We are searching data for your request:
Upon completion, a link will appear to access the found materials.
The Dual N-Back task is the only task I'm aware of that has empirical support showing that it improves working memory. It appears to improve working memory through multitasking, although this multitasking must fit certain criteria, namely that you can't learn the task so well that you can eventually "automate" it without going into working memory.
Thus, I'm wondering whether other forms of multi-tasking such as playing Starcraft II (see my earlier thoughts) and driving with a cellphone can also improve working memory?
A possible follow-up question: if not, what differentiates the type of multitasking found in the Dual N-Back task from other types of multi-tasking?
In general, I'd hypothesise that "memory-training" programs will not lead to domain-general increases in fluid intelligence nor working memory.
As general background, you might want to check out the literature on expert memory.
- Practice is very effective at improving performance on the practised task. Transfer is real and does exist, but it is often small in effect.
- Let's take the study by Ericsson, Chase, and Faloon (1980). After 230 hours of practice a participant, SF, was trained to increase their digit span (i.e., the sequence of random numbers they could recall) from 7 to 79 numbers. One might think that SF had increased his working memory. However, talk aloud protocols suggest that SF was using sophisticated domain specific mnemonic strategies (e.g., linking running times to random numbers). The skill did not generalise to other memory stimuli.
- The expertise literature is filled with examples of experts seemingly defying the limits of human ability. Yet, such achievements tend to be domain specific and achieved through substantial practice. For an excellent review of the expertise literature, see Ericsson, Krampe, and Tesch-Romer (1993).
- It is very difficult to modify very large domain general abilities.
Thus, my general advice is for people wanting to improve their working memory or fluid intelligence is to instead focus on what domain specific ability they want to improve and focus on practicing that.
- Ericsson, K. A., Chase, W. G., and Faloon, S. (1980). Acquisition of a memory skill. Science, 208(4448):1181-1182. FREE PDF
- Ericsson, K. A., Krampe, T., and Tesch-Romer, C. (1993). The role of deliberate practice in the acquisition of expert performance. Psychological Review, 100:363-406.
In my mind the dual-n-back (aka Jaeggi training) is still controversial. A meta-analysis by Jaeggi and co-authors (2015) found in favor of their original paper, but others have gone on record to disagree with their conclusions, Melby-Lervåg & Hulme (2016) in particular who had conducted a meta-analysis concluding the opposite in 2013; Jaeggi and co-authors disagree with that criticism, of course. Dougherty et al. (2016) were also unconvinced of the meta-analysis of the Jaeggi group.
There's a 2016 editorial in Nature on the state of affairs on this line of research, with the subheading:
Conflicting results are expected in a young field, but what do you do when even the meta-analyses do not agree?
An empirical study by Lawlor-Savage and Goghari (2016) published after those meta-analyses and which (according to its introduction) took methodological concerns outlined in the meta-analysis into consideration for its own experimental design, also failed to find improvements from dual-n-back. I'm not saying this is the last word on the issue, just that replication has proven difficult.
So I think it's somewhat premature to ask what else works like this, when we're not sure this works.
I think the "automate" problem is caused by several reasons in my experience.
1: brainworkshop's design on fixed time intervals between trials is bad. Some of the time, we increase our n-back level without some seconds more to recall the letter or position, which ruins the rest of the trials. And so, we chose to remember letters and positions and finally "automate" this bad practice.
2: It is bad to have too many letters in brainworkshop. If you can, imagine how we tried to remember a long word while we were still children. So we suck the long chain of letters into our minds instead of recalling what we saw.
3: Positions are not the best way to practice n-back, just like in the last reason. The human brain can easily recognize this kind of 3x3 spatial pattern. While we are training, we can't avoid learning the 3x3 patterns eventually. How hard to stick a 3 position set and a 3 letter set in mind, right? But while we are sticking those, we have less parts of our minds available to recall memories.
A New Integrated Model for Multitasking during Web Searching ☆
There have been many studies in the field of information behavior. Recently Du and Spink (2011) presented a model, which simulates multitasking, cognitive coordination and cognitive shifts on web. However, this model does not incorporate personal variables and the impact of task or web design. This research addresses this gap. Information and psychological scientists have shown that information behavior (IR) is affected by: the affective domain, cognitive attributes, psychological factors, personality dimensions and sociological factors. This study investigates how IR is affected by working memory (wm), cognitive coordination, cognitive shifts and various artifacts and task variables influenced by the PAT model (Personal, Artifact and Task characteristics) of flow. The research is exploratory and takes a pragmatic, mixed method approach. Thirty University students will participate. The research tools include: pre and post questionnaires, working memory tests, Flow State Scale test, think aloud data, observations, audio-visual data, web search logs and use of the Critical Decision Method. The qualitative data will be coded and analyzed thematically and will be related to the quantitative data. This study is expected to identify the impact of all these variables on multitasking IR in the web and provide a new integrated framework, which is not only going to help information scientists to better understand this behavior but also web companies to develop more effective web products.
Multiple processing limitations underlie multitasking costs
Human multitasking is typically defined as the practice of performing more than one task at the same time (dual task) or rapidly alternating between multiple tasks (task switching). The majority of research in multitasking has been focusing on individual paradigms, with surprisingly little effort in understanding their relationships. We adopted an individual-difference approach to reveal the limitations underlying multitasking costs measured in different paradigms. Exploratory factor analyses revealed not a general multitasking factor but instead three different processing limitations associated with response selection, retrieval and maintenance of task information, and task-set reconfiguration. The three factors were only weakly correlated with and thus not reducible to common measures of processing speed, working memory capacity and fluid intelligence. Males and females excelled in different aspects of multitasking, demonstrating the benefit of using a multifaceted view of multitasking competency in group comparison. Findings of the current study help resolve conflicting results between studies using different paradigms, and form the basis of more comprehensive measurement tools and training protocols covering different aspects of multitasking limitations. The study will also help future integration of multitasking abilities into the theoretical framework of executive function.
This is a preview of subscription content, access via your institution.
'Brain training' may boost working memory, but not intelligence
Brain training games, apps, and websites are popular and it's not hard to see why—who wouldn't want to give their mental abilities a boost? New research suggests that brain training programs might strengthen your ability to hold information in mind, but they won't bring any benefits to the kind of intelligence that helps you reason and solve problems.
The findings are published in Psychological Science, a journal of the Association for Psychological Science.
"It is hard to spend any time on the web and not see an ad for a website that promises to train your brain, fix your attention, and increase your IQ," says psychological scientist and lead researcher Randall Engle of Georgia Institute of Technology. "These claims are particularly attractive to parents of children who are struggling in school."
According to Engle, the claims are based on evidence that shows a strong correlation between working memory capacity (WMC) and general fluid intelligence. Working memory capacity refers to our ability to keep information either in mind or quickly retrievable, particularly in the presence of distraction. General fluid intelligence is the ability to infer relationships, do complex reasoning, and solve novel problems.
The correlation between WMC and fluid intelligence has led some to surmise that increasing WMC should lead to an increase in both fluid intelligence, but "this assumes that the two constructs are the same thing, or that WMC is the basis for fluid intelligence," Engle notes.
To better understand the relationship between these two aspects of cognition, Engle and colleagues had 55 undergraduate students complete 20 days of training on certain cognitive tasks. The students were paid extra for improving their performance each day to ensure that they were engaged in the training. Students in the two experimental conditions trained on either complex span tasks, which have been consistently shown to be good measures of WMC, or simple span tasks. With the simple span tasks, the students were asked to recall items in the order they were presented for complex span tasks, the students had to remember items while performing another task in between item presentations. A control group trained on a visual search task that, like the other tasks, became progressively harder each day.
The researchers administered a battery of tests before and after training to gauge improvement and transfer of learning, including a variety of WMC measures and three measures of fluid intelligence.
The results were clear: Only students who trained on complex span tasks showed transfer to other WMC tasks. None of the groups showed any training benefit on measures of fluid intelligence.
"For over 100 years, psychologists have argued that general memory ability cannot be improved, that there is little or no generalization of 'trained' tasks to 'untrained' tasks," says Tyler Harrison, graduate student and lead author of the paper. "So we were surprised to see evidence that new and untrained measures of working memory capacity may be improved with training on complex span tasks."
The results suggest that the students improved in their ability to update and maintain information on multiple tasks as they switched between them, which could have important implications for real-world multitasking:
"This work affects nearly everyone living in the complex modern world," says Harrison, "but it particularly affects individuals that find themselves trying to do multiple tasks or rapidly switching between complex tasks, such as driving and talking on a cell phone, alternating between conversations with two different people, or cooking dinner and dealing with a crying child."
Despite the potential boost for multitasking, the benefits of training didn't transfer to fluid intelligence. Engle points out that just because WMC and fluid intelligence are highly correlated doesn't mean that they are the same:
"Height and weight in human beings are also strongly correlated but few reasonable people would assume that height and weight are the same variable," explains Engle. "If they were, gaining weight would make you taller and losing weight would make you shorter—those of us who gain and lose weight periodically can attest to the fact that that is not true."
The researchers plan to continue this research to better understand how training specific aspects of cognition can lead to positive transfer to other tasks, both in the lab and in the real world.
Can intense multitasking improve fluid intelligence/working memory? - Psychology
Working memory predicts multitasking over-and-above other cognitive, personality, and experience-based variables.
However, many previous studies selected tasks to measure working memory that were dual-tasks themselves.
The current work indicates single- and dual-task memory span tasks equally predict individual differences in synthetic work.
Recent research has identified working memory as a critical component of multitasking ability. These studies showed that working memory accounted for multitasking variance over-and-above that predicted by other cognitive, personality, and experience-based variables. However, a limitation of these previous studies was that the tasks selected to measure working memory were dual-tasks themselves. The purpose of the current research was to determine if working memory measures must be dual-tasks to predict multitasking performance, or if other types of working memory measures that do not rely upon the dual-task methodology predict multitasking just as well, if not better. Three different serial order memory span tasks (one dual-task and two single-task) and one multitask were administered to a sample of healthy young adults. The results showed that single- and dual-task working memory measures predicted multitasking to a similar degree. The results indicate there is something fundamental about working memory's relationship with multitasking ability.
