ROI-based decoding analysis in Haxby et al. dataset

In this script we reproduce the data analysis conducted by Haxby et al.[1].

Specifically, we look at decoding accuracy for different objects in three different masks: the full ventral stream (mask_vt), the house selective areas (mask_house) and the face selective areas (mask_face), that have been defined via a standard GLM-based analysis.

# Fetch data using nilearn dataset fetcher
from nilearn import datasets
from nilearn.plotting import show

Load and prepare the data

# by default we fetch 2nd subject data for analysis
haxby_dataset = datasets.fetch_haxby()
func_filename = haxby_dataset.func[0]

# Print basic information on the dataset
print(
    "First subject anatomical nifti image (3D) located is "
    f"at: {haxby_dataset.anat[0]}"
)
print(
    f"First subject functional nifti image (4D) is located at: {func_filename}"
)

# load labels
import pandas as pd

# Load nilearn NiftiMasker, the practical masking and unmasking tool
from nilearn.maskers import NiftiMasker

labels = pd.read_csv(haxby_dataset.session_target[0], sep=" ")
stimuli = labels["labels"]
# identify resting state labels in order to be able to remove them
task_mask = stimuli != "rest"

# find names of remaining active labels
categories = stimuli[task_mask].unique()

# extract tags indicating to which acquisition run a tag belongs
run_labels = labels["chunks"][task_mask]

# apply the task_mask to  fMRI data (func_filename)
from nilearn.image import index_img

task_data = index_img(func_filename, task_mask)

[fetch_haxby] Dataset found in /home/runner/nilearn_data/haxby2001
First subject anatomical nifti image (3D) located is at: /home/runner/nilearn_data/haxby2001/subj2/anat.nii.gz
First subject functional nifti image (4D) is located at: /home/runner/nilearn_data/haxby2001/subj2/bold.nii.gz

Decoding on the different masks

The classifier used here is a support vector classifier (svc). We use Decoder and specify the classifier.

import numpy as np

# Make a data splitting object for cross validation
from sklearn.model_selection import LeaveOneGroupOut

from nilearn.decoding import Decoder

cv = LeaveOneGroupOut()

We use Decoder to estimate a baseline.

import warnings

mask_names = ["mask_vt", "mask_face", "mask_house"]

mask_scores = {}
mask_chance_scores = {}

for mask_name in mask_names:
    print(f"Working on {mask_name}")
    # For decoding, standardizing is often very important
    mask_filename = haxby_dataset[mask_name][0]
    masker = NiftiMasker(mask_img=mask_filename, verbose=1)
    mask_scores[mask_name] = {}
    mask_chance_scores[mask_name] = {}

    for category in categories:
        print(f"Processing {mask_name} {category}")
        classification_target = stimuli[task_mask] == category
        # Specify the classifier to the decoder object.
        # With the decoder we can input the masker directly.
        # We are using the svc_l1 here because it is intra subject.
        decoder = Decoder(
            estimator="svc_l1",
            cv=cv,
            mask=masker,
            scoring="roc_auc",
            verbose=1,
        )

        with warnings.catch_warnings():
            # ignore warnings thrown because the ROI mask we are using
            # are much smaller than the whole brain.
            warnings.filterwarnings(action="ignore", category=UserWarning)
            decoder.fit(task_data, classification_target, groups=run_labels)

        mask_scores[mask_name][category] = decoder.cv_scores_[1]
        mean = np.mean(mask_scores[mask_name][category])
        std = np.std(mask_scores[mask_name][category])
        print(f"Scores: {mean:1.2f} +- {std:1.2f}")

        dummy_classifier = Decoder(
            estimator="dummy_classifier",
            cv=cv,
            mask=masker,
            scoring="roc_auc",
            verbose=1,
        )

        with warnings.catch_warnings():
            # ignore warnings thrown because the ROI mask we are using
            # are much smaller than the whole brain.
            warnings.filterwarnings(action="ignore", category=UserWarning)
            dummy_classifier.fit(
                task_data, classification_target, groups=run_labels
            )

        mask_chance_scores[mask_name][category] = dummy_classifier.cv_scores_[
            1
        ]

