Would you look where Berlusconi is looking?

"You should see what I'm looking at!"

Our attention is pulled in the direction of where we see other people looking. It happens so automatically that experts have assumed it's a reflex response, impervious to conscious factors, such as the particular identity of the gazer whose line of sight we're following. But a recent monkey study challenged this interpretation: high-status macaques were observed following the gaze of other high-status, but not low-status, monkeys. Inspired by this result, a team of Italian psychologists have examined whether our attention is influenced more by the gaze of politicians whose political persuasion matches our own.

Marco Tullio Liuzza and his colleagues recorded the eye movements of 28 participants sat at a computer screen. Approximately half were right-wing and half were left-wing. The task was to make a fast leftward saccade if a central square turned blue, or a fast rightward saccade if it turned orange. The interesting twist was that this central square was located in the middle of the eyes of a politician, who was shown staring straight ahead. Just before the central square changed colour, the politician's eyes shifted direction either in the same direction indicated by the square (potentially facilitating the participants' own eye movement) or in the opposite direction. The faces that were used belonged to Silvio Berlusconi, the right-wing Italian Prime Minister; Bruno Vespa (right-wing commentator); Antonio Di Pietro (current left-wing leader); and Romano Prodi (former left-wing Prime Minister).

The gaze shift by the political faces made no difference to the speed of the participants' own eye movements, but did affect their accuracy. For right-wing participants, their accuracy was influenced far more by the gaze shifts of Berlusconi and Vespa (the right-wingers) than by Pietro or Prodi. The influence of Berlusconi's and Vespa's eyes was similar in magnitude. By contrast, the left-wing participants were not influenced by the gaze direction of the political faces, left-wing or right-wing. Another detail was that Berlusconi's gaze direction had a stronger influence on those participants who considered their own personality to be similar to his.

A potential confound is that Berlusconi wasn't just a right-wing character, he was also Prime Minister at the time of the study, and his party was the leading party. His influence (and perhaps Vespa's, by association) on right-wing participants might therefore have been related to his position of authority, not just his political leanings. Certainly past research has shown that conservatives are more sensitive to authority than liberals.

Specifically on the reason why left-wingers weren't influenced by Berlusconi's gaze - Liuzza and his team said this was consistent with "studies showing that Italian left-wing voters detest the right-wing leader."

The researchers concluded that their study shows how: "... a sophisticated blend of situational and dispositional factors underlies the capture of reflexive gaze-following exerted on voters by the gaze of politicians. Future studies on the plasticity of this effect may provide new insights in the fundamental aspect of the human tendency to coalesce in large groups and complex societies."
_________________________________

Liuzza, M., Cazzato, V., Vecchione, M., Crostella, F., Caprara, G., and Aglioti, S. (2011). Follow My Eyes: The Gaze of Politicians Reflexively Captures the Gaze of Ingroup Voters. PLoS ONE, 6 (9) DOI: 10.1371/journal.pone.0025117

Post written by Christian Jarrett for the BPS Research Digest.

How a simple point of grammar could affect our voting decisions

A simple tweak in the tense of a verb could make the difference between electoral victory and defeat, according to a study by US researchers.

Imagine you encountered the following text: "Timmy Tucker is a senior politician. Last year Timmy championed human rights, and was fiddling his expenses."

Now compare with this version: "Timmy Tucker is a senior politician. Last year Timmy was championing human rights, and fiddled his expenses."

How does each version affect your view of Timmy Tucker? New findings from Caitlin Fausey and Teenie Matlock suggest that the first version is more likely to damage Timmy's re-election prospects.

The researchers found that the imperfect tense (e.g. "was fiddling") exacerbates the effect of a negative claim about a politician, compared with the perfect tense (e.g. "fiddled"). Fausey and Matlock aren't certain why this is, but they think the imperfect tense gives the sense that an action is ongoing, whereas the perfect tense brings closure.

For an initial study, 354 participants were split into four groups, with each reading one of four versions of a description of a politician who was up for re-election. Participants who read a version in which the man was described as last year "taking hush money" were more confident that he wouldn't be re-elected and estimated that he'd taken more money, as compared with participants who read a version in which it was written that last year "he took hush money". This subtle change in verb tense made no difference to the verdict of participants who read a positive account of the politician ("was collecting donations" vs. "collected donations").

