On the sentient level of how dogs may fair, in terms of ‘high’ intelligence corresponding with in normal or excessive intelligence away from the norm: : validate Adorno/Lukatcs designation-deification, as maintained by Habeas’ note of Nietzche’s irony?
)))((()(((
| Trauma catastrophic |
−0.10 |
0.03 |
0.90 |
−3.70 |
2.20E-04 |
54,472 |
56,797 |
51.04 |
5640 |
5,554 |
49.62 |
| Social isolation |
−0.16 |
0.04 |
0.85 |
−4.28 |
1.83E-05 |
210,478 |
23,110 |
9.89 |
14634 |
1,414 |
8.81 |
| Trauma childhood stressors |
−0.27 |
0.03 |
0.76 |
−8.92 |
4.62E-19 |
73,178 |
38,076 |
34.22 |
8141 |
3,053 |
27.27 |
| Trauma adulthood stressors |
−0.37 |
0.03 |
0.69 |
−12.43 |
1.83E-35 |
65,178 |
45,654 |
41.19 |
7608 |
3,565 |
31.91 |
| General anxiety |
−0.37 |
0.05 |
0.69 |
−7.18 |
7.00E-13 |
61,127 |
11,118 |
15.39 |
6925 |
847 |
10.90 |
| Post-traumatic stress disorder |
−0.40 |
0.07 |
0.67 |
−5.77 |
7.70E-09 |
103,826 |
7,022 |
6.33 |
10730 |
441 |
3.95 |
| Trauma childhood abuse |
−0.48 |
0.05 |
0.62 |
−9.18 |
4.14E-20 |
98,544 |
12,688 |
11.41 |
10374 |
819 |
7.32 |
| Neuroticism score |
−0.12 |
0.01 |
|
−10.17 |
2.87E-24 |
|
|
|
|
|
|
Open in a new tab
Note. Neuroticism estimate is a standardized beta.
Abbreviations: g-factor, general intelligence; high g-factor, participants with a g-factor score 2 SD above the mean; average g-factor, participants with a g-factor score between ±2 SD from the mean; Cannabis use, never used versus used at least once; E, evening; M, morning.
Somatic phenotypes
Compared to individuals in the average g-factor group, the odds of having some type of allergy increased by 13% for individuals in the high g-factor group (OR = 1.13, 95% CI [1.08;1.18]). This was explained by their greater propensity to having eczema and other allergies (e.g., food; respectively, 1.25 and 1.33 times more likely), which were included in the general allergy diagnosis (Table 2). The odds of being myopic increased by 93% in the high g-factor group (OR = 1.93, 95% CI [1.82;2.04]) and this remained significant when controlling for educational attainment (OR = 1.75; Section S2.2 of the Supplementary Material).
Trauma
Compared to individuals in the average g-factor group, the odds of experiencing catastrophic trauma, adulthood stressors, childhood abuse, and childhood stressors decreased by 10% (OR = 0.90, 95% CI [0.87;0.94]), 31% (OR = 0.69, 95% CI [0.66;0.72]), 38% (OR = 0.62, 95% CI [0.58;0.67]), and 24% (OR = 0.76, 95% CI [0.73;0.79]; Table 2) in the high g-factor group, respectively.
Traits
Compared to individuals in the average g-factor group, the odds of feeling more socially isolated decreased by 15% in the high g-factor group (OR = 0.85, 95% CI [0.80;0.90]), whereas the odds of having an evening-like chronotype, ever engaging in same-sex behavior, and ever using cannabis increased by 15% (OR = 1.15, 95% CI [1.11;1.19]), 23% (OR = 1.23, 95% CI [1.14;1.33]), and 25% (OR = 1.25, 95% CI [1.20;1.31]), respectively, in the high g-factor group. Supplementary Figure S4 shows that there are more individuals with a higher-than-average g-factor that have used cannabis 1 to 100 times and that over 100 times there are no group differences. The high g-factor group had a lower neuroticism score than individuals in the average g-factor group (β = −0.12, SE = 0.01; Table 2).
Phenotypes with group differences between both the high and average and the low and average g-factor groups
Childhood stressors, childhood abuse, adulthood stressors, PTSD, and social isolation were more prevalent in the low compared to the average g-factor group (Supplementary Table S2) and were more prevalent in the average compared to the high g-factor group (Table 2), suggesting that the prevalence of these phenotypes decreases with an increasing g-factor. The low g-factor group had a higher neuroticism score than the average g-factor group (β = 0.22, SE = 0.02), which had a higher neuroticism score than the high g-factor group. The odds of ever trying cannabis and having an evening-like chronotype, respectively, decreased by 41% (OR = 0.59, 95% CI [0.53;0.65]) and 14% (OR = 0.86, 95% CI [0.82;0.90]) in the low compared to the average g-factor group.
Discussion
We examined differences in the prevalence of mental health disorders, somatic disorders, and certain traits between individuals with high (2 SD above mean) and average g-factor scores (within 2 SD of the mean) in the UK Biobank (N ≃ 7,266–252,249). We contrasted these results with differences observed between individuals with low and average g-factor scores.
