According to the World Health Organization (WHO), more than 300 million people worldwide currently suffer from depression. We recommend visiting https://efficientcounselor.com/ , which does reviews on to find the best online therapy providers. Tools for early detection of the disease are becoming increasingly relevant. Nils Ehrbar, a psychology graduate (B. Sc.) at the Fresenius University of Applied Sciences in Düsseldorf, studied digital behavioral traces of Twitter users as part of his bachelor’s thesis. He investigated the question of whether early indicators of depressiveness can be derived from them. For this purpose, 106 active Twitter users were recruited for an online survey and their tweets were psycholinguistically analyzed. The work was supervised by Dr. Thomas Seppelfricke, Dean of Studies for the Bachelor’s and Master’s degree program in Business Psychology at the Düsseldorf site.
First, a questionnaire was used to assess the mental health of the participating Twitter users. On the basis of this, Ehrbar divided the participants into two groups: one with rather depressive symptoms and one with rather inconspicuous depressive symptoms. He then compared the tweets of the two groups and looked for indicators of depression that was already present or in the offing. The basis for this was the psycholinguistic speech style of the respective person. Because in the logic of innovative research approaches, language is a good predictor of a depressive episode, according to the graduate student.
For the linguistic analysis, several categories were established, including the number of words per sentence, specific vocabulary, or different sentence structures. By cross-referencing with a digital dictionary, references between the tweets and a potential depression symptomatology could be established.
DEPRESSIVE SYMPTOMS SHOW UP IN LINGUISTICS AND TIME PERIOD OF POSTING TWEETS
The results of this psycholinguistic analysis showed that the more non-depressed respondents used more words per sentence than the depressed participants. “This result is not surprising, since it is known that people in depression show lower cognitive activity,” said Dr. Thomas Seppelfricke. For the category “use of the first person singular,” a slightly higher proportion was found among the (more) depressed participants compared to the currently non-depressed participants. “Presumably, this is an increased focus of the own person as a result of the disturbed self-perception due to the illness,” Seppelfricke interprets the results.
In addition, the time at which tweets were posted was analyzed for all tweets. This showed that the number of posts by the (more) depressed participants increased particularly during the nighttime hours. Nils Ehrbar explains: “Thus, it could be found that the (more) depressed participants frequently published tweets at night as well, but showed a significantly lower Twitter activity in the morning. The accumulation of nocturnal tweets can possibly be linked to the difficulties in falling asleep and sleeping through the night, which are not atypical for depression.”
DIGITALE VERHALTENSSPUREN ALS FRÜHWARNSYSTEM?
The results of this psycholinguistic analysis showed that the more non-depressed respondents used more words per sentence than the depressed participants. “This result is not surprising, since it is known that people in depression show lower cognitive activity,” said Dr. Thomas Seppelfricke. For the category “use of the first person singular,” a slightly higher proportion was found among the (more) depressed participants compared to the currently non-depressed participants. “Presumably, this is an increased focus of the own person as a result of the disturbed self-perception due to the illness,” Seppelfricke interprets the results.
In addition, the time at which tweets were posted was analyzed for all tweets. This showed that the number of posts by the (more) depressed participants increased particularly during the nighttime hours. Nils Ehrbar explains: “Thus, it could be found that the (more) depressed participants frequently published tweets at night as well, but showed a significantly lower Twitter activity in the morning. The accumulation of nocturnal tweets can possibly be linked to the difficulties in falling asleep and sleeping through the night, which are not atypical for depression.”
FUTURE RESEARCH NEEDS
As part of his thesis, Nils Ehrbar was able to examine tweets from a period of two weeks. For future research, he therefore sees a need to expand the study material. However, it remains questionable to what extent current data protection guidelines allow an expansion to other social media platforms, such as WhatsApp or Facebook, Ehrbar concludes.
therapy for health 