In a constantly changing world, it is critical to update prior beliefs and memories in light of new circumstances. As new information arrives, we often need to update previously encoded information in the brain retrospectively. Imagine discovering that a longtime friend has lied to you about something important. You might automatically start looking back and reinterpreting their behavior, perhaps finding different motives for their past actions. This updated understanding of the past will assist you in your future interactions with that friend. Importantly, updating representations of real-world events does not necessarily involve rewriting or erasing the content of the previous memory for the event – it can also include adding new information that alters one’s overall interpretation of what happened. In this paper, we use the term ‘memory updating’ to refer to this process of updating representations of past events based on new information. To effectively support ‘memory updating’, the episodic memory system must be capable of modifying stored representations in light of new incoming information. Under this framework, memories are dynamic entities that can be reorganized or reconstructed even after encoding takes place (Bartlett and Burt, 1933; Conway and Pleydell-Pearce, 2000; Hassabis and Maguire, 2007; Schacter et al., 1998; Schacter, 2012).

Research in the last few decades suggests that memories are malleable to modification when they are reactivated (Przybyslawski and Sara, 1997), and relevant new information is presented (Besnard et al., 2012; Hupbach et al., 2015; Nader and Einarsson, 2010; Sinclair and Barense, 2019). Behavioral paradigms using a retroactive interference design have been widely used to study post-encoding changes in human memory (e.g. Lee et al., 2017; Hupbach et al., 2015; Samide and Ritchey, 2020; Scully et al., 2017). Only a subset of studies, however, have investigated changes in the content of memory, as opposed to the weakening or strengthening of old memories (Dongaonkar et al., 2013; Hupbach et al., 2007). At the neural level, changes in the functional connectivity of mPFC and amygdala circuitry have been associated with post-retrieval fear extinction (Feng et al., 2016; Schiller et al., 2013). These experimental studies have clinical significance and provide valuable insight into the behavioral and neural substrates of memory updating in humans. However, it is unclear how findings obtained using tightly controlled paradigms and isolated stimuli generalize to memory updating in everyday life (Nastase et al., 2020). In the present work, we introduce a naturalistic interference-based design that resembles our real-world experiences where new information obtained post-encoding is not compatible with previously encoded events. Using an audiovisual movie and verbal recall, we aim to utilize recent advances in naturalistic neuroimaging to study how memories are reshaped to incorporate new incoming information.

The brain’s default mode network (DMN)—comprising the posterior medial cortex, medial prefrontal cortex, temporoparietal junction, and parts of anterior temporal cortex—was originally described as an intrinsic or ‘task-negative’ network, activated when participants are not engaged with external stimuli (Raichle et al., 2001; Buckner et al., 2008). This observation led to a large body of work showing that the DMN is an important hub for supporting internally driven tasks such as memory retrieval, imagination, future planning, theory of mind, and creating and updating situation models (Svoboda et al., 2006; Addis et al., 2007; Hassabis and Maguire, 2007; Hassabis and Maguire, 2009; Schacter and Addis, 2007; Szpunar et al., 2007; Spreng et al., 2009; Koster-Hale and Saxe, 2013; Ranganath and Ritchey, 2012). However, it is not fully understood how this network contributes to these varying functions, and in particular—the focus of the present study—memory processes. Activation of this network during ‘offline’ periods has been proposed to play a role in the consolidation of memories through replay (Kaefer et al., 2022). Interestingly, prior work has also shown that the DMN is reliably engaged during ‘online’ processing (encoding) of continuous rich dynamic stimuli such as movies and audio stories (Stephens et al., 2013; Hasson et al., 2008). Regions in this network have been shown to have long ‘temporal receptive windows’ (Hasson et al., 2008; Lerner et al., 2011; Chang et al., 2022), meaning that they integrate and retain high-level information that accumulates over the course of extended timescales (e.g. scenes in movies, paragraphs in text) to support comprehension. This combination of processing characteristics suggests that the DMN integrates past and new knowledge, as regions in this network have access to incoming sensory input, recent active memories, and remote long-term memories or semantic knowledge (Yeshurun et al., 2021; Hasson et al., 2015). These integration processes feature in many of the ‘constructive’ processes attributed to DMN such as imagination, future planning, mentalizing, and updating situation models (Schacter and Addis, 2007; Ranganath and Ritchey, 2012). Notably, constructive processes are highly relevant to real-world memory updating, which involves selecting and combining the relevant parts of old and new memories. Recent work has shown that neural patterns during encoding and recall of naturalistic stimuli (movies) are reliably similar across participants in this network (Chen et al., 2017; Oedekoven et al., 2017; Zadbood et al., 2017; see Bird, 2020 for a review of recent naturalistic studies on memory), and the DMN displays distinct neural activity when listening to the same story with different perspectives (Yeshurun et al., 2017). Building on this foundation of prior work on the DMN, we asked whether we could find neural evidence for the retroactive influence of new knowledge on past memories.

In the current work, using a novel naturalistic paradigm intended to simulate a real-life situation of adaptive memory updating, we asked how new information changes the neural representations in the DMN during the recall of prior events. To answer this question, we used a popular Hollywood-style film titled ‘The Sixth Sense’ (M. Night Shyamalan, 1999), which contains a dramatic twist in the final scene. [Spoiler alert!] The movie depicts the story of a clinical psychologist treating a child who claims to see ghosts. In the final scene, it is revealed that the doctor was in fact, a ghost himself throughout the movie. Therefore, there are two coherent interpretations of the movie: the Doctor (or naive) interpretation (labeled D in Figure 1), which is typically held by viewers up until they encounter the ‘twist ending’; and the Ghost (or spoiled) interpretation (labeled G in Figure 1), which is held by viewers after they learn about the twist. In this setting, memory updating is operationalized as the transition from the Doctor (D) interpretation to the Ghost (G) interpretation.

Figure 1

Download asset Open asset

Our study design hinges on the hypothesis that participants who received the twist and are aware that the doctor is a ghost might have distinct neural representations of the events from those who encoded the movie while ignorant of the twist. Importantly, we predicted that encountering the twist after encoding the movie would initiate a retrospective update in the interpretation of the encoded movie and that this update would be reflected in both verbal recall and patterns of brain activity during remembering. In contrast, the neural representations of the events in the movie will remain unchanged during recall in subjects who do not need to update their memories during recall (i.e. in subjects in the ‘no-twist’ condition who are only aware of the D interpretation, or subjects in the ‘spoiler’ condition who knew all along about the G interpretation).

In a large set of regions in the DMN, we found that context changed how the movie was encoded into memory. In other words, the neural representations for each event in the movie were different for viewers who believed the doctor was alive versus viewers who believed the doctor was a ghost. Furthermore, in several DMN regions, we found that neural representations were updated during recall for viewers who learned that the doctor was a ghost after watching the movie. Together these results suggest that areas in the default mode network are actively updating the neural representations as they integrate incoming information with prior knowledge.

