Are Atheists Mutants? The Left Hand of Daftness

As discussed in two previous posts, a recent paper (Dutton, Madison, & Dunkel, 2017) argues that belief in a moral god has been selected for by evolution in civilized societies, and that deviations from this belief, such as atheism are likely to be caused adverse genetic mutations that have become prevalent in modern times due to relaxation of selective pressures. The authors argued that these “deviations” should be correlated with markers of mutational load, including poor health, left-handedness, autism, and fluctuating asymmetry. However, the evidence they provide does not support their claims. My previous blogpost discussed the apparent relationship between religiosity and health and showed that, as this was dependent on environmental and cultural factors, it was unlikely to be an indicator of mutation load. Furthermore, the authors presented no evidence that atheism as such is associated with poorer health compared to being religious. In this post, I discuss the other supposed indicators, particularly left-handedness, and show once more that the evidence does not support the authors’ claims.

As I noted in a previous post, Dutton et al. begin their paper with the rather unusual step of citing passages from the Bible that supposedly suggest that believers are healthier and more attractive, while, in contrast “Those inspired by other gods or by Satan are, in contrast, autistic (Mark 9:25) and even left-handed (Matt 25: 41). They subtly imply that there might be some reason that the authors of these books mentioned these features specifically. They then go on to state that the theory they are proposing “may be in line with some of these observations.” Specifically, “that there has been deleterious mutation accumulation since the Industrial Revolution, as a function of relaxed selection on many health and fitness-related dimensions.” Ironically, some of the Bible verses they have selected do not even mention the features they focus on. For example, consider Matt 25: 41, which supposedly says something about left-handed people. This is the verse in question: “Then shall he say also unto them on the left hand, Depart from me, ye cursed, into everlasting fire, prepared for the devil and his angels” (King James version). Note that this verse does not say anything about being left-handed, it simply refers to people located to the left side of the speaker, presumably because the left is symbolically associated with evil and misfortune. Additionally, I noted in a previous post that the verse supposedly referring to autism actually refers to a person suffering from demonic possession. Interpreting this as saying something about autism (considering the symptoms described, the person might have had epilepsy for all we know) or general characteristics of “those inspired by other gods or by Satan” is a wildly speculative interpretation. As I will show, the authors handling of scientific evidence is not that much better than how they interpret Bible verses.

Even in the realm of science, one must be on the lookout for sharp practices...

Source: Wikimedia Commons

The authors suggest that left-handedness is associated with developmental instability (i.e. problems with a person’s physical and mental development), and argue that because left-handedness has a heritability of .25, it “might therefore also be associated with mutation load.” Note that the heritability of most human characteristics such as personality traits and intelligence, has been estimated at around .49 (Polderman et al., 2015), so heritability of .25 is rather low by comparison and indicates there is a more substantial influence of the environment. The authors present data from a previous study that found that there was a weak negative correlation between religiosity and left-handedness. That is, people who are left-handed were somewhat less likely to be religious. Based on this, they conclude that the results support their theory that atheism is associated with developmental instability and therefore linked to adverse genetic mutations. This conclusion is invalid for several reasons.

Firstly, the religiosity measure used in the study basically measured intensity of religious commitment. As I noted in a previous post, there is a qualitative difference between religious believers who are low in commitment to their faith and those who are not religious believers at all. Hence, the correlational results only showed that there was an inverse relationship between left-handedness and religious commitment, but do not necessarily say anything about atheism as such. Furthermore on this point, the study also included comparisons by religious affiliation, e.g., Catholics, Baptists, Methodists, Jewish, agnostics, and atheists. The only significant difference found was that Jews scored higher on left-handedness than Methodists and Catholics; there were no significant differences between atheists or agnostics and the other groups. Hence, this study says nothing conclusive at all about atheists.

However, there is a deeper problem with the interpretation of this study. Dutton et al.’s claim that left-handedness is a marker of deleterious mutation load is based on several findings linking left-handedness to a range of adverse outcomes, such as physical and psychological problems, which they assume are indicators of underlying genetic problems. They use these findings to argue that “left-handedness is a reflection of an atypical brain,” and “reflects a brain that undergoes developmental instability due to a combination of elevated numbers of deleterious mutations and an acutely inauspicious environment.” Therefore, “In other words, left-handedness is a reflection of mutations which under conditions of natural selection would be damaging and would likely be selected against.” For example, they cite a paper by Goodman (2014) that shows that left-handedness is associated with a small cognitive disadvantage compared to right-handedness, which results in lower average earnings and lower socioeconomic status. Contrary to what Dutton et al. assert, Goodman clearly states that left-handedness is frequently caused by environmental factors, such as birth complications rather than genetics. Furthermore, when disadvantages associated with left-handedness are present, they are largely or entirely due to environmental insults such as birth complications rather than genetics. Goodman goes on to state, “Those likely to be left-handed due to genetics show smaller or no deficits relative to righties, suggesting the importance of environmental shocks as the source of disadvantage.” Hence, left-handedness is not a valid proxy for deleterious mutation load! Furthermore, while it may be the case that certain clinical groups have high rates of left-handedness, this does not mean that left-handedness in non-clinical samples is related to any kind of pathology, as “normal” left-handedness may have a different etiology.

