Fluid and crystallized intelligence
According to the hypothesis, put forward in 1943 by the psychologist Raymond Cattell, which he subsequently developed into a theory, published in 1971, general intelligence (g) is subdivided into
- Fluid intelligence (gf)—the ability to induce primary relational abstractions by contemplating data elements simultaneously held in the working memory.
- Crystallized intelligence (gc)—the ability to deduce secondary relational abstractions by applying primary relational abstractions to each other. But the deduced relations among relations must be checked by fluid intelligence.
Fluid intelligence (which is intelligence proper) is determined by the capacity of the working memory, localized in the prefrontal cortex, which degenerates faster than other cortical regions in the course of aging and encephalopathies. Fluid intelligence peaks at age 20 and then declines, so that the human becomes increasingly pseudointelligent and machinelike.
Crystallized intelligence (which is machinelike pseudointelligence) reaches a plateau by age 28 and begins to decline at age 53 (crystallized intelligence declines a few times slower than fluid intelligence).
Fluid and crystallized intelligence were originally identified by Raymond Cattell. Concepts of fluid and crystallized intelligence were further developed by Cattell's student, John L. Horn. Since Cattell's and Horn's publications, the concepts of fluid and crystallized intelligence have become so ingrained in the field of intelligence that they are no longer routinely attributed to Cattell or Horn—much as Cattell's scree plot became ingrained in the practice of factor analysis or Freud's concept of the subconscious is ingrained in psychology and in the public's perceptions of the mind.
Fluid versus crystallized
Fluid intelligence is the capacity for the intuitive (irrational) feeling of the relations between objects by focusing attention on those objects not consecutively but simultaneously (which requires a sufficiently large working memory). The observer mentally merges with the observed system of objects, and is thus able to feel their interrelations. The direct eduction of relations is also known as grokking.
Holding a complex relational model in the working memory requires a deep isolation from the decohering influence of the environment. That is why a superior fluid intelligence is an attribute of individuals with Asperger syndrome, to which belong Isaac Newton and all those who have a talent for understanding complex systems.
Each type of crystallized intelligence is independent of the other (increasing a student's proficiency in Latin does not increase the student's proficiency in algebra), but to a high degree dependent on the individual's fluid intelligence (students proficient in Latin tend to be proficient in algebra, too, because people with a high gf tend to acquire more gc-knowledge and at faster rates), and, to a lesser degree, on the quality of the learning environment (e.g., the quality of available books).
Some researchers have linked the theory of fluid and crystallized intelligences to Piaget's concept of figurative (statically intuitive) and operative (stepwisely logical) intelligences (see Piaget's theory of cognitive development#Pre-operational stage). Fluid intelligence and Piaget's figurative intelligence both concern inductive reasoning (the statically intuitive eduction of relations). Crystallized intelligence and Piaget's operative intelligence both concern deductive reasoning (the stepwisely logical manipulation of relations). But in Piaget's theory, statically intuitive figurative intelligence is subordinate to stepwisely logical operative intelligence, whereas in Cattell's theory, stepwisely logical crystallized intelligence is disparaged as machinelike pseudointelligence.
Fluid intelligence generally correlates with measures of abstract reasoning and puzzle solving. Crystallized intelligence correlates with abilities that depend on knowledge and experience, such as vocabulary, general information, and analogies. Paul Kline identified a number of factors that shared a correlation of at least r=.60 with Gf and Gc. Factors with [clarify] of greater than 0.6 on Gf included induction, visualization, quantitative reasoning, and ideational fluency. Factors with median loadings of greater than 0.6 on Gc included verbal ability, language development, reading comprehension, sequential reasoning, and general information. It may be suggested that tests of intelligence may not be able to truly reflect levels of fluid intelligence. Some authors have suggested that unless an individual was truly interested in the problem presented, the cognitive work required may not be performed because of a lack of interest. These authors contend that a low score on tests which are intended to measure fluid intelligence may reflect more a lack of interest in the tasks rather than inability to complete the tasks successfully.
Measurement of fluid intelligence
There are various measures that assess fluid intelligence. The Cattell Culture Fair IQ test, the Raven Progressive Matrices (RPM), and the performance subscale of the WAIS are measures of Gf. The RPM is one of the most commonly used measures of fluid abilities. It is a non-verbal multiple choice test. Participants have to complete a series of drawings by identifying relevant features based on the spatial organization of an array of objects, and choosing one object that matches one or more of the identified features. This task assesses the ability to consider one or more relationships between mental representations or relational reasoning. Propositional analogies and semantic decision tasks are also used to assess relational reasoning.
