David Halpern
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
David Halpern has a total of 57 co-authors. Here are the top co-authors by number of shared publications:
- Michael J. Kahana
University of Pennsylvania
10 shared publications Shannon Tubridy
Department of Psychology, New York University, New York, NY 10003
4 shared publicationsTodd Matthew Gureckis
4 shared publications- Bradley C. Lega
Department of Neurological Surgery, University of Texas Southwestern Medical Center, Dallas, TX, USA
3 shared publications - Chengyuan Wu
Department of Neurological Surgery, Thomas Jefferson University Hospital, Philadelphia, Pennsylvania, USA
3 shared publications Lila Davachi
Department of Psychology, Columbia University, New York, NY 10027
3 shared publications
Latest Publications
- Open Access
- 17 November 2026
- Published by Center for Open Science in PsyArXiv
Memorable Decision Cues are More Associable With Value
When making decisions, we must use the sensory cues available at the moment of choice to predict the values of our options. Yet little is known about how properties of these cues affect the choice process, particularly our ability to retrieve decision-relevant memories. Here, we test whether and how image memorability, the population likelihood of recognizing a particular image, influences memory-guided decisions. Across five pre-registered experiments, memorable cue images were not chosen more overall; instead, they predicted more accurate retrieval of associated values and more value-sensitive choices. We account for this pattern of behavior with a formal exemplar model in which distinctive cues suffer less interference from competing memories during retrieval of associated values. The model predicts, and our data confirm, two interactions: choices are more sensitive to value differences between memorable options, and a difference in memorability biases choices toward the more memorable option when both are high in value, but away from it when both are low. Finally, we find that the same pattern held for choices between common foods, suggesting that shared associative memory processes are recruited for choices involving both novel and highly familiar options. Together, our results suggest that more memorable cues are more associable with value, and provide a process-level account for how memorability influences downstream decisions.
- 17 November 2026
- Published by Center for Open Science in PsyArXiv
Open AccessMemorable Decision Cues are More Associable With Value
When making decisions, we must use the sensory cues available at the moment of choice to predict the values of our options. Yet little is known about how properties of these cues affect the choice process, particularly our ability to retrieve decision-relevant memories. Here, we test whether and how image memorability, the population likelihood of recognizing a particular image, influences memory-guided decisions. Across five pre-registered experiments, memorable cue images were not chosen more overall; instead, they predicted more accurate retrieval of associated values and more value-sensitive choices. We account for this pattern of behavior with a formal exemplar model in which distinctive cues suffer less interference from competing memories during retrieval of associated values. The model predicts, and our data confirm, two interactions: choices are more sensitive to value differences between memorable options, and a difference in memorability biases choices toward the more memorable option when both are high in value, but away from it when both are low. Finally, we find that the same pattern held for choices between common foods, suggesting that shared associative memory processes are recruited for choices involving both novel and highly familiar options. Together, our results suggest that more memorable cues are more associable with value, and provide a process-level account for how memorability influences downstream decisions.
- Open Access
- 23 September 2026
- Published by MIT Press in Imaging Neuroscience
- Vol. 4
Getting blood from a stone: Improving brain–behavior inferences without brain data
In recent years, the cognitive neuroscience literature has come under criticism for containing many low-powered studies, limiting the ability to make reliable statistical inferences. Typically, the suggestion for increasing power is to collect more data with neural signals. However, many studies in cognitive neuroscience use parameters estimated from behavioral data in order to make inferences about neural signals (such as fMRI BOLD signal). In this paper, we explore how cognitive neuroscientists can learn more about their neuroimaging signal by collecting data on behavior alone and using alternative estimators designed to leverage this information. We demonstrate through simulation and mathematical derivations that knowing more about the marginal distribution of behavior can improve inferences about the mapping between cognitive processes and neural data. We analyze the magnitude of this benefit, finding that it depends on the desired estimand and several underlying study parameters. While in many cases the absolute gains in precision can be modest, our results demonstrate that, in realistic settings, additional behavioral data can lead to the same improvement in the precision of inferences more cheaply and easily than collecting additional data from subjects in a neuroimaging study. This means that when conducting a neuroimaging study, researchers now have another knob to turn in a design analysis: the number of subjects collected in the scanner and the number of behavioral subjects collected outside the scanner (in the laboratory or online).