Working memory training: Improving intelligence – Changing brain activity
The main objectives of the study were: to investigate whether training on working memory (WM) could improve fluid intelligence, and to investigate the effects WM training had on neuroelectric (electroencephalography – EEG) and hemodynamic (near-infrared spectroscopy – NIRS) patterns of brain activity. In a parallel group experimental design, respondents of the working memory group after 30 h of training significantly increased performance on all tests of fluid intelligence. By contrast, respondents of the active control group (participating in a 30-h communication training course) showed no improvements in performance. The influence of WM training on patterns of neuroelectric brain activity was most pronounced in the theta and alpha bands. Theta and lower-1 alpha band synchronization was accompanied by increased lower-2 and upper alpha desynchronization. The hemodynamic patterns of brain activity after the training changed from higher right hemispheric activation to a balanced activity of both frontal areas. The neuroelectric as well as hemodynamic patterns of brain activity suggest that the training influenced WM maintenance functions as well as processes directed by the central executive. The changes in upper alpha band desynchronization could further indicate that processes related to long term memory were also influenced.
► We investigated the influence of working memory training on intelligence and brain activity. ► Working memory training increased individuals performance on tests of intelligence. ► No gains on test performance were observed in individuals of an active control group. ► Working memory training influenced neuroelectric and hemodynamic patterns of brain activity.
Measuring Human Capabilities: An Agenda for Basic Research on the Assessment of Individual and Group Performance Potential for Military Accession (2015)
Committee Conclusion: The constructs of fluid intelligence (novel reasoning), working memory capacity, executive attention, and inhibitory control are important to a wide range of situations relevant to the military, from initial selection, selection for a particular job, and training regimes to issues having to do with emotional, behavioral, and impulse control in individuals after accession. These constructs reflect a range of cognitive, personality, and physiological dimensions that are largely unused in current assessment regimes. The committee concludes that these topics merit inclusion in a program of basic research with the long-term goal of improving the Army&rsquos enlisted accession system.
The committee considers the areas of fluid intelligence, working memory capacity, executive attention, and inhibitory control as offering new constructs for the Army&rsquos consideration, even though some aspects of these ideas have been studied for decades. The newer research brings these several heretofore separate topics together and extends the relevance of the constructs beyond performance on specific tasks to broader issues of cognitive and emotional control. These topics are presented in a single chapter because there is considerable evidence that they overlap in terms of their theoretical motivations and definitions, their measurement, their variance, and their patterns of prediction. These topics are also brought together because, at the same time these overlaps are evident, future research must determine whether these various constructs reflect a single common mechanism or highly related but separate psychological mechanisms that might play different roles in the regulation of behavior, thought, and emotion.
If the latter hypothesis is supported, then a second issue is whether much more specific assessment of those separate mechanisms can add predictive validity for performance in the jobs for which potential military recruits are assessed.
Each section of the chapter begins with a brief history about one or more of the constructs listed in the title, focusing on how research on these constructs has converged and diverged over time. It then presents findings from various researchers who have studied these issues most recently, describes the evidence for the validity of the constructs in predicting performance of real-world tasks, and discusses the transition of what has been fairly basic research agenda on these topics to a more testing-oriented agenda. The sections end with a discussion of questions that should be addressed in future projects.
The idea that intelligence could be thought of as a general and therefore domain-free variable dates back at least to Spearman (1904). However, the idea that fluid and crystallized intelligence were separable was proposed by Spearman&rsquos student Raymond Cattell (1941) and elaborated by Cattell and his student John Horn (Horn and Cattell, 1966a, 1966b). As described in Cattell&rsquos biography by the website Human Intelligence: 1
Fluid abilities (Gf) drive the individual&rsquos ability to think and act quickly, solve novel problems, and encode short-term memories. They have been described as the source of intelligence that an individual uses when he or she doesn&rsquot already know what to do. Fluid intelligence is grounded in physiological efficiency, and is thus relatively independent of education and acculturation (Horn, 1967). The other factor, encompassing crystallized abilities (Gc), stems from learning and acculturation, and is reflected in tests of knowledge, general information, use of language (vocabulary) and a wide variety of acquired skills (Horn and Cattell, 1967). Personality factors, motivation and educational and cultural opportunity are central to its development, and it is only indirectly dependent on the physiological influences that mainly affect fluid abilities.
Fluid intelligence (Gf) is important for reasoning and novel problem solving, and there is strong and emerging evidence that it represents the heritable and biological aspect of intelligence (Plomin et al., 2008 Wright et al., 2007). Longitudinal and cross-sectional studies across the life span have repeatedly shown that, while crystallized intelligence&mdashthe culturally derived knowledge aspect of intelligence&mdashremains high and even increases
over the life span, Gf declines over age (Horn and Cattell, 1967). In addition, individual differences in fluid intelligence (i.e., rank-order differences) appear to be quite stable over the life span (Deary et al., 2009, 2012). For example, Deary and his colleagues in the Lothian cohort studies made use of the fact that over 150,000 11-year-olds in the Lothian region of Scotland were tested for intelligence (IQ scores) more than 50 years ago and many of those individuals have been available for testing in recent years. Recently, Deary and colleagues (2012) conducted a genome-wide complex trait analysis on this sample and found a genetic correlation of 0.62 between intelligence in childhood and in old age. Furthermore, it appears that this relationship is higher for the lower quartile of abilities than for the upper quartile, which suggests that a more complete understanding of this relationship would be important for the selection and assignment of enlisted personnel.
The validity of fluid measures has been demonstrated for military-related tasks such as air traffic control (Ackerman and Cianciolo, 2002) and multitasking (Hambrick et al., 2010, 2011). The long-term stability and validity of fluid measures have been demonstrated in a sustained program of studies by David Lubinski and Camilla Benbow (2000, 2006). They started with a sample of 13-year-olds identified as being in the top 1 percent of individuals on measures of verbal and mathematical reasoning and tracked those individuals into middle adulthood (Lubinski and Benbow, 2006). Scores on these measures substantially predicted accomplishments in a wide array of domains in middle adulthood. Even at the highest levels, the scores obtained at age 13 predicted the number of patents, academic publications, and achievement in science and business at later ages.
The distinction between fluid and crystallized abilities becomes critically important in selection for the military. Recent papers have suggested that the Armed Services Vocational Aptitude Battery (ASVAB) is largely crystallized and that incremental validity can be added with measures of working memory capacity and fluid intelligence. The ASVAB does include a spatial ability subtest (Assembling Objects) which reflects a fluid ability in the typical examinee population (see Chapter 4, Spatial Abilities, for further discussion). Roberts and colleagues (2000) reported two studies, with a total of 7,100 subjects, showing that the ASVAB largely reflects acculturated learning and minimally reflects fluid abilities (Gf). Hambrick and colleagues (2011) had Navy sailors perform a synthetic work task that simulated the multitasking demands of many different jobs. While the ASVAB did predict performance on this task, the ability to update working memory accounted for even more variance in the prediction of multitasking and synthetic work. Future research will be important to improve understanding of the mechanisms underlying fluid abilities and the differences between the
mechanisms of working memory and fluid intelligence, including measures of these constructs as potential supplemental tests to the ASVAB.
There is ongoing military interest in and research on measures of fluid abilities. An expert panel charged with a review of the ASVAB recommended consideration of existing and new measures of fluid abilities as potential additions to the ASVAB (Drasgow et al., 2006). Alderton and colleagues (1997) examined a battery of tests in the spatial ability and working memory domains, administered in conjunction with the ASVAB. Their data show that Assembling Objects has a substantial loading on a general factor, as well as loading on a specific spatial ability factor. Thus, although it does indeed reflect a measure in the fluid abilities domain, it is likely not the best measure of fluid intelligence. Nonverbal reasoning tests, such as matrix tests, commonly produce very high general factor loadings, and a matrix test will be administered to all military applicants starting in April 2015 (see Russell et al., 2014).
The psychological and biological mechanisms reflected in standard tests of fluid intelligence and responsible for individual differences in the construct have been largely ignored in the psychometric literature and only recently have been addressed in the cognitive psychology and neuroscience literature. This lack of understanding of the specific cognitive abilities and the underlying biomarkers reflected in fluid intelligence is a gap in knowledge that it is important to fill to maximize the benefits of such assessments. If, for example, fluid intelligence is a composite of several underlying specific cognitive abilities it would be extremely useful to know whether those abilities are differentially related to various criterion measures and whether they might interact in some way that would be important to assess.
Measures of memory span (short-term memory) have been used to study memory abilities since Ebbinghaus (see Dempster, 1981). The first publication of a study using memory span as a measure (Jacobs, 1887) reported a strong relationship between a child&rsquos memory span and rank in class, and Francis Galton himself (1887) observed that few mentally deficient individuals could recall more than two items in a span test. Simple memory span tasks have been included in most large-scale tests of intelligence. Thus, from the beginning, what came to be called short-term memory appeared to reflect important individual differences in higher-order cognitive functions. The emergence of short-term memory as a major construct in cognitive psychology was predicated largely on research using span-like tasks, meaning that most of the work was done using serial recall of short lists of digits, letters, or words and with the same pool of items used over and over across lists. Crowder (1982), in a paper titled &ldquoThe De-
mise of Short-term Memory,&rdquo argued against two separate memory stores, and one of his arguments was based on the lack of relationship between measures of short-term memory and measures of real-world cognition. If short-term memory was important to real-world cognition, then individual differences in measures of that memory should correspond to individual differences in reading, learning, decision making, etc., and there was little evidence supporting that conclusion.
The picture clarified substantially when complex span measures were shown to have quite substantial correlations with reading and listening comprehension (Daneman and Carpenter, 1980 Engle and Kane, 2004). Examples of two complex spans alongside a simple letter span task, all of which require manipulation and remembering of verbal materials, are shown in Figure 2-1. In the reading span task, the subject is to read aloud the sentence and decide whether the sentence makes sense. That is followed by a letter to recall. In the operation span task, the subject is to calculate whether the equation is correct and then see a letter to recall. After two to seven such items, the subject is shown a set of question marks and asked to recall the to-be-remembered items.
Complex tasks may also involve the manipulation and remembering of nonverbal information such as the tasks in Figure 2-2. These tasks require the subject to make a decision about a pattern such as whether the rotated
FIGURE 2-1 Example of a simple span task, a reading span task, and an operation span task.
NOTE: WMC = working memory capacity.
SOURCE: Engle, Randall W. (2010). Role of working memory capacity in cognitive control. Current Anthropology, 51(S1):S17&ndashS26. Reproduced by permission of and published by The University of Chicago Press.