Three distinct experimental groups watched concise versions of a popular Hollywood-style film titled “The Sixth Sense” (M. Night Shyamalan, 1999) in the fMRI scanner (Figure 1B, left column). Following the movie viewing all three groups were asked to freely recall the movie in the scanner (Figure 1B, left column). Participants in the main group (the ‘twist’ condition, Figure 1B, middle row), watched the movie with the twist scene at the end. Therefore, they watched the movie naive to the true nature of the doctor (Movie-Doctor or M_D). During their recall, however, they were aware of the twist information and could use it to update their memory (Recall-Ghost or R_G). In order to identify interpretation-specific neural patterns, we needed two comparison conditions: the Movie-Ghost (M_G) condition during viewing, and the Recall-Doctor (R_D) condition during recall. Therefore, we introduced two other groups to the study: participants in one group (the ‘spoiled’ condition; Figure 1B, top row) were exposed to the twist at the beginning of the movie. This group watched and recalled the movie knowing that the doctor was a ghost (M_G and R_G). The other group (the ‘no-twist’ condition; Figure 1B, bottom row) never received the twist information throughout encoding and remained naive to the true nature of the doctor in both their encoding and recall (M_D and R_D). This design allowed us to compare the behavioral and neural patterns of response in participants across the two interpretations. We compared the patterns of neural responses in the ‘twist’ group with the patterns in the ‘spoiled’ and ‘no-twist’ groups during encoding and recall. We predicted that the ‘twist’ group would be more similar to the ‘no-twist’ group during encoding (both having the Doctor interpretation) but more similar to the ‘spoiled’ group during recall (both having the Ghost interpretation). Moreover, we asked whether the memory updating would make the recall of the ‘twist’ group more similar to the encoding of the ‘spoiled’ group (see the ‘prediction legends’ in Figures 2 and 3).

We used intersubject pattern similarity analysis (intersubject pattern correlation: pISC, see Methods) to compare the neural event representations between groups. This analytic approach is motivated by prior work showing that slowly-evolving activity patterns in DMN represent event-level information (Baldassano et al., 2018) and that pISC captures scene-specific pattern similarity between groups who watched the same movie, groups who verbally recalled the same story (in their own words), and across viewing and recall of the same scenes (Chen et al., 2017; Zadbood et al., 2017). In this analysis, scene-specific neural patterns during encoding of the movie were obtained by averaging data across time within each scene in each subject (Chen et al., 2017; Zadbood et al., 2017). For the cued recall data, we ran a GLM analysis (Mumford et al., 2012) to capture responses corresponding to the recall of single events. fMRI responses averaged across time points within an event (or estimated from the GLM) for each ROI served as the spatial response patterns (i.e. neural event representations) that were then compared across groups using pISC. We then compared pairs of pattern similarity correlations based on our hypotheses. For example, we hypothesized that the updated recall in the ‘twist’ group would be more similar to the recall of the ‘spoiled’ group (as both groups have the Ghost interpretation during the recall) than of the ‘no-twist’ group (who have the Doctor interpretation). To test this hypothesis, in each ROI, we measured pISC once between ‘twist’ and ‘spoiled’ groups and once between ‘twist’ and ‘no-twist’ groups. We then compared these two sets of pattern similarity values to quantify which two groups’ neural event representations were more similar. We focused our analyses on a predetermined selection of movie scenes (i.e. 7 ‘critical scenes’ out of 18 total scenes) in which the Doctor or Ghost interpretation of the main character in the movie would dramatically change the overall interpretation of those scenes. Selection of these scenes was based on ratings from five raters asked to quantify the influence of the twist on the interpretation of each scene (see Methods).

After watching the movie, participants performed a cued-recall task in which they watched a few seconds of the beginning of all movie scenes (18 scenes) and were asked to describe what happened next in that scene. The recall task was identical across the three experimental conditions. Participants were highly accurate in recognizing the corresponding scenes from the movie cues (94% accuracy in the ‘twist’ group, 93% in the ‘spoiled’ group, and 97% in the ‘no-twist’group). Only the scenes that were correctly recalled were included in the neural analyses. The content of recall was evaluated using two separate measures assigned by human raters. Memory score assessed the quality and detail of memory. Twist score assessed whether the twist information was incorporated into the recall and ranged from 1 (the recall purely reflected the Doctor interpretation) to 5 (the recall purely reflected the Ghost interpretation). Memory score and twist score were expected to capture different aspects of the recall behavior; for example a detailed recall of the original scene about the doctor treating the child (high memory score) may not include information about the doctor being a ghost (low twist score). Indeed, there was no significant correlation between memory scores and twist scores across participants (r=0.07, p=0.56). If participants were unaware of the twist or did not incorporate it into their recall at all, we would expect the average twist score of the critical scenes to be approximately equal to 1 (‘purely reflects the Doctor interpretation’). In the main experimental group (‘twist’ group), 14 out of 19 participants scored above 2 (median score = 3.25) on the twist score, indicating that they incorporated the new interpretation into their recall. Importantly, the ‘twist’ group (twist score: M=3.16, SD = 1.03) exhibited a significantly higher twist score (t(37) = 6.37, p<0.001) than the ‘no-twist’ group (twist score: M=1.65, SD = 0.22). Note that these two groups had no knowledge of the twist when they encoded the movie. Therefore, this result confirms that participants in the ‘twist group’ updated their memories of the movie to incorporate the twist. No significant difference (t(35) = 1.46, p=0.15) was observed between the twist score of the ‘twist’ group and the ‘spoiled’ group (twist score: M=2.72, SD = 0.74). This finding suggests that the ‘twist’ group recalled the movie more similarly to the group that knew the twist while watching the movie.

Neural representation of the twist information during movie-viewing

First, we set out to test how contextual knowledge about the twist modifies the neural patterns in the DMN during the encoding of the movie into memory. As encoding conditions are directly compared, we refer to this analysis as encoding-encoding throughout the paper. We compared the spatially distributed neural activity patterns elicited during movie-viewing (encoding) in the ‘twist’ group (M_D) to the activity patterns obtained during encoding in the ‘no-twist’ group (M_D) and the ‘spoiled’ group (M_G). We hypothesized that within the regions of the brain that are sensitive to different interpretations, the pattern similarity between the ‘twist’ group (M_D) and the ‘no-twist’ group (M_D) should be higher than the similarity between the ‘twist’ group (M_D) and the ‘spoiled’ group (M_G) (Figure 2A, prediction legends).

Figure 2

Download asset Open asset

Indeed, there was significantly greater intersubject pattern correlation in parts of the DMN between the ‘twist’ and ‘no-twist’ experimental groups (who had a similar M_D interpretation of the movie during encoding) than across experimental groups with opposing interpretations (M_D versus M_G). These areas included the dorsal and lateral PFC, left precuneus, left retrosplenial cortex, left angular gyrus, middle temporal cortex, left superior temporal cortex, and left temporal pole (Figure 2A). These results fit with previous findings demonstrating that the timecourse of brain responses in DMN regions reflects different perspectives when listening to a spoken narrative (Yeshurun et al., 2017). Our results extend these findings by showing that different interpretations are discriminable in spatial response patterns measured while viewing audiovisual movie stimuli.

Relationship between the neural representations during encoding and recall

To provide further neural evidence for the shift from Doctor interpretation during encoding to Ghost interpretation during recall in the ‘twist’ group, we directly compared the brain responses elicited during encoding and recall (encoding-recall analyses). Chen et al., 2017 have demonstrated that, across free recall of a movie, neural patterns are reinstated in DMN. In addition, these scene-specific neural patterns changed between encoding and recall in a systematic manner across individuals (Chen et al., 2017). We hypothesized that updating one’s interpretation to incorporate twist information might alter the neural representations during recall, such that they become more similar to the neural patterns elicited during encoding of the spoiled movie.

We tested this hypothesis in two ways. First, we predicted that (Figure 3A, prediction legend) the neural pattern similarity between recall in the ‘twist’ group and encoding in the ‘spoiled’ group (R_G to M_G) would be higher than the pattern similarity between recall in the ‘no-twist’ group and encoding of the ‘spoiled’ group (R_D to M_G). Our analysis confirmed this prediction in the left angular gyrus, left dorsomedial PFC, and right middle temporal cortex (Figure 3A).

