Weather and behaviour

In the 1989 Spike Lee film Do the Right Thing tempers fray as the temperature climbs. The movie depicts one day in a heatwave in Brooklyn as hotheads become hotter, the heat of the day culminating in a riot provoked by seemingly not much at all other than the weather.

Several studies have found a relationship between high temperature and increased aggressive behaviour. In the United States, crime of all sorts increases on hot days, particularly violent crime (murder, assault, robbery, and rape). In agreement with this finding, violent crime is more common in the warmer southern states – in contrast, non-violent crime is less common further south. A similar pattern has been found in France, Spain, and Italy, countries with large north to south variations. The frequency of civil conflicts in the period 1950–2004 in tropical regions is related to the El Niño and La Niña phenomena, with conflicts being twice as likely in El Niño years. The El Niño is the warm phase of the ENSO, the El Niño Southern Oscillation, when a large area of warm water develops in the eastern equatorial Pacific coast off South America, leading to increasing temperatures in the region, as well as drought in some areas. These events happen every three to seven years, and last around a year or longer. The corresponding cold phase is called the La Niña, and civil conflicts in the region are less common during these events.

As you might guess by now, it isn't clear what the real reason for this relationship is. There are probably several. When it's cold people are less likely to go out and engage in social interactions in the streets; in Do the Right Thing trouble got going when the crowd congregated. People start drinking more alcohol on hot summer nights, and we know excessive alcohol can fuel violence. Unpleasant conditions increase discomfort and irritability. Heat makes us more aroused and more likely to act.

However, the relationship between temperature and tendency to violence isn't a simple straight line but is instead what we call curvilinear, broadly in the shape of an inverted U. A study of the number of aggravated assaults in Dallas, a southern city with plenty of hot days to provide data, showed that the number of assaults rises as the temperature rises to approximately 30˚C, but as it goes above that, the amount of violence goes down again. It's as though above a certain temperature it's just too darned hot to do anything. We also observe different patterns in the night and day: the number of violent incidents is related to the temperature by a simple straight line in the somewhat cooler hours of night. How can we explain these patterns? The best-known account is Baron and Bell's negative affect escape (NAE) model, which states that the discomfort caused by moderately high and low temperatures facilitates aggression, but extreme temperatures arouse competing motives, such as a desire to escape, which conflict with and thereby reduce aggression.

We don't observe an increase just in violent crime; the incidence of less drastic aggression, such as road rage, increases too. In one study carried out in Phoenix, Arizona, a researcher sat at a road junction every Saturday from April to August, during which period the temperature ranged from 29°C to 42°C (and yes, 29˚ was the cool end of the study). The researcher sat in her car at the same junction, a single lane controlled by traffic lights, and when the lights turned green she very socially just sat there. The researchers measured the frequency and duration of honks from the cars behind her, and honks became more frequent and lasted longer the hotter the weather. The relationship was linear – the graph of honk number versus temperature forms a straight line. The researchers also noted that the drivers of cars with their windows down honked the most – because, they argued, those cars didn't have air conditioning. Even supposedly friendly games are affected: sportspeople are more cantankerous as the mercury rises, too. Baseball pitchers (throwers) retaliate more when it's hot. A revenge pitch is when a pitcher perceives a teammate to have been deliberately hurt by a pitch, and seeks revenge with his or her own throw. The number of revenge pitches increases on hot days. It isn't down to a factor such as hot weather making pitches less accurate because throws are just accurate on hot days as cold days, so on hot days it's cold-blooded revenge. Pitchers retaliated 22% of the time on days that peaked at 13°C, whereas their rate of retaliation rose to 27% when the temperature reached 35°C. The authors of the baseball study suggest two things are going on in hot weather: inhibitions against retaliating are lower, and pitchers are more likely to consider the actions of others as being hostile in intent.

There is mixed news about less pleasant weather: crime goes down when it's raining, but the incidence of domestic violence goes up.

Weather and cognition

I find it difficult to concentrate on warm, sunny days; I just want to be outside, enjoying the sunshine. On the other hand I find it difficult to focus when it's snowing because snow is relatively rare and very beautiful. Rainy days have a fascination of their own too, and I am frequently distracted by checking the rainfall total for the day, or at least the hourly rainfall rate. But I am unusually obsessed by the weather and therefore probably not typical. Are there effects of the weather on cognition in the general population?