FIGURE 2-2 Three different spatial tasks.
NOTE: WMC = working memory capacity
SOURCE: Kane et al. (2004, p. 196).
block letter would be a correct letter when upright or whether the figure is symmetrical around a vertical axis. Each decision is followed by an item to be remembered such as the arrow pointing in one of eight directions and being one of two lengths, or a cell in a matrix.
One might think that tasks that differ as widely as those in Figures 2-1 and 2-2 would yield very different predictive validity for higher level tasks, but that is not the case. As shown in Figure 2-3, a huge array of such tasks has been shown to reflect a coherent latent factor. Further, that latent factor, typically called &ldquoworking memory capacity&rdquo (WMC), has a very high relationship to the construct for fluid intelligence.
The wide array of WMC tasks have been shown to be quite valid in predicting performance on a huge variety of real-world cognitive tasks. Quoting from Engle and Kane (2004, p. 153):
Scores on WMC tasks have been shown to predict a wide range of higher-order cognitive functions, including: reading and listening comprehension (Daneman and Carpenter, 1983), language comprehension (King and Just, 1991), following directions (Engle et al., 1991), vocabulary learning (Daneman and Green, 1986), note-taking (Kiewra and Benton, 1988), writing (Benton et al., 1984), reasoning (Barrouillet, 1996 Kyllonen and
Christal, 1990), bridge-playing (Clarkson-Smith and Hartley, 1990), and computer-language learning (Kyllonen and Stephens, 1990 Shute, 1991). Recent studies have begun to demonstrate the importance of WMC in the domains of social/emotional psychology and in psychopathology, either through individual-differences studies or studies using a working memory load during the performance of a task (Feldman-Barrett et al., in press ). For example, low WMC individuals are less good at suppressing counterfactual thoughts, that is, those thoughts irrelevant to, or counter to, reality.
FIGURE 2-3 Path model for structural equation analysis of the relation between working memory capacity and reasoning factors.
SOURCE: Kane et al. (2004, p. 205).
The expert panel charged with a review of the ASVAB, described in the previous discussion of fluid abilities, also recommended consideration of working memory measures as potential additions to the ASVAB (Drasgow et al., 2006). Previously, Alderton and colleagues (1997) examined a battery of tests that included working memory measures, administered in conjunction with the ASVAB. Sager and colleagues (1997) offered evidence of the validity of working memory measures in this battery for predicting military training outcomes. Furthermore, a working memory test from this battery is currently being administered to Navy applicants (see Russell et al., 2014). Working memory measures were also explored in Project A, the Army&rsquos large-scale exploration of the relationship between a broad array of individual-differences constructs and various performance domains (Russell and Peterson, 2001 Russell et al., 2001).
Although the construct under discussion here is typically referred to as working memory capacity, there is strong and emerging evidence that the critical factor for regulation of thought and emotion is the ability to control one&rsquos attention, often referred to as executive attention (EA). EA refers to the ability to prevent attention capture by both endogenous and exogenous events (Engle and Kane, 2004). Individuals assessed to have lower EA are thought to be more likely to allow internally or externally generated events to capture their attention from tasks currently being performed. Thus, studies will often use the same tasks developed to measure WMC but will refer to the construct as Executive Attention.
There is a strong connection between the measures of WMC described above and measures of attention such as the Stroop task, antisaccade task, dichotic listening, and the flanker task. In an example of the antisaccade task, subjects stare at a fixation point on a computer screen while there are two boxes 11 degrees to each side of the fixation. At some point, one of the boxes will flicker and the subject is to look at the box on the opposite side of the screen. The flickering box affords movement, and evolution has predisposed us to look at that box since things that move have possible survival consequences. Performance can be measured either by eye movement analysis or by having the subject identify a briefly presented item in the box opposite to the flickering box (Kane et al., 2001 Unsworth et al., 2004) in both cases low WMC individuals are nearly twice as likely to make an error and glance at the flickering box. In the dichotic listening task, low WMC individuals are more than three times more likely than high WMC individuals to hear their name in the to-be-ignored ear.
The strong relationship of performance on these low-level attention tasks to the WMC tasks suggests that EA is likely to play a crucial role in both types of tasks. We do note that although EA is conceptualized as a cognitive ability, the pattern of relationships among various WMC tasks may also result from differences across participants in the degree of en-
gagement with the tasks. Attributing relationships to EA differences alone requires the assumption of a common level of task motivation (usually a high level is assumed).
The concept of individual differences in WMC/EA has been used in explanations of psychopathologies such as alcoholism and schizophrenia. For example, Finn (2002) proposed a cognitive-motivational theory of vulnerability to alcoholism in which one key factor is WMC/EA. He argued that greater WMC allows an individual to better manipulate, monitor, and control the behavioral tendencies resulting from alcoholism, and that this directly affects the ability to resist a prepotent behavior such as taking a drink in spite of being aware that such behavior is ultimately maladaptive. Individual differences in WMC/EA have also been shown to be important in emotion regulation (Hofmann et al., 2011). Thus, assessment of whether individuals are likely to be more or less able to control impulses and self-destructive thoughts would benefit from inclusion of WMC measures.
The linkages between EA and impulse control suggest that examinations of EA may benefit from examining relations with self-control measures in the personality domain to determine the degree of overlap and potential incremental validity of one over the other. Recent studies have shown that the tendency to mind-wander during performance of a critical task is highly associated with measures of WMC (McVay and Kane, 2009, 2012a, 2012b). These researchers used a variety of techniques to measure what they called task-unrelated thoughts during performance of complex tasks. In one study (Kane et al., 2007), subjects carried a Palm Pilot 2 and were alerted eight random times over the course of their day to answer questions about the tasks they were currently performing, their level of concentration, how challenging the task was, how much effort they were expending, and whether their mind had wandered in the last few minutes. The results in Figure 2-4 show clearly that low and high WMC individuals differed greatly in their tendency to mind-wander and that the differences grew as more concentration was required in the task and the task became more challenging. Low WMC individuals are more likely to mind-wander as a task increases in challenge and effort level required. One question that could be investigated through future research would be the cause or effect related to whether mind wandering is a consequence of task difficulty and WMC or a predictor of WMC (suggesting that mind wandering is a consequence rather than a cause of WMC performance). These differences in performance would seem to be generalizable to a wide range of tasks performed in the Army across the full spectrum of operations from peacetime to combat situations.
2 Palm Pilot was an early personal digital assistant that could be set up with multiple alarms and short interactive response-entry actions.
FIGURE 2-4 High versus low WMC individuals and task-unrelated thoughts in daily life.
NOTE: Values on the y-axis represent the mind wandering dependent variable, scored on each questionnaire as either 1 (for mind wandering) or 2 for on-task thoughts lower values thus indicate more mind wandering. Values on the x-axis represent group-centered ratings for (a) concentration (&ldquoI had been trying to concentrate on what I was doing&rdquo), (b) challenge (&ldquoWhat I&rsquom doing right now is challenging&rdquo), and (c) effort (&ldquoIt takes a lot of effort to do this activity&rdquo).
SOURCE: Kane, J.J., L.H. Brown, J.C. McVay, I. Myin-Germeys, P.J. Silva, and T.R. Kwapil. (2007). For whom the mind wanders, and when: An experience-sampling study of working memory and executive control in daily life. Psychological Science, 18(7):167. Reproduced by permission of SAGE Publications.
While a general mental abilities (i.e., Gf) approach is useful and has been considered the gold standard for predicting job performance (Schmidt and Hunter, 1998), recent work in this area suggests the importance of WMC in such predictions. In particular, WMC has been found to capture specific aptitudes beyond general mental abilities (Bosco and Allen, 2011 Hambrick et al., 2010 König et al., 2005). A recent study by König and colleagues (2005) testing 122 college students found that WMC was the best predictor of multitasking (similar conclusions were supported by Damos, 1993 Hambrick et al., 2010, 2011 and Stankov et al., 1989). These studies also showed WMC remained predictive of multitasking performance after controlling for fluid intelligence. In hierarchical regression analyses, WMC demonstrated the highest correlations with several measures of multitasking and predicted the most unique variance (Hambrick et al., 2010, 2011). Other research has found that WMC and Gf are distinct but strongly related (Kane et al., 2005).
Another perspective on assessments of WMC and EA is that, although they have great validity in predicting performance in real-world job situations, some research indicates they produce smaller mean racial/ethnic
group differences than do measures of crystallized ability. Subgroup differences contribute to adverse impact, a violation of Title VII of the 1964 Civil Rights Act. Under that statute, a violation of Title VII 3 may be demonstrated by showing that an employment practice or policy has a disproportionately adverse effect on members of the protected class as compared with nonmembers of the protected class. Such impact is only acceptable to the extent that the practice is proven to be germane to the job being selected for. In other words, a test that has good validity and low adverse impact against a protected class is preferred over one that has good validity but has higher adverse impact.
A series of studies (Bosco and Allen, 2011) compared the EA battery developed by the Engle lab (Engle and Kane, 2004) with the Wonderlic test in terms of ability to predict job performance and associated adverse impact due to race (i.e., different mean scores for the two racial groups on the test). In three different studies, respectively involving college students, MBA students, and individuals working in a large financial firm, Bosco and Allen found that the EA battery accounted for greater variance in task or job performance than the Wonderlic test and had substantially less adverse impact. The EA battery predicted an additional 7.2 percent of the variance beyond the Wonderlic on the job simulation task, as well as an additional 5.2 percent of the variance in supervisor ratings of job performance. The reduced adverse impact for the EA battery was also found for supervisory ratings of managers in the workplace environment.
These findings are intriguing enough to mention however, they are based on modest sample sizes, and additional replication is needed to solidify the basis of these findings. Verive and McDaniel (1996) report a meta-analysis of short-term memory tests on nearly 28,000 subjects and found that the black-white difference was less than half what it is on typical general cognitive ability tests, and yet the validity estimates remained high: .41 for job performance and .49 for training performance. Again, although interesting, the committee does not view these results as definitive. For example, the meta-analysis relies on untested assumptions about the degree of range restriction in the samples, and there is variance associated with these meta-analytic mean estimates that deserves to be understood.
Because short-term memory tests have been shown to be relatively unreliable and have reduced validity compared to measures of working memory capacity and executive attention (Engle et al., 1999a, 1999b), one might expect the latter measures to be even more resistant to adverse impact. This is consistent with recent work by Redick and colleagues (2012) in which gender differences were shown to be minimal on working memory complex span tasks over a sample size of 6,000 young adults.