Figure 3

Download asset Open asset

Second, if participants in the ‘twist’ group were to fully update their interpretation at recall from Doctor to Ghost, we would expect activity patterns during recall in the ‘twist’ group to be more similar to encoding in the ‘spoiled’ group (R_G to M_G) compared to encoding in their own (‘twist’) group (R_G to M_D) (Figure 3B, prediction legends). When we looked for regions showing this effect, we found weak effects in the predicted direction in the left angular gyrus, left frontal pole, and right anterior temporal ROIs (note that all of these comparisons were performed across participants; see Methods for details); however, these effects did not survive correction for multiple comparisons at an FDR-corrected p<0.05 (Figure 3B). The most straightforward interpretation of these weak effects is, in general, ‘twist’ group participants did not fully update their interpretations; that is, there may have been some lingering memory of the Doctor interpretation in the ‘twist’ group in some participants even after they were exposed to Ghost interpretation and updated their memory.

To test this hypothesis, we ran an exploratory analysis where we correlated neural pattern change (i.e. the degree to which the neural pattern at recall matched the Doctor or Ghost encoding pattern) with behavioral twist scores (i.e. how much each subject discussed the twist during recall) across participants in the ‘twist’ group, in each DMN ROI (Appendix 1—figure 1). If weak neural pattern change effects are due to incomplete memory updating, we would expect to see a positive correlation between these measures. We observed a positive correlation between the neural and behavioral indices of memory update in posterior regions of the DMN, including precuneus and angular gyrus. The right precuneus ROI exhibited a notable relationship (r=0.62); however, this did not survive FDR correction across ROIs.

The role of scene content

In the prior analyses, we focused on ‘critical scenes’, selected based on ratings from five raters who quantified the influence of the twist on the interpretation of each scene (see Methods). An independent post-experiment analysis of the verbal recall behavior of the fMRI participants yielded ‘twist scores’ that were also highest for these scenes; that is, the expected and perceived effect of twist information on recall behavior were found to match. In our next analysis, we asked whether the neural event representations reflect these differences in the twist-related content of the scenes. In other words, are the ‘critical scenes’ with highly twist-dependent interpretations truly critical for our observed effects?

To answer this question, we re-ran our main encoding-encoding and recall-recall pISC analysis in each DMN ROI (Figures 2–3). We calculated interaction indices (Figure 2C) first by including all scenes, and second by including only the 11 non-critical scenes. To better compare the effect of including different subsets of scenes to our original results, in Figure 4 we show the results in 15 ROIs that exhibited meaningful effects in our main analyses (Figure 2C). Figure 4A demonstrates that ‘critical scenes’ yielded higher interaction indices compared to all scenes or non-critical scenes across all ROIs. The interaction score across all DMN ROIs was significantly higher in ‘critical scenes’ than all scenes (t(23) = 7.19, p=2.53 x 10^–7) and non-critical scenes (t(23) = 7.3, p=1.95 x 10^–7). These results show that critical scenes are indeed responsible for the observed pISC differences across groups.

Figure 4

Download asset Open asset

Next, to determine whether scene-specific neural event representations—as opposed to coarser differences in general mental state across all scenes with similar interpretations—drive our observed pISC differences, we shuffled the labels of critical scenes within each group before calculating and comparing pISC across groups. By repeating this procedure 1000 times and recalculating the interaction index at each iteration, we constructed a null distribution of interaction indices for shuffled critical scenes (light magenta distributions in Figure 4B). In 12 out of 24 DMN regions, interaction indices were statistically significant based on the shuffled-scene distribution (p<0.025, FDR controlled at q<.05). All of these 12 regions were among the ROIs that showed meaningful effects in our original analysis (Figure 2C). Regions with significant scene-specific interaction effects are marked as blue dots with black borders in Figure 4B. Overall, the findings from this analysis confirm that our results are driven by changes to scene-specific representations.

To further evaluate the relationship between scene-specific twist information in the brain and behavior, we ran an exploratory analysis which was focused on the changes in the neural event representations during recall of the ‘twist’ group and their corresponding recall behavior. We ran the same pISC procedure described in Figures 2B and 3B. However, we did not average the pISC differences across scenes. In the recall-recall analysis, this procedure yielded 18 values (one for each scene, averaged across participants) indicating whether the neural event representations during updated recall were more similar to the spoiled recall or the naïve recall for a given scene. We then correlated these values with the twist score data (based on ratings of verbal recall) for each scene averaged across participants (Appendix 1—figure 4A). None of these correlations were significant after correction for multiple tests; however, the four ROIs with significant effects in the main recall-recall analysis (Figure 2B) all showed positive correlations, particularly left mPFC. We repeated the same procedure to compute the relationship between scene-level neural event representations and behavioral twist scores in the encoding-recall analysis (Figure 3B). In the original analysis (Figure 3B), we evaluated whether the neural representations of updated recall in the ‘twist’ group were more similar to encoding in the ‘spoiled’ group than their own naïve encoding. Here, we followed the same procedure of correlating scene-level outputs with scene-level twist scores (Appendix 1—figure 4B). Again, none of these correlations were significant after correcting for multiple tests, but the three regions with (uncorrected) effects in the original analysis (Figure 3B) all displayed positive correlations with twist score. As we did not find strong results in this exploratory set of analyses, we refrain from overinterpreting them. However, they appear to match the direction of our main analyses; with greater statistical power analyses of this sort may provide insights into how neural event representations are updated in a scene-specific manner.

Using a novel naturalistic paradigm that prompted participants to update their previously-encoded memories, we studied how new information can retrospectively change the event representations in the default mode network. At encoding, a widespread network of frontal, parietal, and temporal regions exhibited significantly higher pattern similarity between groups in which participants had the same interpretation of the movie (naive to the twist; see Figure 2A). This result demonstrates how a belief about the identity of the doctor (which can broadly be construed as the context or the state of mind of the observer) can shape the encoding processes of new information (the same movie) into memory. But information is not only shaped by context during encoding, as stored representations must also be amenable to change as the context changes at a later stage. Indeed, our unique paradigm allows us to see how the patterns of stored representations change, as we learn about the twist in the movie. In particular, the neural patterns during recall changed in the twist condition to better match the neural patterns in the spoiled condition observed during recall in the ventromedial PFC, right precuneus, and temporal cortex (see Figure 2B). Furthermore, numerous areas throughout the DMN showed a significant interaction whereby neural patterns in the ‘twist’ group became relatively more similar to patterns from the ‘spoiled’ Ghost group (compared to the ‘no-twist’ Doctor group) at recall (compared to encoding; Figure 2C).

We also found evidence for memory updating by directly comparing patterns from encoding and retrieval. In the left angular gyrus, left dorsomedial PFC, and right middle temporal cortex, viewing the twist at the end of the movie (vs. not viewing the twist) resulted in neural patterns at recall becoming more similar to the ‘spoiled’ Ghost encoding patterns (Figure 3A). In some regions, this updating effect led to ‘twist’ recall patterns being numerically more similar to the ‘spoiled’ encoding patterns than to encoding patterns from the ‘twist’ condition, but this effect did not survive multiple comparisons correction (Figure 3B). We suggested that the weakness of this effect may be attributable to some participants not fully discarding the Doctor interpretation when they update their interpretation; in line with this, an exploratory analysis showed that—in some DMN ROIs—the degree of neural change was nominally correlated (across participants) with behavioral ‘twist scores’ capturing how strongly a participant’s recall was influenced by the twist (Appendix 1—figure 1; these exploratory correlations did not survive multiple comparisons correction). Taken together, our results provide further evidence for the involvement of DMN regions in integrating new information with prior knowledge to form distinct high-level event representations. In particular, we suggest a subset of core DMN regions are implicated in representing changes in event interpretations during memory updating.

