Our understanding of the world is shaped by inferences about underlying structure. For example, at the gym, you might notice that the same people tend to arrive around the same time and infer that they are friends that work out together. Consistent with this idea, after participants are presented with a temporal sequence of objects that follows an underlying community structure, they are biased to infer that objects from the same community share the same properties. Here, we used fMRI to measure neural representations of objects after temporal community structure learning and examine how these representations support inference about object relationships. We found that community structure learning affected inferred object similarity: When asked to spatially group items based on their experience, participants tended to group together objects from the same community. Neural representations in perirhinal cortex predicted individual differences in object grouping, suggesting that high-level object representations are affected by temporal community learning. Furthermore, participants were biased to infer that objects from the same community would share the same properties. Using computational modeling of temporal learning and inference decisions, we found that inductive reasoning is influenced by both detailed knowledge of temporal statistics and abstract knowledge of the temporal communities. The fidelity of temporal community representations in hippocampus and precuneus predicted the degree to which temporal community membership biased reasoning decisions. Our results suggest that temporal knowledge is represented at multiple levels of abstraction, and that perirhinal cortex, hippocampus, and precuneus may support inference based on this knowledge.

The human brain constantly anticipates the future based on memories of the past. Encountering a familiar situation reactivates memory of previous encounters, which can trigger a prediction of what comes next to facilitate responsiveness. However, a prediction error can lead to pruning of the offending memory, a process that weakens its representation in the brain and leads to forgetting. Our goal in this study was to evaluate whether memories are spared from such pruning in situations that allow for accurate predictions at the categorical level, despite prediction errors at the item level. Participants viewed a sequence of objects, some of which reappeared multiple times (“cues”), followed always by novel items. Half of the cues were followed by new items from different (unpredictable) categories, while others were followed by new items from a single (predictable) category. Pattern classification of fMRI data was used to identify category-specific predictions after each cue. Pruning was observed only in unpredictable contexts, while encoding of new items was less robust in predictable contexts. These findings demonstrate that how associative memories are updated is influenced by the reliability of abstract-level predictions in familiar contexts.

Events with overlapping elements can be encoded as two separate representations or linked into an integrated representation, yet we know little about the conditions that promote one form of representation over the other. Here, we tested the hypothesis that the proximity of overlapping events would increase the probability of integration. Participants first established memories for house–object and face–object pairs; half of the pairs were learned 24 hr before an fMRI session, and the other half 30 min before the session. During scanning, participants encoded object–object pairs that overlapped with the initial pairs acquired on the same or prior day. Participants were also scanned as they made inference judgments about the relationships among overlapping pairs learned on the same or different day. Participants were more accurate and faster when inferring relationships among memories learned on the same day relative to those acquired across days, suggesting that temporal proximity promotes integration. Evidence for reactivation of existing memories—as measured by a visual content classifier—was equivalent during encoding of overlapping pairs from the two temporal conditions. In contrast, evidence for integration—as measured by a mnemonic strategy classifier from an independent study [Richter, F. R., Chanales, A. J. H., & Kuhl, B. A. Predicting the integration of overlapping memories by decoding mnemonic processing states during learning. Neuroimage, 124, 323–335, 2016]—was greater for same-day overlapping events, paralleling the behavioral results. During inference itself, activation patterns further differentiated when participants were making inferences about events acquired on the same day versus across days. These findings indicate that temporal proximity of events promotes integration and further influences the neural mechanisms engaged during inference.

Despite the importance of learning and remembering across the lifespan, little is known about how the episodic memory system develops to support the extraction of associative structure from the environment. Here, we relate individual differences in volumes along the hippocampal long axis to performance on statistical learning and associative inference tasks—both of which require encoding associations that span multiple episodes—in a developmental sample ranging from ages 6 to 30 years. Relating age to volume, we found dissociable patterns across the hippocampal long axis, with opposite nonlinear volume changes in the head and body. These structural differences were paralleled by performance gains across the age range on both tasks, suggesting improvements in the cross-episode binding ability from childhood to adulthood. Controlling for age, we also found that smaller hippocampal heads were associated with superior behavioral performance on both tasks, consistent with this region's hypothesized role in forming generalized codes spanning events. Collectively, these results highlight the importance of examining hippocampal development as a function of position along the hippocampal axis and suggest that the hippocampal head is particularly important in encoding associative structure across development.

