PMC

Overview of the object memory testing protocols discussed in this review. Mice are first habituated to an empty arena prior to beginning behavioral training (Habituation). (A) In novel object recognition (OR), mice are then allowed to explore two identical novel objects placed in the arena (Training). Finally, retention is tested by presenting mice with one novel and one familiar object (Testing). Mice who remember the familiar object spend more time exploring the novel object relative to the familiar object or to chance. (B) Object placement (OP) uses the same apparatus and general procedure, but during testing, one training object moves to a new location in the arena, rather than being replaced with a new object. (C) Object pairs used in our laboratory’s OR and OP protocols.

Voxel-wise encoding models were used to assess whether our statistical model of visual object context (object2vec) could reliably explain variance in the fMRI responses to the experimental object categories. Linear regression was used to map the representations of object2vec onto fMRI responses. We assessed the out-of-sample prediction accuracy of these regression models through a 9-fold cross-validation procedure. Each fold of the cross-validation design contains a set of object categories that do not appear in any other fold. These folds were shown in separate fMRI runs. Parameters for the voxel-wise linear regression models were estimated using the fMRI data for 8 folds of the object categories and the learned regression weights were then applied to the held-out object categories in the remaining fold to generate a set of predicted fMRI responses. This procedure was repeated for each fold of the cross-validation design, and the predicted fMRI responses from each fold were concatenated together. Prediction accuracy was assessed by calculating the voxel-wise correlations of the predicted and actual fMRI responses across all object categories.

Procedure for learning stable object values. A, Stable object–reward association task. Among a set of eight fractal objects, four were assigned as “high-valued” objects and the other four were assigned as “low-valued” objects. On each trial, one of the eight objects was presented at one of four positions, and the monkey made a saccade to it. If the object was a high-valued object, a large reward was delivered. If the object was a low-valued object, a small reward was delivered. For each set of visual objects, the learning procedure was done repeatedly across many daily sessions, during which each fractal remained to be either a high-valued object or a low-valued object. B, Well learned fractals (≥5 learning sessions) used for monkey G. Each row of eight fractals (shown on each of left and right sides) were used as a set of objects. Among them, the left four fractals were high-valued objects (associated with a large reward) and the right four fractals were low-valued objects (associated with a small reward). Monkey G learned 648 fractals, among which 288 (shown here) were well learned. Monkey D learned 696 fractals, among which 280 were well learned. Many of the fractals were shared by both monkeys.

(a) Days 1–3: acclimatization. (b) Days 4–7: spatial change. (c) Days 8–9: novel-object recognition. (d–e) Percent time spent in object contact for WT (d) and KO mice (e) after object 2 had been moved to a new location. An omnibus RMANOVA found main effects of objects [F_3,153 = 37.668, P<0.001], and the objects by genotype [F_3,153 = 41.226, P<0.001] and the days by objects by genotype interactions to be significant [F_9,153 = 3.761, P<0.007]. Bonferroni corrected pair-wise comparisons found that on day 4, WT mice increased exploration of object 2 relative to objects 1 and 3 and to the old location of object 2 (Ps<0.038). This behavior was maintained though day 7 (Ps<0.050) and the percent time spent in the previous location of object 2 decreased significantly on days 5–7 relative to that on day 4 (Ps<0.027). By comparison, percent time spent with object 2 was not increased for KO mice. Instead, mutants spent more time exploring the old location of object 2 than objects 1–3 (Ps<0.021). Exploration of object 2 by mutants increased compared to object 1 (Ps<0.026) on days 5–7 and to object 3 on days 5–6 (Ps<0.051). Unlike WT controls, ECE-2 mice increased exploration of the old location of object 2 on days 4–7 (Ps<0.001) and spent less time exploring object 2 in its new location on days 4–5 (Ps<0.020). (f–g) Percent time in object contact for WT (f) and KO mice (g) after object 3 was substituted with a novel object. An omnibus RMANOVA revealed main effects of objects [F_3,51 = 103.112, P<0.001], and the objects by genotype [F_3,51 = 25.778, P<0.001] and days by objects by genotype interaction to be significant [F_3,51 =17.778, P<0.001]. Bonferroni comparisons showed that on day 8, WT mice increased exploration of the novel relative to objects 1 and 2 (Ps<0.001). However, by day 9 the percent time spent exploring the novel object was no different than that for the two familiar objects. On both days, the percent time spent in the old location was reduced relative to that for object 2 (Ps<0.001). By comparison, KO mice failed to show any increase in the percent time spent exploring the novel object on day 8, but by day 9 there was a marked increase in exploration of the novel compared to the two familiar objects (Ps<0.001). Nonetheless, mutants spent more time in the old location of object 2 (Ps<0.011) on days 8–9 than WT controls. n = 9 WT and 10 KO mice. *P<0.05, WT compared to KO animals; ^P<0.05, percent total time spent with objects on days 2 and 3 compared to day 1, or days 5–7 compared to day 4; +P<0.05, object 2 new location versus its old location; #P<0.05, novel object versus object 1 and object 2 in its new location.