The default mode network, traditionally known to support internally oriented processes, is now considered a major hub for actively processing incoming external information and integrating it with prior knowledge in the social world (Yeshurun et al., 2021). Our experimental design targets naturalistic event representations unfolding over seconds to minutes. There have been many studies to date corroborating the discovery of a cortical hierarchy of increasing temporal receptive windows where high-level event representations are encoded at the top of the hierarchy—in the DMN (Hasson et al., 2008; Lerner et al., 2011; Hasson et al., 2015; Baldassano et al., 2018; etc). This network is involved in episodic encoding and retrieval (Rugg and Vilberg, 2013) and constructive memory-related tasks such as imagining fictitious scenes and future events (Addis et al., 2007; Hassabis and Maguire, 2007; Hassabis and Maguire, 2007; Rugg and Vilberg, 2013; Schacter and Addis, 2007; Schacter and Addis, 2007). Our design relies on an event-level correspondence between the encoding (viewing) and verbal recall of movie scenes. Previous research has localized modality-independent representations of movie scenes (Zadbood et al., 2017) and their similarity during encoding and recall (Chen et al., 2017) to the DMN. These characteristics make this network a good candidate to contribute to memory updating—a constructive process in which new information is integrated into past event memories in service of better guiding behavior. Our findings support this idea by showing the shift in neural representations during updated recall in a subset of regions in this network.

At encoding, a widespread set of areas including dorsal and lateral PFC, left precuneus, left retrosplenial cortex, and left angular gyrus had differentiable neural patterns across the two interpretations of the movie. These results are consistent with previous work that showed the time course of brain responses in DMN distinguishes between groups when participants are prompted to take two different perspectives before listening to an audio story (Yeshurun et al., 2017). We extend these results to an audiovisual movie, and provide evidence that interpretative perspective is also encoded in spatially distributed neural response patterns for narrative events, averaged across minutes-long scenes. Interestingly, the difference in neural responses measured by Yeshurun and colleagues was not significant between the two perspectives of the story in the ventral portion of mPFC. Similarly, vmPFC ROIs did not exhibit a significant difference between the Doctor and Ghost representations during the encoding phase in our experiment. Previous research has implicated mPFC in processing schematic information and integration of new information into prior knowledge (Gilboa and Marlatte, 2017; Schlichting and Preston, 2017; van Kesteren et al., 2012). Using naturalistic clips as schematic events, it has been shown that response patterns in mPFC are particularly dependent on intact and predictable schemas (Baldassano et al., 2018). Together, these results suggest that our manipulation (Doctor and Ghost interpretations) may not have substantially altered the schemas that participants were using during movie-viewing (e.g. during a restaurant scene, participants will need to use their ‘restaurant’ schema to interpret it, regardless of whether the doctor is alive or a ghost)—although we interpret these null results with caution.

Even though groups had different knowledge/perspectives during encoding, we found higher pattern similarity across groups if they had similar twist knowledge during recall in vmPFC, right precuneus, and parts of temporal cortex. Previous findings suggest mPFC is involved in not just encoding but retrieval of memories in relation to prior knowledge (Brod et al., 2015; van Kesteren et al., 2010) and retrieval of overlapping representations to support integration and organization of related memories (Tompary and Davachi, 2017). Our observations during recall fit with these findings and suggest that shifting toward a more similar perspective during recall leads to higher neural similarity in mPFC. However, during encoding, we did not observe a significant pattern correlation between groups that held the same interpretation of the movie. Furthermore, vmPFC was significant in our interaction analysis (Figure 2C), indicating that the similarity structure of vmPFC patterns across conditions was significantly different at encoding versus retrieval. Together, these results suggest vmPFC is differently implicated in encoding and recall of story-specific representations during processing of naturalistic events. In addition to mPFC, right precuneus and parts of temporal cortex exhibited significantly higher pattern similarity in the ‘twist’ and ‘spoiled’ groups who recalled the movie with the same interpretation. Precuneus is a core region in the posterior medial network, which is hypothesized to be involved in constructing and applying situation models (Ranganath and Ritchey, 2012). Our findings support a role for precuneus in deploying interpretation-specific situation models when retrieving event memories. In particular, we suggest that the posterior medial network may encode a shift in the situation model of the ‘twist’ group in order to accommodate the new Ghost interpretation.

We performed two targeted analyses to look for evidence of memory updating across encoding and recall: the interaction analysis (Figure 2C) and the encoding-recall analysis (Figure 3). We hypothesized that a shift in direction of pISC difference would occur when neural representations during recall in the ‘twist’ group start to reflect the Ghost interpretation. The interaction analysis probed this shift indirectly by taking into account the effects of both encoding-encoding and recall-recall analyses. Unlike the interaction analysis, in the encoding-recall analysis, we directly compared neural event representations during encoding and recall. Interestingly, all regions exhibiting an effect across the two encoding-recall analyses, excluding left anterior temporal cortex, were present in the interaction results. Among these regions, the left angular gyrus/TPJ exhibited an effect across all three analyses. As a core hub in the mentalizing network, temporo-parietal cortex has been implicated in theory of mind through perspective-taking, rationalizing the mental state of someone else, and modeling the attentional state of others (Frith and Frith, 2006; Guterstam et al., 2021; Saxe and Kanwisher, 2003). The motivations behind some actions of the main character in the movie heavily depend on whether the viewer perceives them as a Doctor or a Ghost, and participants may focus on this during both encoding and recall. We speculate that neural event representations in AG/TPJ in the current experiment may be related to mentalizing about the main character’s actions. Under this interpretation, the updated event representations during recall following the twist would be more closely aligned to the ‘spoiled’ encoding representations, as a consequence of memory updating in the ‘twist’ group.

Our findings are consistent with the view that DMN synthesizes incoming information with one’s prior beliefs and memories (Yeshurun et al., 2021). We add to this framework by providing evidence for the involvement of DMN regions in updating prior beliefs in light of new knowledge. Across our different encoding and recall analyses, we observe memory updating effects in a varied subset of DMN regions that do not cleanly map onto a specific subsystem of DMN (Robin and Moscovitch, 2017; Ranganath and Ritchey, 2012; Ritchey and Cooper, 2020). Rather than being divergent, these results might be reflecting inherent differences between the processes of encoding and recall of naturalistic events. It has been proposed that neural representations corresponding to encoding of events are systematically transformed during recall of those events (Chen et al., 2017; Favila et al., 2020; Musz and Chen, 2022). While we provide evidence for reinstatement of memories in DMN, our findings also support a transformation of neural representation during recall, as encoding-recall results were weaker in some areas than recall-recall findings. This transformation could affect how different regions and sub-systems of DMN represent memories, and suggests that the concerted activity of multiple subsystems and neural mechanisms might be at play during encoding, recall and successful updating of naturalistic event memories.