The fact that one should be careful when comparing results from people with some kind of pathology with those in non-clinical samples is worth remembering when considering the relationship between atheism and autism. Dutton et al. cite a study (Caldwell-Harris et al., 2011) that found that a sample of people with autism had higher rates of atheism (26%) and agnosticism (17%) than a sample of “neurotypical” people (17% and 10%, respectively). Caldwell-Harris et al. attributed this to autistic people having difficulty with “mentalizing,” i.e., the ability to imagine other minds. They considered that mentalizing may be important in conceptualizing the existence of a personal god, and that autistic people might be less likely to believe in god for this reason. This is a fair point, but I think it is worth pointing out that people become atheists for many different reasons, which may be unconnected with deficiencies in mentalizing. Hence, it would not be appropriate to generalize from samples of autistic people to atheists as a group. By analogy, people with schizophrenia tend to have high rates of religious belief. However, one should not generalize from this that religious belief is a sign of schizophrenia.

In addition to poor health, left-handedness, and autism, the other marker of mutation load that Dutton et al. considered was fluctuating asymmetry, which refers to small random deviations from perfect symmetry in bilaterally paired structures. Interestingly, although they predicted that religious people would have lower fluctuating asymmetry than atheists, they admit that they could find any research evidence for fluctuating asymmetry related to atheism. However, there is a piece of evidence indirectly relating to fluctuating asymmetry that they basically dismiss, which is that atheists tend to have slightly higher intelligence compared to religious believers (I discussed this in a previous post). They do not mention this in their literature review and only in the Discussion section do they acknowledge that this is inconsistent with their views. Quite reasonably, they consider that intelligence would have been selected for in preindustrial times, yet they also assert without evidence that it should be positively associated with religiousness because of the latter's alleged inverse association with mutation load. Dutton et al. propose several conflicting solutions to this supposed paradox. First, they assert that “intelligence has been found to be only extremely weakly influenced by mutational load.” Then they suggest that an optimum level of religiousness and intelligence are selected for, and that very high intelligence may actually be a maladaptive mutation, which seems to contradict their previous assertion. Finally, they make this puzzling statement, which is not supported by any references or evidence: “Indeed, it could be proposed that, with the reduction in stress brought about by industrialisation, a level of intelligence which was once associated with religiousness is now associated with atheism. In other words, the environment has changed, rendering intelligence less adaptive than it once was.”

I think this last assertion is not only wildly speculative, but hopelessly confused and seems like an attempt by the authors to have their cake and eat it. That is, they seem to be trying to assert simultaneously that intelligence is not associated with mutation load, except when it is, and that it used to be adaptive but it no longer is, and all of this without providing any substantive evidence.

However, despite the confused assertions of Dutton et al., there is substantial evidence that intelligence is an adaptive trait, even in modern times, and that it is associated with genetic fitness and low mutation load. For example, a meta-analysis of 14 studies showed that there is a moderate but robust negative correlation between intelligence and fluctuating asymmetry, ranging from −0.12 to −0.20 (Deary, 2011). That is, people with higher intelligence test scores tend to be more symmetrical, indicating lower mutation load. Additionally, there is a very large body of evidence, derived from studies involving thousands of people around the world, linking intelligence with reduced risk of early death, and death from cardiovascular disease, suicide, homicide, and accidents. Furthermore, intelligence has been linked to a wide range of physical and mental health outcomes, and there is evidence that intelligence is an index of general bodily “system integrity” (Deary, 2011). All of this evidence suggests that high intelligence is associated with health because they both are influenced by genetic quality.

Intelligence Essential Reads

How Do You Know What You Know?

How Gifted Children Change in the Teenage Years

Considering that atheism tends to be modestly associated with higher intelligence, if one were to use the kind of logic that Dutton et al. use, then would have to conclude that atheism is associated with lower not higher mutation load. However, to remain scientific, we need to be cautious before drawing such a conclusion, as the evidence for this is only indirect. More specifically, atheism has a rather weak connection with higher intelligence, and intelligence in turn has a moderate connection with reduced fluctuating asymmetry, which signifies reduced mutation load. Hence, one cannot conclusively state that atheism is associated with mutation load just because it is associated with something else that is associated with this; more direct measures would be needed to link the two.

As to whether atheism is adaptive or whether being religious is adaptive in an evolutionary sense, the most likely answer is that it depends on the prevailing conditions of a given time and place. People who consider personality traits from an evolutionary perspective are inclined to think that for any personality trait there is no set level that is always adaptive, and that relatively high or low levels of a given trait might be adaptive in different environments or circumstances (Penke, Denissen, & Miller, 2007). The same might apply for the adaptiveness of religious belief vs. disbelief. Furthermore, the specific contents of a person’s beliefs are probably much more malleable depending on individual circumstances than levels of personality traits, which seem to be under stronger genetic influences (Eaves et al., 2012). Rather than assuming, as Dutton et al. do, that a specific kind of religiosity is the norm and that any deviation from the same is an aberration that indicates maladaptive mutations, it might be more scientific to consider that human variation is a normal and recurring feature of human society.

© Scott McGreal. Please do not reproduce without permission. Brief excerpts may be quoted provided a link to the original article is provided.