Back

Lightweight Adaptive and Recoverable Implants (LARI) for Chronic Single and Dual Neuropixels 2.0 Recordings for Naturalistic Behaviors in Feature-Rich Environments

Singh, M. R.; Gregg, C.

2026-01-14 neuroscience
10.64898/2026.01.13.699139 bioRxiv
Show abstract

Understanding the neural mechanisms underlying naturalistic decision-making requires chronic, high-quality neural recordings from distributed brain regions in freely behaving animals. However, existing silicon probe implants for mice are limited by weight, fragility, lack of independent probe adjustment, and poor suitability for long-term recordings in complex environments. Here we introduce LARI-1.0 (Lightweight Adjustable and Reusable Implant) and LARI-mini, two Neuropixels 2.0 implant systems designed for stable, recoverable, and independently adjustable single- and dual-probe recordings in mice. LARI-1.0 enables independent vertical and lateral positioning of two probes, while LARI-mini provides a lighter single-probe alternative, reducing implant mass by ~20%. Fully assembled weights of ~2.3 g (LARI-1.0) and ~1.5 g (LARI-mini) remain within 10% and 7.5% of adult male mouse body weight, respectively. We validated implant performance in a custom feature-rich naturalistic foraging environment that elicited full-body movements, rapid elevation changes, and ethological foraging behaviors across fed, fasted, and high-fat diet states. Both implants exhibited excellent mechanical stability, minimal motion-related artifacts, and probe drift comparable to previous studies. High-quality single-unit activity with stable waveform features, 5-6x signal-to-noise ratios, and no systematic loss of well-isolated units was maintained for up to 180 days. These results demonstrate that LARI-1.0 and LARI-mini provide lightweight, durable, and customizable implant solutions enabling long-term, high-fidelity neural recordings during naturalistic behavior.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.