No abstract

Recent findings suggest that the contents of memory encoding and retrieval can be decoded from the angular gyrus (ANG), a subregion of posterior lateral parietal cortex. However, typical decoding approaches provide little insight into the nature of ANG content representations. Here, we tested whether complex, multidimensional stimuli (faces) could be reconstructed from ANG by predicting underlying face components from fMRI activity patterns in humans. Using an approach inspired by computer vision methods for face recognition, we applied principal component analysis to a large set of face images to generate eigenfaces. We then modeled relationships between eigenface values and patterns of fMRI activity. Activity patterns evoked by individual faces were then used to generate predicted eigenface values, which could be transformed into reconstructions of individual faces. We show that visually perceived faces were reliably reconstructed from activity patterns in occipitotemporal cortex and several lateral parietal subregions, including ANG. Subjective assessment of reconstructed faces revealed specific sources of information (e.g., affect and skin color) that were successfully reconstructed in ANG. Strikingly, we also found that a model trained on ANG activity patterns during face perception was able to successfully reconstruct an independent set of face images that were held in memory. Together, these findings provide compelling evidence that ANG forms complex, stimulus-specific representations that are reflected in activity patterns evoked during perception and remembering.

No abstract

The hippocampus is thought to compare predicted events with current perceptual input, generating a mismatch signal when predictions are violated. However, most prior studies have only inferred when predictions occur without measuring them directly. Moreover, an important but unresolved question is whether hippocampal mismatch signals are modulated by the degree to which predictions differ from outcomes. Here, we conducted a human fMRI study in which subjects repeatedly studied various word-picture pairs, learning to predict particular pictures (outcomes) from the words (cues). After initial learning, a subset of cues was paired with a novel, unexpected outcome, whereas other cues continued to predict the same outcome. Critically, when outcomes changed, the new outcome was either "near" to the predicted outcome (same visual category as the predicted picture) or "far" from the predicted outcome (different visual category). Using multivoxel pattern analysis, we indexed cue-evoked reactivation (prediction) within neocortical areas and related these trial-by-trial measures of prediction strength to univariate hippocampal responses to the outcomes. We found that prediction strength positively modulated hippocampal responses to unexpected outcomes, particularly when unexpected outcomes were close, but not identical, to the prediction. Hippocampal responses to unexpected outcomes were also associated with a tradeoff in performance during a subsequent memory test: relatively faster retrieval of new (updated) associations, but relatively slower retrieval of the original (older) associations. Together, these results indicate that hippocampal mismatch signals reflect a comparison between active predictions and current outcomes and that these signals are most robust when predictions are similar, but not identical, to outcomes.

Memory retrieval can strengthen, but also distort memories. Parietal cortex is a candidate region involved in retrieval-induced memory changes as it reflects retrieval success and represents retrieved content. Here, we conducted an fMRI experiment to test whether different forms of parietal reactivation predict distinct consequences of retrieval. Subjects studied associations between words and pictures of faces, scenes, or objects, and then repeatedly retrieved half of the pictures, reporting the vividness of the retrieved pictures ("retrieval practice"). On the following day, subjects completed a recognition memory test for individual pictures. Critically, the test included lures highly similar to studied pictures. Behaviorally, retrieval practice increased both hit and false alarm (FA) rates to similar lures, confirming a causal influence of retrieval on subsequent memory. Using pattern similarity analyses, we measured two different levels of reactivation during retrieval practice: generic "category-level" reactivation and idiosyncratic "item-level" reactivation. Vivid remembering during retrieval practice was associated with stronger category- and item-level reactivation in parietal cortex. However, these measures differentially predicted subsequent recognition memory performance: whereas higher category-level reactivation tended to predict FAs to lures, item-level reactivation predicted correct rejections. These findings indicate that parietal reactivation can be decomposed to tease apart distinct consequences of memory retrieval.