Standardized IQ tests such as those used in psychoeducational assessment also include tests of fluid intelligence. In the Woodcock-Johnson Tests of Cognitive Abilities, Gf is assessed by two tests: Concept Formation (Test 5) in the Standard Battery and Analysis Synthesis (Test 15) in the Extended Battery. On Concept Formation tasks, the individual has to apply concepts by inferring the underlying "rules" for solving visual puzzles that are presented in increasing levels of difficulty. Individuals at the preschool level have to point to a shape that is different from others in a set. As the level of difficulty increases, individuals increasingly demonstrate an understanding of what constitutes a key difference (or the "rule") for solving puzzles involving one to one comparisons, and on later items identifying common differences among a set of items. For more difficult items, individuals need to understand the concept of "and" (e.g. solution must have some of this and some of that) and the concept of "or" (e.g. to be inside a box, the item must be either this or that). The most difficult items require fluid transformations and cognitive shifting between the various types of concept puzzles that the examinee has worked with previously.
Concept Formation tasks assess inductive reasoning ability. In the Analysis-Synthesis test, the individual has to learn and orally state the solutions to incomplete logic puzzles that mimic a miniature mathematics system. The test also contains some of the features involved in using symbolic formulations in other fields such as chemistry and logic. The individual is presented with a set of logic rules, a "key" that is used to solve the puzzles. The individual has to determine the missing colors within each of the puzzles using the key. Complex items present puzzles that require two or more sequential mental manipulations of the key to derive a final solution. Increasingly difficult items involve a mix of puzzles that require fluid shifts in deduction, logic, and inference. Analysis Synthesis tasks assess general sequential reasoning.
In the Wechsler Intelligence Scale for Children-IV (WISC IV), the Perceptual Reasoning Index contains two subtests that assess Gf: Matrix Reasoning, which involves induction and deduction, and Picture Concepts, which involves induction. In the Picture Concepts task, children are presented a series of pictures on two or three rows and asked which pictures (one from each row) belong together based on some common characteristic. This task assesses the child's ability to discover the underlying characteristic (e.g. rule, concept, trend, class membership) that governs a set of materials. Matrix Reasoning also tests this ability as well as the ability to start with stated rules, premises, or conditions and to engage in one or more steps to reach a solution to a novel problem (deduction). In the Matrix Reasoning test, children are presented a series or sequence of pictures with one picture missing. Their task is to choose the picture that fits the series or sequence from an array of five options. Since Matrix Reasoning and Picture Concepts involve the use of visual stimuli and do not require expressive language, they are considered to be non-verbal tests of Gf.
Within the corporate environment, fluid intelligence is a predictor of a person's capacity to work well in environments characterised by complexity, uncertainty, and ambiguity. The Cognitive Process Profile (CPP) measures a person's fluid intelligence and cognitive processes. It maps these against suitable work environments according to Elliott Jacques Stratified Systems Theory.
Development and physiology
Fluid intelligence, like reaction time, typically peaks in young adulthood and then steadily declines. This decline may be related to local atrophy of the brain in the right cerebellum. Other researchers have suggested that a lack of practice, along with age-related changes in the brain may contribute to the decline. Crystallized intelligence typically increases gradually, stays relatively stable across most of adulthood, and then begins to decline after age 65. The exact peak age of cognitive skills remains elusive, depending on the skill measurement as well as on the survey design. Cross-sectional data shows typically an earlier onset of cognitive decline in comparison with longitudinal data. The former may be confounded due to cohort effects while the latter may be biased due to prior test experiences.
Improving fluid intelligence with training on working memory
According to David Geary, Gf and Gc can be traced to two separate brain systems. Fluid intelligence involves both the dorsolateral prefrontal cortex, the anterior cingulate cortex, and other systems related to attention and short-term memory. Crystallized intelligence appears to be a function of brain regions that involve the storage and usage of long-term memories, such as the hippocampus.
Some researchers question whether the results of training are long lasting and transferable, especially when these techniques are used by healthy children and adults without cognitive deficiencies. A meta-analytical review conducted by researchers from the University of Oslo in 2012 concluded that "memory training programs appear to produce short-term, specific training effects that do not generalize."
In a study using four individual experiments, 70 participants (36 of them female, all with a mean age of 25.6) recruited from the University of Bern community, Susanne M. Jaeggi and her colleagues at the University of Michigan found that healthy young adults who practiced a demanding working memory task (dual n-back, a task that has strong face validity, has received some criticism regarding its construct validity and is in widespread use as a measure of working memory) approximately 25 minutes per day for between 8 and 19 days had statistically significant increases in their scores on a matrix test of fluid intelligence taken before and after the training than a control group who did not do any training at all.