Thus, the WMC/EA approach to assessment appears to provide substantial incremental validity for specific job situations and yet is less influenced by race or ethnic group. This tentative finding would seem to be particularly important for the modern Army situation but clearly needs further study and development, including research into cost-effective large-scale testing mechanisms suitable for administration in mobile or other non-laboratory settings without compromising validity, reliability, or test security. (See Section 5 of this report, Methods and Methodology, for further discussion of research topics to facilitate such developments.) In developing a future research program, it is important to recognize that although much research has been conducted on the constructs of fluid intelligence, WMC, and EA, research on the relationship between WMC and fluid intelligence is a relatively new and incomplete endeavor that combines two typically parallel research approaches: experimental and differential. Bringing these research approaches under one roof will improve the identification and understanding of the mechanisms responsible for the constructs of WMC, fluid intelligence, and EA, thus making significant contributions to the basic understanding of individual differences.
Research Recommendation: Fluid Intelligence, Working Memory Capacity, and Executive Attention
The U.S. Army Research Institute for the Behavioral and Social Sciences should support research to understand the psychological, cognitive, and neurobiological mechanisms underlying the constructs of fluid intelligence (novel reasoning), working memory capacity, and executive attention.
- A. Research should be conducted to ascertain whether these constructs reflect a common mechanism or are highly related but distinct mechanisms.
- B. Assessments reflecting the results of research into the commonality versus distinctness of these constructs should be developed for purposes of validity investigations.
- C. Ultimately, the basic research results from items A and B above should be used to inform research into time-efficient, computer-automated assessment(s).
The research on WMC/EA described above illustrates how measures based on tasks conducted in the laboratory (&ldquolab task measures&rdquo) can be used to index individual differences in cognitive control or executive
Data from the large, long-running U.S. Health and Retirement Study found that healthy cognition characterized most of the people with at least a college education into their late 80s, while those who didn’t complete high school had good cognition up until their 70s.
The study found that those who had at least a college education lived a much shorter time with dementia than those with less than a high school education: an average of 10 months for men and 19 months for women, compared to 2.57 years (men) and 4.12 years (women).
The data suggests that those who graduated high school can expect to live (on average) at least 70% of their remaining life after 65 with good cogntion, compared to more than 80% for those with a college education, and less than 50% for those who didn't finish high school.
The analysis was based on a sample of 10,374 older adults (65+ average age 74) in 2000 and 9,995 in 2010.
More education linked to better cognitive functioning later in life
Data from around 196,000 subscribers to Lumosity online brain-training games found that higher levels of education were strong predictors of better cognitive performance across the 15- to 60-year-old age range of their study participants, and appear to boost performance more in areas such as reasoning than in terms of processing speed.
Differences in performance were small for test subjects with a bachelor's degree compared to those with a high school diploma, and moderate for those with doctorates compared to those with only some high school education.
But people from lower educational backgrounds learned novel tasks nearly as well as those from higher ones.
Youthful cognitive ability strongly predicts mental capacity later in life
Data from more than 1,000 men participating in the Vietnam Era Twin Study of Aging revealed that their cognitive ability at age 20 was a stronger predictor of cognitive function later in life than other factors, such as higher education, occupational complexity or engaging in late-life intellectual activities.
All of the men, now in their mid-50s to mid-60s, took the Armed Forces Qualification Test at an average age of 20. The same test of general cognitive ability (GCA) was given in late midlife, plus assessments in seven cognitive domains.
GCA at age 20 accounted for 40% of the variance in the same measure at age 62, and approximately 10% of the variance in each of the seven cognitive domains. Lifetime education, complexity of job and engagement in intellectual activities each accounted for less than 1% of variance at average age 62.
The findings suggest that the impact of education, occupational complexity and engagement in cognitive activities on later life cognitive function simply reflects earlier cognitive ability.
The researchers speculated that the role of education in increasing GCA takes place primarily during childhood and adolescence when there is still substantial brain development.
Guerra-Carrillo, B., Katovich, K., & Bunge, S. A. (2017). Does higher education hone cognitive functioning and learning efficacy? Findings from a large and diverse sample. PLOS ONE, 12(8), e0182276. https://doi.org/10.1371/journal.pone.0182276
A study involving 218 participants aged 18-88 has looked at the effects of age on the brain activity of participants viewing an edited version of a 1961 Hitchcock TV episode (given that participants viewed the movie while in a MRI machine, the 25 minute episode was condensed to 8 minutes).
While many studies have looked at how age changes brain function, the stimuli used have typically been quite simple. This thriller-type story provides more complex and naturalistic stimuli.
Younger adults' brains responded to the TV program in a very uniform way, while older adults showed much more idiosyncratic responses. The TV program (“Bang! You're dead”) has previously been shown to induce widespread synchronization of brain responses (such movies are, after all, designed to focus attention on specific people and objects following along with the director is, in a manner of speaking, how we follow the plot). The synchronization seen here among younger adults may reflect the optimal response, attention focused on the most relevant stimulus. (There is much less synchronization when the stimuli are more everyday.)
The increasing asynchronization with age seen here has previously been linked to poorer comprehension and memory. In this study, there was a correlation between synchronization and measures of attentional control, such as fluid intelligence and reaction time variability. There was no correlation between synchronization and crystallized intelligence.
The greatest differences were seen in the brain regions controlling attention (the superior frontal lobe and the intraparietal sulcus) and language processing (the bilateral middle temporal gyrus and left inferior frontal gyrus).
The researchers accordingly suggested that the reason for the variability in brain patterns seen in older adults lies in their poorer attentional control — specifically, their top-down control (ability to focus) rather than bottom-up attentional capture. Attentional capture has previously been shown to be well preserved in old age.
Of course, it's not necessarily bad that a watcher doesn't rigidly follow the director's manipulation! The older adults may be showing more informed and cunning observation than the younger adults. However, previous studies have found that older adults watching a movie tend to vary more in where they draw an event boundary those showing most variability in this regard were the least able to remember the sequence of events.
The current findings therefore support the idea that older adults may have increasing difficulty in understanding events — somthing which helps explain why some old people have increasing trouble following complex plots.
The findings also add to growing evidence that age affects functional connectivity (how well the brain works together).
It should be noted, however, that it is possible that there could also be cohort effects going on — that is, effects of education and life experience.
There's been a lot of talk in recent years about the importance of mindset in learning, with those who have a “growth mindset” (ie believe that intelligence can be developed) being more academically successful than those who believe that intelligence is a fixed attribute. A new study shows that a 45-minute online intervention can help struggling high school students.
The study involved 1,594 students in 13 U.S. high schools. They were randomly allocated to one of three intervention groups or the control group. The intervention groups either experienced an online program designed to develop a growth mindset, or an online program designed to foster a sense of purpose, or both programs (2 weeks apart). All interventions were expected to improve academic performance, especially in struggling students.
The interventions had no significant benefits for students who were doing okay, but were of significant benefit for those who had an initial GPA of 2 or less, or had failed at least one core subject (this group contained 519 students a third of the total participants). For this group, each of the interventions was of similar benefit interestingly, the combined intervention was less beneficial than either single intervention. It's plausibly suggested that this might be because the different messages weren't integrated, and students may have had some trouble in taking on board two separate messages.
Overall, for this group of students, semester grade point averages improved in core academic courses and the rate at which students performed satisfactorily in core courses increased by 6.4%.
GPA average in core subjects (math, English, science, social studies) was calculated at the end of the semester before the interventions, and at the end of the semester after the interventions. Brief questions before and after the interventions assessed the students' beliefs about intelligence, and their sense of meaningfulness about schoolwork.
GPA before intervention was positively associated with a growth mindset and a sense of purpose, explaining why the interventions had no effect on better students. Only the growth mindset intervention led to a more malleable view of intelligence only the sense-of-purpose intervention led to a change in perception in the value of mundane academic tasks. Note that the combined intervention showed no such effects, suggesting that it had confused rather than enlightened!
In the growth mindset intervention, students read an article describing the brain’s ability to grow and reorganize itself as a consequence of hard work and good strategies. The message that difficulties don't indicate limited ability but rather provide learning opportunities, was reinforced in two writing exercises. The control group read similar materials, but with a focus on functional localization in the brain rather than its malleability.
In the sense-of-purpose interventions, students were asked to write about how they wished the world could be a better place. They read about the reasons why some students worked hard, such as “to make their families proud” “to be a good example” “to make a positive impact on the world”. They were then asked to think about their own goals and how school could help them achieve those objectives. The control group completed one of two modules that didn't differ in impact. In one, students described how their lives were different in high school compared to before. The other was much more similar to the intervention, except that the emphasis was on economic self-interest rather than social contribution.
The findings are interesting in showing that you can help poor learners with a simple intervention, but perhaps even more, for their indication that such interventions are best done in a more holistic and contextual way. A more integrated message would hopefully have been more effective, and surely ongoing reinforcement in the classroom would make an even bigger difference.
Because this is such a persistent myth, I thought I should briefly report on this massive study that should hopefully put an end to this myth once and for all (I wish! Myths are not so easily squashed.)
This study used data from 377,000 U.S. high school students, and, agreeing with a previous large study, found that first-borns have a one IQ point advantage over later-born siblings, but while statistically significant, this is a difference of no practical significance.
The analysis also found that first-borns tended to be more extroverted, agreeable and conscientious, and had less anxiety than later-borns, — but those differences were “infinitesimally small”, amounting to a correlation of 0.02 (the correlation between birth order and intelligence was .04).
The study controlled for potentially confounding factors, such as a family's economic status, number of children and the relative age of the siblings at the time of the analysis.
A separate analysis of children with exactly two siblings and living with two parents, enabled the finding that there are indeed specific differences between the oldest and a second child, and between second and third children. But the magnitude of the differences was again “minuscule”.
Perhaps it's not fair to say the myth is trounced. Rather, we can say that, yeah, sure, birth order makes a difference — but the difference is so small as not to be meaningful on an individual level.
Data from 1.1 million young Swedish men (conscription information taken at age 18) has shown that those with poorer cardiovascular fitness were 2.5 times more likely to develop early-onset dementia later in life and 3.5 times more likely to develop mild cognitive impairment, while those with a lower IQ had a 4 times greater risk of early dementia and a threefold greater risk of MCI. A combination of both poor cardiovascular fitness and low IQ entailed a more than 7 times greater risk of early-onset dementia, and more than 8 times greater risk of MCI.
