We are notoriously bad at assessing risk. One problem is that we confuse ‘scary’ with ‘dangerous.’ Many scary things are not in fact very dangerous. Airplane flights are a good example of this. Many dangerous things, on the other hand, fail to scare us—think French fries and global warming.
Moreover, in weighing competing scenarios and the threat they embody, our brain is wired to privilege certain calculations over others. As Tversky and Kahneman have shown in their Nobel Prize winning research, humans think in heuristics—mental short cuts—that, while often useful, contain built-in blind spots that may lead us astray.
The other is the ‘representativeness heuristic,’ which makes us judge that what looks, walks, and sounds like a duck must be a duck, when this is, in fact, not likely true if there aren’t many ducks around. Thus, the introvert with the aptitude for numbers drawn randomly from a group of 100 people—five accountants and 95 Marines—is most likely a Marine, not an accountant, contrary to what most people would predict.
Fear- and risk-related heuristics are wired into our genetic hardware because they carry adaptive value. Fears of snakes, water, and heights are common all around the world because those of our ancestors who were predispose to acquiring them survived and reproduced more often, transmitting those genes to us.
Of course, our risk and fear judgments are also influenced by the socio-cultural context. Those schooled hard in the might of God will fear his wrath greatly. Our assessments will also depend on our personal experience: once bitten, twice shy, as the saying goes. Immediate circumstances will affect our risk judgment as well: surrounded by a cheering crowd egging us on, we are more likely to attempt something we would judge as too risky on our own.
The media we consume may also affect our risk assessment, as it primes certain fears over others. The odds an American will die of a terrorist attack are smaller than the odds of being crushed by falling furniture. But terrorists are scarier, in part because images and narratives of their horrific deeds are constantly on TV, social networks, and in the papers.
Risk judgments often emerge from the interaction between these multiple influences. For example, after 9/11, the vivid coverage and visual images of the planes crashing into the towers worked to activate people’s pre-existing biologically wired fear of flying, leading many to opt for driving (thus, ironically, placing themselves in greater actual danger, since driving overall is riskier than flying).
Our difficulty in accurately weighing risk informs every aspect of our lives, including sex. A series of studies by Terry Conley and her colleagues, published recently in the International Journal of Sexual Health, explores how our assessment of sex-related risks may be uniquely biased and inaccurate.
In the first study, 681 participants were asked to estimate the number of people out of 1000 who would be expected to die in one of two scenarios: (a) a 300-mile drive or (b) as the result of contracting an HIV infection from a single act of unprotected sex. Participants over-estimated the death risk in both scenarios, but they rated the HIV infection death risk as much higher than the car crash death risk. Of course, in reality one is 20 times more likely to die from a car accident than from HIV contracted during one act of unprotected sex.
In the second study, 310 adult volunteers received one of two scenarios. In one, a woman goes to a party after noticing some symptoms and disregarding them as a urinary tract infection. She has unprotected sex with a guy. Later it turned out she had Chlamydia and gave it to him. The second scenario involves a woman who feels somewhat under the weather; dismissing her symptoms as allergies, she goes to a party. She has unprotected sex with a guy. It turns out she had swine flu, and gave it to him. The participants were asked to rate the woman in each scenario on how selfish, dumb, and risky her behavior was. Participants rated the first woman more selfish, dumb, and risky than the second.
In Conley’s third study, 1158 participants were given similar STI and swine flu transmission scenarios as in the second study, but the researchers varied the severity of the outcome associated with acquiring each disease (mild or moderate in the case of chlamydia, which is nonfatal; mild, moderate or severe in the case of swine flu, which can be fatal). The researchers found that participants rated the person transmitting the STI that led to a mild outcome (doctor visit and a course of antibiotics) more negatively than they rated the person who transmitted swine flu that resulted in a severe outcome (partner’s death).
The fourth, archival study examined Department of Public Health websites and Department of Motor Vehicle websites for each of the 50 states for information about STI prevention or car accident prevention, respectively. Of the 43 states mentioning STI prevention, 72% specifically mentioned that abstinence is the most effective means of preventing the spread of STIs. None of the 39 DMV websites mentioning accident prevention recommended the elimination of driving as a means to avoid accidents.