A second study with a further 127 participants was similar except this time they all read a version that featured both a positive and negative claim about the politician. Those participants who read the description featuring a negative claim in the imperfect tense with the positive claim in the perfect tense ("was removing homes and extended roads") were less likely to say he would be re-elected (40 per cent vs. 56 per cent), compared with those who read the same claims with the tenses the other way around ("removed homes and extending roads")*.

"Because scandals involving political candidates are a hot topic in media coverage and campaign ads, insight into the power of the grammar used to communicate negative information will likely improve our understanding about how linguistic media shapes voting patterns," the researchers said.
_________________________________

Fausey, C., and Matlock, T. (2011). Can Grammar Win Elections? Political Psychology, 32 (4), 563-574 DOI: 10.1111/j.1467-9221.2010.00802.x

*In this example, extending roads is assumed by the researchers to be a positive activity - environmentally minded readers might not agree with that assumption!

This post was written by Christian Jarrett for the BPS Research Digest.

Is God always on the right?

This former Republican President claims to read the bible daily 

Holding conservative values tends to go hand in hand with being more religious, at least in the United States. Indeed, the idea that the US is divided between liberal atheists versus religious conservatives is at the heart of the country's so-called 'Culture Wars'. This has led some psychologists to suggest that there's a deep-seated link between conservatism and religiosity, such that the same innate attitudes and motivations that drive one also drive the other. A new study challenges this claim, providing evidence instead that the conservative/religiosity overlap is created largely by the contemporary political discourse, as peddled in the media. Among people who don't follow political news and commentary, Ariel Malka and his colleagues found that the typical conservative/religiosity link was largely absent.

Malka's team analysed data collected from 7,056 US citizens between 1996 and 2008. The participants answered questions about their degree and denomination of religiosity, their favoured political party (Democrat or Republican), their position on various political issues such as gun control and immigration, and their level of political engagement - for example whether they read the newspaper or followed political news.

The usual link between religiosity and conservatism was found. But crucially, this association was strongly moderated by political engagement. So, for those who were highly politically engaged, religiosity tended to go hand in hand with nearly every conservative characteristic that was measured, including party identification and views on gender roles, gun control and homosexuality. The only exceptions were immigration and the death penalty, for which religiosity predicted a more liberal view. In contrast, for people who weren't politically engaged, the religiosity/conservatism link was profoundly diminished, to just four of the twelve conservative characteristics that were measured.

Malka and his colleagues conceded that the remaining conservatism/religiosity link, even among those low in political engagement, suggests that there may be some truth in the idea of an organic link between the two belief systems. 'However,' they added, 'when considering the full range of preferences and values associated with "conservatism" nowadays, engagement with political communication seems to be the predominant factor that drives the alignment of religiosity and political orientation.'

What this suggests in simple terms is that politically engaged, religious Americans watch the news and listen to political commentaries and this leads them to shift towards more conservative values. Or, alternatively, it means politically engaged, conservative folk watch the news and listen to the commentaries, and this encourages them towards religion. Either way, or both ways, it seems these people are being swayed by the contemporary political discourse in the United States.

One final, alternative interpretation is that political engagement is affected by people's religiosity/political match-up, rather than affecting it. By this account, when people's politics and religion don't match, they choose to disengage from politics. Longitudinal research is needed to test this. 'We hope that the present analyses are supplemented with cross-national, time series, longitudinal, and experimental analyses to enhance understanding of how context of information influences the relation between these two socially significant constructs,' the researchers said.
_________________________________

Ariel Malka, Yphtach Lelkes, Sanjay Srivastava, Adam Cohen, and Dale Miller (2011). Religiosity, political engagement, and political conservatism. Political Psychology. In Press. pdf via author website.

Do political scandals really distract us from important issues?

Barely a day goes by without some political scandal or other splashed across the papers. Critics argue this obsession with tittle-tattle distracts the electorate from more important policy issues. '...a fiercely independent media is the guarantor of democracy,' Will Hutton wrote in 2000, before warning that the British media's obsession with scandal 'paradoxically, may be beginning to endanger it [democracy]'.

A new study by Beth Miller at the University of Missouri-Kansas City challenges the assumption that scandal is a distraction. Every two days, she presented 413 undergrads with a newspaper article containing information about a policy position held by a mayoral candidate. Then, 1 to 14 days later, she tested the students' memory for the candidate's policies. The important twist was that for half the participants, the fourth of five newspaper articles, rather than being about a policy, was about a scandal involving the candidate - in particular, his confession to an extra-marital affair.