The increased risk remained even when controlled for other risk factors, such as heredity, medical history, and social-economic circumstances.
The development of early-onset dementia was taken from national disease registries. During the study period, a total of 660 men were diagnosed with early-onset dementia.
A further study of this database, taken from 488,484 men, of whom 487 developed early-onset dementia (at a median age of 54), found nine risk factors for early-onset dementia that together accounted for 68% of the attributable risk. These factors were alcohol intoxication, stroke, use of antipsychotics, depression, father's dementia, drug intoxication other than alcohol, low cognitive function at age 18, low stature at age 18, and high blood pressure at age 18.
By using brain scans from 152 Vietnam veterans with a variety of combat-related brain injuries, researchers claim to have mapped the neural basis of general intelligence and emotional intelligence.
There was significant overlap between general intelligence and emotional intelligence, both in behavioral measures and brain activity. Higher scores on general intelligence tests and personality reliably predicted higher performance on measures of emotional intelligence, and many of the same brain regions (in the frontal and parietal cortices) were found to be important to both.
More specifically, impairments in emotional intelligence were associated with selective damage to a network containing the extrastriate body area (involved in perceiving the form of other human bodies), the left posterior superior temporal sulcus (helps interpret body movement in terms of intentions), left temporo-parietal junction (helps work out other person’s mental state), and left orbitofrontal cortex (supports emotional empathy). A number of associated major white matter tracts were also part of the network.
Two of the components of general intelligence were strong contributors to emotional intelligence: verbal comprehension/crystallized intelligence, and processing speed. Verbal impairment was unsurprisingly associated with selective damage to the language network, which showed some overlap with the network underlying emotional intelligence. Similarly, damage to the fronto-parietal network linked to deficits in processing speed also overlapped in places with the emotional intelligence network.
Only one of the ‘big five’ personality traits contributed to the prediction of emotional intelligence — conscientiousness. Impairments in conscientiousness were associated with damage to brain regions widely implicated in social information processing, of which two areas (left orbitofrontal cortex and left temporo-parietal junction) were also involved in impaired emotional intelligence, suggesting where these two attributes might be connected (ability to predict and understand another’s emotions).
It’s interesting (and consistent with the growing emphasis on connectivity rather than the more simplistic focus on specific regions) that emotional intelligence was so affected by damage to white matter tracts. The central role of the orbitofrontal cortex is also intriguing – there’s been growing evidence in recent years of the importance of this region in emotional and social processing, and it’s worth noting that it’s in the right place to integrate sensory and bodily sensation information and pass that onto decision-making systems.
All of this is to say that emotional intelligence depends on social information processing and general intelligence. Traditionally, general intelligence has been thought to be distinct from social and emotional intelligence. But humans are fundamentally social animals, and – contra the message of the Enlightenment, that we have taken so much to heart – it has become increasingly clear that emotions and reason are inextricably entwined. It is not, therefore, all that surprising that general and emotional intelligence might be interdependent. It is more surprising that conscientiousness might be rooted in your degree of social empathy.
It’s also worth noting that ‘emotional intelligence’ is not simply a trendy concept – a pop quiz question regarding whether you ‘have a high EQ’ (or not), but that it can, if impaired, produce very real problems in everyday life.
Emotional intelligence was measured by the Mayer, Salovey, Caruso Emotional Intelligence Test (MSCEIT), general IQ by the Wechsler Adult Intelligence Scale, and personality by the Neuroticism-Extroversion-Openness Inventory.
One of the researchers talks about this study on this YouTube video and on this podcast.
What underlies differences in fluid intelligence? How are smart brains different from those that are merely ‘average’?
Brain imaging studies have pointed to several aspects. One is brain size. Although the history of simplistic comparisons of brain size has been turbulent (you cannot, for example, directly compare brain size without taking into account the size of the body it’s part of), nevertheless, overall brain size does count for something — 6.7% of individual variation in intelligence, it’s estimated. So, something, but not a huge amount.
Activity levels in the prefrontal cortex, research also suggests, account for another 5% of variation in individual intelligence. (Do keep in mind that these figures are not saying that, for example, prefrontal activity explains 5% of intelligence. We are talking about differences between individuals.)
A new study points to a third important factor — one that, indeed, accounts for more than either of these other factors. The strength of the connections from the left prefrontal cortex to other areas is estimated to account for 10% of individual differences in intelligence.
These findings suggest a new perspective on what intelligence is. They suggest that part of intelligence rests on the functioning of the prefrontal cortex and its ability to communicate with the rest of the brain — what researchers are calling ‘global connectivity’. This may reflect cognitive control and, in particular, goal maintenance. The left prefrontal cortex is thought to be involved in (among other things) remembering your goals and any instructions you need for accomplishing those goals.
The study involved 93 adults (average age 23 range 18-40), whose brains were monitored while they were doing nothing and when they were engaged in the cognitively challenging N-back working memory task.
Brain activity patterns revealed three regions within the frontoparietal network that were significantly involved in this task: the left lateral prefrontal cortex, right premotor cortex, and right medial posterior parietal cortex. All three of these regions also showed signs of being global hubs — that is, they were highly connected to other regions across the brain.
Of these, however, only the left lateral prefrontal cortex showed a significant association between its connectivity and individual’s fluid intelligence. This was confirmed by a second independent measure — working memory capacity — which was also correlated with this region’s connectivity, and only this region.
In other words, those with greater connectivity in the left LPFC had greater cognitive control, which is reflected in higher working memory capacity and higher fluid intelligence. There was no correlation between connectivity and crystallized intelligence.
Interestingly, although other global hubs (such as the anterior prefrontal cortex and anterior cingulate cortex) also have strong relationships with intelligence and high levels of global connectivity, they did not show correlations between their levels of connectivity and fluid intelligence. That is, although the activity within these regions may be important for intelligence, their connections to other brain regions are not.
So what’s so important about the connections the LPFC has with the rest of the brain? It appears that, although it connects widely to sensory and motor areas, it is primarily the connections within the frontoparietal control network that are most important — as well as the deactivation of connections with the default network (the network active during rest).
This is not to say that the LPFC is the ‘seat of intelligence’! Research has made it clear that a number of brain regions support intelligence, as do other areas of connectivity. The finding is important because it shows that the left LPFC supports cognitive control and intelligence through a mechanism involving global connectivity and some other as-yet-unknown property. One possibility is that this region is a ‘flexible’ hub — able to shift its connectivity with a number of different brain regions as the task demands.
In other words, what may count is how many different connectivity patterns the left LPFC has in its repertoire, and how good it is at switching to them.
An association between negative connections with the default network and fluid intelligence also adds to evidence for the importance of inhibiting task-irrelevant processing.
All this emphasizes the role of cognitive control in intelligence, and perhaps goes some way to explaining why self-regulation in children is so predictive of later success, apart from the obvious.
A large long-running New Zealand study has found that people who started using cannabis in adolescence and continued to use it for years afterward showed a significant decline in IQ from age 13 to 38. This was true even in those who hadn’t smoked marijuana for some years.
The study has followed a group of 1,037 children born in 1972-73. At age 38, 96% of the 1004 living study members participated in the latest assessment. Around 5% were regularly smoking marijuana more than once a week before age 18 (cannabis use was ascertained in interviews at ages 18, 21, 26, 32, and 38 years, and this group was not more or less likely to have dropped out of the study).
This group showed an average decline in IQ of 8 points on cognitive tests at age 38 compared to scores at age 13. Such a decline was not found in those who began using cannabis after the age of 18. In comparison, those who had never used cannabis showed a slight increase in IQ. The effect was dose-dependent, with those diagnosed as cannabis dependent on three or more occasions showing the greatest decline.
While executive function and processing speed appeared to be the most seriously affected areas, impairment was seen across most cognitive domains and did not appear to be statistically significantly different across them.
The size of the effect is shown by a further measure: informants (nominated by participants as knowing them well) also reported significantly more attention and memory problems among those with persistent cannabis dependence. (Note that a decline of 8 IQ points in a group whose mean is 100 brings it down to 92.)
The researchers ruled out recent cannabis use, persistent dependence on other drugs (tobacco, alcohol, hard drugs), and schizophrenia, as alternative explanations for the effect. The effect also remained after years of education were taken into account.
The finding supports the view that the adolescent brain is vulnerable to the effects of marijuana, and that these effects are long-lasting and significant.
Some numbers for those interested: Of the 874 participants included in the analysis (those who had missed at least 3 interviews in the 25 years were excluded), 242 (28%) never used cannabis, 479 (55%) used it but were never diagnosed as cannabis-dependent, and 153 (17%) were diagnosed on at least one of the interviews as cannabis-dependent. Of these, 80 had been so diagnosed on only one occasion, 35 on two occasions, and 38 on three or more occasions. I note that the proportion of males was significantly higher in the cannabis-dependent groups (39% in never used 49% in used but never diagnosed 70%, 63%, 82% respectively for the cannabis-dependent).
Grasp of fractions and long division predicts later math success
One possible approach to improving mathematics achievement comes from a recent study finding that fifth graders' understanding of fractions and division predicted high school students' knowledge of algebra and overall math achievement, even after statistically controlling for parents' education and income and for the children's own age, gender, I.Q., reading comprehension, working memory, and knowledge of whole number addition, subtraction and multiplication.
The study compared two nationally representative data sets, one from the U.S. and one from the United Kingdom. The U.S. set included 599 children who were tested in 1997 as 10-12 year-olds and again in 2002 as 15-17-year-olds. The set from the U.K. included 3,677 children who were tested in 1980 as 10-year-olds and in 1986 as 16-year-olds.
You can watch a short video of Siegler discussing the study and its implications at http://youtu.be/7YSj0mmjwBM.
Spatial skills improve children’s number sense
More support for the idea that honing spatial skills leads to better mathematical ability comes from a new children’s study.
The study found that first- and second-graders with the strongest spatial skills at the beginning of the school year showed the most improvement in their number line sense over the course of the year. Similarly, in a second experiment, not only were those children with better spatial skills at 5 ½ better on a number-line test at age 6, but this number line knowledge predicted performance on a math estimation task at age 8.