Working on mask_vt
Processing mask_vt scissors
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.2s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    1.1s finished
Scores: 0.69 +- 0.21
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.1s finished
Processing mask_vt face
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    1.2s finished
Scores: 0.68 +- 0.24
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.1s finished
Processing mask_vt cat
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.8s finished
Scores: 0.70 +- 0.19
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.1s finished
Processing mask_vt shoe
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    1.0s finished
Scores: 0.66 +- 0.17
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.1s finished
Processing mask_vt house
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    1.3s finished
Scores: 0.66 +- 0.19
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.1s finished
Processing mask_vt scrambledpix
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.8s finished
Scores: 0.77 +- 0.21
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.1s finished
Processing mask_vt bottle
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.9s finished
Scores: 0.74 +- 0.21
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.1s finished
Processing mask_vt chair
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    1.4s finished
Scores: 0.60 +- 0.32
\[Decoder.fit] Mask volume = 22837.5mm^3 = 22.8375cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 464 features.
\[Decoder.fit] The decoding model will be trained on 464 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.1s finished
Working on mask_face
Processing mask_face scissors
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.4s finished
Scores: 0.72 +- 0.14
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_face face
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.6s finished
Scores: 0.55 +- 0.18
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_face cat
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.6s finished
Scores: 0.58 +- 0.11
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_face shoe
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.8s finished
Scores: 0.65 +- 0.16
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_face house
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.6s finished
Scores: 0.67 +- 0.11
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_face scrambledpix
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.5s finished
Scores: 0.80 +- 0.09
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_face bottle
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.5s finished
Scores: 0.76 +- 0.11
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_face chair
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.8s finished
Scores: 0.58 +- 0.15
\[Decoder.fit] Mask volume = 1476.56mm^3 = 1.47656cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 30 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Working on mask_house
Processing mask_house scissors
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.9s finished
Scores: 0.69 +- 0.17
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_house face
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.3s finished
Scores: 0.64 +- 0.27
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_house cat
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.6s finished
Scores: 0.72 +- 0.14
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_house shoe
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.5s finished
Scores: 0.63 +- 0.18
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_house house
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.7s finished
Scores: 0.34 +- 0.17
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_house scrambledpix
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.1s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.7s finished
Scores: 0.69 +- 0.17
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_house bottle
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.7s finished
Scores: 0.77 +- 0.16
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished
Processing mask_house chair
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.5s finished
Scores: 0.73 +- 0.13
\[Decoder.fit] Mask volume = 5807.81mm^3 = 5.80781cm^3
\[Decoder.fit] Standard brain volume = 1.88299e+06mm^3
\[Decoder.fit] Original screening-percentile: 20
\[Decoder.fit] Corrected screening-percentile: 100
\[Decoder.fit] The decoding model will be trained on 118 features.
\[Decoder.fit] The decoding model will be trained on 118 features.
[Parallel(n_jobs=1)]: Using backend SequentialBackend with 1 concurrent workers.
[Parallel(n_jobs=1)]: Done   1 out of   1 | elapsed:    0.0s remaining:    0.0s
[Parallel(n_jobs=1)]: Done  12 out of  12 | elapsed:    0.0s finished

We make a simple bar plot to summarize the results

import matplotlib.pyplot as plt

plt.figure(constrained_layout=True)

tick_position = np.arange(len(categories))
plt.xticks(tick_position, categories, rotation=45)

for color, mask_name in zip("rgb", mask_names, strict=False):
    score_means = [
        np.mean(mask_scores[mask_name][category]) for category in categories
    ]
    plt.bar(
        tick_position, score_means, label=mask_name, width=0.25, color=color
    )

    score_chance = [
        np.mean(mask_chance_scores[mask_name][category])
        for category in categories
    ]
    plt.bar(
        tick_position,
        score_chance,
        width=0.25,
        edgecolor="k",
        facecolor="none",
    )

    tick_position = tick_position + 0.2

plt.ylabel("Classification accuracy (AUC score)")
plt.xlabel("Visual stimuli category")
plt.ylim(0.3, 1)
plt.legend(loc="lower right")
plt.title("Category-specific classification accuracy for different masks")

show()

References

[1]

James V. Haxby, M. Ida Gobbini, Maura L. Furey, Alumit Ishai, Jennifer L. Schouten, and Pietro Pietrini. Distributed and overlapping representations of faces and objects in ventral temporal cortex. Science, 293(5539):2425–2430, 2001. doi:10.1126/science.1063736.

Estimated memory usage: 1320 MB