While our main goal in this paper was to examine how neural representations of naturalistic events change in the DMN, we also examined visual and somatosensory networks. Aside from the encoding-encoding analysis in which some visual and somatosensory regions showed stronger similarity between two groups with the same interpretation of the movie, we did not find any regions with significant effects in these two networks in the other analyses. Unlike the recall phase where each participant has their unique utterance with their own choice of words and concepts to describe the movie, the encoding (move-watching) stimulus is identical across all groups. Therefore, the effects observed during encoding-encoding analysis in sensory regions could reflect similarity in perception of the movie guided by similar attentional state while watching scenes with the same interpretation (e.g. similarity in gaze location, paying attention to certain dialogues, or small body movements while watching the movie with the same Doctor or Ghost interpretations). In our whole brain analysis, these regions did not have significant interaction effects, which suggests that the effects were isolated to encoding. In the whole-brain analysis, we also observed a significant encoding-encoding and interaction effects in anterior cingulate cortex, as well as recall-recall and interaction effects in dlPFC. These results suggest that both the ‘spoiled’ manipulation and the ‘twist’ may recruit top-down control and conflict monitoring processes during naturalistic viewing and recall.

Our findings provide further insight into the functional role of the DMN. However, these results have been obtained using only one movie. While naturalistic paradigms better capture the complexity of real life and provide greater ecological generalizability than highly-controlled experimental stimuli and tasks (Nastase et al., 2020), they are still limited by the properties of the particular naturalistic stimulus used. For example, this movie—including the twist itself—hinges on suspension of disbelief about the existence of ghosts. Future work is needed to extend our findings about updating event memories to a broader class of naturalistic stimuli: for example, movies with different kinds of (non-supernatural) plot twists, spoken stories with twist endings, or using autobiographical real-life situations where new information (e.g. discovering a longtime friend has lied about something important) triggers re-evaluation of the past (e.g. reinterpreting their friend’s previous actions). Moreover, our current method relies on averaging spatially coarse activity patterns across subjects (and time points within an event). Future extensions of this work may benefit from using functional alignment methods (Haxby et al., 2020; Chen et al., 2015) to capture more fine-grained event representations which are shared across participants.

During recall, many participants recounted both the old and new interpretations (Ghost and Doctor) of movie scenes. This behavior indicated that they maintained both representations in parallel (possibly competing), rather than overwriting the old representation with new information. The simultaneous presence of these representations poses an interesting theoretical question for future studies: When does updating the memory cause us to lose traces of the old interpretation, and when do the old and new interpretations end up co-existing in memory? Previous studies have shown that old and new memory traces are simultaneously reactivated in the brain, leading to competition (e.g. Kuhl et al., 2012), and this competition can trigger learning processes that resolve the competition; for example, by weakening one of the memories or by restructuring the memories so they can coexist (Ritvo et al., 2019). Understanding how competition between interpretations plays out over time is an important topic for future work; existing research on memory revaluation suggests that updating may be a temporally-extended process driven by successive replays of the new information, rather than taking place all at once (see, e.g., Momennejad et al., 2018). In clinical settings, methods inspired by reconsolidation and memory updating are extensively used to treat maladaptive memories (Phelps and Hofmann, 2019). In these clinical contexts, it will be especially important to understand the factors that influence the ‘end state’ of this competition between interpretations (in terms of our study: who ends up fully adopting the Ghost interpretation and who ends up with lingering traces of the Doctor interpretation).

In summary, our findings show that in a movie with a dramatic twist ending, the new information introduced by the twist causes a new (Ghost) interpretation of past events to take root in participants’ brains. Overall, these results highlight the importance of DMN regions in updating naturalistic memories and suggest new approaches to studying real-world memory modification in both experimental and clinical treatment settings.

Code available at: https://github.com/azadbood/sixthsense, (copy archived at swh:1:rev:113b3203722573033b5cc4535f6523c1fca5c1d7) Data available at: https://doi.org/10.18112/openneuro.ds004359.v1.0.0.

1. 1. Zadbood A
   2. Nastase SA
   3. Chen J
   4. Norman KA
   5. Hasson U

   (2022) OpenNeuro

   SixthSense.

   https://doi.org/10.18112/openneuro.ds004359.v1.0.0

1. 1. Addis DR
   2. Wong AT
   3. Schacter DL

   (2007) Remembering the past and imagining the future: common and distinct neural substrates during event construction and elaboration

   Neuropsychologia 45:1363–1377.

   https://doi.org/10.1016/j.neuropsychologia.2006.10.016

   • PubMed
   • Google Scholar
2. Software

   1. Asieh Z

   (2022) SixthSense, version swh:1:rev:113b3203722573033b5cc4535f6523c1fca5c1d7

   Software Heritage.

   https://archive.softwareheritage.org/swh:1:dir:7f0444cb2867810520afa3a060e1c9683964b8aa;origin=https://github.com/azadbood/sixthsense;visit=swh:1:snp:1a323a91bc3f2153abac45a071620e2ddeaef0e1;anchor=swh:1:rev:113b3203722573033b5cc4535f6523c1fca5c1d7
3. 1. Avants BB
   2. Epstein CL
   3. Grossman M
   4. Gee JC

   (2008) Symmetric diffeomorphic image registration with cross-correlation: evaluating automated labeling of elderly and neurodegenerative brain

   Medical Image Analysis 12:26–41.

   https://doi.org/10.1016/j.media.2007.06.004

   • PubMed
   • Google Scholar
4. 1. Baldassano C
   2. Hasson U
   3. Norman KA

   (2018) Representation of real-world event schemas during narrative perception

   The Journal of Neuroscience 38:9689–9699.

   https://doi.org/10.1523/JNEUROSCI.0251-18.2018

   • PubMed
   • Google Scholar
5. 1. Bartlett FC
   2. Burt C

   (1933) Remembering: a study in experimental and social psychology

   British Journal of Educational Psychology 3:187–192.

   https://doi.org/10.1111/j.2044-8279.1933.tb02913.x

   • Google Scholar
6. 1. Benjamini Y
   2. Hochberg Y

   (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing

   Journal of the Royal Statistical Society 57:289–300.

   https://doi.org/10.1111/j.2517-6161.1995.tb02031.x

   • Google Scholar
7. 1. Besnard A
   2. Caboche J
   3. Laroche S

   (2012) Reconsolidation of memory: a decade of debate

   Progress in Neurobiology 99:61–80.

   https://doi.org/10.1016/j.pneurobio.2012.07.002

   • PubMed
   • Google Scholar
8. 1. Bird CM

   (2020) How do we remember events?