Hasty answers may make boys better at math
A study following 311 children from first to sixth grade has revealed gender differences in their approach to math problems. The study used single-digit addition problems, and focused on the strategy of directly retrieving the answer from long-term memory.
Accurate retrieval in first grade was associated with working memory capacity and intelligence, and predicted a preference for direct retrieval in second grade. However, at later grades the relation reversed, such that preference in one grade predicted accuracy and speed in the next grade.
Unlike girls, boys consistently preferred to use direct retrieval, favoring speed over accuracy. In the first and second grades, this was seen in boys giving more answers in total, and more wrong answers. Girls, on the other hand, were right more often, but responded less often and more slowly. By sixth grade, however, the boys’ practice was paying off, and they were both answering more problems and getting more correct.
In other words, while ability was a factor in early skilled retrieval, the feedback loop of practice and skill leads to practice eventually being more important than ability — and the relative degrees of practice may underlie some of the gender differences in math performance.
The findings also add weight to the view being increasingly expressed, that mistakes are valuable and educational approaches that try to avoid mistakes (e.g., errorless learning) should be dropped.
Infants can’t compare big and small groups
Our brains process large and small numbers of objects using two different mechanisms, seen in the ability to estimate numbers of items at a glance and the ability to visually track small sets of objects. A new study indicates that at age one, infants can’t yet integrate those two processes. Accordingly, while they can choose the larger of two sets of items when both sets are larger or smaller than four, they can’t distinguish between a large (above four) and small (below four) set.
In the study, infants consistently chose two food items over one and eight items over four, but chose randomly when asked to compare two versus four and two versus eight.
The researchers suggest that educational programs that claim to give children an advantage by teaching them arithmetic at an early age are unlikely to be effective for this reason.
Previous research has pointed to a typical decline in our sense of control as we get older. Maintaining a sense of control, however, appears to be a key factor in successful aging. Unsurprisingly, in view of the evidence that self-belief and metacognitive understanding are important for cognitive performance, a stronger sense of control is associated with better cognitive performance. (By metacognitive understanding I mean the knowledge that cognitive performance is malleable, not fixed, and strategies and training are effective in improving cognition.)
In an intriguing new study, 36 older adults (aged 61-87, average age 74) had their cognitive performance and their sense of control assessed every 12 hours for 60 days. Participants were asked questions about whether they felt in control of their lives and whether they felt able to achieve goals they set for themselves.
The reason I say this is intriguing is that it’s generally assumed that a person’s sense of control — how much they feel in control of their lives — is reasonably stable. While, as I said, it can change over the course of a lifetime, until recently we didn’t think that it could fluctuate significantly in the course of a single day — which is what this study found.
Moreover, those who normally reported having a low sense of control performed much better on inductive reasoning tests during periods when they reported feeling a higher sense of control. Similarly, those who normally reported feeling a high sense of control scored higher on memory tests when feeling more in control than usual.
Although we can’t be sure (since this wasn’t directly investigated), the analysis suggests that the improved cognitive functioning stems from the feeling of improved control, not vice versa.
The study builds on an earlier study that found weekly variability in older adults’ locus of control and competency beliefs.
Assessment was carried out in the form of a daily workbook, containing a number of measures, which participants completed twice daily. Each assessment took around 30-45 minutes to complete. The measures included three cognitive tests (14 alternate forms of each of these were used, to minimize test familiarity):
- Letter series test: 30 items in which the next letter in a series had to be identified. [Inductive reasoning]
- Number comparison: 48 items in which two number strings were presented beside each other, and participants had to identify where there was any mismatch. [Perceptual speed]
- Rey Auditory Verbal Learning Task: participants have to study a list of 15 unrelated words for one minute, then on another page recall as many of the words as they could. [Memory]
Sense of control over the previous 12 hours was assessed by 8 questions, to which participants indicated their agreement/disagreement on a 6-point scale. Half the questions related to ‘locus of control’ and half to ‘perceived competence’.
While, unsurprisingly, compliance wasn’t perfect (it’s quite an arduous regime), participants completed on average 115 of 120 workbooks. Of the possible 4,320 results (36 x 120), only 166 were missing.
One of the things that often annoys me is the subsuming of all within-individual variability in cognitive scores into averages. Of course averages are vital, but so is variability, and this too often is glossed over. This study is, of course, all about variability, so I was very pleased to see people’s cognitive variability spelled out.
Most of the variance in locus of control was of course between people (86%), but 14% was within-individual. Similarly, the figures for perceived competence were 88% and 12%. (While locus of control and perceived competence are related, only 26% of the variability in within-person locus of control was associated with competence, meaning that they are largely independent.)
By comparison, within-individual variability was much greater for the cognitive measures: for the letter series (inductive reasoning), 32% was within-individual and 68% between-individual for the number matching (perceptual speed), 21% was within-individual and 79% between-individual for the memory test, an astounding 44% was within-individual and 56% between-individual.
Some of this within-individual variability in cognitive performance comes down to practice effects, which were significant for all cognitive measures. For the memory test, time of day was also significant, with performance being better in the morning. For the letter and number series tests, previous performance also had a small effect on perceived competence. For the number matching, increase in competence subsequent to increased performance was greatest for those with lower scores. However, lagged analyses indicated that beliefs preceded performance to a greater extent than performance preceding beliefs.
While it wasn’t an aspect of this study, it should also be noted that a person’s sense of control may well vary according to domain (e.g., cognition, social interaction, health) and context. In this regard, it’s interesting to note the present findings that sense of control affected inductive reasoning for low-control individuals, but memory for high-control individuals, suggesting that the cognitive domain also matters.
Now this small study was a preliminary one and there are several limitations that need to be tightened up in subsequent research, but I think it’s important for three reasons:
- as a demonstration that cognitive performance is not a fixed attribute
- as a demonstration of the various factors that can affect older adults’ cognitive performance
- as a demonstration that your beliefs about yourself are a factor in your cognitive performance.
This is another demonstration of stereotype threat, which is also a nice demonstration of the contextual nature of intelligence. The study involved 70 volunteers (average age 25 range 18-49), who were put in groups of 5. Participants were given a baseline IQ test, on which they were given no feedback. The group then participated in a group IQ test, in which 92 multi-choice questions were presented on a monitor (both individual and group tests were taken from Cattell’s culture fair intelligence test). Each question appeared to each person at the same time, for a pre-determined time. After each question, they were provided with feedback in the form of their own relative rank within the group, and the rank of one other group member. Ranking was based on performance on the last 10 questions. Two of each group had their brain activity monitored.
Here’s the remarkable thing. If you gather together individuals on the basis of similar baseline IQ, then you can watch their IQ diverge over the course of the group IQ task, with some dropping dramatically (e.g., 17 points from a mean IQ of 126). Moreover, even those little affected still dropped some (8 points from a mean IQ of 126).
Data from the 27 brain scans (one had to be omitted for technical reasons) suggest that everyone was initially hindered by the group setting, but ‘high performers’ (those who ended up scoring above the median) managed to largely recover, while ‘low performers’ (those who ended up scoring below the median) never did.
Personality tests carried out after the group task found no significant personality differences between high and low performers, but gender was a significant variable: 10/13 high performers were male, while 11/14 low performers were female (remember, there was no difference in baseline IQ — this is not a case of men being smarter!).
There were significant differences between the high and low performers in activity in the amygdala and the right lateral prefrontal cortex. Specifically, all participants had an initial increase in amygdala activation and diminished activity in the prefrontal cortex, but by the end of the task, the high-performing group showed decreased amygdala activation and increased prefrontal cortex activation, while the low performers didn’t change. This may reflect the high performers’ greater ability to reduce their anxiety. Activity in the nucleus accumbens was similar in both groups, and consistent with the idea that the students had expectations about the relative ranking they were about to receive.
It should be pointed out that the specific feedback given — the relative ranking — was not a factor. What’s important is that it was being given at all, and the high performers were those who became less anxious as time went on, regardless of their specific ranking.
There are three big lessons here. One is that social pressure significantly depresses talent (meetings make you stupid?), and this seems to be worse when individuals perceive themselves to have a lower social rank. The second is that our ability to regulate our emotions is important, and something we should put more energy into. And the third is that we’ve got to shake ourselves loose from the idea that IQ is something we can measure in isolation. Social context matters.
Benefits of high quality child care persist 30 years later
Back in the 1970s, some 111 infants from low-income families, of whom 98% were African-American, took part in an early childhood education program called the Abecedarian Project. From infancy until they entered kindergarten, the children attended a full-time child care facility that operated year-round. The program provided educational activities designed to support their language, cognitive, social and emotional development.
The latest data from that project, following up the participants at age 30, has found that these people had significantly more years of education than peers who were part of a control group (13.5 years vs 12.3), and were four times more likely to have earned college degrees (23% vs 6%).
They were also significantly more likely to have been consistently employed (75% had worked full time for at least 16 of the previous 24 months, compared to 53% of the control group) and less likely to have used public assistance (only 4% received benefits for at least 10% of the previous seven years, compared to 20% of the control group). However, income-to-needs ratios (income taken into account household size) didn’t vary significantly between the groups (mainly because of the wide variability on the face of it, the means are very different, but the standard deviation is huge), and neither did criminal involvement (27% vs 28%).
See their website for more about this project.
Evidence that more time at school raises IQ
It would be interesting to see what the IQs of those groups are, particularly given that maternal IQ was around 85 for both treatment and control groups. A recent report analyzed the results of a natural experiment that occurred in Norway when compulsory schooling was increased from seven to nine years in the 1960s, meaning that students couldn’t leave until 16 rather than 14. Because all men eligible for the draft were given an IQ test at age 19, statisticians were able to look back and see what effect the increased schooling had on IQ.
They found that it had a substantial effect, with each additional year raising the average IQ by 3.7 points.
While we can’t be sure how far these results extend to other circumstances, they are clear evidence that it is possible to improve IQ through education.
Why children of higher-income parents start school with an advantage
Of course the driving idea behind improved child-care in the early years is all about the importance of getting off to a good start, and you’d expect that providing such care to children would have a greater long-term effect on IQ than simply extending time at school. Most such interventions have looked at the most deprived strata of society. An overlooked area is that of low to middle income families, who are far from having the risk factors of less fortunate families.
A British study involving 15,000 five-year-olds has found that, at the start of school, children from low to middle income families are five months behind children from higher income families in terms of vocabulary skills and have more behavior problems (they were also 8 months ahead of their lowest income peers in vocabulary).
