This inference was drawn out, in terms of human thinking behaviour, through a detailed analysis of seven-billion words that formed 800-million tweets sent and received across Twitter.
To achieve this monumental task of analysis, scientists from the University of Bristol, U.K., deployed artificial intelligence. Initially the researchers aggregated and anonymised twitter content, which was sampled each hour over a period of four years. The tweets were drawn from users residing in 54 of the U.K.’s biggest cities, by population.
From this analysis the researchers sought to answer the research question and hence determine if human thinking modes change collectively. Exactly how many thinking modes there are is disputed between psychologists, with the consensus range falling between four and six.
The analysis produced several interesting findings based on variations in language, especially the association of key words, at different points in time together with 73 psychometric indicators.
For example, the findings revealed that at 6 am, analytical thinking tended to peak, with the words used correlating with a logical way of thinking. This included a high use of nouns, articles and prepositions. The early morning period demonstrated general concerns with achievement and power.
By the time evenings began, this form of thinking faded, and the thinking style become far more emotional. Here language tended to be more impulsive, social, and emotional.
The essential finding was that human language changes significantly between night and day, which seems to reflect changes in human concerns, which reveals underlying cognitive and emotional processes. These shifts coincide with what is happening to human physiology, matching changes in neural activity and hormonal levels, and thus there appears to be a connection with the circadian clock.
Circadian clocks are the central mechanisms that drive circadian rhythms, which are biological process that displays an endogenous, entrainable oscillation of about 24 hours. Disrupting these cycles, such as shift work, has been associated with ill-health effects.
The research has been published in the journal PLoS One, with the paper titled “Diurnal variations of psychometric indicators in Twitter content.”
