Is intelligence testing, intelligent?


We are all familiar with the term intelligence. We constantly use the word in our daily lives, and it appears to be prevalent across centuries, paradigms and cultures, accessible to young and old alike, and its possession valued greatly among society. Given its universality, accessibility and implications to every day life, it is no surprise that intelligence is also one of the most vastly researched psychological constructs. This has given rise to hundreds of tests aiming to normally distribute and differentiate across different levels of intelligence. However, a paradoxical qualm arises, as the definite conceptualization of intelligence still remains allusive to the human mind. Surprisingly, we are vastly better at measuring intelligence than at conceptualizing it! (1)

Unfortunately, throughout the history of empirical psychological measurement, intelligence has evoked an operational definition. Since its inception, its conceptualization emerged from the way in which it was measured and not the extremely more useful real definition of the construct. This has led to a current state of affaires, where several theories of intelligence are in combat, leading to different ways of measuring their definition of “intelligence”. There seems to have evolved a prevalent debate regarding the underlying breakdown of an individual’s intelligence, and whether there appears to be one fundamental factor determining all aspects of ability (as Spearman’s G proposes) or many different forms of intelligences all independent of each other (Gardner’s 7 initial intelligences). (2) Despite this ambiguity of definitions and subsequent tests, one prevalent predisposition arises among all of them, and that is their unfortunate focus on verbal and logical-mathematical skills. I believe this fixation has vastly defined the academic curriculum and has classified individuals for over a century now on one of the many facets of intelligence, a somewhat unintelligent view of measuring the construct.

Furthermore, as with any test, results suffer from the strength of its underlying construct and conceptualization. However, there are a limited number of psychological constructs that have experienced such divergent and seemingly almost contradicting definitions as intelligence (refer here for a list of definitions). As such, it is imperative to take any result on an intelligence test (mainly in the form of IQ) with a grain of salt. One must recognize that it represents a certain context, culture and often a limited view of what it means to be “intelligent”. I recognize that there is extensive evidence regarding the predictions that higher IQ have in terms of positive life outcomes, but I believe our past held beliefs about intelligence and its measurement methods have predisposed us to a sub-optimal fixation, negating progress in getting to the core of intelligence itself, and not how it has been measured. (3)

This arises the next logical question: Can you really measure intelligence? Luckily, in recent years we have evolved from Boring’s classical definition of “intelligence is what is measured by the test” to more contemporary views such as those of Pinter, who defines it is an ability to adapt. (4) This new definition arises important considerations, particularly in the field of measurement. If intelligence is defined as the ability to adapt and learn, wouldn’t the completion of a test by an individual render it forever obsolete? Furthermore, if learning occurs through experience, wouldn’t it be useless to assume an equivalent level of education or experience for everyone?


Most importantly, if we let intelligence be defined by the ways in which it is measured, and directly correlate intelligence to a test grade, are we not indirectly limiting the scope and breadth of knowledge a certain generation should acquire by solely valuing learning in a strictly defined (test) domain? These questions were not necessary based on our past constructs, but as we adopt new beliefs, we must also adopt new testing methods. Most importantly, are we trying to amass several different aspects of an individual under one global term, and as a result, failing to comprehend it?

Lastly, technology has also provided a unique opportunity to reexamine our definitions of intelligence, as we enable the possibility of non-sentient entities to acquire artificial intelligence, surpassing in some cases that of humans, and increasing our own capabilities. If our capabilities increase via AI does our intelligence do so as well? As a result, does the term intelligence mean the same thing for a human to a non-human machine?

Our society is changing at an accelerating state, and those skills that were once considered useful are no longer so. Technology is playing a key role in this process and is transforming the ways in which we think and interact. If our constructs are evolving with these views, our tests must not fall behind. The 21st century should prove the be the era in which, for the first time, intelligence testing becomes intelligent.


— — — — — — — — — — — — — — — — — — — — — — — — — — — —


(1) Robert J. Gregory (2016). Psychological Testing, Updated 7th edition, p. 15

(2) “Can You Really Measure Intelligence?” Udara Jay. N.p., 15 Feb. 2016. Web. 21 Mar. 2016. <>.

(3) “What Does IQ Really Measure?” What Does IQ Really Measure? N.p., 25 Apr. 2011. Web. 21 Mar. 2016. <>.

(4) Bulk Collection of Signals Intelligence.” (2015): n. pag. Web. <>