Low-middle income (LMI) households are defined by the Resolution Foundation (who funded this research) as members of the working-age population in income deciles 2-5 who receive less than one-fifth of their gross household income from means-tested benefits (see their website for more detail on this).
Now the difference in home environment between LMI and higher income households is often not that great — particularly when you consider that it is often a difference rooted in timing. LMI households are more common in this group of families with children under five, because the parents are usually at an early stage of life. So what brings about this measurable difference in language and behavior development?
This is a tricky thing to derive from the data, and the findings must be taken with a grain of salt. And as always, interpretation is even trickier. But with this caveat, let’s see what we have. Let’s look at demographics first.
The first thing is the importance of parental education. Income plus education accounted for some 70-80% of the differences in development, with education more important for language development and income more important for behavior development. Maternal age then accounted for a further 10%. Parents in the higher-income group tended to be older and have better education (e.g., 18% of LMI mothers were under 25 at the child’s birth, compared to 6% of higher-income mothers 30% of LMI parents had a degree compared to 67% of higher-income parents).
Interestingly, family size was equally important for language development (10%), but much less important for behavior development (in fact this was a little better in larger families). Differences in ethnicity, language, or immigration status accounted for only a small fraction of the vocabulary gap, and none of the behavior gap.
Now for the more interesting but much trickier analysis of environmental variables. The most important factor was home learning environment, accounting for around 20% of the difference. Here the researchers point to higher-income parents providing more stimulation. For example, higher-income parents were more likely to read to their 3-year-olds every day (75% vs 62% 48% for the lowest-income group), to take them to the library at least once a month (42% vs 35% vs 26%), to take their 5-year-old to a play or concert (86% vs 75% vs 60%), to a museum/gallery (67% vs 48% vs 36%), to a sporting activity at least once a week (76% vs 57% vs 35%). Higher-income parents were also much less likely to allow their 3-year-olds to watch more than 3 hours of TV a day (7% vs 17% vs 25%). (I know the thrust of this research is the comparison between LMI and higher income, but I’ve thrown in the lowest-income figures to help provide context.)
Interestingly, the most important factor for vocabulary learning was being taken to a museum/gallery at age 5 (but remember, these correlations could go either way: it might well be that parents are more likely to take an articulate 5-year-old to such a place), with the second most important factor being reading to 3-year-old every day. These two factors accounted for most of the effects of home environment. For behavior, the most important factor was regular sport, followed by being to a play/concert, and being taken to a museum/gallery. Watching more than 3 hours of TV at age 3 did have a significant effect on both vocabulary and behavior development (a negative effect on vocabulary and a positive effect on behavior), while the same amount of TV at age 5 did not.
Differences in parenting style explained 10% of the vocabulary gap and 14% of the behavior gap, although such differences were generally small. The biggest contributors to the vocabulary gap were mother-child interaction score at age 3 and regular bedtimes at age 3. The biggest contributors to the behavior gap were regular bedtimes at age 5, regular mealtimes at age 3, child smacked at least once a month at age 5 (this factor also had a small but significant negative effect on vocabulary), and child put in timeout at least once a month at age 5.
Maternal well-being accounted for over a quarter of the behavior gap, but only a small proportion of the vocabulary gap (2% — almost all of this relates to social support score at 9 months). Half of the maternal well-being component of the behavior gap was down to psychological distress at age 5 (very much larger than the effect of psychological distress at age 3). Similarly, child and maternal health were important for behavior (18% in total), but not for vocabulary.
Material possessions, on the other hand, accounted for some 9% of the vocabulary gap, but none of the behavior gap. The most important factors here were no internet at home at age 5 (22% of LMIs vs 8% of higher-incomes), and no access to a car at age 3 (5% of LMIs had no car vs 1% of higher incomes).
As I’ve intimated, it’s hard to believe we can disentangle individual variables in the environment in an observational study, but the researchers believe the number of variables in the mix (158) and the different time points (many variables are assessed at two or more points) provided a good base for analysis.
Washbrook, E., & Waldfogel, J. (2011). On your marks : Measuring the school readiness of children in low-to-middle income families. Resolution Foundation, December 2011.
Short-term adaptive cognitive training based on the n-back task is reported to increase scores on individual ability tests, but the key question of whether such increases generalize to the intelligence construct is not clear. Here we evaluate fluid/abstract intelligence (Gf), crystallized/verbal intelligence (Gc), working memory capacity (WMC), and attention control (ATT) using diverse measures, with equivalent versions, for estimating any changes at the construct level after training. Beginning with a sample of 169 participants, two groups of twenty-eight women each were selected and matched for their general cognitive ability scores and demographic variables. Under strict supervision in the laboratory, the training group completed an intensive adaptive training program based on the n-back task (visual, auditory, and dual versions) across twenty-four sessions distributed over twelve weeks. Results showed that this group had the expected systematic improvements in n-back performance over time this performance systematically correlated across sessions with Gf, Gc, and WMC, but not with ATT. However, the main finding showed no significant changes in the assessed psychological constructs for the training group as compared with the control group. Nevertheless, post-hoc analyses suggested that specific tests and tasks tapping visuospatial processing might be sensitive to training.
An Overview of Research into the Cognitive Basis of Intelligence
Abstract. This paper provides an overview of the research into the cognitive basis of intelligence. This research explains cognitive abilities in terms of cognitive units or properties of such units. Furthermore, this research is characterized by the application of so-called elementary cognitive tasks. The various approaches of this research originate from the concepts of cognitive psychology: Mental (and perceptual) speed, attention, working memory, memory access, and learning. All the approaches led to measures which correlate with measures of intelligence. The enormous importance of the cognitive basis is highlighted by the observation that predictors taken from the various approaches explain approximately 50% of the variance of intelligence. At the latent level the rate of explanation seems to surmount the 70% barrier. Furthermore, the problems and perspectives of the approaches are addressed.
Ackerman, P.L. (2000) . Domain-specific knowledge as the “dark matter” of adult intelligence: Gf/Gc, personality and interest correlates . Journal of Gerontology: Psychological Sciences , 55B, 69–84 First citation in articleCrossref, Google Scholar
Acton, G.S., Schroeder, D.H. (2001) . Sensory discrimination as related to general intelligence . Intelligence , 29, 263–271 First citation in articleCrossref, Google Scholar
Anderson, J.R. (1982) . Acquisition of cognitive skills . Psychological Review , 89, 369–406 First citation in articleCrossref, Google Scholar
Anderson, J.R., Fincham, J.M., Douglass, S. (1997) . The role of examples and rules in the acquisition of a cognitive skill . Journal of Experimental Psychology: Learning, Memory, & Cognition , 23, 932–945 First citation in articleCrossref, Google Scholar
Atkinson, R.C., Shiffrin, R.M. (1968) . Human memory: A proposed system and its control processes . In K.W. Spence & J.T. Spence (Eds.), The psychology of learning and motivation: Advances in research and theory (Vol. 2, pp. 90-195). New York: Academic Press First citation in articleGoogle Scholar
Baddeley, A.D. (1986) . Working memory . Oxford: Clarendon Press First citation in articleGoogle Scholar
Bayliss, D.M., Jarrold, C., Gunn, D., Baddeley, A. (2003) . The complexities of complex span: Explaining individual differences in working memory in children and adults . Journal of Experimental Psychology: General , 132, 71–92 First citation in articleCrossref, Google Scholar
Bucik, V. (1993) . “Speed,” “Power,” or both? Speed of information processing and Spearman's g . Poster presented at the 3rd European Conference on Psychology, Tampere, Finland First citation in articleGoogle Scholar
Carpenter, P.A., Just, M.A., Shell, P. (1990) . What one intelligence test measures: A theoretical account of the processing in the Raven Progressive Matrices Test . Psychological Review , 97, 404–431 First citation in articleCrossref, Google Scholar
Carroll, J.B. (1993) . Human cognitive abilities . New York: Cambridge University Press First citation in articleCrossref, Google Scholar
Cattell, R.B. (1963) . Theory of fluid and crystallized intelligence: A critical experiment . Journal of Educational Psychology , 54, 1–22 First citation in articleCrossref, Google Scholar
Colom, R., Rebollo, I., Palacios, A., Juan-Espinosa, M., Kyllonen, P.C. (2004) . Working memory is (almost) perfectly predicted by g . Intelligence , 32, 277–296 First citation in articleCrossref, Google Scholar
Coull, J.T. (1998) . Neural correlates of attention and arousal: Insights from electrophysiology, functional neuroimaging, and psychopharmacology . Progress in Neurobiology , 55, 343–361 First citation in articleCrossref, Google Scholar
Deary, I. (2000) . Simple information processing and intelligence . In R.J. Sternberg (Ed.), Handbook of intelligence (pp. 267- 284). Cambridge, MA: Cambridge University Press First citation in articleCrossref, Google Scholar
Deary, I. (2001a) . Human intelligence differences: A recent history . Trends in Cognitive Sciences , 5, 127–130 First citation in articleCrossref, Google Scholar
Deary, I. (2001b) . Human intelligence differences: Toward a combined experimental-differential approach . Trends in Cognitive Sciences , 5, 164–170 First citation in articleCrossref, Google Scholar
Engle, R.W., Tuholski, S.W., Laughlin, J.E., Conway, A.R.A. (1999) . Working memory, short-term memory, and general fluid intelligence: A latent-variable approach . Journal of Experimental Psychology: General , 128, 309–331 First citation in articleCrossref, Google Scholar
