Reply to ‘The core language network separated from other networks during primate evolution’
Related Articles
A unified acoustic-to-speech-to-language embedding space captures the neural basis of natural language processing in everyday conversations
This study introduces a unified computational framework connecting acoustic, speech and word-level linguistic structures to study the neural basis of everyday conversations in the human brain. We used electrocorticography to record neural signals across 100 h of speech production and comprehension as participants engaged in open-ended real-life conversations. We extracted low-level acoustic, mid-level speech and contextual word embeddings from a multimodal speech-to-text model (Whisper). We developed encoding models that linearly map these embeddings onto brain activity during speech production and comprehension. Remarkably, this model accurately predicts neural activity at each level of the language processing hierarchy across hours of new conversations not used in training the model. The internal processing hierarchy in the model is aligned with the cortical hierarchy for speech and language processing, where sensory and motor regions better align with the model’s speech embeddings, and higher-level language areas better align with the model’s language embeddings. The Whisper model captures the temporal sequence of language-to-speech encoding before word articulation (speech production) and speech-to-language encoding post articulation (speech comprehension). The embeddings learned by this model outperform symbolic models in capturing neural activity supporting natural speech and language. These findings support a paradigm shift towards unified computational models that capture the entire processing hierarchy for speech comprehension and production in real-world conversations.
Language measures correlate with other measures used to study emotion
Researchers are increasingly using language measures to study emotion, yet less is known about whether language relates to other measures often used to study emotion. Building on previous work which focuses on associations between language and self-report, we test associations between language and a broader range of measures (self-report, observer report, facial cues, vocal cues). Furthermore, we examine associations across different dictionaries (LIWC-22, NRC, Lexical Suite, ANEW, VADER) used to estimate valence (i.e., positive versus negative emotion) or discrete emotions (i.e., anger, fear, sadness) in language. Associations were tested in three large, multimodal datasets (Ns = 193–1856; average word count = 316.7–2782.8). Language consistently related to observer report and consistently related to self-report in two of the three datasets. Statistically significant associations between language and facial cues emerged for language measures of valence but not for language measures of discrete emotions. Language did not consistently show significant associations with vocal cues. Results did not tend to significantly vary across dictionaries. The current research suggests that language measures (in particular, language measures of valence) are correlated with a range of other measures used to study emotion. Therefore, researchers may wish to use language to study emotion when other measures are unavailable or impractical for their research question.
Land use conversion increases network complexity and stability of soil microbial communities in a temperate grassland
Soils harbor highly diverse microbial communities that are critical to soil health, but agriculture has caused extensive land use conversion resulting in negative effects on critical ecosystem processes. However, the responses and adaptations of microbial communities to land use conversion have not yet been understood. Here, we examined the effects of land conversion for long-term crop use on the network complexity and stability of soil microbial communities over 19 months. Despite reduced microbial biodiversity in comparison with native tallgrass prairie, conventionally tilled (CT) cropland significantly increased network complexity such as connectivity, connectance, average clustering coefficient, relative modularity, and the number of species acting at network hubs and connectors as well as resulted in greater temporal variation of complexity indices. Molecular ecological networks under CT cropland became significantly more robust and less vulnerable, overall increasing network stability. The relationship between network complexity and stability was also substantially strengthened due to land use conversion. Lastly, CT cropland decreased the number of relationships between network structure and environmental properties instead being strongly correlated to management disturbances. These results indicate that agricultural disturbance generally increases the complexity and stability of species “interactions”, possibly as a trade-off for biodiversity loss to support ecosystem function when faced with frequent agricultural disturbance.
Configural processing as an optimized strategy for robust object recognition in neural networks
Configural processing, the perception of spatial relationships among an object’s components, is crucial for object recognition, yet its teleology and underlying mechanisms remain unclear. We hypothesize that configural processing drives robust recognition under varying conditions. Using identification tasks with composite letter stimuli, we compare neural network models trained with either configural or local cues. We find that configural cues support robust generalization across geometric transformations (e.g., rotation, scaling) and novel feature sets. When both cues are available, configural cues dominate local features. Layerwise analysis reveals that sensitivity to configural cues emerges later in processing, likely enhancing robustness to pixel-level transformations. Notably, this occurs in a purely feedforward manner without recurrent computations. These findings with letter stimuli successfully extend to naturalistic face images. Our results demonstrate that configural processing emerges in a naíve network based on task contingencies, and is beneficial for robust object processing under varying viewing conditions.
Evolving adeno-associated viruses for gene transfer to the kidney via cross-species cycling of capsid libraries
The difficulty of delivering genes to the kidney has limited the translation of genetic medicines, particularly for the more than 10% of the global population with chronic kidney disease. Here we show that new variants of adeno-associated viruses (AAVs) displaying robust and widespread transduction in the kidneys of mice, pigs and non-human-primates can be obtained by evolving capsid libraries via cross-species cycling in different kidney models. Specifically, the new variants, AAV.k13 and AAV.k20, were enriched from the libraries following sequential intravenous cycling through mouse and pig kidneys, ex vivo cycling in human organoid cultures, and ex vivo machine perfusion in isolated kidneys from rhesus macaques. The two variants transduced murine kidneys following intravenous administration, with selective tropism for proximal tubules, and led to markedly higher transgene expression than parental AAV9 vectors in proximal tubule epithelial cells within human organoid cultures and in autotransplanted pig kidneys. Following ureteral delivery, AAV.k20 efficiently transduced kidneys in pigs and macaques. The AAV.k13 and AAV.k20 variants are promising vectors for therapeutic gene-transfer applications in kidney diseases and transplantation.
Responses