Conley and her fellow researchers concluded that, “The stigmatization of STIs is beyond the degree of severity (relative to other diseases) and viewed as unjustifiably risky (relative to other risky activities).” In other words, social stigma surrounding sex causes people to over-estimate sex related risk, judge sex-related behavior more negatively, and propose draconian risk control measures that are not applied to non-sexual risks of greater probability (like driving).
This is an intriguing, thought provoking claim. Yet it is a bit empirically risky (!) and should be entertained cautiously, as the results are open to multiple alternative interpretations.
First, it is important to note that the ‘between-subjects’ design, used in the first three studies, did not allow participants to see both scenarios for direct comparison. Thus, the groups estimated risk in entirely different contexts. Context affects judgment. For example, participants in the first study who saw the ‘unprotected sex’ scenario may have judged it as more risky because they associated it with its normative counterpart, ‘safe sex.’ Those who saw the “driving” scenario had no equivalent counterpart for mental comparison. A better design would have equated non-normative sex (‘unprotected sex’) with non-normative driving, such as ‘driving without a seatbelt,’ or ‘buzzed driving;’ results, one predicts, would have been different.
Similarly, in the second and third studies, it is unclear whether the participants’ negative judgments were informed by the stigma around STI or by their views on unprotected sex. Unknowingly giving someone an STI through unprotected sex is indeed dumber and more selfish than giving them swine flu, insofar as the former outcome can be easily avoided by using a condom.
Thus, the researchers’ claim that, “people respond in unduly negative ways to people who transmit sexual diseases,” while perhaps factually true, is not demonstrated by their data. Their results may just as well be interpreted as showing that people respond more negatively to people who behave more recklessly. A failure to prevent a negative outcome that can be easily avoided is judged more negatively than a failure to prevent a negative outcome that’s hard to avoid, independent of the severity of the outcome is. In the same way, social psychology research has shown that our decision to help a victim of misfortune will depend heavily on whether we perceive them as blameless, independent of the severity of their misfortune.
More generally, judging the transmission of a non-lethal STI as more negative than that of lethal flu, as the third study’s participants did, is not particularly irrational or surprising if considered in a broader context. We are social animals above all. Thus, we often judge social risks as more severe than physical risks. Loss of reputation or status is often experienced as a fate “worse than death.” This is not unique to sex or STIs. Students often rate their fear of public speaking as greater than their fear of death. Soldiers often choose to risk death rather than risk being labeled cowards and losing face with their fellow soldiers.
Moreover, there is no reason to assume, as the researchers do, that, “stigmatisation of STIs specifically triggers the biased perceptions of sexual risk.” Stigma, on the whole, is not necessary to produce over judgment. People view spiders as more dangerous than bathtubs. Snake phobia is more common and easily acquired than bathtub phobia. Alas, in urban USA more people die from slipping in the tub (one in 11,469) than by snake bite (one in fifty million). Yet there is no social stigma against spiders. Similarly, people routinely over estimate the risk of flying compared to the risk of driving, yet there’s no stigma attached to flying.
Stigma is unnecessary for over judging risk. It is also insufficient. If stigma and risk assessment are linked, we may just as rightly assume that stigma will lead to under assessment of risk. Incest is highly stigmatized. Yet its prevalence is routinely under-estimated.
Additionally, if stigma and over judgment of risk are indeed causally linked, the link may not follow, as the researchers suggest, from stigma to over judgment. Over judgment of risk may just as well be the cause, rather than the result, of stigma. How would that work? Well, it is possible that rather than being a software problem (product of social stigma) risk judgment may in fact be more of a hardware problem (biological programming product). As mentioned earlier, our brains are biologically wired to notice and magnify certain risks more readily than others. Specifically, risks that were present in our ancestral environment (such as anything related to sex) invoke a stronger aversive response than new risks (such as risks relating to driving). Such a strong aversive response may cause us to stigmatize (view negatively) any behavior that provokes it.
In sum, STIs and other sexual risks may well be commonly over-judged, and such errors need to be pointed out and corrected. Social stigma surrounding STIs and other sexual topics may well need to be addressed and eased to improve public health (although, as Brian Earp notes in The Atlantic, the effects of stigma on behavior are not uniform).
Still, Conley’s work here, while provocative, does little to truly demonstrate—or clarify the nature of—the potential link between over judgment of risk and social stigma, sexual or otherwise.