What email address or phone number would you like to use to sign in to Docs.com?
If you already have an account that you use with Office or other Microsoft services, enter it here.
Or sign in with:
Signing in allows you to download and like content, and it provides the authors analytical data about your interactions with their content.
Embed code for: 1-s2.0-S0160289613001256-main
Select a size
Testing the hypothesized effect of dysgenic fertility on intelligence with existing reaction time data: A comment on Woodley, te Nijenhuis, and Murphy (2013) Irwin W. Silverman⁎ Department of Psychology, Bowling Green State University, Bowling Green, OH 43403, USA a r t i c l e i n f o a b s t r a c t Article history: Received 23 July 2013 Received in revised form 14 August 2013 Accepted 23 August 2013 Available online 26 September 2013 Dysgenic fertility has supposedly resulted in a decline in general intelligence (g) over time. In light of evidence that simple visual reaction time (RT) is inversely related to IQ, Woodley et al. (2013) tested the hypothesized dysgenic effect by subjecting to a meta-regression simple visual RT data collected over 100 years in 15 studies. This analysis found that RT had significantly increased according to a linear function over this time period. Woodley et al. then used this result to estimate the rate at which g had declined over the same period. The present comment points out that there are large gaps in the distribution of RTs analyzed by Woodley et al. with respect to year tested, and that RT in males did not vary as a function of year in the 13 studies published from 1941 on. It is concluded that although existing data are consistent with the idea that g has been adversely affected by dysgenic fertility, it cannot be determined at what rate g has fallen over time. © 2013 Elsevier Inc. All rights reserved. Keywords: Dysgenic fertility Reaction time General intelligence Contents References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 666 Lynn (2011) and others (e.g., Herrnstein & Murray, 1994) claim that as the birth rate fell in developed countries, fertility becamedysgenic,i.e.,peoplewithlowerIQstendedtohavemore childrenthanpeoplewithhigherIQs,and,asaresult,intelligence declined in these countries. Contrary to this claim, scores on IQ tests increased on average over the past century in a substantial number of developed countries (Flynn, 1984, 1987). However, it has beenargued thatthese increases owe not torealincreases in intelligence, but rather to people having become familiar with thecontentofIQteststhroughexposuretoricheropticaldisplays (Neisser, 1997) and to tasks that require manipulating abstract concepts (Flynn, 2007). In that conventional IQ tests are content bound, so to speak, much effort has been devoted to finding “purer” measures of intelligence, i.e., measures theorized to reflect the underlying neurophysiological processes that determine intelligence, or, more precisely, g, the hypothetical general factor in intelligence. Jensen (2006) has argued that a major determinant of g is speed of information processing, as shown in research which has found that individual differences in information processing, as measured in a wide variety of tasks, are correlated with IQ (see reviews by Jensen, 2006; Sheppard & Vernon, 2008). Of these information processing tasks, the one that has the longest history is the simple reaction time (RT) task. Thus, results with the simple RT task would seem to provide a ready means for testing the hypothesis that intelligence has Intelligence 41 (2013) 664–666 ⁎ Tel.: +1 561 752 9150. E-mail address: email@example.com. 0160-2896/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.intell.2013.08.008 Contents lists available at ScienceDirect Intelligence declined over time. That is, if simple RT and g are determined by the same process and if g has declined over time, then we should observe that simple RT has increased over time. SimpleRThasbeenstudiedacrossmuchoftheagespectrum, but the most extensive amount of research on this task has been with young adults. Several years ago, I reviewed the research findings for simple visual RT in young adults living in Western countries (Silverman, 2010). This review revealed that the simple RTs reported in 13 studies published from 1941 on were, with one exception, slower than the RTs found in a study conducted in the late 1800s by the eminent scientist Francis Galton. Although I did not relate these results to the dysgenic effect under discussion, I did suggest that the increase in RT over this period of time could be due to a decline in infant and child death rates, which could have resulted in more people reaching adulthood in less than robust health. I buttressed this argument by citing research showing that RT is inversely related to health (Emery, Huppert, & Schein, 1995) and to life expectancy (Deary & Der, 2005; Shipley, Der, Taylor, & Deary, 2006). This expla- nation for the increase in RT is compatible with the idea that intelligence has declined across generations given the validity of three premises: that RT is inversely related to IQ, that RT is inversely related to health; and that more people in poor health werewinnowedfrom thepopulationinthe19thcenturythanin the 20th century. The question naturally arises whether it is possible to provide an estimate as to the rate at which RT has increased over time. In my article, I argued against making such an estimate based on the fact that the studies of RT that I reviewed were