Article Figures & Data
Figures
- Figure 1.
Histology. Top, Photomicrographs of stained coronal sections through the area of the intended lesion in the 3 animals with sdlPFC lesions (sdlPFC1 to sdlPFC3) alongside drawings of the intended extent of the lesions on drawings of representative coronal sections (left). Bottom, Photomicrographs of stained coronal sections through the area of the intended lesion in the 3 animals with OFC lesions (OFC1 to OFC3) alongside drawings of the intended extent of the lesions on drawings of representative coronal sections (left). Top row of each panel represents reconstructions of the area of cortex lesioned on drawings of representative lateral and ventral surfaces. Numerals indicate distance in millimeters from the interaural plane. This figure is based on data from, and a supplementary figure in, Buckley et al. (2009).
- Figure 2.
DMP tasks. Each trial consisted of a sample (red cross) stage, a distractor (blue square) stage, a delay, and then a probe/choice (2 red crosses) stage. Temporally taxing DMP: five levels of delay interval between distractor and probes (1, 2, 4, 8, or 16 s); spatially taxing variant DMP: four levels of separation between two red crosses in probe (visual angles of either 4.8°, 8.6°, 15.2°, or 21.7°, which are equivalent to 23, 16, 9, and 5 cm, respectively, on screen; all delay fixed at 1 s). S+, Target; S−, foil. Gray dotted circle in the figure is invisible to the animal and is just shown to illustrate the two choices are always equidistant to the distractor to obviate proximity bias.
- Figure 3.
Differential deficits in meta-indices and accuracy confidence interaction in sdlPFC group in spatially demanding recognition. Meta-performance for the three monkey groups (OFC, sdlPFC, CON). Metacognitive accuracy in sdlPFC group was lower than CON group for spatial-variant task, but not for the temporal-variant task: (A) SDT meta-d′/d′; (B) hierarchical-model meta-d′/d′; and (C). Horizontal axes represent the two spatial recognition tasks (temporal-variant; spatial-variant). Vertical axes represent the three meta-indices. D, E, Accuracy in high confidence trials is usually higher than low confidence trials (for both CON and OFC monkeys in both tasks), but such effects were disrupted in the sdlPFC monkeys, especially in the spatial-variant task. “×” indicates significant group × confidence interaction (p < 0.05). *p < 0.05. Colored dots represent individuals. Error bars indicate SEM.
- Figure 4.
RT distributions for correct and incorrect responses of individual monkeys. x axis, RT; y axis, density. Green and red lines indicate the kernel density estimation. Blue lines indicate the medial RT of each monkey. Bin size for the histograms is set at 200 ms.
- Figure 5.
sdlPFC and OFC lesions did not result in recognition impairment. Memory task performance was intact in both tasks: (A) temporal-variant percentage correct; (B) temporal-variant RT; (C) spatial-variant percentage correct; and (D) spatial-variant RT. Error bars indicate SEM.
- Figure 6.
Meta-memory deficits could not be explained away by speed-and-accuracy trade-off. Inverse efficiency (RT/percentage correct) shows no main effect of group in (A) temporal-variant task and (B) spatial-variant task. Error bars indicate SEM.
- Figure 7.
No meta-memory deficit following dlPFC lesion in WCST analog. Meta-performance for the two monkey groups (sdlPFC, CON) in the two spatial recognition tasks (temporal-variant; spatial-variant) and WCST analog. Vertical axes represent the three meta-indices: (A) SDT meta-d′/d′; (B) hierarchical-model meta-d′/d′; and (C) Metacognitive accuracy in the sdlPFC group was lower than CON group for spatial-variant task (see also Fig. 4A–C), but not for WCST analog or temporal-variant task. Error bars indicate SEM. *p < 0.05. Colored dots represent individual monkeys.
- Figure 8.
Strong correlations among the three metacognitive metrics. Pearson correlations computed among the three indices were all statistically significant (all p values 0.001). Colored dots represent individual data points collapsed across monkey groups, with each monkey shown twice (temporal and spatial variants).
Tables
- Table 1.
Percentiles of each monkey's actual meta-scores compared with the simulated data for all three tasks in which no impairment was founda
Group ID Temporal variant φ Temporal variant meta d′/d′ Temporal variant H-model meta d′/d′ Spatial variant φ Spatial variant meta d′/d′ Spatial variant H-model meta d′/d′ WCST φ WCST meta d′/d′ WCST H-model meta d′/d′ CON 1 99.5 99.5 99.5 99.5 99 99 99.5 99.5 99.5 CON 2 99.5 99.5 99.5 89 89 93 99.5 99.5 99 CON 3 99.5 99.5 99.5 76.5 76.5 83 99.5 91 85 CON 4 99 99 99 96.5 96.5 96 99.5 99.5 99 CON 5 99 99 99 83 83 88 99.5 99.5 99.5 CON 6 99.5 99.5 99.5 75.5 75.5 84 99.5 99.5 97 CON 7 99.5 99.5 99.5 Statistics for CON 0.017 < 0.001 0.017 < 0.001 0.017 < 0.001 0.2456 < 0.001 0.2456 < 0.001 0.176 < 0.001 0.0056 < 0.001 0.096 < 0.001 0.156 < 0.001 OFC 1 99 93 93 94 94 84 99.5 99.5 99.5 OFC 2 93 99.5 99.5 94 99 99 99.5 99.5 99.5 OFC 3 99.5 99 99 89 98 98 99.5 99.5 99.5 Statistics for OFC 0.073 < 0.001 0.073 < 0.001 0.073 < 0.001 0.113 < 0.001 0.063 < 0.001 0.163 = 0.0040 0.0053 < 0.001 0.0053 < 0.001 0.0053 < 0.001 sdlPFC 1 99.5 99 99 48b 1b 1b 99.5 99.5 99.5 sdlPFC 2 96 81 89 68b 83b 83b 99.5 99.5 99 sdlPFC 3 99.5 99.5 99.5 48b 48b 48b 99.5 99 99 Statistics for sdlPFC 0.043 < 0.001 0.193 = 0.006 0.113 = 0.0013 0.523 = 0.140b 0.993 = 0.970b 0.993 = 0.970b 0.0053 < 0.001 0.013 < 0.001 0.013 < 0.001
