelligence, higher IQ folks show lower levels of activity in the PFC than do lower IQ folks. That is, higher luid IQ folks seem to be able to solve problems using fewer neural resources, or more eficiently, than do their lower luid IQ counterparts. For these types of problems, they may also be able to recruit visual attention areas in the parietal lobes more reliably. However, this relationship is somewhat complex, as the difficulty of these types of problems increases, high-fluid-IQ folks show increased neural activation, while their low-IQ counterparts seem to "tune out" a bit, utilizing fewer resources.

h2 and and c2 Estimate

Across all studies, h2 estimates vary between about .4 and .8. And c2 is approximately .25 (or a bit less). It should be noted, however, h2 tends to increase and c2 tends to decrease with age. Why? As individuals get older, they become increasingly able to choose their own environments (presumably those that are compatible with their genes). And, of course, the effects of their original home environments begin to wane. An important caveat to these estimates for h2 and c2 is that the values that both assume depend to a large extent on the relative variability of genes and environments in the populations studied. a restricted range (or reduced variability) of either variable will tend to reduce its estimated coefficient. So, it is not particularly meaningful to talk of a single estimate of h2 or c2.

Class and Race Differences in Socialization for Intellectual Abilities

Adoption studies show that adoption into a high SES vs. a low SES environment adds about 12 IQ points to a kid. Why is this so? Hart and Risley (1995) found that by the age of 3, kids of professional parents are exposed to 50% more words than kids of working-class parents, and 200% (three times) more words than the kids of unemployed African-American moms. Moreover, kids of professional parents were the most likely to receive a high ratio of encouragements to reprimands, and kids of unemployed African-American moms were the most likely to receive a high ratio of reprimands to encouragements. Using a complex way of measuring environments, Hart and Risley have found that a 1 SD increase in environmental stimulation is associated with an increase of 9 IQ points. However, in these studies, it may have been that genes and environments were confounded, making it impossible to eliminate a genetic explanation for the observed differences (e.g., professional parents exposed their smarter kids to more words because both parents and kids had better genes).

Influential Approaches to Intelligence- Biological Approaches (which to some degree overlaps with psychometric and multiple-forms approaches).

Advances in functional imaging of the brain (e.g., MRI and PET) have made it possible to more precisely identify the neural structures and mechanisms involved in intelligent functioning. 1. Brain Structure and Crystallized and Fluid Intelligence. Scores on measures of luid intelligence tend to be markedly reduced when folks suffer some sort of damage to the prefrontal cortex Conversely, damage to the same PFC areas tends to spare crystallized intelligence Consistently, many neuro-imaging studies implicate PFC activity in the performance of tasks thought to involve fluid intelligence So, when you think fluid intelligence, think PFC There is strikingly little consistency in the research linking neural structures with individual differences in crystallized intelligence

Sternberg makes the following distinctions between analytical and practical problems.

Analytical Problems: Are formulated by others Are clearly defined Come with the information needed to solve them Have a single right answer Are dis embedded from everyday experience, and Have little intrinsic interest Practical Problems Must be recognized by the person Are poorly defined Require information seeking Have various acceptable answers Are embedded in everyday experience, and Require personal motivation and involvement

Responsibilities of Test Users and Constructors (cont)

And you can't claim ignorance in the event that something goes horribly wrong. E. Test constuctors should: Provide a test manual with sufficient data to judge appropriate use of test. Validity, reliability, scoring and administration standards, description of normative sample. Lots of tests get published that don't meet any of these standards.

African American Culture

Boykin (1994) hold that African American culture emphasizes a lively and expressive interchange of ideas and feelings, harmonious group relationships, and loosely defined time constraints. Traditional educational culture, on the other hand, emphasizes a highly structured and static work environment in which individual achievement and consumption of objective knowledge within rigidly defined time periods is the norm. Because these cultures are in many ways mutually incompatible, African American kids find it hard to fit in at school. This depresses both achievement and IQ. Chronic release of unusually large volumes of stress hormones has been found to be injurious to many parts of the brain critical for intelligent behavior African American kids may be disproportionately exposed to the types of stressors that release these hormones.

