{
  "paper": "bouyeure-2026-fear-rsa",
  "claims": 30,
  "panels": {
    "total": [
      "fig1a",
      "fig1b",
      "fig1c",
      "fig1d",
      "fig1e",
      "fig2a",
      "fig2b",
      "fig3a",
      "fig3b",
      "fig3c",
      "fig3d",
      "fig4a",
      "fig4b",
      "fig5a",
      "fig5b",
      "fig5c",
      "fig5d"
    ],
    "orphan": [
      "fig1a",
      "fig1b",
      "fig1c",
      "fig1d",
      "fig1e",
      "fig3b",
      "fig4b",
      "fig5a",
      "fig5c",
      "fig5d"
    ],
    "figure_level": [
      "fig1",
      "fig4",
      "fig5"
    ],
    "phantom": [],
    "pct": 41.2
  },
  "statistics": {
    "total": 15,
    "orphan": [
      "p <0.025",
      "p <0.0125",
      "t(253)=-3.01",
      "t(253)=-2.7"
    ],
    "pct": 73.3
  },
  "spans": {
    "segmented": 320,
    "obligations": 59,
    "textual": 261,
    "accounted": 43,
    "pct": 72.9,
    "orphans": [
      {
        "uid": "introduction-038",
        "section": "introduction",
        "text": "(A) Example structure of a trial.",
        "stats": [],
        "panels": [
          "fig1a"
        ]
      },
      {
        "uid": "introduction-041",
        "section": "introduction",
        "text": "( B ) Paradigm structure with four different experimental phases (rows) and four different cue types (columns).",
        "stats": [],
        "panels": [
          "fig1b"
        ]
      },
      {
        "uid": "introduction-043",
        "section": "introduction",
        "text": "( C ) Conditioned stimulus (CS) items (left) and context videos (right).",
        "stats": [],
        "panels": [
          "fig1c"
        ]
      },
      {
        "uid": "introduction-046",
        "section": "introduction",
        "text": "( D ) Representational Similarity Matrices (RSMs) for each experimental phase, shown here from the dorsal anterior cingulate cortex (ACC) for illustrative purposes.",
        "stats": [],
        "panels": [
          "fig1d",
          "fig1a",
          "fig1c"
        ]
      },
      {
        "uid": "introduction-049",
        "section": "introduction",
        "text": "( E ) Top: Cue generalization mask for the representational similarity analysis (RSA) matrices estimated within each searchlight.",
        "stats": [],
        "panels": [
          "fig1e"
        ]
      },
      {
        "uid": "results-029",
        "section": "results",
        "text": "( B ) Cue representations during reversal.",
        "stats": [],
        "panels": [
          "fig3b"
        ]
      },
      {
        "uid": "results-043",
        "section": "results",
        "text": "We first compared CS++ and CS-- cues, corresponding to the CS+ vs. CS- contrast during fear acquisition ( Figure 3B ).",
        "stats": [],
        "panels": [
          "fig3b"
        ]
      },
      {
        "uid": "results-054",
        "section": "results",
        "text": "However, item stability was higher for CS+- vs. CS ++cues in a middle temporal cluster during test new ( Figure 3E ), and for CS++ compared to CS-- cues in an inferior temporal cluster during test old ( Figure 3F ; Supplementary file 1d ).",
        "stats": [],
        "panels": [
          "fig3e",
          "fig3f"
        ]
      },
      {
        "uid": "results-080",
        "section": "results",
        "text": "We also compared this measure across experimental phases and related it to the representational geometries of cues ( Figure 5A ).",
        "stats": [],
        "panels": [
          "fig5a"
        ]
      },
      {
        "uid": "results-082",
        "section": "results",
        "text": "( A ) Calculation of context specificity as the difference of within-context similarity and between-context similarity.",
        "stats": [],
        "panels": [
          "fig5a"
        ]
      },
      {
        "uid": "results-085",
        "section": "results",
        "text": "( C ) Calculation of item reinstatement and generalized reinstatement (similarities of item representations across different phases; left) and context specificity (difference between acquisition and reversal; right).",
        "stats": [],
        "panels": [
          "fig5c"
        ]
      },
      {
        "uid": "results-087",
        "section": "results",
        "text": "( D ) Higher context specificity during reversal predicted reinstatement during test old , as a function of CS type: ( Di ) Higher reversal context specificity predicted more pronounced generalized reinstatement of CS+- vs. CS-+ acquisition memory traces in anterior cingulate cortex (ACC)/superior frontal gyrus (SFG) (left), and reversely, more pronounced generalized reinstatement of CS-+ vs. CS+- acquisition memory traces in the precuneus (right).",
        "stats": [],
        "panels": [
          "fig5d",
          "fig5a",
          "fig5c"
        ]
      },
      {
        "uid": "results-096",
        "section": "results",
        "text": "We extracted subject-wise measures of context specificity from the PFC cluster in Figure 4B and compared them to both item reinstatement and generalized reinstatement.",
        "stats": [],
        "panels": [
          "fig4b"
        ]
      },
      {
        "uid": "results-099",
        "section": "results",
        "text": "We used LME models with \u2018context specificity\u2019 and \u2018cue type\u2019 as predictors and \u2018item reinstatement\u2019 or \u2018generalized reinstatement\u2019 as the dependent variable ( Figure 5C ).",
        "stats": [],
        "panels": [
          "fig5c"
        ]
      },
      {
        "uid": "discussion-035",
        "section": "discussion",
        "text": "Interestingly, representations from the first test phase were less reinstated during the second test phase (despite the fact that both test phases occurred during the second experimental day), compared with acquisition and reversal traces formed on the first experimental day ( Figure 4E ).",
        "stats": [],
        "panels": [
          "fig4e"
        ]
      },
      {
        "uid": "discussion-048",
        "section": "discussion",
        "text": "This suggests that for these cues, the more distinct the contextual encoding during reversal, the more strongly the original fear memory trace resurfaced, likely reflecting a return of fear ( Figure 5D ).",
        "stats": [],
        "panels": [
          "fig5d"
        ]
      }
    ]
  }
}