Conditional Co-Recall by Category

Measure whether recalled items pull same-category neighbors into recall more than different-category neighbors.

Conditional co-recall by category measures how the probability that a neighbor was recalled depends on whether it belongs to the same or a different category as a recalled anchor item. For a given reference category, each recalled anchor from that category contributes to two bins: same-category neighbors and different-category neighbors.

A higher same-category probability relative to different-category indicates within-category clustering in recall, even after controlling for the anchor being recalled.

Workflow

import osimport warningsfrom jaxcmr.analyses.conditional_corec_by_cat import plot_conditional_corec_by_catfrom jaxcmr.helpers import find_project_root, generate_trial_mask, load_data, save_figurewarnings.filterwarnings("ignore")

dataset_name = "CohenKahana2022_noELI"
figure_dir = "results/figures"
figure_str = ""
trial_query = "data['subject'] > 0"
category_field = "valence"
category_value = -1
relation_labels = ["Neg->Neg", "Neg->Other"]
confidence_level = 0.95

project_root = find_project_root()
figure_dir = os.path.join(project_root, figure_dir)
data_path = os.path.join(project_root, "data", f"{dataset_name}.h5")
data = load_data(data_path)
trial_mask = generate_trial_mask(data, trial_query)

plot_conditional_corec_by_cat(
    data,
    trial_mask,
    category_field=category_field,
    category_value=category_value,
    relation_labels=relation_labels,
    confidence_level=confidence_level,
)
save_figure(figure_dir, figure_str)

Interpretation

The plot compares co-recall probabilities for same-category and different-category neighbors of recalled anchor items. Key patterns:

• Same > Different: recalled anchors preferentially pull same-category neighbors into recall, indicating category clustering.
• Same ≈ Different: category membership does not predict co-recall beyond baseline.

API Details

Notebook parameters

• dataset_name — name of the dataset (used to construct the HDF5 path as data/{dataset_name}.h5).
• figure_dir — directory for saving figures.
• figure_str — base filename for the saved figure. Leave empty to display without saving.
• trial_query — a Python expression evaluated against the dataset to select trials.
• category_field — name of the field in the dataset containing category labels.
• category_value — the category value defining the anchor group.
• relation_labels — list of two labels for the legend: [same_label, different_label].
• confidence_level — confidence level for subject-wise error bars.