In general gloomy weather lowers our mood, which helps us focus more and think more deeply. Some studies have found that memory is better on cloudy, rainy days than sunny days. One typical study tested the ability of Australian shoppers to recall the identity and location of several small items (e.g., model cars, a piggy bank) placed at random places in a small shop; the design was an "ambush" study, where randomly selected participants were stopped on leaving the store, across different times and with therefore different weather conditions. The researchers found that shoppers could recall on average three times as many items on cloudy days than sunny days. The proposed mechanism is that mood might be better on sunny days. There are then several reasons why good mood might lead to worse memory: people might be less motivated to remember, either preferring simply to hurry up and get on with enjoying themselves or worrying that putting a lot of effort into remembering something will spoil their good mood. Different moods might lead us to engage in different types of cognitive processing, with one prominent proposal being that negative mood makes us focus on bottom-up processing, while positive mood encourages us to focus on top-down processing. Emphasis on bottom-up processing might be more advantageous in simple memory tasks. It's all a bit speculative.

In general the effects are complex and several things are most likely going on. The meteorological variables that have the greatest effect on mood and memory are hours of sunshine, temperature, and humidity. High levels of humidity lowered scores on concentration tasks and increased reports of sleepiness. Rising temperatures lowered anxiety. The number of hours of sunshine predicts how optimistic we are. Pleasant weather, with higher temperature and higher barometric pressure, results in better mood, better memory, and "broadened" cognitive style, leading to people being more open and creative. However, these effects are found only during the spring and depend on the amount of time spent outside – as you might expect, the more time you spend outside, the greater the effect of these meteorological variables on how you think. Perhaps in line with the research on seasonal affective disorder, we might observe these effects only in spring because people have been deprived of pleasant weather in the winter, and it's the contrast that's important.

In summer cloudy days help us think more clearly in a focused way, perhaps because our mood is slightly lower, while sunny weather improves creativity and receptivity to new ideas. The lack of sunlight associated with rainy days also causes serotonin levels to dip, and as serotonin levels decrease, carbohydrate cravings increase, so we become hungrier for starchy foods on cloudy days. Excessive carbohydrate consumption promotes production of the amino acid tryptophan, which increases tiredness, thereby lowering concentration. So your concentration might benefit from your slightly lower mood on cloudy days, but that benefit will be lost if you try to compensate by eating carbs.

The weather also affects what we value. The researcher Uri Simonsohn has studied how weather affects college admissions. One study looked at 682 university applicants and how the weather affected the way interviewers perceived them: on cloudy days the interviewers had a preference for applicants with strong academic profiles, while on sunny days non-academic attributes, such as being an athlete or taking part in extra-curricular activities, were taken much more into account. As Simonsohn says, clouds make nerds look good. It works both ways: on cloudy days applicants are more likely to prefer institutions with strong academic records at the cost of non-academic strengths. Simonsohn examined the behaviour of 1,284 prospective students at a college known for its "academic strengths and recreational weaknesses", which is a nice and kind way of putting it. Cloud cover increased the probability of enrolment, by quite a margin: an increase in cloudiness on a particular day by one standard deviation led to almost a 10% increase in the number of applications. Of course the students didn't consciously reason "It's cloudy – I like this place"; the mechanism of action is presumably through non-conscious priming of academic activities – sunshine makes you think of the advantages of being outside, while dull days make you think about being indoors studying. Which university you go to has massive effects on your subsequent life, and I find it rather frightening that our lives might be determined by how cloudy it is on a particular day.

Given these findings, it is not surprising that the weather should have effects on schooling. Although I know of no data on the subject, it is probably the case that children find it easier to concentrate and remember things slightly better on cloudy days. There is some evidence that falling barometric pressure increases misbehaviour, and we get rapidly falling pressure when it is going to rain, is raining, and if the depression is still approaching, when it's very windy. There is an old lay belief that falling pressure affects the blood flow to the brain, but this belief isn't supported by any evidence. As boys are generally more likely to misbehave (or at least be caught), anything influencing naughtiness might be more apparent with boys.