These estimates, however

Depend critically upon the particular populations studied! It may be that their values will be either higher or lower in different populations. For example, in different populations the effects of genes may be negligible and the effects of shared environments may be quite powerful. Perhaps the most robust effect of population on heritability estimates for IQ is that, in the US at least, for low SES children the effects of c2 are quite high and the effects of h2 are quite low. For high SES American children, on the other hand, the pattern is reversed, with shared environments accounting for very little variation in IQ, and shared genes accounting for the lion's share of it. One implication of the American findings, however, is that greater inclusion of poorer and less educated kids in relevant studies would increase c2 estimates for IQ.

Hypothesized Effect of Stereotype Threat?

Diminished task performance in the stereotyped domain. Why? Attentional, cognitive, and other resources need for successful task performance are diverted to intrusive thoughts relevant to stereotype confirmation (e.g., self-doubt, future disappointment, effects on group perceptions).

Labeling (con)

Moreover, as noted by the great personality researcher George Kelly, once applied, a label tends to place the onus on the person to prove that he or she is or is not characterized by the label. Even when the label is not applicable to the person. E.G., A boy may be diagnosed with/labeled by attention deficit hyperactivity disorder when he's really just distracted by the cute girl who sits in front of him

Caste-like minorities.

Ogbu (1994) designates as "caste-like" minorities those disadvantaged subgroups in a population whose status is involuntary. Those born in caste-like minorities do not generally expect that they will be able to assume positions of power, authority, and wealth within a society. Examples are the Maori of New Zealand, the "untouchables" of India, and the Burakumin of Japan. Ogbu contends that members of caste-like minorities tend not to believe that they can improve their situation through effort and commitment. That is, they lack "effort optimism." It may be that this lack of effort optimism directly and indirectly results in low IQ test scores.

Mechanisms of Stereotype Threat

Research has shown that the conditions that are conducive to stereotype threat impair performance by: (1) increasing anxiety . (2) increasing physiological arousal. (3) impairing working memory. (4) diverting cognitive resources to defensive attempts and counterproductive attempts at thought suppression. (5) increasing worries about attempts to avoid failure. among possibly other processes (this list may not be comprehensive).

Shared Environment Effects in Childhood and Adulthood

Shared environmental effects are those that distinguish different families from each other: they are shared among the same family members but not individuals from different families. Their values range between 0 and 1.0 (interpreted like r2). Although these effects are frequently erroneously reported to be zero, for both kids and adults, they range between about .07 and .26, with values close to .20 being the most common. iii. Within-Family Effects (non-shared environments): Within-family effects are those that result from different environments experienced by different family members There is precious little literature on within-family effects beyond the inconsistent literature on birth-order.

Moral Issues (stuff)

So, the ethical right not to be tested is of trivial practical significance. However, morally test takers should know: 1. Test scores, 2. Interpretations, And •3. Bases of any decision that affects their lives. Unfortunately, test publishers often copyright important info. about their tests Content of tests is reasonably protected to prevent test invalidation. However, no excuse not to publish data on psychometric adequacy and potentially biased nature of test.

Biological Variables

The second type of environmental influence that we will explore is biological rather than social. For example, individuals differ in the extent to which they are adequately nourished and are exposed to environmental toxins.

Stereotype Threat, Women, and Math: Spencer et al. (1999):

The subjects in this study were women and men who were good at math and saw it as relatively self-defining. All subjects attempted to solve very difficult problems taken from the quantitative GREs. For half of the subjects, stereotype threat was activated by claiming that these types of problems tended to be solved more successfully by men. The other half of the subjects were explicitly assured that these problems never showed gender differences. Results: When they were led to believe that these problems were better solved by males, the gals did worse than the guys. However, when they were led to expect that these problems showed no gender differences, the lass's performance did not suffer vis-à-vis the lads.

Stereotype Threat, African Americans, and Intellectual Performance. Steele and Aronson (1995):

The subjects were African American (AA) and European American (EA) Stanford kids. All of them attempted to answer difficult questions from the GRE verbal test. However, half of each group were led to believe that EA students typically outperformed AA students on this test. The other half were led to believe that there were no such race differences. Results: AAs performed worse than EAs only when led to expect race differences.

Labeling

Use of tests often involves application of a label. Labeling often involves stigma and may cause others to respond negatively. example, knowing that artist Vincent Van Gogh clearly suffered from severe psychopathology (he cut off his own ear, for example), it becomes difficult not to think of him, at least partially, in terms of this disorder. And we may forget that he was primarily a remarkably gifted artist. Labels often instill a passive attitude on the part of the person and others with regard to life circumstances, and make it easier for everyone to "give up" on the labeled person.