   Current Opinion in Behavioral Sciences 32:120–125.

   https://doi.org/10.1016/j.cobeha.2020.01.020

   • Google Scholar
9. 1. Brod G
   2. Lindenberger U
   3. Werkle-Bergner M
   4. Shing YL

   (2015) Differences in the neural signature of remembering schema-congruent and schema-incongruent events

   NeuroImage 117:358–366.

   https://doi.org/10.1016/j.neuroimage.2015.05.086

   • PubMed
   • Google Scholar
10. 1. Buckner RL
    2. Andrews-Hanna JR
    3. Schacter DL

    (2008) The brain’s default network: anatomy, function, and relevance to disease

    Annals of the New York Academy of Sciences 1124:1–38.

    https://doi.org/10.1196/annals.1440.011

    • PubMed
    • Google Scholar
11. Preprint

    1. Chang CHC
    2. Nastase SA
    3. Hasson U

    (2022) Information Flow across the Cortical Timescales Hierarchy during Narrative Construction

    bioRxiv.

    https://doi.org/10.1101/2021.12.01.470825

    • Google Scholar
12. Conference

    1. Chen PH
    2. Chen J
    3. Yeshurun Y
    4. Hasson U
    5. Haxby J
    6. Ramadge PJ

    (2015) A Reduced-Dimension fMRI Shared Response Model

    Advances in Neural Information Processing Systems 28.

    https://papers.nips.cc/paper/2015/hash/b3967a0e938dc2a6340e258630febd5a-Abstract.html

    • Google Scholar
13. 1. Chen J
    2. Leong YC
    3. Honey CJ
    4. Yong CH
    5. Norman KA
    6. Hasson U

    (2017) Shared memories reveal shared structure in neural activity across individuals

    Nature Neuroscience 20:115–125.

    https://doi.org/10.1038/nn.4450

    • PubMed
    • Google Scholar
14. 1. Conway M
    2. Pleydell-Pearce C

    (2000) The construction of autobiographical memories in the self-memory system

    Psychological Review 107:261–288.

    https://doi.org/10.1037/0033-295x.107.2.261

    • PubMed
    • Google Scholar
15. 1. Cox RW

    (1996) AFNI: software for analysis and visualization of functional magnetic resonance neuroimages

    Computers and Biomedical Research, an International Journal 29:162–173.

    https://doi.org/10.1006/cbmr.1996.0014

    • PubMed
    • Google Scholar
16. 1. Dongaonkar B
    2. Hupbach A
    3. Gomez R
    4. Nadel L

    (2013) Effects of psychosocial stress on episodic memory updating

    Psychopharmacology 226:769–779.

    https://doi.org/10.1007/s00213-013-2998-8

    • PubMed
    • Google Scholar
17. 1. Esteban O
    2. Markiewicz CJ
    3. Blair RW
    4. Moodie CA
    5. Isik AI
    6. Erramuzpe A
    7. Kent JD
    8. Goncalves M
    9. DuPre E
    10. Snyder M
    11. Oya H
    12. Ghosh SS
    13. Wright J
    14. Durnez J
    15. Poldrack RA
    16. Gorgolewski KJ

    (2019) FMRIPrep: a robust preprocessing pipeline for functional MRI

    Nature Methods 16:111–116.

    https://doi.org/10.1038/s41592-018-0235-4

    • PubMed
    • Google Scholar
18. 1. Favila SE
    2. Lee H
    3. Kuhl BA

    (2020) Transforming the concept of memory reactivation

    Trends in Neurosciences 43:939–950.

    https://doi.org/10.1016/j.tins.2020.09.006

    • PubMed
    • Google Scholar
19. 1. Feng P
    2. Zheng Y
    3. Feng T

    (2016) Resting-state functional connectivity between amygdala and the ventromedial prefrontal cortex following fear reminder predicts fear extinction

    Social Cognitive and Affective Neuroscience 11:991–1001.

    https://doi.org/10.1093/scan/nsw031

    • PubMed
    • Google Scholar
20. 1. Fisher RA

    (1915) Frequency distribution of the values of the correlation coefficient in samples from an indefinitely large population

    Biometrika 10:507.

    https://doi.org/10.2307/2331838

    • Google Scholar
21. 1. Fonov V
    2. Evans A
    3. McKinstry R
    4. Almli C
    5. Collins D

    (2009) Unbiased nonlinear average age-appropriate brain templates from birth to adulthood

    NeuroImage 47:S102.

    https://doi.org/10.1016/S1053-8119(09)70884-5

    • Google Scholar
22. 1. Frith CD
    2. Frith U

    (2006) The neural basis of mentalizing

    Neuron 50:531–534.

    https://doi.org/10.1016/j.neuron.2006.05.001

    • PubMed
    • Google Scholar
23. 1. Gilboa A
    2. Marlatte H

    (2017) Neurobiology of schemas and schema-mediated memory

    Trends in Cognitive Sciences 21:618–631.

    https://doi.org/10.1016/j.tics.2017.04.013

    • PubMed
    • Google Scholar
24. 1. Gorgolewski K
    2. Burns CD
    3. Madison C
    4. Clark D
    5. Halchenko YO
    6. Waskom ML
    7. Ghosh SS

    (2011) Nipype: a flexible, lightweight and extensible neuroimaging data processing framework in python

    Frontiers in Neuroinformatics 5:13.

    https://doi.org/10.3389/fninf.2011.00013

    • PubMed
    • Google Scholar
25. 1. Greve DN
    2. Fischl B

    (2009) Accurate and robust brain image alignment using boundary-based registration

    NeuroImage 48:63–72.

    https://doi.org/10.1016/j.neuroimage.2009.06.060

    • PubMed
    • Google Scholar
26. 1. Guterstam A
    2. Bio BJ
    3. Wilterson AI
    4. Graziano M

    (2021) Temporo-parietal cortex involved in modeling one’s own and others’ attention

    eLife 10:e63551.

    https://doi.org/10.7554/eLife.63551

    • PubMed
    • Google Scholar
27. 1. Hassabis D
    2. Maguire EA

    (2007) Deconstructing episodic memory with construction

    Trends in Cognitive Sciences 11:299–306.

    https://doi.org/10.1016/j.tics.2007.05.001

    • PubMed
    • Google Scholar
28. 1. Hassabis D
    2. Maguire EA

    (2009) The construction system of the brain

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 364:1263–1271.

    https://doi.org/10.1098/rstb.2008.0296

    • PubMed
    • Google Scholar
29. 1. Hasson U
    2. Yang E
    3. Vallines I
    4. Heeger DJ
    5. Rubin N

    (2008) A hierarchy of temporal receptive windows in human cortex

    The Journal of Neuroscience 28:2539–2550.

    https://doi.org/10.1523/JNEUROSCI.5487-07.2008

    • PubMed
    • Google Scholar
30. 1. Hasson U
    2. Chen J
    3. Honey CJ

    (2015) Hierarchical process memory: memory as an integral component of information processing

    Trends in Cognitive Sciences 19:304–313.

    https://doi.org/10.1016/j.tics.2015.04.006

    • PubMed
    • Google Scholar
31. 1. Haxby JV
    2. Guntupalli JS
    3. Nastase SA
    4. Feilong M

    (2020) Hyperalignment: modeling shared information encoded in idiosyncratic cortical topographies

    eLife 9:e56601.

    https://doi.org/10.7554/eLife.56601

    • PubMed
    • Google Scholar
32. 1. Hupbach A
    2. Gomez R
    3. Hardt O
    4. Nadel L

    (2007) Reconsolidation of episodic memories: a subtle reminder triggers integration of new information

    Learning & Memory 14:47–53.

    https://doi.org/10.1101/lm.365707

    • PubMed
    • Google Scholar
33. Book

    1. Hupbach A
    2. Gomez R
    3. Nadel L

    (2015) Memory reconsolidation

    In: Hupbach A, editors. The Wiley Handbook on the Cognitive Neuroscience of Memory. John Wiley & Sons, Ltd. pp. 244–264.

    https://doi.org/10.1002/9781118332634.ch12

    • Google Scholar
34. 1. Jenkinson M
    2. Bannister P
    3. Brady M
    4. Smith S

    (2002) Improved optimization for the robust and accurate linear registration and motion correction of brain images

    NeuroImage 17:825–841.

    https://doi.org/10.1016/s1053-8119(02)91132-8

    • PubMed
    • Google Scholar
35. 1. Kaefer K
    2. Stella F
    3. McNaughton BL
    4. Battaglia FP

    (2022) Replay, the default mode network and the cascaded memory systems model

    Nature Reviews. Neuroscience 23:628–640.