heterogeneous with respect to several variables that might affect RT, including composition of the samples, testing procedure, and handling of data. I further stated that “In principle it would be possible to uncover the rate at which RT increased [over time] by controlling for potentially confounding variables in a multiple regression analysis. However, this requires that each of these variables be represented by multiple data points, but this requirement cannot be met by the present dataset. Accurately describing change over time also requires that both ends of the temporal dimension be well represented in the dataset and that the dataset be free of outliers (Cohen, Cohen, West, & Aiken, 2003); neither of theserequirements can be met in the present dataset” (Silverman, 2010, pp. 44–45). DespitemyadmonitionnottousethedatathatIsummarized to estimate the rate of change in RT over time, this is precisely what Woodley, te Nijenhuis, and Murphy (2013) did in a study recently reported in this journal. Having embraced the dysgenic fertility hypothesis, their purpose was to use the findings for the rateatwhichRThadincreasedovertimetodrawaninferenceas to the rate at which g had declined over time. Their study involvedtwoparts.Inpartone,theyconductedarandom-effects meta-regression analysis of the RTs that I summarized, plus the RTs obtained in another early study (Thompson, 1903). The meta-regression showed that RT had increased linearly over time, with the estimated rate of increase being .71 ms per year from 1884 on. In part two, with the preceding result in hand, and with Deary, Der, and Ford (2001) having found in a large Scottish sample that simple visual RT was inversely related to IQ, r = −.31, they estimated the rate at which g declined per decade over the same period. (The estimate was madeafterbothRTandIQwerecorrectedformeasurementerror and restriction of range.) The figure arrived at was 1.16 points per decade. The purpose of this comment is to point out two reasons for discounting the results reported by Woodley et al. (2013). The first is that Woodley et al. overlooked large gaps in the RT dataset that they analyzed. To be specific, the dataset consists of two RTs for each sex obtained in studies conducted in the late 1800s; 14 RTs for males obtained in 13 studies published from 1941 to 2004; and 7 RTs for females obtained in 7 studies published from 1970 to 2004. Thus, in this dataset, there is a gap spanning about four decades for males and a gap spanning about seven decades for females. (For females, this constitutes about 60 percent of the entire span of years covered by the dataset.) With such large gaps in the temporal distribution of RT, I submit that one should be extremely cautious about drawing conclusions about the shape of the overall function that relates RT to year tested. Thus, instead of RThavingchangedaccordingtoalinearfunctionofyeartested,it is possible that RT changed according to a curvilinear function of year tested. Indeed, as will be seen, the possibility exists that a stepwise function describes the relation between RT and year tested. A possible counterargument to my objection is that even when there is a large gap in a temporal distribution, it is not unreasonable to make an inference about the shape of the overall temporal distribution, given that the results show a definite trend before or after the gap. In view of this counterargument, I decided to determine whether RT in males increased across year tested according to a linear function for the 13 studies published from 1941 on. (I did not perform a parallel analysis for females because, as noted above, there were only seven RTs for females in the studies published from 1970 on.) To make this determination, I performed a random-effects meta-regression on the RTs listed in Table 1 of Woodley et al. (2013) with the exception of the RTs reported by Seashore, Starmann, Kendall, and Helmick (1941). To explain, Seashore et al. compared performance in the same subjects under two conditions: when the visual stimulus was signaled and when it was not. As inclusion of the RTs for both conditions would have violated the assumption in regression analysis that errors are uncorrelated, I averaged the means and SDs across condi- tions. (The difference between conditions was in fact not significant, t(46) = .45.) Thus, the present meta-regression was performed on a total of 13 RTs, with each study mean being weighed by the standard error of the mean (SEM), the latter being calculated by dividing the standard deviation for each study by the sample size, i.e., SEM = SD/ ﬃﬃﬃﬃ Np . The meta-regression itself was conducted using software avail- able at StatsToDo.com. The meta-regression yielded a beta coefficient for year tested of .46 with the 95% confidence interval extending from −.57 to 1.49. As this confidence interval includes zero, this indicates that year tested was not significantly related to RT inmalesinstudiespublishedfrom1941on.Hence,asmentioned above, RT might have changed over time according to a linear or a curvilinear function, or even according to a stepwise function. Toconclude,Ibelievethattheonlythingcanbesaidwithany degree of confidence about changes in RT over time is what I previously concluded, viz., that the RTs obtained in modern studiesofRTareslowerthanthosethatwereobtainedbyGalton 665I.W. Silverman / Intelligence 41 (2013) 664 –666 inthelate1800s.