A further examination of these findings was published in 2008 in the Proceedings of the Nation Academy of Sciences of the United States of America. Summarizing findings in the study as evidence that demonstrates that "fluid intelligence is trainable to a significant and meaningful degree."
Attention is drawn to the limitations of these results and the need for specific follow up inquiery. Robert J. Sternberg comments that"it is unclear to what extent the results can be generalized to other working-memory tasks" and states "it would be useful to show that the training transfers to success in meaningful behaviours that extend beyond the realm of psychometric testing". Sternberg asserts that ability level of the test participants is not necessarily examining a wide range of ability levels, or "address whether the training is durable over extended periods of time [and not only] "fleeting."
A second study conducted at the University of Technology in Hangzhou, China, supports Jaeggi's results independently. After student subjects were given a 10-day training regimen based on the dual n-back working memory theory, the students were tested on Raven's Standard Progressive Matrices. Their scores were found to have increased significantly.
Subsequent studies on n-back, namely by Chooi & Thompson and Redick et al., do not support the findings of the Jaeggi study. Although participants' performance on the training task improved, these studies showed no significant improvement in the mental abilities tested, especially fluid intelligence and working memory capacity.
- 21st century skills
- CHC theory
- Deeper learning
- General intelligence factor
- Three stratum theory
- Malleability of intelligence
- Outline of human intelligence
- Raymond Cattell
- Spatial intelligence (psychology)
- Sternberg, Robert J. Handbook of Human Intelligence. CUP, 1982, p. 75. "A different approach to the problem was implicit in the work of R. B. Cattell (1971). Some years ago, Cattell (1943) called attention to a possible distinction between what he called ‘fluid’ intelligence and ‘crystallized’ intelligence, the former representing basic capacity and the latter representing abilities acquired through learning, practice, and exposure to education."
- Cattell, R. B. (1971). Abilities: Their structure, growth, and action. New York: Houghton Mifflin. ISBN 0-395-04275-5.
- Cattell, R. B. Intelligence: Its Structure, Growth and Action. Elsevier, 1987, p. 294. "Their tendency to show up as two distinct primary factors, and for men to be better than machines on induction, is presumably due to the special processes involved in the initial ‘eduction of relations’ between given fundaments. Once these relationships are educed, i.e., abstracted, they can be manipulated, as relationships just as in deductive reasoning. Indeed, as we have seen, the manipulation of these deductive relationships can be relatively mechanical, so that in respect to comprehensiveness of conclusions and infallibility of inference a child taught the rules of algebra can apply them to reach conclusions more quickly and accurately than a wise adult unschooled in algebra. One sees here the rise of what is behaviorally a pseudointelligence, and anatomically a capacity to handle complex relations with very little neural storage mass."
- Cattell, R. B. Intelligence: Its Structure, Growth and Action. Elsevier, 1987, p. 294. "For this reason, we would predict that inductive reasoning should load gf more than gc, relative to deductive reasoning. However, the results of deductive reasoning, including those of mathematics, though mechanically reached, are in fact never accepted at face value without checking them in other ways. The mathematician, as distinct from the idiot savant of rapid calculation, is never far in his formulae from his degree of direct insight. Indeed, as Poincare (1914) pointed out, he often reaches a conclusion first by ‘intuitive’ insights and then builds up the mechanics of formal derivation afterward. Consequently, we should not expect the crystallized intelligence of man, by any mechanical manipulation of its stored relational abstractions, far to exceed its fluid intelligence capacity for direct relation-perception. At least this would be rare enough not to upset the usual correlation picture in the general population. With the machine, on the other hand, which has no reluctance to exceed its insight, crystallized intelligence could acquire more of a life of its own, and mechanically proliferate relations among relations. Since the laws of the manipulation are only known for lower levels, much of the product might be nonsense."
- Dehn, Milton J. Long-Term Memory Problems in Children and Adolescents. John Wiley & Sons, 2010, p. 69. "The functioning of core working memory processes, namely executive processes, is thought to reside in the prefrontal cortex."
- Fuster, Joaquin. The Prefrontal Cortex. Elsevier, 2008, p. 44. "In the prefrontal cortex of the normal human subject, involutional signs usually appear in the seventh or eighth decade of life. They include volume loss (of gray matter, as well as individual neurons), atrophy of dendrites, and loss of synaptic spines. The prefrontal cortex leads most cortical areas in morphological aging. It is also one of the most vulnerable to the neurodegenerative changes that take place in dementias."