    https://doi.org/10.1038/s41583-022-00620-6

    • PubMed
    • Google Scholar
36. 1. Koster-Hale J
    2. Saxe R

    (2013) Theory of mind: A neural prediction problem

    Neuron 79:836–848.

    https://doi.org/10.1016/j.neuron.2013.08.020

    • PubMed
    • Google Scholar
37. 1. Kuhl BA
    2. Bainbridge WA
    3. Chun MM

    (2012) Neural reactivation reveals mechanisms for updating memory

    The Journal of Neuroscience 32:3453–3461.

    https://doi.org/10.1523/JNEUROSCI.5846-11.2012

    • PubMed
    • Google Scholar
38. 1. Lee JLC
    2. Nader K
    3. Schiller D

    (2017) An update on memory reconsolidation updating

    Trends in Cognitive Sciences 21:531–545.

    https://doi.org/10.1016/j.tics.2017.04.006

    • PubMed
    • Google Scholar
39. 1. Lerner Y
    2. Honey CJ
    3. Silbert LJ
    4. Hasson U

    (2011) Topographic mapping of a hierarchy of temporal receptive windows using a narrated story

    The Journal of Neuroscience 31:2906–2915.

    https://doi.org/10.1523/JNEUROSCI.3684-10.2011

    • PubMed
    • Google Scholar
40. 1. Momennejad I
    2. Otto AR
    3. Daw ND
    4. Norman KA

    (2018) Offline replay supports planning in human reinforcement learning

    eLife 7:e32548.

    https://doi.org/10.7554/eLife.32548

    • PubMed
    • Google Scholar
41. 1. Mumford JA
    2. Turner BO
    3. Ashby FG
    4. Poldrack RA

    (2012) Deconvolving BOLD activation in event-related designs for multivoxel pattern classification analyses

    NeuroImage 59:2636–2643.

    https://doi.org/10.1016/j.neuroimage.2011.08.076

    • PubMed
    • Google Scholar
42. 1. Musz E
    2. Chen J

    (2022) Neural signatures associated with temporal compression in the verbal retelling of past events

    Communications Biology 5:489.

    https://doi.org/10.1038/s42003-022-03418-5

    • PubMed
    • Google Scholar
43. 1. Nader K
    2. Einarsson EO

    (2010) Memory reconsolidation: an update

    Annals of the New York Academy of Sciences 1191:27–41.

    https://doi.org/10.1111/j.1749-6632.2010.05443.x

    • PubMed
    • Google Scholar
44. 1. Nastase SA
    2. Gazzola V
    3. Hasson U
    4. Keysers C

    (2019) Measuring shared responses across subjects using intersubject correlation

    Social Cognitive and Affective Neuroscience 14:667–685.

    https://doi.org/10.1093/scan/nsz037

    • PubMed
    • Google Scholar
45. 1. Nastase SA
    2. Goldstein A
    3. Hasson U

    (2020) Keep it real: rethinking the primacy of experimental control in cognitive neuroscience

    NeuroImage 222:117254.

    https://doi.org/10.1016/j.neuroimage.2020.117254

    • PubMed
    • Google Scholar
46. 1. Oedekoven CSH
    2. Keidel JL
    3. Berens SC
    4. Bird CM

    (2017) Reinstatement of memory representations for lifelike events over the course of a week

    Scientific Reports 7:14305.

    https://doi.org/10.1038/s41598-017-13938-4

    • PubMed
    • Google Scholar
47. 1. Phelps EA
    2. Hofmann SG

    (2019) Memory editing from science fiction to clinical practice

    Nature 572:43–50.

    https://doi.org/10.1038/s41586-019-1433-7

    • PubMed
    • Google Scholar
48. 1. Power JD
    2. Mitra A
    3. Laumann TO
    4. Snyder AZ
    5. Schlaggar BL
    6. Petersen SE

    (2014) Methods to detect, characterize, and remove motion artifact in resting state fmri

    NeuroImage 84:320–341.

    https://doi.org/10.1016/j.neuroimage.2013.08.048

    • PubMed
    • Google Scholar
49. 1. Przybyslawski J
    2. Sara SJ

    (1997) Reconsolidation of memory after its reactivation

    Behavioural Brain Research 84:241–246.

    https://doi.org/10.1016/s0166-4328(96)00153-2

    • PubMed
    • Google Scholar
50. 1. Raichle ME
    2. MacLeod AM
    3. Snyder AZ
    4. Powers WJ
    5. Gusnard DA
    6. Shulman GL

    (2001) A default mode of brain function

    PNAS 98:676–682.

    https://doi.org/10.1073/pnas.98.2.676

    • PubMed
    • Google Scholar
51. 1. Ranganath C
    2. Ritchey M

    (2012) Two cortical systems for memory-guided behaviour

    Nature Reviews. Neuroscience 13:713–726.

    https://doi.org/10.1038/nrn3338

    • PubMed
    • Google Scholar
52. 1. Ritchey M
    2. Cooper RA

    (2020) Deconstructing the posterior medial episodic network

    Trends in Cognitive Sciences 24:451–465.

    https://doi.org/10.1016/j.tics.2020.03.006

    • PubMed
    • Google Scholar
53. 1. Ritvo VJH
    2. Turk-Browne NB
    3. Norman KA

    (2019) Nonmonotonic plasticity: how memory retrieval drives learning

    Trends in Cognitive Sciences 23:726–742.

    https://doi.org/10.1016/j.tics.2019.06.007

    • PubMed
    • Google Scholar
54. 1. Robin J
    2. Moscovitch M

    (2017) Details, GIST and schema: hippocampal–neocortical interactions underlying recent and remote episodic and spatial memory

    Current Opinion in Behavioral Sciences 17:114–123.

    https://doi.org/10.1016/j.cobeha.2017.07.016

    • Google Scholar
55. 1. Rugg MD
    2. Vilberg KL

    (2013) Brain networks underlying episodic memory retrieval

    Current Opinion in Neurobiology 23:255–260.

    https://doi.org/10.1016/j.conb.2012.11.005

    • PubMed
    • Google Scholar
56. 1. Samide R
    2. Ritchey M

    (2020) Reframing the past: role of memory processes in emotion regulation

    Cognitive Therapy and Research 45:848–857.

    https://doi.org/10.1007/s10608-020-10166-5

    • Google Scholar
57. 1. Saxe R
    2. Kanwisher N

    (2003) People thinking about thinking people. The role of the temporo-parietal junction in “theory of mind”

    NeuroImage 19:1835–1842.

    https://doi.org/10.1016/s1053-8119(03)00230-1

    • PubMed
    • Google Scholar
58. 1. Schacter DL
    2. Norman KA
    3. Koutstaal W

    (1998) The cognitive neuroscience of constructive memory

    Annual Review of Psychology 49:289–318.

    https://doi.org/10.1146/annurev.psych.49.1.289

    • PubMed
    • Google Scholar
59. 1. Schacter DL
    2. Addis DR

    (2007) The cognitive neuroscience of constructive memory: remembering the past and imagining the future

    Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences 362:773–786.

    https://doi.org/10.1098/rstb.2007.2087

    • PubMed
    • Google Scholar
60. 1. Schacter DL

    (2012) Constructive memory: past and future

    Dialogues in Clinical Neuroscience 14:7–18.

    https://doi.org/10.31887/DCNS.2012.14.1/dschacter

    • PubMed
    • Google Scholar
61. 1. Schaefer A
    2. Kong R
    3. Gordon EM
    4. Laumann TO
    5. Zuo XN
    6. Holmes AJ
    7. Eickhoff SB
    8. Yeo BTT