The weather and creativity

Nobel laureates are not randomly distributed across the globe. The Dutch psychologist Evert Van de Vliert has observed that there are disproportionately more of them towards the poles and around the equator. Countries such as the US, Britain, Scandinavia (particularly Sweden), Russia, Western European countries, Australia, and New Zealand do very well. Britain, for example, with a population of around 60 million, had 120 laureates by 2013.

You might think that such inequality is not surprising, and simply reflects income and poverty, but it is possible to use statistical techniques to remove the effect of these variables. The inequality in distribution remains, and yes, it is due to climate and the weather.

Van de Vliert has been interested in what underlies creativity in general, and has used several indicators, including the number of patents generated by a country, as well as the number of Nobel Prize laureates per capita. He has speculated that extremes are important: cold stress and heat stress promote creativity, as long as the society has the resources to promote research. Cold and heat promote creativity in wealthier populations, but hinder it in poorer populations, controlling as much as possible for several other variables, including intellectualisation (those IQ differences between counties), industrialisation, and urbanisation. As we have observed, climate also affects the spread of human-to-human parasites, which are much more common in hotter climates, and which have an effect on creativity. As might be expected, differences in creativity across the wold are moderated by thermal climate, wealth resources, and parasite load. The amount of precipitation (whether it's a dry or wet climate) doesn't seem to matter.

The data suggest that sunshine is associated with creativity. We find that in the US, studies have shown that the value of patents correlates with the annual sunshine average of the region. These are rather coarse measures of both creativity and weather, but they are suggestive. Furthermore creatively successful people tend to move to sunnier climates, which is perhaps far from surprising, but those who do then experience an increase in their creativity. In contrast, those who relocate to less sunny places experience a decrease in output. As a specific example (and we should be wary in general of specific examples) we might note the outpouring of creativity of the French post-impressionist artist Paul Gauguin, who moved to French Polynesia for much of the last ten years of his life.

Not everyone agrees that sunshine is good for creativity. We have noted that sunshine is bad for focused cognition, and some researchers are adamant that sunshine is just as bad for creativity. The economic anthropologist Adam Alter argues that sunshine makes us less reflective, less likely to take risks, and less creative. Sunshine turns us – or at least many of us – into sun-seeking zombies. As Alter points out, when the warm sunshine comes out, we don't think about buckling down to the hard work that's the basis of creativity, but of heading to the beach or lying on the grass. Hard work and creativity make us feel good in the way that sunshine does, and is a substitute for sunshine, not a result of it.

These contradictory findings can be explained because they are generalisations – we all have different cognitive styles and, as we saw in our examination of how weather influences mood, different preferences. A sun-lover is likely to be affected very differently by a heatwave from someone who detests hot, sunny weather; the first sort of people might give up work and head for the beach, while the latter might shut themselves in a cool dark room and compose a great symphony. I know a famous researcher who spent a sabbatical in southern California, and hated it. He did some of the best work of his life in air-conditioned gloom while there. On the other hand I spent some time in San Diego and got much, much less done than I had anticipated. Goethe complained that "excellent personalities", modestly including his own, "suffer most from the adverse effects of the atmosphere". Handel and Mahler were also affected by the seasons, producing many of their great works in autumn and spring.

The seasons and climate can affect the content of creative work. It is difficult to imagine Gauguin's pictures of Polynesia carrying the same power if he had not actually been there. Similarly Cézanne's landscapes scream intense, bleaching French sunshine. Van Gogh's artworks were swayed by the seasons, being dominated by ominous clouds and darkness in the winter months, and showing more optimistic sunshine, light, and the stars during the summer months. Some argue that his bold, aggressive brushstrokes laden with paint were more frenzied in the winter months, while his summer pictures were less intense.

Working, stock markets, and shopping

Given these differences, it is perhaps not surprising that workers are more productive when the weather is bad. One impressive study examined employee productivity in a bank in Japan, an online workforce in the US, and performance in a laboratory experiment. All showed that people work more and better when the weather isn't so good compared with a nice, sunny, warm day. The researchers argued that there are more distractions in good weather; workers spend time and effort wishing they were enjoying the sun rather than the task that should have been at hand. This idea that for many people cognitive distractibility increases on nice days is a recurrent one with strong explanatory powers. The authors of this study suggest that not all is hopeless; they argue that where possible managers should assign work that requires more sustained attention on wet days, and work that allows more flexibility on sunny days. Most people and most jobs, however, don't give you any choice.

