We’re only one month into the new year and I may have come across the most ridiculous paper I think I’ll read all year. The paper is titled Knowledge of resting heart rate mediates the relationship between intelligence and the heartbeat counting task. They state that ‘intelligence’ is related to heartbeat counting task (HCT), and that HBC is employed as a measure of interoception—which is a ‘sense’ that helps one understand what is going on in their body, sensing the body’s internal state and physiological changes (Craig, 2003; Garfinkel et al, 2015).
Though, the use of HCT as a measure of interoception is controversial (Phillips et al, 1999; Brener and Ring, 2016) mostly because it is influenced by prior knowledge of one’s resting heart rate. The concept of interoception has been around since 1906, with the term first appearing in scientific journals in the 1942 (Ceunen, Vlaeyen, and Dirst, 2016). It’s also interesting to note that interoceptive accuracy is altered in schizophrenics (who had an average IQ of 101.83; Ardizzi et al, 2016).
Murphy et al (2018) undertook two studies: study one demonstrated an association with ‘intelligence’ and HCT performance whereas study 2 demonstrated that this relationship is mediated by one’s knowledge of resting heart rate. I will briefly describe the two studies then I will discuss the flaws (and how stupid the idea is that ‘intelligence’ partly is responsible for this relationship).
In both studies, they measured IQ using the Wechsler intelligence scales, specifically the matrix and vocabulary subtests. In study 1, they had 94 participants (60 female, 33 female, and one ‘non-binary’; gotta always be that guy eh?). In this study, there was a small but positive correlation between HCT and IQ (r = .261).
In study 2, they sought to again replicate the relationship between HCT and IQ, determine how specific the relationship is, and determine whether higher IQ results in more accurate knowledge of one’s heart rate which would then improve their scores. They had 134 participants for this task and to minimize false readings they were asked to forgo caffeine consumption about six hours prior to the test.
As a control task, participants were asked to complete a timing accuracy test (TAT) in which they were asked to count seconds instead of heartbeats. The correlation with HCT performance and IQ was, again, small but positive (r = -.211) with IQ also being negatively correlated with the inaccuracy of resting heart rate estimations (r = .363), while timing accuracy was not associated with the inaccuracy of heart rate estimates, IQ or HCT. In the end, knowledge of average resting heart rate completely mediated the relationship between IQ and HCT.
This study replicated another study by Mash et al (2017) who show that their “results suggest that cognitive ability moderates the effect of age on IA differently in autism and typical development.” This new paper then extends this analysis showing that it is fully mediated by prior knowledge of average resting heart rate, and this is key to know.
This is simple: if one has prior knowledge of their average resting heart rate and their fitness did not change from the time they were aware of their average resting heart rate then when they engage in the HCT they will then have a better chance of counting the number of beats in that time frame. This is very simple! There are also other, easier, ways to estimate your heart rate without doing all of that counting.
Heart rate (HR) is a strong predictor of cardiorespiratory fitness. So it would follow that those who have prior knowledge of their HRs would more fitness savvy (the authors don’t really say too much about the subjects if there is more data when the paper is published in a journal I will revisit this). So Murphy et al (2018) showed that 1) prior knowledge of resting heart rate (RHR) was correlated—however low—with IQ while IQ was negatively correlated with the inaccuracy of RHR estimates. So the second study replicated the first and showed that the relationship was specific (HCT correlated with IQ, not any other measure).
The main thing to keep in mind here is that those who had prior knowledge of their RHR scored better on the task; I’d bet that even those with low IQs would score higher on this test if they, too, had prior knowledge of their HRs. That’s, really, what this comes down to: if you have prior knowledge of your RHR and your physiological state stays largely similar (body fat, muscle mass, fitness, etc) then when asked to estimate your heart rate by, say, using the radial pulse method (placing two fingers along the right side of the arm in line just above the thumb), they, since they have prior knowledge, will more accurately guess their RHR, if they had low or high IQs, regardless.
I also question the use of the HCT as a method of interoception, in line with Brener and Ring (2016: 2) who write “participants with knowledge about heart rate may generate accurate counting scores without detecting any heartbeat sensations.” So let’s say that HCT is a good measure of interoception, then it still remains to be seen whether or not manipulating subjects’ HRs would change the accuracy of the analyses. Other studies have shown that testing HR after one exercises, people underestimate their HR (Brener and Ring, 2016: 2). This, too, is simple. To get your max HR after exercise, subtract your age from 220. So if you’re 20 years old, your max HR would be 200, and after exercise, if you know you’re body and how much energy you have expended, then you will be able to estimate better with this knowledge.
Though, you would need to have prior knowledge, of course, of these effects and knowledge of these simple formulas to know about this. So, in my opinion, this study only shows that people who have a higher ‘IQ’ (more access to cultural tools to score higher on IQ tests; Richardson, 2002) are also more likely to, of course, go to the doctor for checkups, more likely to exercise and, thusly, be more likely to have prior knowledge of their HR and score better than those with lower IQs and less access to these types of facilities where they would have access to prior knowledge and get health assesments to have prior knowledge like those with higher IQs (which are more likely to be middle class and have more access to these types of facilities).
I personally don’t think that HCT is a good measure of interoception due to the criticisms brought up above. If I have prior knowledge of my HR (average HR for a healthy person is between 50-75 BPM depending on age, sex, and activity (along with other physiological components) (Davidovic et al, 2013). So, for example,if my average HR is 74 (I just checked mine last week and I checked it in the morning, and averaged 3 morning tests one morning was 73, the other morning was 75 and the third was 74 for an average of 74 BPM), and I had this prior knowledge before undergoing this so-called HCT interoception task, I would be better equipped to score better than one who does not have the same prior knowledge of his own heart rate as I do.
In conclusion, in line with Brener and Ring (2016), I don’t think that HCT is a good measure for interoception, and even if it were, the fact that prior knowledge fully mediates this relationship means that, in my opinion, other methods of interoception need to be found and studied. The fact that if someone has prior knowledge of their HR can and would skew things—no matter their ‘IQ’—since they know that, say, their HR is in the average range (50-75 BPM). I find this study kind of ridiculous and it’s in the running for most ridiculous things I have read all year. Prior knowledge (both with RHR and PEHR; post-exercise heart rate) of these variables will have you score better and, since IQ is a measure of social class then with the small correlation between HCT and IQ found by Murphy et al (2018), some (but most is not) is mediated by IQ, which is just largely tests for skills found in a narrow social class, so it’s no wonder that they corrrlate—however low—and the reason why the relationship was found is obvious, especially if you have some prior knowledge of this field.