(ThisconclusionalsoholdsfortheRTsobtained in the late 1800s by Thompson. Although Thompson obtained RTs that were somewhat slower than those obtained by Galton, Thompson's RTs are faster than those obtained in the great bulk of the studies published from 1941 on.) One caveat to the forgoing conclusion is that it applies primarily to males. This limitation stems from my decision not to use the female RTs to make comparisons over time. In contrast, Woodley et al. (2013) made full use of the female RTs.This,Isubmit,wasamistake.Toexplain,Woodleyetal., in comparing RTs over time, created a single RT for each of the samples in their dataset referring to each of them as an “effect size.” When both sexes were represented in the sample, the effect size was the weighted average of the mean RTs for the two sexes. But when only males were represented in the sample, the effect size was taken to be the male mean RT. This method for dealing with studies withoutfemaleswouldbeacceptableifthemeanRTsforthe two sexes were equal. However, as seen in Table 1 of Woodley et al., male RTs were faster than female RTs, the difference ranging from 3.6 to 30.0 ms, with an unweighted mean difference of 16.11. Thus, readers should be aware of the possible distortions in the results reported by Woodley et al. due to the indiscriminate mixing of male and female data. We are left with a state of affairs that is likely not very satisfying to proponents of the dysgenic fertility hypothesis. No doubt they would dearly love to be able to describe the shape of the function that relates RT to year tested so as to use that function to infer the rate at which g has declined over time. But that is not possible to do with existing data. Of course, there may be undiscovered RT studies that could change the conclusions that can be drawn about the secular trend in RT. However, even if such studies can be found, the lack of standardization in the way in which RT has been measured and the way in which the resulting data were handled will be a great impediment to being able to describe the shape of the function that relates RT to year tested, and thus being able to describe how g has changed over time. References Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple regression/correlation analysis for the behavioral; sciences (3rd ed.) Mahwah, NJ: Erlbaum. Deary, I. J., & Der, G. (2005). Reaction time explains IQ’s association with death Psychological Science, 16, 64 –69. Deary, I. J., Der, G., & Ford, G. (2001). Reaction time and intelligence differences. Psychological Science, 29, 389–399. Emery, C. F., Huppert, F. A., & Schein, R. I. (1995). Relationships among age, exercise, health, and cognitive function in a British sample. Gerontologist, 35, 378–385. Flynn, J. R. (1984). The mean IQ of Americans: Massive gains 1932 to 1978. Psychological Bulletin, 95, 29 –51. Flynn, J. R. (1987). Massive IQ gains in 14 nations: What IQ tests really measure. Psychological Bulletin, 101, 171–191. Flynn, J. R. (2007). What is intelligence?: Beyond the Flynn effect, 2009. Cambridge, UK: Cambridge University Press. Herrnstein, R., & Murray, C. (1994). The bell curve: Intelligence and class structure in American life. New York: The Free Press. Jensen, A. R. (2006). Clocking the mind: Mental chronometry and individual differences. Oxford, UK: Elsevier. Lynn, R. (2011). Dysgenics: Genetic determination in modern populations (revised ed.)London, UK: Ulster Institute for Social Research. Neisser, U. (1997). Rising scores on intelligence tests. American Scientist, 85, 440–447. Seashore, R. H., Starmann, R., Kendall, W. E., & Helmick, J. S. (1941). Group factors in simple and discrimination reaction times. Journal of Experimental Psychology, 29, 346–394. Sheppard, L. D., & Vernon, P. A. (2008). Intelligence and speed of information-processing: A review of 50 years of research. Personality and Individual Differences, 44, 535–551. Shipley, B. A., Der, G., Taylor, M. D., & Deary, I. J. (2006). Cognition and all-cause mortality across the entire adult age range: Health and lifestyle survey. Psychosomatic Medicine, 68, 17 –24. Silverman, I. W. (2010). Simple reaction time: It is not what it used to be. The American Journal of Psychology, 129, 39 –50. Thompson, H. B. (1903). The mental traits of sex: An experimental investigation of the normal mind in men and women. Chicago, IL: The University of Chicago Press. Woodley,M.A.,teNijenhuis,J.,&Murphy,R.(2013).WeretheVictorianscleverer than us? The decline in general intelligence estimated from a meta-analysis of the slowing of simple reaction time. Intelligence. http://dx.doi.org/ 10.1016/j.intee.2013.04.006. 666 I.W. Silverman / Intelligence 41 (2013) 664–666 referring to each of them as an “effect size.” When both sexes were represented in the sample, the effect size was the weighted average of the mean RTs for the two sexes. But when only males were represented in the sample, the effect size was taken to be the male mean RT. This method for dealing with studies withoutfemaleswouldbeacceptableifthemeanRTsforthe two sexes were equal. However, as seen in Table 1 of Woodley et al., male RTs were faster than female RTs, the difference ranging from 3.6 to 30.0 ms, with an unweighted mean difference of 16.11. Thus, readers should be aware of the possible distortions in the results reported by Woodley et al. due to the indiscriminate mixing of male and female data. We are left with a state of affairs that is likely not very satisfying to proponents of the dysgenic fertility hypothesis. No doubt they would dearly love to be able to describe the shape of the function that relates RT to year tested so as