- Cacioppo, John T.; Freberg, Laura. Discovering Psychology: The Science of Mind. Cengage Learning, 2012, p. 448
- Papalia, Diane E. Instructor's manual to accompany human development. McGraw-Hill, 1978, p. 207. "Many studies of intelligence have shown that intellectual performance peaks at the age of 20 and then declines."
- Cattell, R. B. (1971). Abilities: Their structure, growth, and action. New York: Houghton Mifflin. ISBN 978-0-395-04275-5.[page needed]
- Google Reveals Blueprint for Quantum Supremacy. MIT Technology Review, 4 October 2017. "But surpassing the limits of conventional computing—achieving quantum supremacy, as physicists call it—has turned out to be more difficult than everyone expected. Quantum states are delicate objects—sneeze and they vanish. Because of this, physicists have become bogged down with the practical challenge of isolating quantum computers and their quantum processing machinery from the outside world."
- Hayashi, Mika; Kato, Motoichiro; Igarashi, Kazue; Kashima, Haruo (2008). "Superior fluid intelligence in children with Asperger's disorder". Brain and Cognition. 66 (3): 306–10. doi:10.1016/j.bandc.2007.09.008. PMID 17980944.
- Soulières, Isabelle; Dawson, Michelle; Gernsbacher, Morton Ann; Mottron, Laurent (2011). Skoulakis, Efthimios M. C (ed.). "The Level and Nature of Autistic Intelligence II: What about Asperger Syndrome?". PLoS ONE. 6 (9): e25372. Bibcode:2011PLoSO...625372S. doi:10.1371/journal.pone.0025372. PMC 3182210. PMID 21991394.
- Dawson, M.; Soulieres, I.; Ann Gernsbacher, M.; Mottron, L. (2007). "The Level and Nature of Autistic Intelligence". Psychological Science. 18 (8): 657–62. doi:10.1111/j.1467-9280.2007.01954.x. PMC 4287210. PMID 17680932.
- Muir, Hazel. Einstein and Newton showed signs of autism. New Scientist, 30 April 2003. "Autism is heritable, and there are clues that the genes for autism are linked to those that confer a talent for grasping complex systems—anything from computer programs to musical techniques. Mathematicians, engineers and physicists, for instance, tend to have a relatively high rate of autism among their relatives."
- Cattell, R. B. Intelligence: Its Structure, Growth and Action. Elsevier, 1987, pp. 138–39
- Papalia, D.; Fitzgerald, J.; Hooper, F. H. (1971). "Piagetian Theory and the Aging Process: Extensions and Speculations". The International Journal of Aging and Human Development. 2: 3–20. doi:10.2190/AG.2.1.b.
- Schonfeld, Irvin S. (1986). "The Genevan and Cattell-Horn conceptions of intelligence compared: Early implementation of numerical solution aids". Developmental Psychology. 22 (2): 204–12. doi:10.1037/0012-16220.127.116.11.
- Voyat, Gilbert. Piaget Systematized. Psychology Press, 1982, p. 5. "This is how and why Piaget distinguishes between these two aspects of thinking: The figurative aspect deals with the description of states, concerns perception, and entails schemata; the operative aspect, on the other hand, provides understanding of pathways from one state to another, referred to as schemes of action and abstraction. Before he has built adequate logical operations, the child thinks only in terms of the states; he thinks in an essentially figurative, preoperational, or intuitive way."
- Kline, P. (1998). The new psychometrics: Science, psychology and measurement. London: Routledge.[page needed]
- Messick, Samuel (1989). "Meaning and Values in Test Validation: The Science and Ethics of Assessment". Educational Researcher. 18 (2): 5–11. doi:10.3102/0013189X018002005. JSTOR 1175249.
- Raven, J.; Raven, J. C.; Court, J. H. (2003) . "Section 1: General Overview". Manual for Raven's Progressive Matrices and Vocabulary Scales. San Antonio, TX: Harcourt Assessment.[page needed]
- Bornstein, Joel C.; Foong, Jaime Pei Pei (2009). "MGluR1 Receptors Contribute to Non-Purinergic Slow Excitatory Transmission to Submucosal VIP Neurons of Guinea-Pig Ileum". Frontiers in Neuroscience. 3: 46. doi:10.3389/neuro.21.001.2009. PMC 2695390. PMID 20582273.
- Wright, Samantha B.; Matlen, Bryan J.; Baym, Carol L.; Ferrer, Emilio; Bunge, Silvia A. (2007). "Neural correlates of fluid reasoning in children and adults". Frontiers in Human Neuroscience. 1: 8. doi:10.3389/neuro.09.008.2007. PMC 2525981. PMID 18958222.
