Two prominent, yet contrasting, theoretical views are available to characterize the underlying drivers of language evolution: on the one hand, task-specific utility maximization; on the other hand, task-agnostic communicative efficiency. The latter has recently been grounded in an information-theoretic tradeoff between communicative complexity and informativeness, known as the Information Bottleneck (IB) principle. Here, we integrate these two views and propose an information-constrained emergent communication framework that trades off utility, informativeness, and complexity. To train agents within our framework, we develop a method, called Vector-Quantized Variational Information Bottleneck (VQ-VIB), that allows agents to interact using information-constrained discrete communication embedded in a continuous vector space. We test this approach in three domains and show that pressure for informativeness facilitates faster learning and better generalization to novel domains. At the same time, limiting complexity yields better alignment with actual human languages. Lastly, we find that VQ-VIB outperforms previously proposed emergent communication methods; we posit that this is due to the semantically-meaningful communication embedding space that VQ-VIB affords. Overall, our work demonstrates the role of cognitively-motivated optimality principles in inducing aspects of human-like communication among artificial agents.

The detailed study of eye movements in reading has shed considerable light into how language processing unfolds in real time. Yet eye movements in reading remain inadequately studied in non-native (L2) readers, even though much of the world’s population is multilingual. Here we present a detailed analysis of the quantitative functional influences of word length, frequency, and predictability on eye movement measures in reading in a large, linguistically diverse sample of non-native English readers. We find many similar qualitative effects as in L1 readers, but crucially also a proficiency-sensitive “ lexicon-context tradeoff ”. The most proficient L2 readers’ eye movements approach an L1 pattern, but as L2 proficiency diminishes, readers’ eye movements become less sensitive to a word’s predictability in context and more sensitive to word frequency, which is context-invariant. This tradeoff supports a rational, experience-dependent account of how context-driven expectations are deployed in L2 language processing.

We present CELER ( C orpus of E ye Movements in L 1 and L2 E nglish R eading), a broad coverage eye-tracking corpus for English. CELER comprises over 320,000 words, and eye-tracking data from 365 participants. Sixty-nine participants are L1 (first language) speakers, and 296 are L2 (second language) speakers from a wide range of English proficiency levels and five different native language backgrounds. As such, CELER has an order of magnitude more L2 participants than any currently available eye movements dataset with L2 readers. Each participant in CELER reads 156 newswire sentences from the Wall Street Journal ( WSJ ), in a new experimental design where half of the sentences are shared across participants and half are unique to each participant. We provide analyses that compare L1 and L2 participants with respect to standard reading time measures, as well as the effects of frequency, surprisal, and word length on reading times. These analyses validate the corpus and demonstrate some of its strengths. We envision CELER to enable new types of research on language processing and acquisition, and to facilitate interactions between psycholinguistics and natural language processing (NLP).

Piazza et al. reported a strong correlation between education and approximate number sense (ANS) acuity in a remote Amazonian population, suggesting that symbolic and nonsymbolic numerical thinking mutually enhance one another over in mathematics instruction. But Piazza et al. ran their task using a computer display, which may have exaggerated the connection between the two tasks, because participants with greater education (and hence better exact numerical abilities) may have been more comfortable with the task. To explore this possibility, we ran an ANS task in a remote population using two presentation methods: (a) a computer interface and (b) physical cards, within participants. If we only analyze the effect of education on ANS as measured by the computer version of the task, we replicate Piazza et al.’s finding. But importantly, the effect of education on the card version of the task is not significant, suggesting that the use of a computer display exaggerates effects. These results highlight the importance of task considerations when working with nonindustrialized cultures, especially those with low education. Furthermore, these results raise doubts about the proposal advanced by Piazza et al. that education enhances the acuity of the approximate number sense.