Horn, J.L. & Blankson, N.. (2006). Foundations for Better Understanding of Cognitive Abilities. In D.P. Flanagan & P.L. Harrison (Eds.), Contemporary Intellectual Assessment: Theories, tests, and issues-2nd Edition (pp. 41-76). New York: Guilford Press.
I learned yesterday that John Horn of CHC theory had passed away. I was saddened by this news, given his significant and influential work related to cognitive theory and assessment. I was looking forward to the opportunity of possibly meeting him at NASP this year and regret that will not come to pass. One way I thought I could pay tribute is to review a few highlights of his contribution to the second edition of Contemporary Intellectual Assessment. There are lots of articles that I’m looking forward to commenting on, but if anyone reads this overview and decides to purchase the book or change their practice in some way for a student, then the post is worthwhile.
- Gf/Gc theory is wrong…this is how the text begins. Horn & Blankson do an excellent job in reminding us that when it comes to cognition, there are variables that we cannot control for and therefore can never engender a true experimental study. We can only look at analyses and associations and make hypotheses and predictions. There can be a tremendous temptation to assert ourselves based on our knowledge and the existing research. Although Horn & Blankson contend Gf/Gc is probably the best of what we have right now, researchers will continue to refine and improve the theoretical framework. I am reminded of the additional abilities that Kevin McGrew described in the chapter on CHC Theory-Past, Present, and Future (in the same volume) and the evolutions of the CHC taxonomy. We must remember to view what we do not dogmatically, but with the notion that with future learning, our goal is to minimize as much of our own error as possible.
- Many of us in our intro psychology books learned about Gf/Gc, specifically that Gf declines with age, but Gc does not. Some great tenets never change. Horn & Blankson expanded on this, focusing on studies that revealed that crystallized intelligence (Gc) and Tertiary Storage and Retrieval (TSR-now known as Glr-Long Term Storage and Retrieval in contemporary CHC theory) only maintain or improve with age. Three other factors (Fluid Reasoning (Gf), Short-Term Memory (SAR-now known as Gsm in contemporary theory) and processing speed (Gs) have been shown to decline with age.
- In addition to the above, Horn & Blankson lump the abilities into three distinct areas: Vulnerable abilities (Gf, SAR, and GS), Expertise Abilities ((Gc, TSR, and Gq), and Sensory-Perceptual Abilities (Gv and Ga).
- To g or not to g: Those who are familiar with CHC theory are likely also familiar with the spirited debate between Carroll’s notion of “g” exists, to Horn’s notion of “g does not exist” (link). Horn & Blankson continue to provide an argument of the nonexistence of ‘g’. They argue thaqt although intercorrelations and commonalities are present between the broad factors Spearman meant the broad abilities to contribute uniquely to ‘g’, without shared variance. Although this has been examined, it has not been able to be established.
- Horn & Blankson also noted that if there was to be such a thing a ‘g’, the closest ability to it is Gf (or fluid reasoning). To quote: “The common factor that was separate from other factors at the second order and identical with a factor identified at the third order in Gustafsson’s (1984) study was interpreted as Gf. This factor corresponds most closely to the construct Spearman described (p. 53).
- Gs, Gf &Gsm-MW, and expertise: Horn & Blankson describe a compelling new way to look at Gs. Although they note that Gs declines with age, measures of responding “as slowly as possible” seem to correlate well with Gs. In other words, people who have trouble performing quickly also have trouble performing deliberately very slowly. The common variable is the amount of attention required to do both. Interestingly, the ability to deliberately be slow also declines with age. Can it be that Gs instead might be related to executive functioning? It’s an intriguing prospect.
- There have been some excellent discussions on the interrelationship of working memory with fluid reasoning (link). Although some might not see working memory as an ability that takes a great deal of intelligence and knowledge, on working memory tasks one must be able to process simultaneous information in order to solve various types of complicated puzzles-particularly those with multiple levels. Horn & Blankson discuss that working memory seems to be much more negatively impacted by age than memory span AND Gf also declines with age--the interrelationship is noteworthy.
- Probably one of the most important notions that Horn & Blankson discuss is that of expertise. They note that although Gc increases with age, most intelligence tests only use survey level knowledge in a variety of areas, but none ever get to the depth of the expert (my own example I might give is that an item that might tap this to the expertise level is to display a painting that is not extremely famous (such as The Mona Lisa or A Starry Night), but one that is a bit more obscure and to have the subject identify the painter, time period, and style of painting.) Interestingly, an English and History double major Master’s level professor may score substantially better than a nuclear engineer!
- Horn & Blankson further the concept of expertise by indicating that the expert’s skills do not necessarily decline with age. They note a study where Gf was related to novice crossword achievement (r = .72) and negatively correlated with age ( r = -.25). However, the expert crossword achiever showed a .24 correlation with age and a near zero correlation with Gf. Furthermore, they note the working memory of the expert to be well preserved depending on how it is tapped (for example, someone who is older may still excel at blindfolded chess, which requires them not only to visualize the pieces, but hold them in working memory.)
- Conclusions: I have to weigh in on the ‘g’ issue, although I worry about being a “loud stream” (users of the WAIS-R will get that one). My own thought is that ‘g’ exists because the whole is greater than the sum of its parts. It does follow well below average ability in a variety of skills will result in an even lower ‘g’ than those skills and the opposite for well above average. Certainly doing well or poorly in most areas seems to relate to a measure of global functioning. However, that said, I still see it as relatively unimportant. At the end of the day, our job is to determine the different strengths and weaknesses our subjects have and how that compares to academic, adaptive, or occupational functioning.
- The chapter by Horn & Blankson is profound in its implications. First, they note that there are a large number of caveats to the long held understanding of which cognitive abilities deteriorate with age. We’ve had a lot of spirited discussions on the IAPCHC listserv about how many of our tests do not tap skills of some of our most intelligence members of society. Horn & Blankson tackle this issue directly. Furthermore, the notion of how the “expert” deviates from the rule of most intelligence tests (given the contrary path their ability trajectories seems to follow over time against nonexpert subjects) gives greater credence to the notion that those things we think we know may not be.
powered by performancing firefox