- 23 September 2026
- Published by MIT Press in Imaging Neuroscience
- Vol. 4
Open AccessGetting blood from a stone: Improving brain–behavior inferences without brain data
In recent years, the cognitive neuroscience literature has come under criticism for containing many low-powered studies, limiting the ability to make reliable statistical inferences. Typically, the suggestion for increasing power is to collect more data with neural signals. However, many studies in cognitive neuroscience use parameters estimated from behavioral data in order to make inferences about neural signals (such as fMRI BOLD signal). In this paper, we explore how cognitive neuroscientists can learn more about their neuroimaging signal by collecting data on behavior alone and using alternative estimators designed to leverage this information. We demonstrate through simulation and mathematical derivations that knowing more about the marginal distribution of behavior can improve inferences about the mapping between cognitive processes and neural data. We analyze the magnitude of this benefit, finding that it depends on the desired estimand and several underlying study parameters. While in many cases the absolute gains in precision can be modest, our results demonstrate that, in realistic settings, additional behavioral data can lead to the same improvement in the precision of inferences more cheaply and easily than collecting additional data from subjects in a neuroimaging study. This means that when conducting a neuroimaging study, researchers now have another knob to turn in a design analysis: the number of subjects collected in the scanner and the number of behavioral subjects collected outside the scanner (in the laboratory or online).
- Open Access
- 20 May 2026
- Published by Center for Open Science in PsyArXiv
A theory of memory for items and associations
We present a retrieved-context theory of memory for items, associations, and their interaction (CMR-IA). Our theory assumes an evolving representation of temporal context that binds to items and associations, allowing the rememberer to make judgments based on the occurrence of a mnemonic target within a particular context. In addition to the assumptions inherited from prior retrieved-context theories, CMR-IA assumes a conjunctive (Gestalt) representation for paired associates, increased attention to rare items, and variable thresholds for recognition decisions. We apply CMR-IA to key findings concerning recognition of items and associations, including effects of recency, similarity, receiver-operating characteristic curves, word frequency, differential forgettings of items and associations, and contiguity effects for successive probes. We also apply CMR-IA to cued recall phenomena, including serial position effects, distribution of correct responses and errors, contiguity effects, associative symmetry, and similarity effects. Finally, we ask whether CMR-IA can account for the dependencies between successive tests of item and associative memory. We show that combining a Gestalt associative mechanism with retrieved-context theory provides a good account for many empirical phenomena concerning item and associative memory. The analysis of successive memory tests highlights the important role of output encoding in our model.
- 20 May 2026
- Published by Center for Open Science in PsyArXiv
Open AccessA theory of memory for items and associations
We present a retrieved-context theory of memory for items, associations, and their interaction (CMR-IA). Our theory assumes an evolving representation of temporal context that binds to items and associations, allowing the rememberer to make judgments based on the occurrence of a mnemonic target within a particular context. In addition to the assumptions inherited from prior retrieved-context theories, CMR-IA assumes a conjunctive (Gestalt) representation for paired associates, increased attention to rare items, and variable thresholds for recognition decisions. We apply CMR-IA to key findings concerning recognition of items and associations, including effects of recency, similarity, receiver-operating characteristic curves, word frequency, differential forgettings of items and associations, and contiguity effects for successive probes. We also apply CMR-IA to cued recall phenomena, including serial position effects, distribution of correct responses and errors, contiguity effects, associative symmetry, and similarity effects. Finally, we ask whether CMR-IA can account for the dependencies between successive tests of item and associative memory. We show that combining a Gestalt associative mechanism with retrieved-context theory provides a good account for many empirical phenomena concerning item and associative memory. The analysis of successive memory tests highlights the important role of output encoding in our model.
