The Network-Based Interpretation of Dreams

By mapping the links between themes that appear in dreams, network scientists reveal the connections between dreams in different cultures for the first time

One of the central figures in the study of dreams is Calvin Hall, an American psychologist who developed the cognitive theory of dreams in the 1950s. Hall’s theory was based on his study of several thousand dream reports from ordinary people living all over the world, which he analysed for common themes and ideas.

His conclusion was that all dreams are based around three common concepts: the idea of the self; the idea of friends, family and other people; and the idea of the social environment.

Since then, the progress in dream research, or oneirology, has been slow. Despite the universal experience of dreaming, psychologists and neuroscientists have little understanding of its purpose or its mechanisms. So new approaches in this field are welcome.

Today, Onur Varol and Filippo Menczer at Indiana University in Bloomington, use network science to study the similarities between dreams reported in English, Chinese and Arabic—research that is funded in part by the Defence Advanced Research Projects Agency or DARPA.

They say their results provide a new window into the cultural links between dreams experienced by people in different parts of the world. What’s more, the work clearly back Hall’s cognitive theory.

Network science has changed the way researchers study a wide range of disparate subjects. The networks formed by the web of links between friends and contacts have become well-known thanks to the explosive emergence of social networking sites.

Less well known is the fact that similar networks exist between characters in many important works of literature, such as the Odyssey in the Iliad, various fairy tales and the mediaeval Icelandic sagas. Network science is also providing unprecedented insight into the way biomolecules interact inside body and hence into the nature of disease; into connectivity between regions in the brain; and even into the links between ingredients in recipes from different parts of the world.

Dreams have recently become amenable to this kind of study because dream reports and their interpretations are now widely available on the web. For example, the DreamBank database compiled at the University of California, Santa Cruz, provides a description of around 1400 symbols that appear in some 20,000 dream reports in English. Examples of symbols include ‘love’, ‘ladder’, ‘quarrel’, ‘voices’ and so on. Similar databases also exist for Chinese and Arabic.

Varol and Menczer began by crawling these databases for symbols and their descriptions. They then measured the similarity between symbols by comparing their descriptions. Next, they created a network for each language in which symbols are nodes and drew links between symbols that were similar.

The result was three networks linking the dream symbols found in English, Chinese and Arabic.

They then analysed each network looking for communities of symbols that are more densely linked. It turns out that these communities contain symbols that share a common theme. For example, one community in English that contains the symbols ‘goal’, ‘hill’ and ‘ladder’ represents achievement after a struggle, say Varol and Menczer. The community with the symbols ‘voices’, ‘mailman’ and ‘message’ represents warnings. And the community with the symbols ‘hiding’, ‘guard’ and ‘raincoat’ represents protection.

Similar communities also emerge in other cultures. “The Chinese community that contains the symbols `rainbow,’ `poor,’ and `quarrel’ tells about good things that will happen in the future, while the one with the nodes `road accident,’ `woman,’ and `illusion’ contains symbols about forewarning and need of life changes,” say Varol and Menczer.

In Arabic, there are distinct communities representing religious characters and objects.

Finally, they compared the structure of the networks in different languages to see how similar symbols are related. “We built a multicultural dream network by identifying interconnections between language layers,” say Varol and Menczer.

It turns out that there are strong similarities between these networks. “These results indicate that…symbols tend to be connected to the same neighbors across languages,” they conclude.

That’s an interesting result that provides good evidence supporting Hall’s cognitive theory of dreams, say Varol and Menczer.

What’s more, it has significant potential for future research. For example, Varol and Menczer say they would like to study things like how often symbols appear in the same dream together and so get a better understanding of the associations between them.

It is also well placed to benefit from forthcoming advances in natural language processing, which should allow more accurate comparison of concepts and sentiment analysis.

Just how this work fits in to DARPA’s broader mission to create breakthrough technologies for national security is rather obscure. But a better understanding of the similarities and differences between cultures is surely in everyone’s interests.

But exactly what kind of insight this work gives into the purpose and mechanisms of dreams, isn’t entirely clear.

Perhaps that’s just a measure of our ignorance about this fascinating but poorly understood subject.

Ref: arxiv.org/abs/1402.2297 : Connecting Dream Networks Across Cultures