It's also been observed that even when you control for possible confounds, stock market returns are slightly higher in warm, sunny weather. It's thought that people are more willing to take risks, and consequently benefit from higher returns. A 2003 study by Hishleifer and Shumway found that sunshine (and only sunshine) was strongly correlated with stock market returns taken from 26 stock markets over 15 years. Note that this behaviour is irrational: sunshine might have a minor effect on some agricultural industries and ice cream production, but the size of the effect seems too big in western industrial societies. It appears that sunshine raises mood a little, and people misattribute this elevation of mood to being optimistic about life in general because the conditions in the world are good. And if the world is in a good place, it makes sense to invest. After you control for sunshine, other weather variables have no effect on stock market returns. This sort of research emphasises first how irrational we are when making complex decisions, and second how we don't think we're irrational, but attribute our beliefs to something more plausible.

The good news for employers is that few weather variables appear to affect how hard their employees try to work. One study showed that the only variable that affected the performance of data input clerks (it was the late 1970s) was a "discomfort index" combining temperature and humidity – people quickly feel uncomfortable in hot, humid conditions. The experiment was carried out in Birmingham, Alabama, and it isn't clear whether the office was air-conditioned. In fact many things aren't clear about this self-admittedly pilot study, so the results should be treated with caution. It is possible that a larger study would show that other factors do have an effect – we know for example that industrial accidents are more likely before a thunderstorm, which would presumably have a dramatic effect on productivity. The range of accidents that can happen to clerks is rather limited. In general it seems that warm, sunny weather decreases what is called our natural risk aversiveness, most people's general reluctance to take risks. While some people take risks and end up in a fight, others take risks on the market. Some have argued that there's some evolutionary benefit to risk taking in good weather, although what that might be isn't clear; perhaps the consequences of making a mistake were less bad for our ancestors. Perhaps. We also need to explain why speculation on the stock market in fine weather tends to pay off – obviously not all risks are good. Clearly something else is going on here; maybe people are more creative in their investments. We don't know.

It's not surprising that the weather can influence our shopping behaviour; of course what we buy is going to be influenced by the climate and weather. There isn't a big demand for raincoats in the Sahara, and few people rush to buy ice cream when it's snowing. Rain increases the sale of umbrellas (I only ever buy them when it's raining, and lose them as soon as the rain stops), and in a heatwave as the temperature soars so does the sale of salad and barbecue food. The weather, though, influences our general shopping behaviour. It seems that sunshine makes us feel more positive and spend more: when the sun is out, people want to shop. One study analysed sales figures from a tea and coffee shop and compared them with the weather over the same period: sales were clearly related to sunshine. Another study exposed some participants to artificial sunlight, and found that these participants said that they were more willing than a control group to buy things from a rather strange list, including green tea, gym membership, newspaper subscriptions, and plane tickets.

This was the finding of a three-part study. In the first study, the investigators analysed sales figures from a retail store that sold tea and tea-related products. They had data across six years of daily sales and daily weather conditions. In the second study, the researchers had participants complete a daily survey, which assessed their mood, how much tea and coffee they bought and consumed, and their total expenses for the day. Participants recorded this information for 20 days in March. The third study manipulated participants' exposure to artificial sunlight, assessed their mood, and questioned their willingness to pay for five products: green tea, juice, a gym membership, an airline ticket, and a newspaper subscription. The researchers concluded that sunshine makes us feel more positive and, in turn, shop more.

Personally I find these results surprising: I can think of few things I'd rather do on a sunny day than go into a sunless store and spend money. On the other hand, looking around the mall near my local gym, it's clear that heavy rain can put people off from going out at all. And that's another likely factor in why people shop more when the weather's fine: people are just more likely to go out.

The nature of shopping is changing, first with a shift to out-of-town shopping and then from physical to online shopping. There is less research on the weather and online shopping, but the data suggests that the reverse happens with online shopping: when the weather's nice, online sales go down. The explanation is probably partly the same, in that when it's nice, people go outdoors, away from their computers. Conversely when it rains, people spend more online, but the extent to which they do so depends on the location. You find more online spending in Marseilles on a wet day than on Paris, probably because people are more used to inclement weather in Paris, and therefore less put off going outside.