- Open Access
- 19 April 2026
- Published by Center for Open Science in PsyArXiv
Organizational Dynamics of Memory Across Days
When individuals repeatedly study and recall information across multiple learning trials their responses exhibit increasing levels of subjective organization. Whereas classic studies investigated the evolution of organization across lists within the short-time span of a single session, here we ask how memory changes over many days. Specifically, we examine how semantic, temporal, and subjective organization during a recall period shapes memory after days of intervening cognitive activity. Analyzing data from two multi-session free recall experiments, we find that subjects demonstrate a strong tendency to cluster recalls based on previous output order, with this effect strengthening across sessions. In line with the idea that thoughts become memories, we show that even false memories produced on a given session tend to re-occur on subsequent days. Our results attest to the crucial role that retrieval plays in shaping long-term episodic memory.
- 19 April 2026
- Published by Center for Open Science in PsyArXiv
Open AccessOrganizational Dynamics of Memory Across Days
When individuals repeatedly study and recall information across multiple learning trials their responses exhibit increasing levels of subjective organization. Whereas classic studies investigated the evolution of organization across lists within the short-time span of a single session, here we ask how memory changes over many days. Specifically, we examine how semantic, temporal, and subjective organization during a recall period shapes memory after days of intervening cognitive activity. Analyzing data from two multi-session free recall experiments, we find that subjects demonstrate a strong tendency to cluster recalls based on previous output order, with this effect strengthening across sessions. In line with the idea that thoughts become memories, we show that even false memories produced on a given session tend to re-occur on subsequent days. Our results attest to the crucial role that retrieval plays in shaping long-term episodic memory.
- 11 March 2025
- Published by Springer Nature in Nature Neuroscience
- Vol. 28 (4) , 883-890
Study-phase reinstatement predicts subsequent recall
Can the brain improve the retrievability of an experience after it has occurred? Systems consolidation theory proposes that item-specific cortical reactivation during post-encoding rest periods facilitates the formation of stable memory representations, a prediction supported by neural evidence in humans and animals. Such reactivation may also occur on shorter timescales, offering a potential account of classic list memory phenomena but lacking in support from neural data. Leveraging the high temporal specificity of intracranial electroencephalography (EEG), we investigate spontaneous reactivation of previously experienced items during brief intervals between individual encoding events. Across two large-scale free-recall experiments, we show that reactivation during these periods, measured by spectral intracranial EEG similarity, predicts subsequent recall. In a third experiment, we show that the same methodology can identify post-encoding reactivation that correlates with subsequent memory, consistent with previous results. Thus, spontaneous study-phase reinstatement reliably predicts memory behavior, linking psychological accounts to neural mechanisms and providing evidence for rapid consolidation processes during encoding.
- 11 March 2025
- Published by Springer Nature in Nature Neuroscience
- Vol. 28 (4) , 883-890
Study-phase reinstatement predicts subsequent recall
Can the brain improve the retrievability of an experience after it has occurred? Systems consolidation theory proposes that item-specific cortical reactivation during post-encoding rest periods facilitates the formation of stable memory representations, a prediction supported by neural evidence in humans and animals. Such reactivation may also occur on shorter timescales, offering a potential account of classic list memory phenomena but lacking in support from neural data. Leveraging the high temporal specificity of intracranial electroencephalography (EEG), we investigate spontaneous reactivation of previously experienced items during brief intervals between individual encoding events. Across two large-scale free-recall experiments, we show that reactivation during these periods, measured by spectral intracranial EEG similarity, predicts subsequent recall. In a third experiment, we show that the same methodology can identify post-encoding reactivation that correlates with subsequent memory, consistent with previous results. Thus, spontaneous study-phase reinstatement reliably predicts memory behavior, linking psychological accounts to neural mechanisms and providing evidence for rapid consolidation processes during encoding.
