bioRxivNeurosciencePreprint
Head movement allocation in freely-moving macaques learning object-context sequences
Saman AbbaspooriD, K. Rahman, Wolf ZinkeiD, Kari L. HoffmaniD
Abstract
Flexible learning and long-term memory are hallmark features of primate cognition that are shaped by interactions with changing environments. A comprehensive understanding of these brain-body-environment interactions requires minimal constraints on behavior, and environments rich with conditional cues that encourage diverse behavioral dynamics. In this study, we introduce a multi-level enclosure (Treehouse) that allows for independent control of spatially distributed touchscreens for studying cognition in freely moving macaques. To test flexible learning, two monkeys completed trials consisting of a regular sequence of conditional object choices (targets) across four distinct spatiotemporal contexts (screens). Both monkeys acquired numerous unique, conditional object-context sequences without detriment to new learning as memoranda accumulated. They also successfully recalled their correct choices after weeks without exposure. To assess how monkeys allocate movement during learning and memory-guided decision making, we tracked their head kinematics and used recurrent neural networks to decode different aspects of their engagement with the environment during the choice epoch of the task. One such model predicted the position of the object about to be touched, including on error trials, tracking overt gaze towards objects. Separate models decoded sequence corner, the ordinal step, and the individual screen stations the monkeys occupied, despite identical stimulus layouts and behavioral demands at each station. Head movements also showed efficiencies that the task did not require: under conditions of uncertainty, monkeys were biased in their initial search locations to the targets that were closest to reach, and across days they made smoother head movements as training progressed, with no change in accuracy. These results demonstrate the potential of immersive 3-D environments, in which body kinematics can shed light on animals' goals, and reveal that situated learning in primates can be rapid, capacious, durable, and flexible.
- organism: non-human primate
- assay: behavioral assay
- research focus: cell type annotation
- approach: deep learning
- Citations
- 4
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- Flexible Learning and Re-ordering of Context-dependent Object Sequences in Nonhuman Primates
Xuan Wen, Adam R. Neumann, … Thilo Womelsdorf·PLoS BiologyPeer-reviewed·Jun 23, 2025
- A naturalistic environment to study natural social behaviors and cognitive tasks in freely moving monkeys
- Structured navigation emerges from self-guided spatial learning in freely moving common marmosets
Nada El Mahmoudi, Francesca Lanzarini, … Jean Laurens·bioRxiv·Mar 22, 2025
- Physical sensorimotor contingencies facilitate learning in monkeys
Su-Jin An, Jack Gabel, … Hansem Sohn·bioRxiv·Nov 4, 2026
- Non-instructed Motor Skill Learning in Monkeys: Insights from Deep Reinforcement Learning Models
L. Carminatti, L. Condro, … E. Dauce·bioRxiv·Dec 4, 2023
- Value and spatial preferences guide habitual reaching and manual object selection in primates
Yun-Hyuk Kim, Jun Park, … Hyoung F. Kim·Scientific ReportsPeer-reviewed·Oct 5, 2026