Google search queries and social media posts provide a means of paying attention to the thoughts, concerns, and expectations of millions of people around the world. Using the right web-scraping bots and Big Data analytics, everyone from marketers to social scientists can analyze this information and use it to make conclusions about what is on the minds of the overwhelming population of users.
Can AI analysis of our dreams help in doing the same thing? This is a bold, intriguing, concept – and one that researchers at Nokia Bell Labs in Cambridge, UK are busy exploring. They have created a tool called “Dreamcatcher” that, they claim, uses the latest natural language processing (NLP) algorithm to identify topics from thousands of written dream reports.
Dreamcatcher is based on an approach to dream analysis called the continuity hypothesis. This hypothesis, supported by strong evidence in research dreams for decades, suggests that our dreams are reflections of the everyday concerns and thoughts of dreamers.
This may sound like common sense. But it is a much different way of thinking about dreams than the more complex interpretations put forth by theorists such as Freud and Jung, who see dreams as windows into hidden libido desires and other usually ambiguous thought processes .
Automated dream analyzer
The AI tool – which Luca Aiello, a senior research scientist at Nokia Bell Labs, described as an “automated dream analyst” – parsed a written description of dreams and then scored them according to an established dream analysis inventory called Hall-Van Day. did. The scale of the palace.
“This inventory consists of a set of scores, which measure from the various elements shown in the dream that some are more or less consistent than some of the standard values established by previous research on dreams,” Aiello said. “These elements include, for example, positive or negative emotions, aggressive interactions between characters, the presence of fictional characters, etc. The scale does not, per se, provide dream interpretation, but it does determine interesting or odd aspects in them. Helps. “
The Dream Report written comes from a collection of 24,000 such records, taken from Dreambank, the largest public collection of English-language dreams yet available. The team’s algorithm is able to pull these reports apart and reassure them in a way that makes sense to the system – for example, “fictional creatures,” “friends,” “male characters,” “female characters. Sorting references into categories like “. And so on. It can then classify these categories by filtering them into groups such as “aggressive”, “favorable”, “sexual” to indicate different types of interactions.
By focusing on the person recording the dream and its contents, researchers can discover some interesting links. A written record might be something like this: “I was in a house. Ezra and a friend were at the computer. This untold thing kept running towards me as soon as I opened the door. There were other strange creatures and were like chickens. They kept trying to attack me. “The Dreamcatcher tool can begin with this description and automatically extract various insights; Eventually filing it “under juvenile concerns and activities”. (The dream was, in fact, recorded by Eazy, the “teenage girl”.)
Aiello said that some of these insights are expected, while others reveal surprising lines of possible future questioning. “For example, a teen’s dreams were characterized by an increase in the frequency of sexual relationships as she approached her adult life,” Aiello said. “More amazingly, we found that visually impaired people have more fictional characters than the ideal, which suggests that our senses affect the way we dream.”
Such analysis is something that even psychologists looking at this data can do – though nowhere near as quick as an AI tool. “It bears witness to the increasing ability of NLP and the increasing ability to capture complex and intangible aspects of language,” Alilo said. “However, it is even more exciting to think that these techniques gave us the ability to do dream analysis on a much larger scale, something that would be impossible with the time-consuming process of unveiling dreams.”
sweet Dreams are made of these
Compared to the Dreamcatcher system calculated by psychologists, the AI algorithm matched 76% of the time. This suggests that further improvements can be made. However, this is a valuable start. Aiello – along with fellow researchers Alessandro Fogli and Danielle Quercia – believe the finished product can have deep applications.
“As more and more people volunteered to share their dreams, we envision the possibility of analyzing the dreams of an entire population – of an entire country – to monitor their psychological well-being over time”
There may be something like a mood-tracking app that asks users to record their dreams, and then fires out recurrent imagery over a certain period of time. Aiello said that such a tool can make reporting a habit a daily dream for people; Rewarding them with on-the-fly dream analysis.
However, the more intriguing concept is the one described at the beginning of this article: a kind of large-scale dream-tracking project that can map the dreams of the world to real events to see how one informs the other. As with many other forms of big data analysis, it will become more useful – and panoramic – as much as it was combined and cross-referenced with other real-world data.
“As more people work voluntarily to share their dreams, we envision the possibility of analyzing the dreams of an entire population – even that of an entire country – over time in their psychological well-being. To monitor, ”Aiello said. “Clearly, this will only be possible with the use of automated tools like ours that make dream analysis possible on a large scale. This occasion will be especially compelling in view of the global challenges that affect everyone’s psyche. Today it is COVID, next year it will be prone to economic crisis, and in three or four years it could be global warming. “
Editors recommendations
