People's everyday language may be more fundamentally rooted in a tendency to prioritize safety over expressing happiness or positive emotions, according to a recent research finding.

On the 24th (local time), citing the science media outlet ScienceDaily, a research team at the University of Vermont analyzed billions of instances of over 20,000 words used in actual sentences and confirmed that a preference for safety appears in human language.

The researchers explained that not only emotion, but also power, danger, and order play crucial roles in understanding the meaning of language. This study was published last May in the international academic journal Science Advances.

Human language has been shown to reflect the characteristic of distinguishing between safety and danger more significantly than positive emotions. (Photo = Clip Art Korea)

In traditional linguistics and psychology, the meanings of words have mainly been explained through three dimensions: valence (positive/negative), arousal (excited/calm), and dominance (submissive/dominant). This is the so-called "VAD model," which has been utilized since the 1950s.

However, the research team determined that this approach alone is insufficient to fully explain the meanings embedded in language. To address this, they discovered a new semantic structure by analyzing words actually used in books, news, social media, and broadcast conversations.

As a result, it was found that explaining word meanings through three dimensions—▲power (weakness and strength), ▲danger (safety and danger), and ▲structure (order and chaos)—is more appropriate.

The new model explained more than 90% of the semantic variance of words measured by the researchers. The explanatory power of the existing VAD model was about 72%.

In particular, the researchers identified a trend where words related to safety are used more frequently than words indicating danger across various linguistic data.

This provides a new interpretation of the existing "Pollyanna principle," which holds that human language uses more positive expressions than negative ones.

The researchers analyzed that this phenomenon may stem from people using language to indicate safe and predictable states rather than simply expressing happiness or positive emotions.

Peter Dodds, a professor at the University of Vermont who led the research, explained that expressions denoting safety play a vital role across human language overall.

The research team also noted the possibility that this tendency is related to human survival. People constantly judge whether their surrounding environment is safe, whether others can be trusted, and what situations are dangerous in daily life. This characteristic may have been reflected in the semantic structure of language as well.

This study does not prove that humans actually value safety more than happiness. What the researchers confirmed is a trend appearing in word meanings and usage frequencies, and further research is needed to determine whether the cause was formed through the evolutionary process.

These results could also impact how artificial intelligence (AI) understands human language.

Currently, many AI systems utilize technology to analyze the positive and negative sentiments of sentences. The research team expects that incorporating factors such as power, danger, and order will enable AI to more sophisticatedly analyze trust, risk, and contextual nuances embedded in sentences.

[Copyright (c) Global Economic Times. All Rights Reserved.]