MacMahon, E. | Search bioRxiv

bioRxiv2025

Monkey See, Model Knew: Large Language Models Accurately Predict Visual Brain Responses in Humans and Non-Human Primates

AO_SCPLOWBSTRACTC_SCPLOWRecent progress in multimodal AI and language-aligned visual representation learning has rekindled debates about the role of language in shaping the human visual system. In particular, the emergent ability of language-aligned vision models (e.g. CLIP) - and even pure language models (e.g. BERT) - to predict image-evoked brain activity has led some to suggest that human visual cortex itself may be language-aligned in comparable ways. But what would we make of this claim if the same procedures could model visual activity in a species without language? Here, we conducted controlled comparisons of pure-vision, pure-language, and multimodal vision-language models in their prediction of human (N=4) and rhesus macaque (N=6, 5:IT, 1:V1) ventral visual activity to the same set of 1000 captioned natural images (the NSD1000). The results revealed markedly similar patterns in model predictivity of early and late ventral visual cortex across both species. This suggests that language model predictivity of the human visual system is not necessarily due to the evolution or learning of language perse, but rather to the statistical structure of the visual world that is reflected in natural language.

neuroscience↗

bioRxiv2022

Chlamydia trachomatis effector Dre1 interacts with dynactin to reposition host organelles during infection

Chlamydia trachomatis is an obligate intracellular pathogen that replicates within a specialized membrane-bound compartment, called the inclusion. Chlamydia species express a unique class of effectors, Incs, which are translocated from the bacteria by a Type III secretion system and are inserted into the inclusion membrane where they modulate the host-bacterium interface. C. trachomatis repositions specific host organelles during infection to acquire nutrients and evade host cell surveillance, however the bacterial and host proteins controlling these processes are largely unknown. Here, we identify an interaction between the host dynactin complex and the C. trachomatis Inc CT192 (CTL0444), hereafter named Dre1 for Dynactin Recruiting Effector 1. We show that dynactin is recruited to the inclusion in a Dre1-dependent manner and that loss of Dre1 diminishes the recruitment of specific host organelles, including the centrosome, mitotic spindle, and Golgi apparatus to the inclusion. Inactivation of Dre1 results in decreased C. trachomatis fitness in cell-based assays and in a mouse model of infection. By targeting particular functions of the versatile host dynactin complex, Dre1 facilitates re-arrangement of certain organelles around the growing inclusion. Our work highlights how C. trachomatis employs a single effector to evoke specific, large-scale changes in host cell organization that establish an intracellular replicative niche without globally inhibiting host cellular function.

microbiology↗