Fernandez-Duque, D., Johnson, M.L. (1998) . Attention metaphors: How metaphors guide the cognitive psychology of attention . Cognitive Science , 23, 83–116 First citation in articleCrossref, Google Scholar
Fernandez-Duque, D., Johnson, M.L. (2002) . Cause and effect theories of attention: The role of conceptual metaphors . Review of General Psychology , 6, 153–165 First citation in articleCrossref, Google Scholar
Grudnik, J.L., Kranzler, J.H. (2001) . Meta-analysis of the relationship between intelligence and inspection time . Intelligence , 29, 523–535 First citation in articleCrossref, Google Scholar
Gustafsson, J.E. (1984) . A unifying model for the structure of intellectual abilities . Intelligence , 8, 179–203 First citation in articleCrossref, Google Scholar
Guthke, J., Beckmann, J.F., Dobat, H. (1997) . Dynamic testing - problems, uses, trends, and evidence of validity . Educational & Child Psychology , 14, 17–32 First citation in articleGoogle Scholar
Hunt, E.B. (1978) . Mechanics of verbal ability . Psychological Review , 85, 109–130 First citation in articleCrossref, Google Scholar
Hunt, E. (1980) . Intelligence as an information-processing concept . British Journal of Psychology , 71, 449–474 First citation in articleCrossref, Google Scholar
Jensen, A.R. (1982) . The chronometry of intelligence . In R.J. Sternberg (Ed.), Advances in research on intelligence (Vol. 1). Hillsdale, NJ: Erlbaum First citation in articleGoogle Scholar
Jensen, A.R. (1987) . Individual differences in the Hick paradigm . In P.A. Vernon (Ed.), Speed of information processing and intelligence (pp. 101-175). Norwood, NJ: Ablex First citation in articleGoogle Scholar
Jensen, A.R. (1989) . The relationship between learning and intelligence . Learning and Individual Differences , 1, 37–62 First citation in articleCrossref, Google Scholar
Jensen, A.R. (1992) . Relation between information-processing time and right/wrong responses . American Journal on Mental Retardation , 97, 290–292 First citation in articleGoogle Scholar
Jensen, A.R. (1998) . The g factor. The science of mental ability . Westport, CT: Praeger First citation in articleGoogle Scholar
Jensen, A.R. (2004) . Obituary . Intelligence , 32, 1–5 First citation in articleCrossref, Google Scholar
Jensen, A.R., Munro, E. (1979) . Reaction time, movement time, and intelligence . Intelligence , 3, 121–126 First citation in articleCrossref, Google Scholar
Johanson, W., Bouchard, T.J., Krueger, R.F., McGue, M., Gottesman, I.I. (2004) . Just one g: Consistent results from three test batteries . Intelligence , 32, 95–107 First citation in articleCrossref, Google Scholar
Kyllonen, P.C., Christal, R.E. (1990) . Reasoning ability is (little more than) working-memory capacity?! . Intelligence , 14, 389– 433 First citation in articleCrossref, Google Scholar
Larson, G.E., Saccuzzo, D.P. (1989) . Cognitive correlates of general intelligence: Toward a process theory of g . Intelligence , 13, 5–13 First citation in articleCrossref, Google Scholar
Lehrl, S., Fischer, B. (1988) . The basic parameters of human information processing: Their role in the determination of intelligence . Personality and Individual Differences , 9, 883–896 First citation in articleCrossref, Google Scholar
Lohman, D.F. (2000) . Complex information processing and intelligence . In R.J. Sternberg (Ed.), Handbook of intelligence (pp. 285-340). Cambridge, MA: Cambridge University Press First citation in articleCrossref, Google Scholar
Miyake, A., Friedman, N.P., Rettinger, D.A., Shah, P., Hegarty, M. (2001) . How are visuospatial working memory, executive functioning, and spatial abilities related? . Journal of Experimental Psychology: General , 130, 621–640 First citation in articleCrossref, Google Scholar
Myors, B., Stankov, L., Oliphant, G.W. (1989) . Competing tasks, working memory, and intelligence . Australian Journal of Psychology , 41, 1–16 First citation in articleCrossref, Google Scholar
Neisser, U. (1967) . Cognitive psychology . New York: Appleton-Century-Crofts First citation in articleGoogle Scholar
Neubauer, A.C. (1995) . Geschwindigkeit der Informationsverarbeitung . [Speed of information processing]. Heidelberg: Springer-Verlag First citation in articleCrossref, Google Scholar
Neubauer, A.C. (1997) . The mental speed approach to the assessment of intelligence . Advances in Cognition and Educational Practice , 4, 149–173 First citation in articleGoogle Scholar
Neubauer, A.C., Knorr, E. (1997) . Elementary cognitive processes in choice reaction time tasks and their correlations with intelligence . Personality and Individual Differences , 23, 715–728 First citation in articleCrossref, Google Scholar
Neubauer, A.C., Knorr, E. (1998) . Three paper-and-pencil tests for speed of information processing: Psychometric properties and correlations with intelligence . Intelligence , 26, 123–15 First citation in articleCrossref, Google Scholar
Posner, M.I., Mitchell, R.F. (1967) . Chronometric analysis of classification . Psychological Review , 74, 392–409 First citation in articleCrossref, Google Scholar
Renkl, A., Schweizer, K. (2000) . Intelligenz in Interaktion mit Lernen und Gedächtnis . [Intelligence in interaction with learning and memory]. In K. Schweizer (Ed.), Intelligenz und Kognition [Intelligence and cognition] (pp. 86-104). Landau: Verlag Empirische Pädagogik First citation in articleGoogle Scholar
Roberts, R.D., Beh, H.C., Stankov, L. (1988) . Hick's Law, competing tasks, and intelligence . Intelligence , 12, 101–120 First citation in articleCrossref, Google Scholar
Roberts, R.D., Stankov, L. (1999) . Individual differences in speed of mental processing and human cognitive abilities: Toward a taxonomic model . Learning and Individual Differences , 11, 1–120 First citation in articleCrossref, Google Scholar
Schmidt-Atzert, L., Bühner, M. (1998) . Types of error in the d2 Test of Attention . Diagnostica , 44, 142–152 First citation in articleGoogle Scholar
Schmidt-Atzert, L., Bühner, M. (2000) . Attention and intelligence . In K. Schweizer (Ed.), Intelligenz und Kognition [Intelligence and cognition] (pp. 125-151). Landau: Verlag Empirische Pädagogik First citation in articleGoogle Scholar
Schweizer, K. (1993a) . The effect of two information-processing skills on the speed-ability relationship . Personality and Individual Differences , 14, 713–722 First citation in articleCrossref, Google Scholar
Schweizer, K. (1993b) . The contribution of access to external information, stimulus complexity, and variability to cognitive abilities . Personality and Individual Differences , 14, 87–95 First citation in articleCrossref, Google Scholar
Schweizer, K. (1993c) . Verbal ability and speed of information-processing . Personality and Individual Differences , 15, 645–652 First citation in articleCrossref, Google Scholar
Schweizer, K. (1994) . Structural diversity of the speed-ability relationship due to information-processing skills . Personality and Individual Differences , 17, 607–616 First citation in articleCrossref, Google Scholar
Schweizer, K. (1995a) . Kognitive Korrelate der Intelligenz . [Cognitive correlates of intelligence]. Göttingen: Hogrefe First citation in articleGoogle Scholar
Schweizer, K. (1996) . The speed-accuracy transition due to task complexity . Intelligence , 22, 115–128 First citation in articleCrossref, Google Scholar
Schweizer, K. (2000a) . Cognitive mechanisms as sources of success and failure in intelligence testing . Psychologische Beiträge , 42, 190–200 First citation in articleGoogle Scholar
Schweizer, K. (2000b) . Was Wahrnehmungsprozesse zur Intelligenz beitragen . [What perceptual processes contribute to intelligence]. In K. Schweizer (Ed.), Intelligenz und Kognition [Intelligence and cognition] (pp. 65-75). Landau: Verlag Empirische Pädagogik First citation in articleGoogle Scholar
Schweizer, K. (2001a) . On the role of mechanisms when the processing complexity is high . European Psychologist , 6, 133–143 First citation in articleLink, Google Scholar
Schweizer, K. (2001b) . Preattentive processes and cognitive ability . Intelligence , 29, 169–186 First citation in articleCrossref, Google Scholar
Schweizer, K., Koch, W. (2001) . A revision of Cattell's Investment Theory: Cognitive properties influencing learning . Learning and Individual Differences , 13, 57–82 First citation in articleCrossref, Google Scholar
Schweizer, K., Koch, W. (2002) . Fluide Intelligenz und mittelfristig erworbenes Wissen in einer spezifischen Wissensdomäne . [Fluid intelligence and medium-term acquired knowledge in a specific domain of knowledge] Zeitschrift für Psychologie , 210, 111–121 First citation in articleLink, Google Scholar
Schweizer, K., Koch, W. (2003) . Perceptual processes and cognitive ability . Intelligence , 31, 211–235 First citation in articleCrossref, Google Scholar
Schweizer, K., Moosbrugger, H. (2004) . Attention and working memory as predictors of intelligence . Intelligence , 32, 329–347 First citation in articleCrossref, Google Scholar
Schweizer, K., Moosbrugger, H., Goldhammer, K. (submitted) . The structure of the relationship between attention and intelligence . First citation in articleGoogle Scholar
Schweizer, K., Zimmermann, P., Koch, W. (2000) . Sustained attention, intelligence, and the crucial role of perceptual processes . Learning and Individual Differences , 12, 271–286 First citation in articleCrossref, Google Scholar
Stankov, L. (1983) . Attention and intelligence . Journal of Educational Psychology , 75, 471–490 First citation in articleCrossref, Google Scholar
Stankov, L. (1988a) . Single tests, competing tasks, and their relationship to the broad factors of intelligence . Personality and Individual Differences , 9, 25–33 First citation in articleCrossref, Google Scholar
Stankov, L. (1988b) . Aging, attention, and intelligence . Psychology & Aging , 3, 59–74 First citation in articleCrossref, Google Scholar
Stankov, L. (1989) . Attentional resources and intelligence: A disappearing link . Personality and Individual Differences , 10, 957–968 First citation in articleCrossref, Google Scholar
Stankov, L. (2000) . Die Beziehung zwischen Komplexität und Intelligenz . [The relationship between complexity and intelligence]. In K. Schweizer (Ed.), Intelligenz und Kognition [Intelligence and cognition] (pp. 37-63). Landau: Verlag Empirische Pädagogik First citation in articleGoogle Scholar
Stankov, L., Myors, B. (1990) . The relationship between working memory and intelligence: Regression and COSAN analyses . Personality and Individual Differences , 11, 1059–1068 First citation in articleCrossref, Google Scholar
Sternberg, S. (1966) . High-speed scanning in human memory . Science , 153, 652–654 First citation in articleCrossref, Google Scholar
Sturm, W., Zimmermann, P. (2000) . Aufmerksamkeitsstörungen . [Attentional deficits]. In W. Sturm, M. Herrmann, & C.-W. Wallesch (Hrsg.), Lehrbuch der klinischen Neuropsychologie [Textbook of clinical neuropsychology] (S. 345-365). Lisse: Swets & Zeitlinger First citation in articleGoogle Scholar