Daylight savings

In spring many regions of the world, particularly in the north, put their clocks one hour forward early in spring, and then put them back one hour in mid-autumn. This shift is called daylight savings time in the US and other areas, and British summer time in the UK. The rationale is that we shift an hour of daylight from the beginning of the day, where we don't need it (few are up and about and gardening at 5 a.m.), to the evening, where we can enjoy a glass of wine outside in the setting sun after a hard day's work. It is a change easily remembered by "spring forward, fall back". There is also the rationale that summer time saves resources; few need electric lights and heating early in the morning, but a bit of light later in the evening saves energy. It is no coincidence that daylight savings was first widely adopted in Europe in the Great War. Germany, Austria-Hungary, and the UK all started a form of daylight savings in 1916, the belief being that summer time conserved resources.

There have been occasional deviations from normal daylight savings. During the Second World War, from 1941 to 1945, Britain was on double summer time, with the winter on summer time and the summer on double summer time, two hours ahead of GMT, again supposedly to spare resources and give more light at the end of the day. I grew up in a period when the UK experimented with British standard time, when the UK stayed on permanent summer time from 1968 to 1971. My memory of that period is that it was perpetually dark. Unfortunately permanent summer time did not mean permanent summer, and the House of Commons on a free vote in December 1970 decided to end the experiment by a large majority.

The British standard time experiment did allow for an analysis of the effects. As you might expect, dark mornings mean more traffic accidents, and this pattern was indeed what was found. On the other hand, substantially fewer accidents and fatalities occurred in the evenings. Overall it was estimated that about 1,000 lives were saved each year in road traffic accidents. However, the UK implemented much more stringent drink-driving laws and checks at the same time, and because people are more likely to consume alcohol in the evening than morning, some or even all of the savings might have been due to this change.

The British Royal Society for the Prevention of Accidents currently campaigns for a return to the wartime summer time and double summer time system, on the grounds that the lighter evenings would result in more saved lives. The proposal is opposed by people who work outside, and people who live further north. We in Scotland are firmly against the idea: it is bad enough when sunrise is around 9 a.m., but 10 a.m. is an unimaginably grim prospect. And that's sunrise, not when the sun actually manages to peek over the hills.

In any case it is not obvious that permanent summer time would save lives. As we have seen, the 1968–1971 savings might have been due to something else entirely. It is also not the case that daylight savings saves as many resources as believed. Adam Alter has observed that much research has shown that daylight savings has ways of increasing consumption rather than reducing it – for example, by increasing the use of air conditioning, which is a voracious consumer of energy.

But most importantly fiddling with time and our body clocks is not good for us. When the clocks change our body clocks have to change too, and that has significant costs, particularly when we lose an hour of sleep in the spring. The effect is the same as giving a nation jet lag, and we all know how bad and unhealthy jet lag is. Adam Alter reports that the road accident rate increases by 7% the day after the change. The US stock market usually fares relatively badly the first business day after the change for no other apparent reason. It takes most people a day or two to adjust, so the increase in accident rates almost certainly lingers.

There might be less obvious but longer-lasting and even more disruptive effects. It has been claimed that daylight savings affects education by putting children and students out of phase with their natural biorhythms for several months of the year. After all, non-daylight savings time is more related to the natural time, so daylight savings is like putting children on permanent jet lag. This claim might sound far-fetched, but there is strong evidence for it. The state of Indiana in the US provides a very neat way of testing these sorts of ideas because some counties in the state observe daylight savings, but not all, which must be very confusing for the local inhabitants, particularly those living near county borders. But given that the only thing that differs between these counties is whether they observe daylight savings (it's a relatively small geographical area without any other substantial differences between counties, such as income, and some of the schools compared are just a few miles apart), Indiana provides a nice test of the effects of daylight savings. Research shows that students in areas that observe daylight savings score 16 points lower on the SAT (originally standing for "Scholastic Aptitude Test") than fellow students nearby in areas that observe standard time all year. Sixteen points is a huge difference; vast sums of money are spent trying to eliminate much smaller regional differences. The researchers Gaski and Gagarin dramatically concluded in 2011 that "DST appears to cause brain damage".