The assumption of many would be that this story would distract participants from the drier, but arguably more important, detail of the politician's policies. Similarly, in psychological terms, it might be argued that the scandalous information would displace the earlier memory traces associated with policies, especially since negative information is known to be particularly memorable and attention-grabbing.

An alternative prediction, however, is that the salience of the scandal would actually benefit all other memories associated with the politician. This is consistent with the idea that memory is an 'associative network' made up of interconnected nodes. By this account, activation of one node - the one representing scandal - will spill over and raise the activation in all related nodes, thus benefiting participants' memory for the mayoral candidate's policies.

Miller found that more policy-related information was recalled by participants who read about the scandal, consistent with the associative-memory account. Moreover, compared with participants in the scandal condition who forgot about it (the scandal), those who remembered it were also more likely to remember policy information - reinforcing the idea that the scandal memory had benefited policy memories. As you might expect, although the scandal benefited participants' memory for policies, it also negatively affected the participants' evaluation of the candidate.

'While these results do not suggest that candidates can engage in scandalous activities without consequence, they do suggest that the depiction of the public as blind to anything but scandalous information seems to be an exaggeration,' Miller said. 'The results ... suggest that exposure to scandalous information ... may have beneficial side-effects not previously explored.'
_________________________________

Miller, B. (2010). The Effects of Scandalous Information on Recall of Policy-Related Information. Political Psychology, 31 (6), 887-914 DOI: 10.1111/j.1467-9221.2010.00786.x

Higher intelligence associated with "thinking like an economist"

As the world economy dusts itself down and edges towards recovery, a provocative new paper claims that people with higher intelligence are more likely to think like economists. That is, they're more likely to be optimistic about the economy; to recognise the economic advantages of markets free from government interference, and the advantages of foreign trade and foreign workers; and to appreciate the economic benefits of achieving greater productivity with less man-power. The lead author is Bryan Caplan, an economics professor at George Mason University. Past essays by him include 'The 4 Boneheaded Biases of Stupid Voters (And we're all stupid voters.)'

Prior research has established that the more time a person spends in education, the more likely their broad economic views are to match that of the typical economist (pdf). Caplan and his colleague Stephen Miller point out that these studies failed to take into account the influence of intelligence. After all, it's known that people with higher IQ tend to spend longer in education and intelligence itself may also directly influence economic beliefs.

To overcome this problem, Caplan and Miller have focused on answers to the General Social Survey, a massive US poll of national opinions performed every two years. Crucially, it includes questions about the economy and a small test of verbal IQ.

Caplan and Miller's finding is that the link between educational background and 'thinking like an economist' is weakened when IQ is taken into account because IQ is the more important factor associated with economic beliefs. It's a complicated picture because IQ and education may be mutually influential. However, if one assumes that education is unable to raise IQ, but that IQ affects time spent in education, then the researchers said 'the net effect on economic beliefs of intelligence is more than double the net effect of education.' Even if one assumes that education can also affect IQ, 'intelligence still has a larger estimated effect [on economic beliefs],' they said.

Does the link between higher intelligence and 'thinking like an economist' mean that economists are generally right and the public wrong? In answer to this question, Caplan and Miller cite Shane Frederick, a decision-making scholar at Yale's School of Management, who's previously argued that it depends on the type of question. For financial issues, he argued, it pays to emulate those 'with higher cognitive abilities'. However, Frederick noted that 'if one were deciding between an apple or an orange, Einstein's preference for apples seems irrelevant.'

Caplan and Miller say they agree with Frederick about this, before concluding boldly: 'The fact that the beliefs of economists and intelligent non-economists dovetail is another reason to accept the "economists are right, the public is wrong" interpretation of lay-expert belief gaps.'
_________________________________

Caplan, B., and Miller, S. (2010). Intelligence makes people think like economists: Evidence from the General Social Survey. Intelligence, 38 (6), 636-647 DOI: 10.1016/j.intell.2010.09.005

It's the way they move - politicians' personalities inferred from their motion patterns

People form impressions about the personality of politicians simply from the way they move, according to a new study. This isn't your typical body-language investigation into double-armed hand-shakes, bitten lips and fidgety fingers. Rather Markus Koppensteiner and Karl Grammer devised a new system for mathematically describing the movement patterns of forty real German politicians giving speeches in parliament. Each 16-second, silent video was converted into a stick figure by using an interactive computer programme to place dots on key landmarks such as the elbows, shoulders and forehead. The movement of these landmarks generated dynamic coordinate data which was then crunched according to mathematical characteristics including the 'turbulence coefficient' (i.e. periods of high activity followed by periods of low activity), and the activity levels of body parts such as the head and arms.

