Every season, hundreds of thousands of hopeful contestants audition for the myriad of television talent shows that exist. Whatever you think you can do, there seems to be a reality talent show where you can try and prove it.

To make it far in these shows, you have to possess an "X Factor", a certain unidentifiable quality that makes you stand out from the rest (in a positive way).

What is this elusive X-Factor? Here I entertain the possibility that the reason why the X-Factor may be so rare, and so elusive, is because of its complex genetic underpinnings.

The X-Factor is actually not one trait or characteristic, but the combination of multiple traits. Even though singing competitions such as American Idol and The X Factor are ostensibly about singing talent, those who have what the judges refer to as the X-Factor go beyond just good singing talent. For instance, on this past season of American Idol, contestants were allowed to play a musical instrument while they sang, and many took the opportunity.

In fact, manifestations of talent in any complex domain depend on multiple traits, and those with the X-Factor in those domains are those with that special combination of these traits. Many traits that make up the X-Factor have genetic underpinnings. Human beings differ quite a bit from each other on many dimensions, such as cognitive capacity, height, physical attractiveness, motivation, personality, and values to name just a few. Behavioral geneticists, using quite sophisticated statistical techniques applied to both adoption and twin data have shown that most of these traits have fairly large heritability coefficients, meaning a sizeable proportion of the variance in a certain trait in a particular population at a particular point in time is the result of genes (note this doesn't rule out the effect of the environment).

(Fun fact: among the traits that have sizeable heritability coefficients are political attitudes, musical tastes, time devoted to watching TV, and religious interests.)

Now how does this elucidate the elusive X-Factor? My esteemed colleague Dean Keith Simonton [1] offers a nuanced genetic model of talent that I think is relevant. Simonton has argued that additive models of talent are too simplistic (see last post for an additive model of music talent). It's too simple to say that practice + music ability + high IQ equals musical ability. No, Simonton says that talent, especially in complex domains, is better represented by a multidimensional and multiplicative model. Let's look at an example to make this more concrete.

For simplicity, let's say that the X-Factor for American idol consists of a combination of 5 traits: 1. singing ability, 2. appearance, 3. personality, 4. stage presence, and 5. charisma. Let's further assume that each trait varies along a scale where 0 represents the complete absence of the trait from the person's genotype. Also, it is assumed that each one of these traits is in itself a result of numerous genes that can be inherited in any combination (what is referred to by behavior geneticists as "polygenic").

Let's then place a value on each trait for each contestant. Here is a hypothetical example using two actual American Idol contestants:

1 2 3 4 5 Total

Ruben Studdard 40 5 20 25 30 300,000

Man dressed as Statue of Liberty 0 20 15 5 10 0

Each of the contestant's value on the five traits that comprise the X-Factor are multiplied together to produce that individual's total X-Factor. This multiplicative nature of the X-Factor means that if any single essential component is missing (has a value of 0), then that person will not have the X-Factor, because the total will be 0, regardless of the value of the rest of the components (this idea has been named emergenic by Lykken [2]). In this example, Ruben Studdard's X-Factor score is 300,000, high enough for him to win the entire competition, whereas the guy dressed as statue of liberty is missing one of the essential components so receives a total X-Factor score of 0 regardless of his score on the other components. We have all seen the American Idol contestant who may have extraordinary singing ability but has a value of 0 on personality. They just stand there and sing, and get chastised by Simon (not to be confused with Simonton). Their utter lack of one crucial ingredient of the X-Factor causes their total X-Factor score to be zero and they are not identified as having the X-Factor. It is also entirely possible that there is a minimum X-Factor score for a participant to make it to the Hollywood round. Whereas a total score of 300,000 may make Ruben a winner, a score of 100,000 may be all the judges are looking for at the early stages of the competition to proceed.

If this model is correct, then you would expect identical twins, who share 100% of the same genes, and thus the same genetic values on each of the essential components of the X-Factor, to be very likely to have similar manifestations of the X-Factor, whereas you would expect fraternal twins (who only share half of their genes) to be very unlikely to have the same X-Factor, because there would be a high probability that at least one essential genetic component is missing.

This is in fact exactly what we find. Take the trait "expressive control", that involves an individual's ability to impress and entertain people, to engage effectively in role-playing, to mimic other persons, and to practice deception successfully. This is clearly an important element of the X-Factor in a variety of entertainment fields. Identical twins (reared apart or together) are correlated a whopping .76 on this trait, whereas fraternal twins are only correlated .16--no more alike than any two people selected randomly from the population [3]! Similar patterns have been found for the ability to influence others (referred to as 'social potency', [4]) and for a creative personality [5], suggesting that many traits that are important components of the X-Factor in many fields are most likely emergenic.

This model may also explain why the X-Factor is so elusive. This is because is it may be the case that the while the components of an X-Factor is theoretically identifiable, there are an infinite amount of ways the components can combine. For American Idol, contestants can vary in their value for each trait, but all have the same X-Factor score. For instance, the ones who make it to Hollywood round may all have the same X-Factor total score, but may all differ in the weights for each trait. Thus, the X-Factor still remains a mystery.

This multiplicative model may also explain why the X-Factor is so rare in the general population. Under a multiplicative model of talent, at the lowest end of the bell curve on the X-Factor would exist the largest proportion of the population. Thus, the hundreds of thousands of hopefully contestants that are sent home immediately. At the upper end would be those few individuals who are several standard deviations above the population mean in their endowment of each of the components. Under a more simple additive model, the X-Factor would be normally distributed in the general population. But this isn't the case.