An official website of the United States government
Abstract Surprisal theory posits that less-predictable words should take more time to process, with word predictability quantified as surprisal, i.e., negative log probability in context. While evidence supporting the predictions of surprisal theory has been replicated widely, much of it has focused on a very narrow slice of data: native English speakers reading English texts. Indeed, no comprehensive multilingual analysis exists. We address this gap in the current literature by investigating the relationship between surprisal and reading times in eleven different languages, distributed across five language families. Deriving estimates from language models trained on monolingual and multilingual corpora, we test three predictions associated with surprisal theory: (i) whether surprisal is predictive of reading times, (ii) whether expected surprisal, i.e., contextual entropy, is predictive of reading times, and (iii) whether the linking function between surprisal and reading times is linear. We find that all three predictions are borne out crosslinguistically. By focusing on a more diverse set of languages, we argue that these results offer the most robust link to date between information theory and incremental language processing across languages.
more »
« less
This content will become publicly available on January 15, 2026
Is musical ability related to second-language acquisition? A meta-analysis
In our multicultural and interconnected world, the ability to learn new languages is important. However, there are significant differences in how successfully adults can learn aspects of non-native languages. Given robust relationships between musical ability and native-language processing, musical ability might also contribute to successful second-language acquisition. However, while several studies have assessed this relationship in various ways, the consistency and robustness of the relationship between musical ability and second-language learning remains unclear. Thus, we synthesized 184 effects across 57 independent studies (n=3181) with a robust variance estimation multivariate meta-analysis, and we narratively summarized partial correlation effects across 12 studies. The available evidence suggests that musical ability is indeed positively related to second-language learning, even after factoring in publication bias revealed by the meta-analysis. Although future work with more diverse participant populations and methodologies is needed to further disentangle this relationship, it is apparent that individuals with better musical ability are generally more successful at second-language learning.
more »
« less
- Award ID(s):
- 2020813
- PAR ID:
- 10581837
- Publisher / Repository:
- The Royal Society
- Date Published:
- Journal Name:
- Royal Society Open Science
- Volume:
- 12
- Issue:
- 1
- ISSN:
- 2054-5703
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Uncovering the genetic underpinnings of musical ability and engagement is a foundational step for exploring their wide‐ranging associations with cognition, health, and neurodevelopment. Prior studies have focused on using twin and family designs, demonstrating moderate heritability of musical phenotypes. The current study used genome‐wide complex trait analysis and polygenic score (PGS) approaches utilizing genotype data to examine genetic influences on two musicality traits (rhythmic perception and music engagement) inN= 1792 unrelated adults in the Vanderbilt Online Musicality Study. Meta‐analyzed heritability estimates (including a replication sample of Swedish individuals) were 31% for rhythmic perception and 12% for self‐reported music engagement. A PGS derived from a recent study on beat synchronization ability predicted both rhythmic perception (β= 0.11) and music engagement (β= 0.19) in our sample, suggesting that genetic influences underlying self‐reported beat synchronization ability also influence individuals’ rhythmic discrimination aptitude and the degree to which they engage in music. Cross‐trait analyses revealed a modest contribution of PGSs from several nonmusical traits (from the cognitive, personality, and circadian chronotype domains) to individual differences in musicality (β= −0.06 to 0.07). This work sheds light on the complex relationship between the genetic architecture of musical rhythm processing, beat synchronization, music engagement, and other nonmusical traits.more » « less
-
Akinlabi, Akinbiyi; Bickmore, Lee; Cahill, Michael; Diercks, Michael; Downing, Laura J.; Essegbey, James; Franich, Katie; McPherson, Laura; Rose, Sharon (Ed.)The tonal nature of many African languages has long raised questions about mu- sical expression and the relationship between language and music. The two main areas of inquiry have been the relationship between tone and melody in vocal mu- sic (tonal textsetting) and the role of tone in musical surrogate languages (e.g. talk- ing drums). However, the degree of similarity between these two genres in terms of tonal adaptation has remained an open question. In this paper, we present a case study comparing the role of tone in two musical traditions from the Sambla ethnic group of Burkina Faso: vocal music and a balafon (xylophone) surrogate lan- guage. We show that the two have different systems of tone-note correspondence and level of phonological encoding, indicating that musical adaptation of tone is not monolithic. We suggest that these different systems of tonal adaptation may stem from functional, structural, and cultural differences between the two musical genres.more » « less
-
The expansion of computer science (CS) into K-12 contexts has resulted in a diverse ecosystem of curricula designed for various grade levels, teaching a variety of concepts, and using a wide array of different programming languages and environments. Many students will learn more than one programming language over the course of their studies. There is a growing need for computer science assessment that can measure student learning over time, but the multilingual learning pathways create two challenges for assessment in computer science. First, there are not validated assessments for all of the programming languages used in CS classrooms. Second, it is difficult to measure growth in student understanding over time when students move between programming languages as they progress in their CS education. In this position paper, we argue that the field of computing education research needs to develop methods and tools to better measure students' learning over time and across the different programming languages they learn along the way. In presenting this position, we share data that shows students approach assessment problems differently depending on the programming language, even when the problems are conceptually isomorphic, and discuss some approaches for developing multilingual assessments of student learning over time.more » « less
-
Successful cross-language clone detection could enable researchers and developers to create robust language migration tools, facilitate learning additional programming languages once one is mastered, and promote reuse of code snippets over a broader codebase. How- ever, identifying cross-language clones presents special challenges to the clone detection problem. A lack of common underlying rep- resentation between arbitrary languages means detecting clones requires one of the following solutions: 1) a static analysis frame- work replicated across each targeted language with annotations matching language features across all languages, or 2) a dynamic analysis framework that detects clones based on runtime behavior. In this work, we demonstrate the feasibility of the latter solution, a dynamic analysis approach called SLACC for cross-language clone detection. Like prior clone detection techniques, we use input/out- put behavior to match clones, though we overcome limitations of prior work by amplifying the number of inputs and covering more data types; and as a result, achieve better clusters than prior at- tempts. Since clusters are generated based on input/output behav- ior, SLACC supports cross-language clone detection. As an added challenge, we target a static typed language, Java, and a dynamic typed language, Python. Compared to HitoshiIO, a recent clone de- tection tool for Java, SLACC retrieves 6 times as many clusters and has higher precision (86.7% vs. 30.7%). This is the first work to perform clone detection for dynamic typed languages (precision = 87.3%) and the first to perform clone detection across languages that lack a common underlying repre- sentation (precision = 94.1%). It provides a first step towards the larger goal of scalable language migration tools.more » « less