Emerging evidence suggests that motivation enhances episodic memory formation through interactions between medial-temporal lobe (MTL) structures and dopaminergic midbrain. In addition, recent theories propose that motivation specifically facilitates hippocampal associative binding processes, resulting in more detailed memories that are readily reinstated from partial input. Here, we used high-resolution fMRI to determine how motivation influences associative encoding and retrieval processes within human MTL subregions and dopaminergic midbrain. Participants intentionally encoded object associations under varying conditions of reward and performed a retrieval task during which studied associations were cued from partial input. Behaviorally, cued recall performance was superior for high-value relative to low-value associations; however, participants differed in the degree to which rewards influenced memory. The magnitude of behavioral reward modulation was associated with reward-related activation changes in dentate gyrus/CA 2,3 during encoding and enhanced functional connectivity between dentate gyrus/CA 2,3 and dopaminergic midbrain during both the encoding and retrieval phases of the task. These findings suggests that, within the hippocampus, reward-based motivation specifically enhances dentate gyrus/CA 2,3 associative encoding mechanisms through interactions with dopaminergic midbrain. Furthermore, within parahippocampal cortex and dopaminergic midbrain regions, activation associated with successful memory formation was modulated by reward across the group. During the retrieval phase, we also observed enhanced activation in hippocampus and dopaminergic midbrain for high-value associations that occurred in the absence of any explicit cues to reward. Collectively, these findings shed light on fundamental mechanisms through which reward impacts associative memory formation and retrieval through facilitation of MTL and ventral tegmental area/substantia nigra processing.

A primary function of the medial temporal lobe (MTL) is to signal prior encounter with behaviorally relevant stimuli. MTL match enhancement—increased activation when viewing previously encountered stimuli—has been observed for goal-relevant stimuli in nonhuman primates during delayed-match-to-sample tasks and in humans during more complex relational memory tasks. Match enhancement may alternatively reflect (a) an attentional response to familiar relative to novel stimuli or (b) the retrieval of contextual details surrounding the past encounter with familiar stimuli. To gain leverage on the functional significance of match enhancement in the hippocampus, high-resolution fMRI of human MTL was conducted while participants attended, ignored, or passively viewed face and scene stimuli in the context of a modified delayed-match-to-sample task. On each “attended” trial, two goal-relevant stimuli were encountered before a probe that either matched or mismatched one of the attended stimuli, enabling examination of the consequences of encountering one of the goal-relevant stimuli as a match probe on later memory for the other (nonprobed) goal-relevant stimulus. fMRI revealed that the hippocampus was insensitive to the attentional manipulation, whereas parahippocampal cortex was modulated by scene-directed attention, and perirhinal cortex showed more subtle and general effects of attention. By contrast, all hippocampal subfields demonstrated match enhancement to the probe, and a postscan test revealed more accurate recognition memory for the nonprobed goal-relevant stimulus on match relative to mismatch trials. These data suggest that match enhancement in human hippocampus reflects retrieval of other goal-relevant contextual details surrounding a stimulus's prior encounter.

The essential role of the medial temporal lobe (MTL) in long-term memory for individual events is well established, yet important questions remain regarding the mnemonic functions of the component structures that constitute the region. Within the hippocampus, recent functional neuroimaging findings suggest that formation of new memories depends on the dentate gyrus and the CA 3 field, whereas the contribution of the subiculum may be limited to retrieval. During encoding, it has been further hypothesized that structures within MTL cortex contribute to encoding in a content-sensitive manner, whereas hippocampal structures may contribute to encoding in a more domain-general manner. In the current experiment, high-resolution fMRI techniques were utilized to assess novelty and subsequent memory effects in MTL subregions for two classes of stimuli—faces and scenes. During scanning, participants performed an incidental encoding (target detection) task with novel and repeated faces and scenes. Subsequent recognition memory was indexed for the novel stimuli encountered during scanning. Analyses revealed voxels sensitive to both novel faces and novel scenes in all MTL regions. However, similar percentages of voxels were sensitive to novel faces and scenes in perirhinal cortex, entorhinal cortex, and a combined region comprising the dentate gyrus, CA 2 , and CA 3 , whereas parahippocampal cortex, CA 1 , and subiculum demonstrated greater sensitivity to novel scene stimuli. Paralleling these findings, subsequent memory effects in perirhinal cortex were observed for both faces and scenes, with the magnitude of encoding activation being related to later memory strength, as indexed by a graded response tracking recognition confidence, whereas subsequent memory effects were scene-selective in parahippocampal cortex. Within the hippocampus, encoding activation in the subiculum correlated with subsequent memory for both stimulus classes, with the magnitude of encoding activation varying in a graded manner with later memory strength. Collectively, these findings suggest a gradient of content sensitivity from posterior (parahippocampal) to anterior (perirhinal) MTL cortex, with MTL cortical regions differentially contributing to successful encoding based on event content. In contrast to recent suggestions, the present data further indicate that the subiculum may contribute to successful encoding irrespective of event content.