Language is the primary medium through which we convey emotions, ideas, and intentions. It is the basis for the complex exchange of information that is crucial for social interaction and community building. And while we recognize the omnipresence of language, it may be less apparent that language serves as more than just a mechanism of communication, it also offers a rich source of information into the minds of the people and groups when they communicate. Whether written or spoken, language reveals profound psychological insights into who we are and how we engage with the world around us.
The field of language psychology (also referred to as psycholinguistics) is dedicated to extracting valuable psychological insights from language data.
Psycholinguistics has a significant history. In the early days of psychology research, Sigmund Freud observed that speech errors, contemporarily nicknamed Freudian slips, could reveal people’s unconscious intentions. This observation laid the foundation for the theory that language serves as a window into the human psyche. The field made significant strides with the emergence of computerized text analysis and the digitization of large collections of written and spoken texts, which made it possible to process and analyze large amounts of text data more efficiently. In 1962, the General Inquirer was developed, a program that employed algorithms to detect emotions and themes in text. During the 1970s and 80s, researchers pioneered language analysis methods that involved counting keyword frequencies in texts to identify linguistic markers of psychological phenomena like mental disorders or depression.
Today, Linguistic Inquiry and Word Count (LIWC) is the most widely used and validated tool in language psychology research. First introduced in 1993, the program includes 100+ psychologically-relevant categories of words, whereby the words contained in each category have psychologically meaningful implications. For example, the pronouns category in LIWC includes words like I, she, and we. LIWC automates text analysis by counting the words that match its categories and then generates scores that reflect the proportion of words in the analyzed text that correspond with each category.
Psycholinguistics recognizes the importance of both content words and function words.
Psycholinguistics derives insights from two primary types of language: content words and function words. Content words capture what people are saying. These words constitute more than 99% of the English vocabulary and are used in sentences to convey context and meaning. For example, in the sentence I had a beautiful and sunny morning, the content words are beautiful, sunny, and morning.
Methods of content word analysis include topic modeling and keyword analysis. Topic modeling typically involves a bottom-up approach, identifying words within text samples that frequently appear together. On the other hand, keyword analysis adopts a top-down approach, beginning with the creation of a list of terms associated with a specific theme or topic of interest.
While content words focus on what people are saying, function words capture how people communicate. Despite constituting only 0.05% of the English vocabulary and possessing limited meaning on their own, function words make up approximately 55% of the words people use in regular communication. In the sentence I had a beautiful and sunny morning, the function words are I, a, and and. When analyzed using psycholinguistics, function words offer insight into the cognitive, personality, and interpersonal-dynamics aspects of communication.
Many of LIWC’s categories analyze the rates at which different categories of function words (e.g., pronouns, prepositions, articles, etc.) occur in language. For example, researchers have demonstrated that high rates of first person singular pronouns (I, me, my) are associated with personality traits such as neuroticism and humility, as self references in language are indicative of increased self-focus (a phenomenon linked to depression and lower group status). Other interesting findings include that people who are more analytical, formal, and structured in their writing and speeches tend to use more articles (a, an, the) and prepositions (to, with, above), function words that help identify the object of the sentence and can provide more detailed and concrete information.
Some analysis methodologies, like our emotions engine (SALLEE), analyzes both function words and content words. For example, SALLEE recognizes the difference in emotional intensity in the sentences I am not really happy and I am really not happy by recognizing the difference in the syntactic order of the content words really and happy and the function word not.
Naturalistic language data is the gold standard in psycholinguistics.
It encompasses a wide range of data sources, including transcribed conversations, social media posts, emails, survey responses, and more—essentially, any form of language data that reflects real-life interactions and is unscripted. This type of data holds immense value for psychological insights because it sheds light on how people truly think and feel.
Nevertheless, the analysis of scripted or extensively edited language can also offer intriguing insights. By studying language such as speeches, company memos, blog posts, and news articles, one can gain insights into individual’s and organizations’ intended persona or brand identity.
Companies and researchers can leverage Receptiviti’s psycholinguistics platform to derive meaningful insights into the psychology of groups and individuals from their language data.
Receptiviti is the leader in understanding the social and psychological meaning behind language. With LIWC as our core science, our technology extracts insights related to personality traits, emotional states, cognitive markers, social dynamics, motivations, and more, all from the words people use. To learn more about how your company can benefit from language psychology insights check out our blog or contact us.
