SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.14389 · 2023

Information-Restricted Neural Language Models Reveal Different Brain Regions' Sensitivity to Semantics, Syntax and Context

arXiv · PDF · Open in the Atlas

Code that ran

5 of 5 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 5 have no contract verdict; 0 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2302.14389.json · badge: /api/ran/2302.14389.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
alexandrepsq/information-restrited-nlms canonical (classifier label)
classifier basis not recorded on the link
5 ran · 0 no recorded run
FunctionStatusWhere it lives
filter_args Ran
ran on a synthesized input
alexandrepsq/information-restrited-nlms/src/irnlm/utils.py
code served (permissive licence) · get_code("674fa3bc13cb4436")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import inspect
import torch

def example_function(x, y, self=None):
    """Example function to filter arguments from."""
    return x + y

def another_func(a, b, c, self=None):
    """Another example function."""
    return a * b + c

# Create a small dictionary with various test values
test_dict = {
    'x': torch.randn(2, 4),
    'y': torch.randn(2, 4),
    'a': 5,
    'b': 10,
    'c': 3,
    'unused_key': 'this should not be included'
}

# Test with example_function
func1 = example_function

# Test with another_func
func2 = another_func

ARGS = {
    'func': example_function,
    'd': test_dict
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
load_pickle Ran
ran on a synthesized input
alexandrepsq/information-restrited-nlms/src/irnlm/utils.py
code served (permissive licence) · get_code("38855fbcd539fb8e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import pickle
import tempfile
import os

# Create a temporary pickle file for testing
temp_dir = tempfile.gettempdir()
test_filename = os.path.join(temp_dir, "test_pickle_file.pkl")

# Create some sample data and save it with PROTOCOL 5
sample_data = {
    "key1": "value1",
    "key2": [1, 2, 3, 4],
    "key3": {"nested": "dict"}
}

with open(test_filename, 'wb') as f:
    pickle.dump(sample_data, f, protocol=5)

# Define arguments for the load_pickle function
ARGS = {
    "filename": test_filename
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
masks_overlap Ran
ran on a synthesized input
alexandrepsq/information-restrited-nlms/src/irnlm/maskers.py
code served (permissive licence) · get_code("4c580cc6918cf329")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned float64, ndim 0, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

class MockNiftiImage:
    """Mock Nifti-Image class that provides get_fdata() method"""
    def __init__(self, data):
        self.data = data
    
    def get_fdata(self):
        return self.data

# Create two simple 3D masks with some overlap
mask1_data = np.zeros((4, 4, 4))
mask1_data[1:3, 1:3, 1:3] = 1.0

mask2_data = np.zeros((4, 4, 4))
mask2_data[2:4, 2:4, 2:4] = 1.0

mask1 = MockNiftiImage(mask1_data)
mask2 = MockNiftiImage(mask2_data)

ARGS = {
    "mask1": mask1,
    "mask2": mask2
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
read_yaml Ran
ran on a synthesized input
alexandrepsq/information-restrited-nlms/src/irnlm/utils.py
code served (permissive licence) · get_code("d726b894b0f1b9c8")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned dict. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import tempfile
import os
import yaml

# Create a temporary YAML file with sample data
temp_dir = tempfile.mkdtemp()
yaml_path = os.path.join(temp_dir, "test_config.yaml")

# Write a simple YAML file
with open(yaml_path, 'w') as f:
    yaml.dump({
        'learning_rate': 0.001,
        'batch_size': 32,
        'epochs': 10,
        'model': 'resnet50'
    }, f)

ARGS = {
    'yaml_path': yaml_path
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
yaml — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type dict — a dict output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a dict output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
specificity_index Ran
ran on a synthesized input
alexandrepsq/information-restrited-nlms/src/irnlm/specificity_index.py
code served (permissive licence) · get_code("7f724966adc291a0")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ndarray, ndim 3, dtype family float, shape (3, 2, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import numpy as np

class MockMasker:
    """Mock masker that implements transform and inverse_transform methods."""
    
    def __init__(self, n_voxels=12):
        self.n_voxels = n_voxels
    
    def transform(self, img):
        """Transform 3D image to 1D array of voxel values."""
        # Flatten the image to 1D
        if isinstance(img, np.ndarray):
            return img.flatten()[:self.n_voxels]
        return np.ones(self.n_voxels)
    
    def inverse_transform(self, data):
        """Transform 1D array back to 3D image."""
        # Reshape 1D data back to 3D
        if len(data) == self.n_voxels:
            side = int(np.cbrt(self.n_voxels))
            if side ** 3 == self.n_voxels:
                return data.reshape(side, side, side)
            else:
                # Fallback for non-cubic voxel counts
                return data.reshape(3, 2, 2)
        return data

# Create test data
np.random.seed(0)

# Create small 3D image arrays (3x2x2 = 12 voxels)
img_sem = np.random.rand(3, 2, 2) * 10
img_syn = np.random.rand(3, 2, 2) * 5

# Create masker instance
masker = MockMasker(n_voxels=12)

# Define the arguments dictionary
ARGS = {
    "img_sem": img_sem,
    "img_syn": img_syn,
    "masker": masker,
    "threshold": 4
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family float, shape (3, 2, 2) — shape (3, 2, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 5 samples lifted from this paper's linked repositories, 5 carry a function identity and 0 do not. Of the 5 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 5 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

A fundamental question in neurolinguistics concerns the brain regions involved in syntactic and semantic processing during speech comprehension, both at the lexical (word processing) and supra-lexical levels (sentence and discourse processing). To what extent are these regions separated or intertwined? To address this question, we trained a lexical language model, Glove, and a supra-lexical language model, GPT-2, on a text corpus from which we selectively removed either syntactic or semantic information. We then assessed to what extent these information-restricted models were able to predict the time-courses of fMRI signal of humans listening to naturalistic text. We also manipulated the size of contextual information provided to GPT-2 in order to determine the windows of integration of brain regions involved in supra-lexical processing. Our analyses show that, while most brain regions involved in language are sensitive to both syntactic and semantic variables, the relative magnitudes of these effects vary a lot across these regions. Furthermore, we found an asymmetry between the left and right hemispheres, with semantic and syntactic processing being more dissociated in the left hemisphere than in the right, and the left and right hemispheres showing respectively greater sensitivity to short and long contexts. The use of information-restricted NLP models thus shed new light on the spatial organization of syntactic processing, semantic processing and compositionality.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2302.14389")
get_code_for_paper("2302.14389")
have("2302.14389")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2302.14389.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2302.14389.svg)](https://syntology.ai/paper/2302.14389)

Connect an agent — have() is free.