So the psychological evidence suggests that we should indeed abolish daylight savings and British summer time, but instead of introducing BST all year round, we should stick with natural time, and have GMT all year round. The price of more barbies finishing in the evening gloaming in the south would be a small one to pay.

Weather and voting

In many countries where voting is not compulsory the turnout at elections can be surprisingly, some say disappointingly, small. In the UK the turnout among eligible voters has gradually declined from a turnout of 83.9% of those eligible to vote in 1950 to 68.7% in 2017; in the US the distribution of turnout over time is more complex, falling to 61.4% in 2016, the lowest since 1996, but the Obama years were marked by high turnouts.

Studies in the US, Spain, and the Netherlands found that wet weather systematically reduces voter turnout, with an estimate that for every 25 mm (about an inch) of rain that falls, turnout is reduced by 1 percentage point. A 2004 study suggested that the value was somewhat larger, with 5% of the turnout being affected by the weather: a reasonable but hardly large amount. A Dutch study found that very sunny weather pushed up turnout by 1%. The effect might be volatile: a subsequent study in Sweden failed to replicate these findings, and the British election expert Professor John Curtice has examined the relationship between voting patterns and weather in all the UK general elections between 1922 and 2010 and found no relationship. February 1974 was the coldest election day recorded (and very wet too), yet voter turnout went up to a very respectable 78.8%.

There is a widespread belief in the UK that rain slightly favours the Conservative Party while warm, sunny weather slightly favours the Labour Party, but there is little evidence to support this belief. There is a similar belief in the same direction in the US, that rain favours the right, but again little evidence to support it, although some have claimed that weather affected the results of the 1960 and 2000 presidential elections; election day in 1960 was very dry across the US, and election day in 2000 was wet in Florida. There is some evidence that the weather has differentially affected turnout in Spain, with high turnout hurting the more right-wing parties more; however, in Spain the beneficiaries of the higher turnout were not the more left-wing parties but miscellaneous smaller parties. The supposed mechanism behind these findings is that voters of higher socioeconomic status are more likely to vote for right-of-centre parties, and they are less dependent on public transport and less likely to be deterred by poor weather. You can imagine if support for one party is more wavering than for the other, bad weather might push some people over the edge and, if they're not that bothered, stop them turning out. However, as the population has generally become wealthier, and most people have cars, it's possible that there was an effect decades ago which is no longer observed.

Of course exceptionally severe weather might interfere with the turnout, but British elections tend to be in late spring or early summer, or occasionally autumn, and American elections in November, when the weather is usually not too extreme. But again some caution is necessary in guessing what people might do: Hurricane Sandy caused great disruption around New York just before the 2012 presidential election, but there was no effect on voting. One possibility here is that disadvantaged groups saw the election as particularly important, and were therefore exceptionally resilient in their behaviour. The weather interacts with our intentions and behaviour in complex ways.

Spring flings

When Christmas passes and the winter comes to an end, the daffodils come out, the lambs gambol, and a young man's fancy turns to fancying. It's a cliché, but is there any support for the notion, and is it just the fancy of young men that's affected? Apparently not: one study in France found that women were more likely to say yes to an offer of a date from an "attractive" stranger on sunny days compared to cloudy days, with the figures being 22% and 14% (acceptance rates which seem high to me whatever the weather). The researchers argued that sunny weather led to better mood, leading to people being more generous in their opinions.

On the whole, though, most of the research has rather surprisingly concluded that it's cold weather that makes the sap rise. In 2004 and 2005 two Polish researchers approached 100 heterosexual men and asked them about their views on female attractiveness by rating pictures of silhouetted women in swimsuits and female breasts of various sizes. Exactly the same images were rated higher in winter than summer. The best explanation is that testosterone levels peak in winter. A study of 1,548 men living in Tromsø, in north Norway, found that testosterone levels were lowest when the temperature was highest and the length of daylight longest; there were two peaks of testosterone, both in fall and winter months, one in February, and the largest peak in October and November. The differences were large, with levels varying by over 30%. These seasonal variations are very sensitive to geography, with the same researchers finding no seasonal variation among 915 men living in San Diego, which has much less pronounced temperature and daylight variations.