A threat in the air everywhere: EA men and stereotype threat in math. Aronson et al. (1999):

When they were led to expect that their math performances would be compared against Asian males, the average math performance of European American men went down fast So, anytime that it appears that one's performance might confirm a relatively unfavorable perception about one's group, stereotype threat effects may occur.

Interventions

a.Schooling. Kids who have been kept out of school have been found to lose several IQ points relative to comparable kids who were in school Possibly due to the relatively enriched intellectual environments found in high SES homes, absence from school (e.g., over the summer) especially hurts lower SES kids Although many prekindergarten educational programs appear to have only short-term effects on IQ, kids who receive further interventions show more long-term benefits And kids who receive the most intensive of these interventions show achievement and education (if not IQ) gains that last until adulthood.

Interventions (cont 3)

e. Cognitive Exercise. The research cited above on interventions to increase g(f) is relevant here. Also, research showing that if men continued to work into their 60s (working more was assumed to involve thinking more), their ability to retain memories for personally relevant events was better maintained. f. Intergeneration Increases in IQ (Flynn Effect) It has been found that since about 1940, the average IQ gain in the Western world is about 3 points per decade. Because genes simply do not change this quickly, these striking differences can only be described as environmental.

Sex Differences (cont.)

i. Spatial and Quantitative Abilities. On mental rotation and other similar tasks, the mean for males tended to be just a bit less than a standard deviation greater than the mean for females That is, at that time a male in about the 20th percentile for the male distribution would be at about the 50th percentile if his score were placed in the female distribution. Although these visuo-spatial differences are found as early as infancy, specific training on them reduces the male advantage markedly Due in part to the fact that females are increasingly encouraged to take math classes, the once-substantial-quantitative-ability sex difference favoring males has reduced significantly over the last few decades.

Sex Differences (cont.)

ii. Verbal Abilities. On verbal fluency tests and the like, the mean for females can often be approximately one standard deviation greater than the mean for males. That is, a female in about the 20th percentile for the female distribution would be at about the 50th percentile if her score were placed in the male distribution. Moreover, males are more likely to both be diagnosed with dyslexia and to stutter than are females. These verbal ability differences favoring females persist in more recent research.

Sex Differences (cont.) 2

iii. Causes. Traditionally, sex differences in the size of the corpus callosum to differences in the hemispheric lateralization of language have been postulated to account for sex differences in cognitive abilities. More recently, sex differences in cerebral white (males have more) and grey (females have more) matter have been looked at to explain these disparities Studies looking at the role of hormones in these cognitive differences have provided weak and inconsistent results Despite these small neuroanatomical differences, Haier et al. (2005) note that "Men and women achieve similar IQ results with different brain regions, suggesting that there is no singular underlying neuroanatomical structure to general intelligence and that different types of brain designs manifest equivalent intellectual performance" That is, men and women find different ways of using their brains to achieve, by and large, the same intellectual results

Ethnic Differences (cont.) 3

iv. African Americans Some three decades ago, the mean IQ score among African Americans was typically about one standard deviation below that for European Americans As early as 1991, there was some evidence that this difference was steadily declining. At about the same time, a similar reduction in the White/Black gap has was also seen in achievement-test scores Dickens and Flynn (2006) found that over the three decades prior to the publication of their paper, African Americans had gained just short of 6 IQ points on their European American counterparts. Although some scholars disagree with this conclusion, the preponderance of the evidence on Black-White differences in IQ points to an environmental explanation. For example, an adoption study by Moore (1986) found that half-European black kids had about the same IQs as did fully black kids, suggesting that there was no advantage to having "white" genes. Moreover, whether fully black or half-European, kids adopted by White families had almost a full standard deviation IQ advantage over the Black-family adopted kids. Clearly, there was something about the White-family environment that was conducive to measured intelligence. Finally meta-analyses of all relevant research suggest that stereotype threat reduces Black kids' measured verbal aptitude by about 1/5th of a standard deviation unit (maybe 3 verbal IQ points).

Stereotype threat is only thought to occur...

when the person to whom the threat is plausibly relevant cares about, or identifies with, the threatened domain. Therefore, stereotype threat is most likely to affect those individuals who possess a relatively high level of skill, competence, and confidence with respect to a given domain (e.g., good mathematicians who care about their grades in math).