- Open Access
- 13 November 2023
- Published by Springer Nature in Research Square
Study-Phase Reinstatement: Encoding Spontaneous Thoughts as Memories
Can the brain improve the retrievability of an experience after it has occurred? Systems consolidation theory proposes that cortical reactivation during extended post-encoding rest periods facilitates the formation of stable memory representations, a prediction supported by neural evidence in humans and animals. Such reactivation may also occur on short time scales as spontaneous thoughts come to mind during encoding, offering a potential account of classic list memory phenomena but lacking in support from neural data. Leveraging the high-temporal specificity of intracranial electroencephalography (iEEG), we investigate spontaneous reactivation of previously experienced items during brief sub-second intervals between individual encoding events. Across two large-scale free recall experiments, we show that reactivation, as measured by spectral iEEG similarity, during these periods predicts subsequent recall. In a third experiment, we show that the same methodology can identify post-encoding reactivation that correlates with subsequent memory, consistent with previous results. Thus, spontaneous study-phase reinstatement reliably predicts memory behavior, linking psychological accounts to neural mechanisms and providing the first such evidence for rapid consolidation processes during encoding.
- 13 November 2023
- Published by Springer Nature in Research Square
Open AccessStudy-Phase Reinstatement: Encoding Spontaneous Thoughts as Memories
Can the brain improve the retrievability of an experience after it has occurred? Systems consolidation theory proposes that cortical reactivation during extended post-encoding rest periods facilitates the formation of stable memory representations, a prediction supported by neural evidence in humans and animals. Such reactivation may also occur on short time scales as spontaneous thoughts come to mind during encoding, offering a potential account of classic list memory phenomena but lacking in support from neural data. Leveraging the high-temporal specificity of intracranial electroencephalography (iEEG), we investigate spontaneous reactivation of previously experienced items during brief sub-second intervals between individual encoding events. Across two large-scale free recall experiments, we show that reactivation, as measured by spectral iEEG similarity, during these periods predicts subsequent recall. In a third experiment, we show that the same methodology can identify post-encoding reactivation that correlates with subsequent memory, consistent with previous results. Thus, spontaneous study-phase reinstatement reliably predicts memory behavior, linking psychological accounts to neural mechanisms and providing the first such evidence for rapid consolidation processes during encoding.
- Open Access
Study-Phase Reinstatement
Can the brain improve the retrievability of an experience after it has occurred? Systems consolidation theory proposes that item-specific cortical reactivation during post-encoding rest periods facilitates the formation of stable memory representations, a prediction supported by neural evidence in humans and animals. Such reactivation may also occur on shorter time scales, offering a potential account of classic list memory phenomena but lacking in support from neural data. Leveraging the high-temporal specificity of intracranial electroencephalography (iEEG), we investigate spontaneous reactivation of previously experienced items during brief intervals between individual encoding events. Across two large-scale free recall experiments, we show that reactivation during these periods, measured by spectral iEEG similarity, predicts subsequent recall. In a third experiment, we show that the same methodology can identify post-encoding reactivation that correlates with subsequent memory, consistent with previous results. Thus, spontaneous study-phase reinstatement reliably predicts memory behavior, linking psychological accounts to neural mechanisms and providing evidence for rapid consolidation processes during encoding.
Open AccessStudy-Phase Reinstatement
Can the brain improve the retrievability of an experience after it has occurred? Systems consolidation theory proposes that item-specific cortical reactivation during post-encoding rest periods facilitates the formation of stable memory representations, a prediction supported by neural evidence in humans and animals. Such reactivation may also occur on shorter time scales, offering a potential account of classic list memory phenomena but lacking in support from neural data. Leveraging the high-temporal specificity of intracranial electroencephalography (iEEG), we investigate spontaneous reactivation of previously experienced items during brief intervals between individual encoding events. Across two large-scale free recall experiments, we show that reactivation during these periods, measured by spectral iEEG similarity, predicts subsequent recall. In a third experiment, we show that the same methodology can identify post-encoding reactivation that correlates with subsequent memory, consistent with previous results. Thus, spontaneous study-phase reinstatement reliably predicts memory behavior, linking psychological accounts to neural mechanisms and providing evidence for rapid consolidation processes during encoding.
