When is a word in good company for learning?

Although identifying the referents of single words is often cited as a key challenge for getting word learning off the ground, it overlooks the fact that young learners consistently encounter words in the context of other words. How does this company help or hinder word learning? Prior investigations into early word learning from children's real-world language input have yielded conflicting results, with some influential findings suggesting an advantage for words that keep a diverse company of other words, and others suggesting the opposite. Here, we sought to triangulate the source of this conflict, comparing different measures of diversity and approaches to controlling for correlated effects of word frequency across multiple languages. The results were striking: while different diversity measures on their own yielded conflicting results, once nonlinear relationships with word frequency were controlled, we found convergent evidence that contextual consistency supports early word learning. RESEARCH HIGHLIGHTS: The words children learn occur in a sea of other words. The company words keep ranges from highly variable to highly consistent and circumscribed. Prior findings conflict over whether variability versus consistency helps early word learning. Accounting for correlated effects of word frequency resolved the conflict across multiple languages. Results reveal convergent evidence that consistency helps early word learning.

Do Large Language Models Know What Humans Know?

Humans can attribute beliefs to others. However, it is unknown to what extent this ability results from an innate biological endowment or from experience accrued through child development, particularly exposure to language describing others' mental states. We test the viability of the language exposure hypothesis by assessing whether models exposed to large quantities of human language display sensitivity to the implied knowledge states of characters in written passages. In pre-registered analyses, we present a linguistic version of the False Belief Task to both human participants and a large language model, GPT-3. Both are sensitive to others' beliefs, but while the language model significantly exceeds chance behavior, it does not perform as well as the humans nor does it explain the full extent of their behavior-despite being exposed to more language than a human would in a lifetime. This suggests that while statistical learning from language exposure may in part explain how humans develop the ability to reason about the mental states of others, other mechanisms are also responsible.

Changes in Adenoma Detection Rate From Full-Spectrum Endoscopy to Standard Forward-Viewing Endoscopy.

GOALS: The aim was to investigate the adenoma detection rate (ADR) of endoscopists who have used full-spectrum endoscopy (Fuse) for 3 years and revert back to traditional forward-viewing endoscopes (R-TFV) at an ambulatory surgical center. BACKGROUND: Traditional forward viewing (TFV) endoscopes have 1 camera and provide an angle of view of 140 to 170 degrees, whereas Fuse provides a 330 degrees view through the addition of 2 side cameras. It has previously been reported that Fuse increased the ADR by 5.4% when compared with previous rates using TFV. Fuse is no longer commercially available. The ADR of endoscopists who revert back to TFV is unknown. STUDY: We conducted a retrospective analysis of data examining the ADR from average risk screening colonoscopies at a 5-room ambulatory surgical center where endoscopists transitioned from TFV to Fuse in April 2014 and then reverted back to TFV in 2016. The primary outcome was ADR. Secondary outcomes were ADR for advanced and right-sided adenomas. RESULTS: A total of 6110 procedures were reviewed. The ADR was 23.70% for TFV, 29.02% for Fuse and 28.88% for R-TFV. The ADR for advanced adenomas was 3.8% for TFV, 6.0% for Fuse and 7.3% for R-TFV. The ADR for right-sided adenomas was 13.0% for TFV, 16.7% for Fuse and 16.0% for R-TFV. The results for all 3 categories showed a statistical difference between TFV and Fuse as well as between TFV and R-TFV. There were no statistical differences between the ADR of Fuse compared with R-TFV. CONCLUSIONS: During R-TFV, endoscopists are able to maintain their increased ability to detect adenomas. This would suggest that there was a change in behavior in endoscopists using Fuse that was durable.

Non-invasive monitoring of cell metabolism and lipid production in 3D engineered human adipose tissues using label-free multiphoton microscopy.

Non-linear optical microscopy methods can characterize over time multiple functional properties of engineered tissues during development. Here, we demonstrate how the combined use of third-harmonic generation (THG) and two-photon excited fluorescence (2PEF) imaging can provide direct quantitative biomarkers of adipogenic stem cell differentiation and metabolic state, respectively. Specifically, we imaged over nine weeks silk scaffolds embedded with human mesenchymal stem cells and exposed to either propagation (PM) or adipogenic differentiation media (AM). THG was employed to visualize the formation of lipid droplets. 2PEF was used to assess the metabolic state of the cells through the redox ratio defined based on the endogenous FAD and NADH fluorescence. The redox ratio of cells in the AM scaffold was significantly lower than that in the PM scaffold during week 5 and 9, and correlated with significant increases in lipid-to-cell volume ratio, and number and size of lipid droplets in the AM scaffold. These findings indicate that the decrease in redox ratio during adipogenic differentiation is associated with fatty acid synthesis and lipid accumulation. Our methods therefore enabled us to identify and measure dynamic correlations between lipid droplet formation and cell metabolic state, while providing insight on the spatial heterogeneity of the observed signals.