- Ferrer, Emilio; O'Hare, Elizabeth D.; Bunge, Silvia A. (2009). "Fluid reasoning and the developing brain". Frontiers in Neuroscience. 3 (1): 46–51. doi:10.3389/neuro.01.003.2009. PMC 2858618. PMID 19753096.
- Woodcock, R. W.; McGrew, K. S.; Mather, N (2001). Woodcock Johnson III. Itasca, IL: Riverside.[page needed]
- Schrank, F. A.; Flanagan, D. P. (2003). WJ III Clinical use and interpretation. Scientist-practitioner perspectives. San Diego, CA: Academic Press.[page needed]
- Wechsler, D. (2003). WISC-IV technical and interpretive manual. San Antonio, TX: Psychological Corporation.[page needed]
- Flanagan, D. P.; Kaufman, A. S. (2004). Essentials of WISC-IV assessment. Hoboken, NJ: John Wiley.[page needed]
- Lee, Jun-Young; Lyoo, In Kyoon; Kim, Seon-Uk; Jang, Hong-Suk; Lee, Dong-Woo; Jeon, Hong-Jin; Park, Sang-Chul; Cho, Maeng Je (2005). "Intellect declines in healthy elderly subjects and cerebellum". Psychiatry and Clinical Neurosciences. 59 (1): 45–51. doi:10.1111/j.1440-1819.2005.01330.x. hdl:10371/27902. PMID 15679539.
- Cavanaugh, J. C.; Blanchard-Fields, F (2006). Adult development and aging (5th ed.). Belmont, CA: Wadsworth Publishing/Thomson Learning. ISBN 978-0-534-52066-3.[page needed]
- Desjardins, Richard; Warnke, Arne Jonas (2012). "Ageing and Skills". OECD Education Working Papers. doi:10.1787/5k9csvw87ckh-en. hdl:10419/57089.
- Kyllonen, Patrick C.; Christal, Raymond E. (1990). "Reasoning ability is (little more than) working-memory capacity?!". Intelligence. 14 (4): 389–433. doi:10.1016/S0160-2896(05)80012-1.
- Geary, D. C. (2005). The origin of mind: Evolution of brain, cognition, and general intelligence. Washington, DC: American Psychological Association.
- Todd W. Thompson; et al. (2013). "Failure of Working Memory Training to Enhance Cognition or Intelligence". PLoS ONE. 8 (5): e63614. doi:10.1371/journal.pone.0063614. PMC 3661602. PMID 23717453.
- Melby-Lervåg, Monica; Hulme, Charles (2012). "Is Working Memory Training Effective? A Meta-Analytic Review". Developmental Psychology. 49 (2): 270–91. doi:10.1037/a0028228. PMID 22612437.
- Jaeggi, Susanne M.; Buschkuehl, Martin; Jonides, John; Perrig, Walter J. (2008). "Improving fluid intelligence with training on working memory". Proceedings of the National Academy of Sciences. 105 (19): 6829–33. Bibcode:2008PNAS..105.6829J. doi:10.1073/pnas.0801268105. JSTOR 25461885. PMC 2383929. PMID 18443283.
- Sternberg, R. J. (2008). "Increasing fluid intelligence is possible after all". Proceedings of the National Academy of Sciences of the United States of America. 105 (19): 6791–6792. doi:10.1073/pnas.0803396105. PMC 2383939. PMID 18474863.
- Qiu, Feiyue; Wei, Qinqin; Zhao, Liying; Lin, Lifang (2009). "Study on Improving Fluid Intelligence through Cognitive Training System Based on Gabor Stimulus". 2009 First International Conference on Information Science and Engineering. pp. 3459–62. doi:10.1109/ICISE.2009.1124. ISBN 978-1-4244-4909-5.
- Chooi, Weng-Tink; Thompson, Lee A. (2012). "Working memory training does not improve intelligence in healthy young adults". Intelligence. 40 (6): 531–42. doi:10.1016/j.intell.2012.07.004.
- Redick, Thomas S.; Shipstead, Zach; Harrison, Tyler L.; Hicks, Kenny L.; Fried, David E.; Hambrick, David Z.; Kane, Michael J.; Engle, Randall W. (2012). "No Evidence of Intelligence Improvement After Working Memory Training: A Randomized, Placebo-Controlled Study". Journal of Experimental Psychology: General. 142 (2): 359–379. doi:10.1037/a0029082. PMID 22708717.