- Open Access
Decoding EEG for optimizing naturalistic memory
Background: Spectral features of human electroencephalographic (EEG) recordings during learning predict subsequent recall variability. New method: Capitalizing on these fluctuating neural features, we develop a non-invasive closed-loop (NICL) system for real-time optimization of human learning. Participants play a virtual navigation and memory game; recording multi-session data across days allowed us to build participant-specific classification models of recall success. In subsequent closed-loop sessions, our platform manipulated the timing of memory encoding, selectively presenting items during periods of predicted good or poor memory function based on EEG features decoded in real time. Results: We observed greater memory modulation (difference between recall rates when presenting items during predicted good vs. poor learning periods) for participants with higher out-of-sample classification accuracy. Comparison with Existing Methods: This study demonstrates greater-than-chance memory decoding from EEG recordings in a naturalistic virtual navigation task with greater real-world validity than basic word-list recall paradigms. Here we modulate memory by timing stimulus presentation based on noninvasive scalp EEG recordings, whereas prior closed-loop studies for memory improvement involved intracranial recordings and direct electrical stimulation. Other noninvasive studies have investigated the use of neurofeedback or remedial study for memory improvement. Conclusions: These findings present a proof-of-concept for using non-invasive closed-loop technology to optimize human learning and memory through principled stimulus timing, but only in those participants for whom classifiers reliably predict out-of-sample memory function.
Open AccessDecoding EEG for optimizing naturalistic memory
Background: Spectral features of human electroencephalographic (EEG) recordings during learning predict subsequent recall variability. New method: Capitalizing on these fluctuating neural features, we develop a non-invasive closed-loop (NICL) system for real-time optimization of human learning. Participants play a virtual navigation and memory game; recording multi-session data across days allowed us to build participant-specific classification models of recall success. In subsequent closed-loop sessions, our platform manipulated the timing of memory encoding, selectively presenting items during periods of predicted good or poor memory function based on EEG features decoded in real time. Results: We observed greater memory modulation (difference between recall rates when presenting items during predicted good vs. poor learning periods) for participants with higher out-of-sample classification accuracy. Comparison with Existing Methods: This study demonstrates greater-than-chance memory decoding from EEG recordings in a naturalistic virtual navigation task with greater real-world validity than basic word-list recall paradigms. Here we modulate memory by timing stimulus presentation based on noninvasive scalp EEG recordings, whereas prior closed-loop studies for memory improvement involved intracranial recordings and direct electrical stimulation. Other noninvasive studies have investigated the use of neurofeedback or remedial study for memory improvement. Conclusions: These findings present a proof-of-concept for using non-invasive closed-loop technology to optimize human learning and memory through principled stimulus timing, but only in those participants for whom classifiers reliably predict out-of-sample memory function.
- Open Access
- 23 March 2023
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 120 (13)
Identifying causal subsequent memory effects
Over 40 y of accumulated research has detailed associations between neuroimaging signals measured during a memory encoding task and later memory performance, across a variety of brain regions, measurement tools, statistical approaches, and behavioral tasks. But the interpretation of these subsequent memory effects (SMEs) remains unclear: if the identified signals reflect cognitive and neural mechanisms of memory encoding, then the underlying neural activity must be causally related to future memory. However, almost all previous SME analyses do not control for potential confounders of this causal interpretation, such as serial position and item effects. We collect a large fMRI dataset and use an experimental design and analysis approach that allows us to statistically adjust for nearly all known exogenous confounding variables. We find that, using standard approaches without adjustment, we replicate several univariate and multivariate subsequent memory effects and are able to predict memory performance across people. However, we are unable to identify any signal that reliably predicts subsequent memory after adjusting for confounding variables, bringing into doubt the causal status of these effects. We apply the same approach to subjects’ judgments of learning collected following an encoding period and show that these behavioral measures of mnemonic status do predict memory after adjustments, suggesting that it is possible to measure signals near the time of encoding that reflect causal mechanisms but that existing neuroimaging measures, at least in our data, may not have the precision and specificity to do so.