We do not yet know why testosterone levels vary seasonally in regions with wide seasonal variations in temperature and length of daylight, but we do know that high levels of testosterone lead to more sexual activity, so unsurprisingly there is a peak in the number of births nine months after the peak in testosterone levels. Given the testosterone peak is in fall, this means the birth peak occurs in summer. One evolutionary possibility is that this mechanism ensures that babies are born at a time of plenty, while in southern latitudes the year-round abundance means that regulation of time of birth is less important. In fact conceptions go down during heatwaves – perhaps sex is less appealing when it's too darned hot. Some recent studies suggest that the length of day is more important in influencing conception in the north, and temperature in the south.

Although this explanation sounds plausible, there are contradictory findings that show that the picture is much more complicated. Other studies have shown that seasonal variations in births are stronger in hotter climates. Another study did find a pronounced seasonal variation in birth rates, in that in warmer southern US states, such as Louisiana and Georgia, birth rates were much lower in April and May, and much higher in August, September, and October. In Louisiana, for example, birth rates were 45% higher in the summer months, which is a dramatic difference. These results are consistent with the idea that conception rates are higher in winter, but inconsistent with the idea that seasonal variations in testosterone are the only thing going on. Several other possibilities come to mind, such as people spending more time indoors or spending more time together, or perhaps sex is just something to do when the weather isn't so great outside. These seasonal variations, though, are declining in strength over time in hot areas, presumably because people are on average spending less time outside, and air conditioning reduces the effect of outside heat and humidity. There is some evidence, though, that seasonal variations in conception are increasing in more northerly, colder regions, for reasons that aren't at all well understood.

It isn't just sex and conception that are affected by the weather and climate. One general finding of psychology that has been particularly demonstrated within social psychology in recent years is how sensitive we are to many influences, many of them outside our awareness. These effects go under the general name of social priming. The classic finding is that thinking about being old makes healthy young students walk more slowly, although this finding has been disputed, with social priming results at the centre of the supposed current replicability crisis in psychology. With such sensitivity to the outside world, it is hardly surprising that our love life is affected by the weather. Another important development is called embodied cognition, an approach that emphasises the way in which the mind is grounded in the body, and in turn the world. So when we think about using a tool, such as a hammer, brain imaging shows that the parts of the brain that would light up if we actually used the hammer become active just by thinking about doing so. Meteorological variables act upon us in just the same way.

In one experiment, student participants met an experimenter in the lobby and travelled up in the lift together to the main psychology rooms. As they did so the experimenter asked the student to hold the experimenter's cup for him or her while he or she jotted down the student's details. The cup contained either hot or iced coffee. As you might guess from the discussion of social priming and embodied cognition, students who held the hot cup later rated the experimenter as more friendly – "warmer" – than those who held the cold cup. Being warm appears to make us think more warmly of people and the world. Similar experiments have shown that just holding something cold can make you feel lonelier, and people who feel socially excluded crave warm drinks and food. So extending these results, being warm will make people feel socially warmer and friendlier, and more likely to interact with other. Researchers have concluded that social warmth and physical warmth are to some extent interchangeable, so warm climates might lead to more warm personalities, more warm interaction, and more romance. It's therefore no surprise that romantic films are more popular when it's warm (and in contrast comedy films are more popular when it's cold, as people want to be cheered – warmed – up).

It isn't just love that is affected by the weather; it also influences how nice we are. As we might expect from the embodied cognition results, people are more helpful when the weather is nice; psychologists say that they're more prosocial. One study showed that drivers in France are more likely to pick up hitchhikers when it's sunny than when it's cloudy. Some studies have found that tipping goes up when it's sunny, although other studies have found no effect. Even participants in experiments and interviews are more helpful when the weather is better. These studies show that it's fine weather in general, not just sunshine, that's important in acting prosocially, and the effects of heat can be offset by high humidity.

- Trevor Harley is Emeritus Professor of Psychology at the University of Dundee. Following a lifelong interest in weather, Trevor now researches at the intersection of psychology and weather, considering himself a psychometeorologist.

This chapter is from 'The Psychology of Weather', from the Routledge series 'The Psychology of Everything'. Read extracts from other books in the series.