Tying it all together

'I've never had any goals in my life, and I think that's just fine'. So opened a funny, witty keynote speech by James Pennebaker, which outlined his journey from physical symptoms, to expressive writing and eventually… Donald Trump. The University of Texas Professor began his prolific research career looking into how people read their bodies, and why some struggle so much with apparent ill health.

In his early work Pennebaker found students who had experienced early sexual trauma had the worst health, and further work revealed any major trauma people kept secret had the same effect. This secrecy, Pennebaker said, seemed to be particularly important. So what might happen if these people were asked to write about it in a lab?

He asked participants to write about a traumatic experience they hadn't spoken about much, for 15 minutes per day for four days. Despite an initial hour or so of feeling down after writing, the measurable results were quite incredible. Pennebaker accessed the records of participants from the University Medical Centre – the experimental group visited the medical centre at half the rate of controls, while further studies saw changes in immune function in people who did expressive writing.

Why might this work, and for whom? Pennebaker suggested the mere act of labelling and acknowledging emotions can help to organise experiences, construct narratives and bring parts of a trauma together. Also when people have a 'still mind' they make better humans and better friends: this in turn brings about positive changes in an individual's social engagement.

Pennebaker became interested in the content of people's expressive writing and he and colleagues developed a computer programme LIWC (Linguistic Inquiry and Word Count) to categorise words within writings into linguistic, psychological and topical categories.

He found a number of writing changes throughout expressive writing which predicted health improvements; the more people used positive, emotional words in writing, the more likely they were to benefit. Even if the positive words appeared in a sentence such as 'I'm not happy, no one cares, there's no joy', that was a start: at least they were somewhere on the continuum.

Later he began looking into the use of content words, which make up about 99 per cent of most people's vocabulary, and function words, the pronouns, prepositions and articles which make up around 1 per cent of vocabulary but about 65 per cent of all the words we say and hear. Pennebaker said these function words are processed by the brain in a distinct way.

They are the hardest to master when learning a second language, and perhaps most importantly these words are social – by analysing them we can get into the head of the author and the relationship between an author and target, he added.

There are many examples of the power of these function words. Surprisingly while we might think that people who use the word 'I' more are raging narcissists, they tend to be people who are depressed. In another surprising finding Pennebaker has found people who use the word 'I' more tend to be lower status: this could reflect a leaders' focus on other members of a group rather than themselves, an attribute Pennebaker calls clout.

In spoken interactions the function words people use, and whether people match their styles of using such words, can be extremely telling. In one study Pennebaker looked at the instant messages of 86 young college couples and could predict whether they'd be together three months later (even though there was no relationship between their own ratings and predictions for the future relationship).

Factor analyses of function words can reveal much about thinking styles too. Pennebaker has also looked at students' college application essays. The high use of nouns, articles, and prepositions reveals someone thinks in an analytical way, while narrative thinking is linked to a lower use of function words and higher use of pronouns, auxiliary verbs and common adverbs. Interestingly analytic thinking correlates with IQ and social class, and better performance at university… it is a marker of complex thinking.

Worryingly, then, Pennebaker revealed he and colleagues have been looking at the words used by the current US President Donald Trump and found he was the lowest on analytical thinking of anyone they had examined. Following this he analysed presidential debates and inaugural addresses going back to George Washington, and found a pattern; since 1980 the candidate with the lower score in analytic thinking has won (apart from Bill Clinton).

In fact, since around 1910 the level of analytic thinking among presidents has been declining in a linear fashion, and the five presidents with the lowest levels of analytic thinking have been last six presidents. Clout, on the other hand, has been increasing in a linear fashion since Woodrow Wilson – Trump and Obama had almost identical levels. Pennebaker concludes voters look for presidents with confidence who don't appear that intelligent. Worryingly, this same pattern can be seen for British Prime Ministers.

Pennebaker took some time to assure people, particularly recent graduates, that while his career seems well planned out, it hasn't been. He said: 'I've been bouncing around and gone this way and that way. For those of you thinking about creating a career, it's really fun doing that… the art form is tying it together to make it sound more coherent afterwards.'

- Read our interview with Professor Pennebaker.

- You can find lots more coverage from the Society's Annual Conference, including more about the online world, and coming up in the July print edition.