I am a cognitive neuroscientist at Unicog. My actual homepage is at pallier.org.

Here are various tools or ressources useful for experiments (or see all my repositories).

• Lexical Databases & Pseudoword generator

  • OpenLexicon: Platform providing access to lexical databases for psycholinguistic research, including Lexique
  • unipseudo-go: Pseudoword generator using trigram Markov chains from real word dictionaries (port of UniPseudo)

• Experiment Frameworks & Stimulus Delivery

  Tools for building and running psychology/neuroscience experiments with precise timing.

  • goxpyriment: General-purpose framework for building behavioral experiments — stimuli, trials, blocks, logging
  • gostim2: Fixed-schedule multimedia stimulus delivery (image, audio, text, video) with millisecond timing and VSYNC sync
  • Audiovis: Audio-visual stimulus presentation based on expyriment

• Stimulus List Preparation

  Tools for generating stimuli and randomizing experimental lists.

  • shuffle-go: Randomize lists with sequential constraints (max repetitions, minimum gap)
  • match-go: Implementation of van Casteren & Davis Mix & Match — match items across conditions on selected variables
  • dot-array-generator-go: Generate non-symbolic number stimuli (dot arrays) with configurable spatial constraints (prot Lauren Aulet's code)
  • images2gv: Convert image sequences into GPU-accelerated video (.gv) with LZ4 compression and efficient frame seeking
  • llm_pseudoword_generator: Neural pseudoword generator trained with a character-level language model
  • jabberwocky: Generates syntactically correct French nonsense text by replacing content words with pseudowords

• Timing & Hardware Interfaces

  • bbtkv3: Capture stimulus/response events using the Black Box ToolKit v3
  • bbtkv2_python: Python module for precise timing acquisition with the Black Box ToolKit v2
  • dlp-io8-g: Send/read TTL triggers via an FTDI chip
  • keyboard_scanner: HID event monitor — captures and timestamps keyboard/mouse input
  • neurospin_meg_response_keys_parallel_port: Record response button presses during MEG experiments via parallel port

• Research Projects

  • The Little Prince (fMRI/MEG/EEG)

    Multi-modal neuroimaging study in which participants listened to The Little Prince audiobook.

    • lpp-paradigms: Experimental stimuli and presentation code across EEG, fMRI, and MEG modalities
    • LePetitPrince: Pipeline computing cross-validated R² maps of how well computational models predict brain activity
    • lpp-scripts3: Full fMRI analysis pipeline: regressor generation, first/second-level analyses, ROI studies

  • Brain Lateralization & Language Models

    Code for:

    • Bonnasse-Gahot, L., & Pallier, C. (2024). fMRI predictors based on language models of increasing complexity recover brain left lateralization. Advances in Neural Information Processing Systems, 37, 125231-125263.

    These codes were written by Laurent Bonnasse-Gahot:

    • llms_brain_lateralization: NeurIPS 2024 — fMRI predictors from language models of increasing complexity recover left lateralization for language
    • llm_training_brain_asym: arXiv:2602.12811 — Left-right brain asymmetry in LLM-based fMRI prediction across OLMo-2 training stages

• Teaching Materials

  • PCBS: Programming for Cognitive and Brain Sciences — full course
  • programming-psychology-experiments: Programming Psychology Experiments (subset of PCBS)
  • statistics_with_R: Basic statistical analyses with R

• Miscellaneous

  • linux-tips: Linux command-line tips and howtos