Next, 150 participants, mostly students, watched these same video clips and rated the personalities of the moving stick figures. The key finding was that there were correlations between the personality characteristics attributed by the students to the stick figures and the motion patterns of those stick figures as identified by the researchers' motion analysis programme. For example, politicians who displayed a high turbulent coefficient (low activity interrupted by periods of high activity) tended to be rated as more agreeable. Those who displayed more head movements were considered less conscientious, and those who showed low turbulence (i.e. consistently high activity levels) with many horizontal and vertical arm movements tended to be considered more extravert.

Koppensteiner told the Digest that these results are preliminary so caution is advised. For example, it's far too soon to consider body language training based on the findings. 'Telling someone to move a lot (which I found to be a signal for high extraversion) would not turn this person into an extravert. Maybe it would just look strange,' he said. Part of the reason for the caution is that these movement patterns were considered all together, so it's not obvious what effect it would have if just one element were changed. Nor is it known whether a trained pattern of movement would provoke the same personality attributions as when that same pattern occurs naturally. Finally, the personality ratings in the current study were made in the absence of any speech so we don't yet know how movement patterns and speech interact.

Notwithstanding those caveats, Koppensteiner said that the findings do have exciting implications - for example, if the technique were refined, it's possible that it may one day be possible to analyse a politician's movement patterns and predict the kind of personality attributions that voters are likely to make about them. Also, if these attributions reflect politicians' actual personalities, then an analysis of their movement patterns could be used as a way to discern their personality. 'Although at an exploratory stage,' Koppensteiner and Grammer concluded, 'the current study hints that charisma is partly linked to individual differences in successfully presenting oneself on a non-verbal level.'
_________________________________

Koppensteiner, M., & Grammer, K. (2010). Motion patterns in political speech and their influence on personality ratings. Journal of Research in Personality, 44 (3), 374-379 DOI: 10.1016/j.jrp.2010.04.002

Why it's time for the media to help our politicians believe they can succeed

A psychology study fresh off the presses shows the importance of positive expectations for the successful resolution of awkward negotiations. The results couldn't be more timely as our senior politicians negotiate over terms for a new coalition British government - the first since the 1970s. The finding suggests that the media has a vital role to play. By fostering optimism in the likely success of the negotiations, the media could help increase the likelihood of a successful resolution.

In an initial experiment, Varda Liberman and colleagues had undergrads negotiate with a postgrad (actually a confederate working for the researchers) over the division of university funds between the undergrad and post-grad student populations. Crucially, half the 34 participants were told that every single previous negotiating pair had managed to reach an agreement (the 'Positive Expectations' condition), whereas the other half of the participants acted as a control and were merely told to try their best to reach a mutually acceptable agreement. The negotiations followed a format whereby the confederate made an offer, the participant responded with a counter-offer, and the confederate replied with a final offer that the participant could either accept or reject. The confederate's offers were the same across the two conditions.

The key finding was that all 17 participants in the Positive Expectations condition accepted the final offer compared with just 5 out of 17 participants in the control condition. The Positive Expectations students also rated the offers of their negotiating partner, the confederate, as fairer and they felt more satisfied with the negotiation outcome.

The tension was raised in a second experiment which involved Jewish Israeli Business School students negotiating with an Arab Israeli woman over the division of funds between Israel and Palestine. Again, the positive expectations of half the students was manipulated by telling them that virtually all previous negotiations of this kind had ended in agreement. This time, 31 out 38 students in the Positive Expectations condition accepted their negotiating partner's final offer compared with just 13 out of 38 students in the neutral, control condition. Moreover, the Positive Expectation students were far happier about the negotiation outcome than the control students.