We found that the high g-factor group did not have more mental health disorders than the average g-factor group and that they were less likely to have general anxiety and PTSD. Individuals with higher intelligence were also less likely to have experienced trauma and stressors, except for adulthood abuse, which may be part of the explanation for the latter finding. The high g-factor group was also less neurotic and felt less socially isolated than the average g-factor group. In contrast, the low g-factor group was more neurotic, felt more socially isolated, and had a greater prevalence of trauma, stressors, and PTSD than the average g-factor group, suggesting that the prevalence of these phenotypes decreases with increasing intelligence. Among the few somatic disorders that were examined, we found that individuals with high intelligence were more myopic and had more allergies, although they had a lower prevalence of hay fever rhinitis and asthma. Individuals with high intelligence were also more likely to present certain traits, such as having an afternoon–evening chronotype, to have ever tried cannabis, or have ever engaged in same-sex behavior, whereas the low g-factor group was less likely to have ever tried cannabis and engaged in same-sex behavior than the average g-factor group. There were no differences between groups in the prevalence of insomnia.
Our results contradicts several studies that reported an increased risk for various psychiatric disorders in individuals with high intelligence (e.g., [2, 5]). These studies were generally based on small samples and suffered from major sampling bias or a lack of a control group [6, 7]. We find that high intelligence is not a risk factor for psychiatric disorders and even a protective factor for general anxiety. Higher intelligence was associated with a decrease in trauma exposure, and consequently PTSD. This is consistent with previous findings [29] and with the association of childhood trauma with lower intelligence [40].
With regard to somatic disorders, we replicate the increased risk of allergies in individuals with high intelligence [2, 41]. One possible explanation for this association is that allergies and intelligence share neural correlates [42]. Another possibility is that more intelligent individuals with a higher g-factor live in more urban areas [43], where allergies are more prevalent [44], or that individuals with high intelligence are more aware of allergic symptoms and have better access to health care. However, the prevalence between groups did not differ across all allergies (e.g., asthma and hay fever rhinitis).
In line with a previous literature review [27], the risk of myopia was greater for individuals with high intelligence. While near-work activities (e.g., reading and computer use) seem to be a risk factor for myopia [45, 46], this association appears to be distinct from that of higher intelligence and education level [22]. Although additional years of education contribute to an increase in the risk of developing myopia [47], most of the evidence points towards shared genetic factors between intelligence and myopia [22], which is consistent with our observation that the risk of myopia associated with a high g-factor only slightly decreased when adjusting for educational attainment.
Our results indicate more afternoon–evening chronotypes in individuals with high intelligence than in individuals with average intelligence, which could be explained by differences in the work schedules of the different g-factor groups [28]. In line with a previous study [30], we find that individuals with high intelligence are more likely to ever have engaged in same-sex behavior. However, we note that our measure reflects sexual exploration rather than sexual orientation. We also found an association between ever trying cannabis and intelligence, but this was only true for individuals who consumed cannabis less than 101 times in their lifetime, suggesting that this measure also reflects a tendency to explore. One possibility is that individuals with higher intelligence, which is positively correlated with the “Openness to Experience” personality trait (r = 0.30; [49]), may be more likely to seek out new experiences and explore alternative behaviors than the average.
This study was first limited in its ability to study neurodevelopmental disorders and some psychotic disorders, which were absent or had few cases in the UK Biobank. We nonetheless investigated phenotypic differences in prevalence between g-factor groups across a greater number of psychiatric disorders than previously done [2, 50] and examined additional phenotypes, including traumatic experiences.
Second, the prevalence of some psychiatric disorders and traits differs from the general population [36] because of the UK Biobank’s “healthy volunteer” bias [35]. However, this should not affect the validity of the group comparisons because the phenotypes were designed with the same criterion across high and average g-factor groups.
Third, while some data may not be missing at random, our participants are more representative of highly intelligent individuals in the general population than those from previous Mensa studies, which did not have a control group and examined individuals that had to take or decided to take an IQ test and join a society for the highly intelligent.
Fourth, although we used self-report diagnoses, which are less precise than professional diagnoses, UK Biobank participants self-reporting any psychiatric diagnosis appear to have an elevated risk of any symptom-based outcome [51]. By using professional and probable diagnoses with validated short mental health tests, we were also able to include individuals with mental health disorders that have not sought or received mental health assistance [52, 53].
Fifth, the prevalence of psychiatric disorders and traits between g-factor groups may differ across the lifespan. However, we were interested in lifetime prevalence, making the UK Biobank, a prospective aging study, a good candidate for the question at hand.
Therefore, while this study is not without limitations, we provide a large scaled analysis of the difference in the prevalence of numerous phenotypes, including mental health and somatic disorders, across individuals of varying intelligence. We find that high intelligence is not a risk factor for psychiatric disorders and is a protective factor for general anxiety and PTSD. Whereas our results support the advantageous nature of higher intelligence [50], the g-factor may still be relevant for psychiatric evaluation: higher intelligence may affect the presentation of symptoms and the available resources for recovery.
Acknowledgments
This research has be
We’ll see!