- 23 March 2023
- Published by National Academy of Sciences in Proceedings of the National Academy of Sciences
- Vol. 120 (13)
Open AccessIdentifying causal subsequent memory effects
Over 40 y of accumulated research has detailed associations between neuroimaging signals measured during a memory encoding task and later memory performance, across a variety of brain regions, measurement tools, statistical approaches, and behavioral tasks. But the interpretation of these subsequent memory effects (SMEs) remains unclear: if the identified signals reflect cognitive and neural mechanisms of memory encoding, then the underlying neural activity must be causally related to future memory. However, almost all previous SME analyses do not control for potential confounders of this causal interpretation, such as serial position and item effects. We collect a large fMRI dataset and use an experimental design and analysis approach that allows us to statistically adjust for nearly all known exogenous confounding variables. We find that, using standard approaches without adjustment, we replicate several univariate and multivariate subsequent memory effects and are able to predict memory performance across people. However, we are unable to identify any signal that reliably predicts subsequent memory after adjusting for confounding variables, bringing into doubt the causal status of these effects. We apply the same approach to subjects’ judgments of learning collected following an encoding period and show that these behavioral measures of mnemonic status do predict memory after adjustments, suggesting that it is possible to measure signals near the time of encoding that reflect causal mechanisms but that existing neuroimaging measures, at least in our data, may not have the precision and specificity to do so.
- Open Access
Neural correlates of memory in a naturalistic spatiotemporal context
We investigated memory encoding and retrieval during a quasi-naturalistic spatial-episodic memory task in which subjects delivered items to landmarks in a desktop virtual environment and later recalled the delivered items. Transition probabilities and latencies revealed the spatial and temporal organization of memory. As subjects gained experience with the town, their improved spatial knowledge led to more efficient navigation and increased spatial organization during recall. Subjects who exhibited stronger spatial organization exhibited weaker temporal organization. Scalp-recorded electroencephalographic (EEG) signals revealed spectral correlates of successful encoding and retrieval. Increased theta power ( T + ) and decreased alpha/beta power ( A − ) accompanied successful encoding, with the addition of increased gamma ( G + ) accompanying successful retrieval. Logistic-regression classifiers trained on spectral features reliably predicted mnemonic success in held-out sessions. Univariate and multivariate EEG analyses revealed a similar spectral T + A − G + of successful memory. These findings extend behavioral and neural signatures of successful encoding and retrieval to a naturalistic task in which learning occurs within a spatiotemporal context.
Open AccessNeural correlates of memory in a naturalistic spatiotemporal context
We investigated memory encoding and retrieval during a quasi-naturalistic spatial-episodic memory task in which subjects delivered items to landmarks in a desktop virtual environment and later recalled the delivered items. Transition probabilities and latencies revealed the spatial and temporal organization of memory. As subjects gained experience with the town, their improved spatial knowledge led to more efficient navigation and increased spatial organization during recall. Subjects who exhibited stronger spatial organization exhibited weaker temporal organization. Scalp-recorded electroencephalographic (EEG) signals revealed spectral correlates of successful encoding and retrieval. Increased theta power ( T + ) and decreased alpha/beta power ( A − ) accompanied successful encoding, with the addition of increased gamma ( G + ) accompanying successful retrieval. Logistic-regression classifiers trained on spectral features reliably predicted mnemonic success in held-out sessions. Univariate and multivariate EEG analyses revealed a similar spectral T + A − G + of successful memory. These findings extend behavioral and neural signatures of successful encoding and retrieval to a naturalistic task in which learning occurs within a spatiotemporal context.