What was going on? Why should the knowledge that comparable prior negotiations ended in success change the way that people negotiate? One reason seemed to be that raised expectations of success led the students to make a more generous counter-offer which meant the gap between their counter-offer and the final offer was smaller. Positive expectations also seemed to change the way that the other party's offers were interpreted. The researchers said that under more pessimistic conditions the other party's offers are interpreted as likely to be in that party's own self interest. By contrast, positive expectations about the negotiation outcome foster a sense that the other party's offers are being made in a more constructive spirit, because they know 'that we need to reach an agreement'.

A problem when it comes to translating the lessons from this research to real life and particularly to the current negotiations among Britain's senior politicians, is that there isn't always a history of success available to inspire optimism. In fact Conservative Prime Minister Edward Heath's attempt to form a coalition with the Liberals in 1974 ended in failure. However, Liberman's team said this needn't be a fatal stumbling block - there are other means of encouraging a sense of optimism and positive expectations for a successful outcome, including 'mutual expressions of goodwill and commitment' or reference to successes in 'previous negotiations between the parties on other, more limited issues.' So far, that's exactly the spirit in which the negotiations seem to be taking place, with politicians on all sides making encouraging comments.

The researchers concluded: 'If our present research gives some basis for optimism about the possibility of bringing theory and research to bear in overcoming barriers to dispute resolution in a strife-worn world, we hope that such optimism will indeed prove to be self-fulfilling, and that practitioners and theorists will be able to find common ground in their efforts to resolve disputes peacefully.'
_________________________________

Liberman, V., Anderson, N., & Ross, L. (2010). Achieving difficult agreements: Effects of Positive Expectations on negotiation processes and outcomes. Journal of Experimental Social Psychology, 46 (3), 494-504 DOI: 10.1016/j.jesp.2009.12.010

Images courtesy of Wikipedia.

How to increase voter turnout

The political parties don't agree on much but what they do all agree on is that the more people who exercise their right to vote, the better. Psychology can help. A new study shows that phone calls to encourage people to vote can be made more effective by a simple strategy - that is, by asking the would-be voter to spell out what time they plan to vote, where they will be coming from prior to voting and what they will have been doing beforehand.

David Nickerson and Todd Rogers targeted 155,669 voters on the electoral roll in Pennsylvania. Frequent voters had been excluded, so these were people who'd chosen to vote only once between 2000 and the time of this study, which took place just prior to the 2008 presidential primary.

Would-be voters received one of three kinds of phone call: either they were encouraged to vote and reminded of their duty; they were asked whether they intended to vote; or they were asked more detailed questions about when, where etc they planned to vote. A control group received no phone call.

A classic study in the 1980s found that simply asking people if they intended to vote ended up making them more likely to vote - a phenomenon known as the 'self-prophecy effect'. However, this effect wasn't replicated here. Would-be voters in the current study, who were simply asked whether they planned to vote or not, were barely more likely to vote than the control group. Same story for the participants who received a call with encouragement to vote. By contrast, would-be voters who were asked questions about the when and where of their voting intentions were, on average, 4.1 per cent more likely to vote than controls.

There's a further twist. Digging deeper the researchers realised that the detailed questions about voting intentions only exerted an influence on would-be voters who were the sole eligible voter in their household. Focusing on just these people, the detailed voting intentions phone call led to an average 9.1 per cent increase in turnout. For people living in a household with multiple eligible voters, by contrast, the same kind of phone call was completely ineffective. Nickerson and Rogers think this difference probably arose because people living in a household with other eligible voters had already had conversations about when and where they planned to vote.

'This research contributes to a growing body of work using behavioural science to facilitate socially important behaviours,' the researchers concluded. 'Campaign professionals can use psychological science more widely to help citizens follow through on their intentions to vote.'
_________________________________

Nickerson DW, & Rogers T (2010). Do you have a voting plan?: implementation intentions, voter turnout, and organic plan making. Psychological science : a journal of the American Psychological Society / APS, 21 (2), 194-9 PMID: 20424044

Link to more Digest posts on the psychology of politics.

How framing affects our thought processes

A take-away restaurant near my house offers customers free home delivery or a ten per cent discount if you pick up. It sounds much better than saying you get no discount for picking up and suffer a ten per cent fee for delivery – this is the power of ‘framing’. Now David Hardisty and colleagues have dug a little deeper into framing, to show first, that these kinds of effects can interact with people's political persuasion, and second, that they can act by altering the order of people's thoughts.

