O.C.E.A.N. / Five Factor Model

Fun Fact: Our first start-up, where we laid the foundation of what is now Mnemonic, had a whale as a logo. Today we dive deep into the ocean. Or better: O.C.E.A.N.

As we use OCEAN today, it is an acronym of different personality traits:

  • Openness to experience
  • Conscientiousness
  • Extraversion
  • Agreeableness
  • Neuroticism

Want to learn more about applications of the five factor model?

Ocean is one of the most, if not the most researched, models to classify psychological traits. Research dates back until the 1960s (Tupes EC, Christal RE (1961). “Recurrent personality factors based on trait ratings”). You can even argue that Hippocrates’s four types of temperament or Galtons lexical hypothesis were a precursor to the big five.

Coincidence sometimes plays a significant role in research. It sometimes seems the time is just right for an idea, see electricity, for example.

In the 1980s and 1990s, four independent research teams tried to find a way to classify psychological traits, Tupes and Cristal, and Goldberg at the Oregon Research Institute, Cattel at the University of Illinois, and Costa and McCrae. These four independent researcher groups used different data and methodology and came up with the same personality traits and five factors to describe their findings.
Even if the name and descriptions differed slightly, the results have been found to be intercorrelated and aligned.

Each of the five traits we know as OCEAN contains two related but opposed aspects. The scales for openness to experience range from Intellect to Openness, Industriousness, and Orderliness for Conscientiousness, Enthusiasm, and Assertiveness for Extraversion, Volatility, and Withdrawal for Neuroticism.

Highly extroverted individuals are assertive and sociable, rather than quiet and reserved. Agreeable individuals are cooperative and polite, rather than antagonistic and rude. Conscientious individuals are task-focused and orderly, rather than distractible and disorganized. Neurotic individuals are prone to experiencing negative emotions, such as anxiety, depression, and irritation, rather than being emotionally resilient. Finally, highly open individuals have a broad rather than narrow range of interests, are sensitive rather than indifferent to art and beauty, and prefer novelty to routine. (Soto & Jackson 2018)

The big five psychological traits / Ocean have been applied and researched even in non-humans. Studies in primate research and dolphins showed some psychological traits in these animals.

In human research, twin studies and heritability are researched, and of course, cultural differences, and gender and how they influence psychological traits (Schmitt DP, Realo A, Voracek M, Allik J (January 2008). “Why can’t a man be more like a woman? Sex differences in Big Five personality traits across 55 cultures”).
Nowadays, over fifty studies related the Five-Factor Model to personality disorders (Widiger TA, Costa PT. Jr. “Five-Factor model personality disorder research.” In: Costa Paul T Jr, Widiger Thomas A., editors. Personality disorders and the five-factor model of personality. 2nd. Washington, DC, US: American Psychological Association; 2002. pp. 59–87. 2002).
Lee Anna Clark, a psychology professor at Notre Dame, even concluded: “the five-factor model of personality is widely accepted as representing the higher-order structure of both normal and abnormal personality traits.”

UT Austin professor of psychology, James W. Pennebaker, focused his research on language according to psychological traits dating back to 1993. The manual to a psychological framework he developed opened with the following passage:
“The ways people use words in their daily lives can provide rich information about their beliefs, fears, thinking patterns, social relationships, and personalities. From the time of Freud’s writings
about slips of the tongue to the early days of computer-based text analysis, researchers began
amassing increasingly compelling evidence that the words we use have tremendous
psychological value” (Gottschalk & Glaser, 1969; Stone, Dunphy, Smith, & Ogilvie, 1966;
Weintraub, 1989).”

His framework consisted of a lexical approach, where semantic and lingual features are mapped regarding psychological traits, emotions, and cognitive components.

Today we move beyond lexical approaches and utilize deep neural network architectures that can pick up the slightest nuances of language and map it towards distinct traits while maintaining information like demographics, click paths, and other rich data that help predict the outcome.

Mnemonic AI researched the possibility of decoding psychological traits for the last few years. After thousands of hours of computation, the result is a deep neural network that uses 455 million parameters to classify psychological traits based on language.
Our system needs no more than 128 characters of input, something is written or said, to deliver on the full OCEAN scale.