    (2018) Local-global parcellation of the human cerebral cortex from intrinsic functional connectivity MRI

    Cerebral Cortex 28:3095–3114.

    https://doi.org/10.1093/cercor/bhx179

    • PubMed
    • Google Scholar
62. 1. Schiller D
    2. Kanen JW
    3. LeDoux JE
    4. Monfils MH
    5. Phelps EA

    (2013) Extinction during reconsolidation of threat memory diminishes prefrontal cortex involvement

    PNAS 110:20040–20045.

    https://doi.org/10.1073/pnas.1320322110

    • PubMed
    • Google Scholar
63. Book

    1. Schlichting ML
    2. Preston AR

    (2017) The hippocampus and memory integration: building knowledge to navigate future decisions

    In: Hannula DE, Duff MC, editors. The Hippocampus from Cells to Systems: Structure, Connectivity, and Functional Contributions to Memory and Flexible Cognition. Springer International Publishing. pp. 405–437.

    https://doi.org/10.1007/978-3-319-50406-3_13

    • Google Scholar
64. 1. Scully ID
    2. Napper LE
    3. Hupbach A

    (2017) Does reactivation trigger episodic memory change? A meta-analysis

    Neurobiology of Learning and Memory 142:99–107.

    https://doi.org/10.1016/j.nlm.2016.12.012

    • PubMed
    • Google Scholar
65. 1. Sinclair AH
    2. Barense MD

    (2019) Prediction error and memory reactivation: how incomplete reminders drive reconsolidation

    Trends in Neurosciences 42:727–739.

    https://doi.org/10.1016/j.tins.2019.08.007

    • PubMed
    • Google Scholar
66. 1. Spreng RN
    2. Mar RA
    3. Kim ASN

    (2009) The common neural basis of autobiographical memory, prospection, navigation, theory of mind, and the default mode: a quantitative meta-analysis

    Journal of Cognitive Neuroscience 21:489–510.

    https://doi.org/10.1162/jocn.2008.21029

    • PubMed
    • Google Scholar
67. 1. Stephens GJ
    2. Honey CJ
    3. Hasson U

    (2013) A place for time: the spatiotemporal structure of neural dynamics during natural audition

    Journal of Neurophysiology 110:2019–2026.

    https://doi.org/10.1152/jn.00268.2013

    • PubMed
    • Google Scholar
68. 1. Svoboda E
    2. McKinnon MC
    3. Levine B

    (2006) The functional neuroanatomy of autobiographical memory: a meta-analysis

    Neuropsychologia 44:2189–2208.

    https://doi.org/10.1016/j.neuropsychologia.2006.05.023

    • PubMed
    • Google Scholar
69. 1. Szpunar KK
    2. Watson JM
    3. McDermott KB

    (2007) Neural substrates of envisioning the future

    PNAS 104:642–647.

    https://doi.org/10.1073/pnas.0610082104

    • PubMed
    • Google Scholar
70. 1. Tompary A
    2. Davachi L

    (2017) Consolidation promotes the emergence of representational overlap in the hippocampus and medial prefrontal cortex

    Neuron 96:228–241.

    https://doi.org/10.1016/j.neuron.2017.09.005

    • PubMed
    • Google Scholar
71. 1. Treiber JM
    2. White NS
    3. Steed TC
    4. Bartsch H
    5. Holland D
    6. Farid N
    7. McDonald CR
    8. Carter BS
    9. Dale AM
    10. Chen CC

    (2016) Characterization and correction of geometric distortions in 814 diffusion weighted images

    PLOS ONE 11:e0152472.

    https://doi.org/10.1371/journal.pone.0152472

    • PubMed
    • Google Scholar
72. 1. Tustison NJ
    2. Avants BB
    3. Cook PA
    4. Zheng Y
    5. Egan A
    6. Yushkevich PA
    7. Gee JC

    (2010) N4ITK: improved N3 bias correction

    IEEE Transactions on Medical Imaging 29:1310–1320.

    https://doi.org/10.1109/TMI.2010.2046908

    • PubMed
    • Google Scholar
73. 1. van Kesteren MTR
    2. Rijpkema M
    3. Ruiter DJ
    4. Fernandez G

    (2010) Retrieval of associative information congruent with prior knowledge is related to increased medial prefrontal activity and connectivity

    Journal of Neuroscience 30:15888–15894.

    https://doi.org/10.1523/JNEUROSCI.2674-10.2010

    • PubMed
    • Google Scholar
74. 1. van Kesteren MTR
    2. Ruiter DJ
    3. Fernández G
    4. Henson RN

    (2012) How schema and novelty augment memory formation

    Trends in Neurosciences 35:211–219.

    https://doi.org/10.1016/j.tins.2012.02.001

    • PubMed
    • Google Scholar
75. 1. Wang S
    2. Peterson DJ
    3. Gatenby JC
    4. Li W
    5. Grabowski TJ
    6. Madhyastha TM

    (2017) Evaluation of field map and nonlinear registration methods for correction of susceptibility artifacts in diffusion MRI

    Frontiers in Neuroinformatics 11:17.

    https://doi.org/10.3389/fninf.2017.00017

    • PubMed
    • Google Scholar
76. 1. Yeshurun Y
    2. Swanson S
    3. Simony E
    4. Chen J
    5. Lazaridi C
    6. Honey CJ
    7. Hasson U

    (2017) Same story, different story: the neural representation of interpretive frameworks

    Psychological Science 28:307–319.

    https://doi.org/10.1177/0956797616682029

    • Google Scholar
77. 1. Yeshurun Y
    2. Nguyen M
    3. Hasson U

    (2021) The default mode network: where the idiosyncratic self meets the shared social world

    Nature Reviews. Neuroscience 22:181–192.

    https://doi.org/10.1038/s41583-020-00420-w

    • PubMed
    • Google Scholar
78. 1. Zadbood A
    2. Chen J
    3. Leong YC
    4. Norman KA
    5. Hasson U

    (2017) How we transmit memories to other brains: constructing shared neural representations via communication

    Cerebral Cortex 27:4988–5000.

    https://doi.org/10.1093/cercor/bhx202

    • PubMed
    • Google Scholar

Author details

1. Asieh Zadbood

   Department of Psychology, Columbia University, New York, United States

   Contribution

   Conceptualization, Data curation, Software, Formal analysis, Investigation, Visualization, Methodology, Writing - original draft, Project administration, Writing - review and editing

   For correspondence

   az2604@columbia.edu

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-9098-0510

2. Samuel Nastase

   Princeton Neuroscience Institute and Department of Psychology, Princeton University, Princeton, United States

   Contribution

   Software, Formal analysis, Investigation, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0001-7013-5275

3. Janice Chen

   Department of Psychological and Brain Sciences, Johns Hopkins University, Baltimore, United States

   Contribution

   Conceptualization, Supervision, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

4. Kenneth A Norman

   Princeton Neuroscience Institute and Department of Psychology, Princeton University, Princeton, United States

   Contribution

   Conceptualization, Supervision, Funding acquisition, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-5887-9682

5. Uri Hasson

   Princeton Neuroscience Institute and Department of Psychology, Princeton University, Princeton, United States

   Contribution

   Conceptualization, Supervision, Funding acquisition, Methodology, Writing - review and editing

   Competing interests

   No competing interests declared

• 2,769

  views

• 414

  downloads

• 6

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.