A Comparison of Danish Listeners’ Processing Cost in Judging the Truth Value of Norwegian, Swedish, and English Sentences

Ocke-Schwen Bohn, Trine Askjær-JørgensenInterspeech 2017

TL;DR

This study investigated how much cognitive effort Danish speakers expend when understanding sentences in their native language versus Norwegian, Swedish, and English. They found that while English comprehension is nearly as effortless as Danish, understanding Norwegian and Swedish requires significantly more processing power, challenging claims of effortless inter-Scandinavian communication.

What Problem Does This Paper Address?

The paper tackles the question of inter-Scandinavian language intelligibility, specifically addressing the claim by the Nordic Council that Danish, Norwegian, and Swedish are so closely related that understanding one requires minimal effort for speakers of another. While these languages share many similarities, previous research has suggested that actual intelligibility might not be as high as claimed. Furthermore, Danes are extensively exposed to English through education and media from a young age, raising the question of whether English comprehension might, paradoxically, be easier for them than understanding their closest linguistic neighbors. The study aims to quantify this "processing cost" – the cognitive effort required to understand – for Danish listeners when processing sentences in these three languages compared to their native Danish.

Methods

The researchers employed a sentence verification task. Native Danish participants listened to simple, true or false statements presented in Danish, Norwegian, Swedish, and English. Their task was to quickly and accurately judge the truth value of each sentence. The "processing cost" was measured using two key metrics:

  • Response Time (RT): The time taken from the end of the sentence to the participant's response. Longer RTs indicate higher processing cost.
  • Proportion of Correct Responses: The accuracy of the truth value judgments. Lower accuracy suggests higher processing cost.

Three separate experiments were conducted, each comparing Danish comprehension against one of the non-native languages (Norwegian, Swedish, or English). Participants were randomly assigned to one of these experimental groups. The presentation order of Danish and non-native language sentences was counterbalanced to control for potential order effects.

Datasets

The study utilized a set of 40 sentences (20 true, 20 false) that were translation equivalents across Danish, Norwegian, Swedish, and English. These sentences were adapted from a previous study by Munro & Derwing [8]. An additional set of 10 familiarization sentences (5 true, 5 false) was also used. The sentences were recorded by native speakers of each language variety (Southern British English and General American for English) and were produced at a normal speaking rate.

Key Insights and Ideas

  • English vs. Scandinavian Languages: The most striking finding is that Danish listeners experience a processing cost for English that is comparable to their native language, while Norwegian and Swedish impose a significantly higher cognitive load. This suggests that extensive exposure and education in English have made it functionally "easier" for Danes to process than their closest linguistic relatives.
  • Challenging the "Nordic Council" Claim: The results directly contradict the notion of effortless inter-Scandinavian communication. The data indicate that understanding Norwegian and Swedish requires substantially more effort for Danish speakers than understanding Danish or English.
  • Processing Cost as a Differentiator: Response time and accuracy serve as robust indicators of cognitive effort. The clear divergence in these metrics between English and the Scandinavian languages highlights the practical implications of language exposure and learning.
  • Potential for Asymmetry: While this study focused on Danish listeners, the findings hint at potential asymmetrical intelligibility within Scandinavia, suggesting that the ease of understanding might not be reciprocal across these languages.

Key Results

  • Response Times: Mean response times for Danish and English sentences were both under 1 second and did not differ significantly. In contrast, response times for Norwegian and Swedish sentences were significantly longer (over 1 second).
  • Correct Responses: The number of correct responses for Danish and English sentences was near the ceiling (maximum 20), indicating high intelligibility. For Norwegian and Swedish sentences, the number of correct responses was significantly reduced, with nearly 25% of these translation equivalents being processed incorrectly.
  • Statistical Significance: Paired t-tests and Wilcoxon Signed Rank tests confirmed significant differences in both response times and accuracy between Danish and Norwegian/Swedish, but not between Danish and English for response times.

Limitations and Future Work

  • Ecological Validity: The authors acknowledge that the study's sentences were presented as "stand-alone" utterances without any linguistic or extra-linguistic context. This limits the ecological validity, as real-world communication is typically richer.
  • Future Directions:
    • Subjective Comprehensibility: The paper mentions a follow-up study where participants rated the difficulty of each non-native sentence. Comparing these subjective ratings with objective measures (RT, accuracy) will provide deeper insights.
  • Fine-grained Analysis: Future work aims to analyze differences in processing cost at the lexical, morphosyntactic, or phonetic/phonological levels to pinpoint the exact sources of difficulty.
  • Reciprocal Studies: The researchers plan to conduct similar studies with Norwegian and Swedish listeners to investigate whether inter-Scandinavian intelligibility is symmetrical or asymmetrical.

Summary updated

BibTeX
@inproceedings{bohn2017comparison,
  author = {Ocke-Schwen Bohn and Trine Askjær-Jørgensen},
  title = {{A Comparison of Danish Listeners’ Processing Cost in Judging the Truth Value of Norwegian, Swedish, and English Sentences}},
  booktitle = {Interspeech},
  year = {2017},
}

Loading related content...

Figure