Hundreds of online participants chose between various flights, computers and so on. In each case they could plump for a cheaper option or a more expensive, greener option, the latter including either a 'tax' to help reduce carbon emissions, or an 'offset' to do the same – depending on how the choice was framed. Whether the expensive option was framed as a tax or offset made no difference to Democrat (left-wing) participants. By contrast, Republicans (right-wing) and Independents were much less likely to choose the more expensive option when it was labelled as a tax.

In a second study the researchers added a technique known as 'concurrent thought listing', which involved the participants sharing their thoughts as they made their product choices.

This process revealed that when the expensive option was labelled as a tax, the Republicans and Independents, but not Democrats, had a consistent tendency to weigh-up the advantages of the cheaper option first before they considered the benefits of the greener choice. This is significant because past research shows that when we appraise options in sequence, the first item we consider tends to be favoured. Consistent with this, the tax frame led Republican participants to not only consider the cheaper option first but also to generate more supporting evidence for it. By contrast, when the expensive, greener option was labelled as an offset, political affiliation was no longer associated with the order in which options were considered, nor the weight of evidence generated for each option.

A final study tested whether the order in which we consider options really does have a causal role in our decision making. Participants of all political persuasions were instructed to consider the benefits of the greener, more expensive option first, whether it was labelled as a tax or offset. Despite this instruction, 54 per cent of Republicans failed to comply (showing just how averse they were to the 'tax' label). However, among those participants who did comply, this instruction had the effect of eliminating the interaction between framing and political affiliation – that is, the Republicans were no longer repelled by the greener, expensive option even when it was labelled as a tax.

‘Policy makers would be wise to note the differential impact that policy labels may have on different groups,’ the researchers concluded. ‘What might seem like a trivial semantic difference to one person can have a large impact on someone else.’
_________________________________

Hardisty, D., Johnson, E., & Weber, E. (2009). A Dirty Word or a Dirty World?: Attribute Framing, Political Affiliation, and Query Theory. Psychological Science, 21 (1), 86-92 DOI: 10.1177/0956797609355572

People really are happier in those US states identified as having better 'quality of life'

With our political leaders giving serious consideration to adopting population well-being, or 'happiness', as their ultimate goal (rather than economic prosperity), there is a greater need than ever to ensure that our scientific measures of the concept are valid. Prior research in this area has tended to involve asking large samples of people how satisfied they are with their lives. But how do we know that their answers are really trustworthy and accurate? Now Andrew Oswald and Stephen Wu have cross-checked an enormous sample of subjective well-being data from the USA to see if it matches up with economists' estimates of where people ought to be happiest based on quality-of-life data for different US states. Their encouraging finding is that there is a most impressive correspondence between the two data sets. Indeed Oswald told the Digest that he almost didn't believe his computer screen when the results came up. The likelihood of the two sets of data corresponding so well by chance is '1 in 10,000' he told me.

Oswald and Wu obtained subjective well-being data for over 1.3 million US citizens using the United States Behavioural Risk Factor Surveillance System. This allowed them to arrive at average well-being or 'happiness' estimates for the different states in the US. They then compared these estimates with quality-of-life data published by Stuart Gabriel in 2003. He and his colleagues looked at about 25 factors, including weather, crime, and commuting time. Whereas a magazine might give these factors equal weighting and calculate a state's desirability by summing its scores across the 25 factors, Gabriel's team used average house prices and wages to ascertain the importance of each factor to potential residents. If a state's house prices and wages remained high despite a wet climate, for example, this would suggest that rain is less important than other factors. Crucially, the subjective data matched the economists' league table of states. People seem to be happier in those places that the economists have identified as having a high quality of life.

'The study's finding suggests that subjective well-being data contain genuine information about the quality of human lives,' the researchers concluded.

So, which states in the US were the most and least happy? Top of the happiness league were Louisiana and Hawaii. Bottom were Connecticut and New York. California didn't fare much better. Commenting on California and New York's dismal showing, Oswald said: 'Many people think these states would be marvellous places to live in. The problem is that if too many individuals think that way, they move into those states, and the resulting congestion and house prices make it a non-fulfilling prophecy.'
_________________________________

AJ Oswald, & S Wu (2009). Objective confirmation of subjective measures of human well-being: Evidence from the USA. Science.