William Hirst - US grants

Memory problems are a sequella of a number of diseases, including herpes encephalitis, stroke, dementia, ischemia and aneurysms. These problems will never be fully understood until organization of the memory system is better understood, both in neurological and psychological terms. The study of amnesia can bear on both the neurological and psychological descriptions of memory. But the neuroscientist or psychologist can only build memory models from this source of evidence when the amnesic syndrome has been adequately characterized. The proposed research is designed to consolidate a recent finding that amnesic recognition is relatively preserved when compared to amnesic recall and to determine whether the amnesic deficit can be characterized not as a failure to encode item information, but as a failure to encode the larger context in which the item is placed, including inter item associations. Experiments designed to test this latter hypothesis, which will be called the Coherence Hypothesis, will investigate the representations of amnesics' direct memories. The Coherence Hypothesis will also be tested in a recognition and recall, SAM. If the results are as expected, the amnesic deficit may not involve the encoding of item information, but the encoding of context and inter item associations. This dichotomy differs from the present, widely accepted dichotomy between direct and indirect memory and may offer an alternative that not only accounts for amnesics' relatively preserved recognition, but also their intact priming and possibly their preserved skill learning.

ABSTRACT
William Hirts/Alexandru Cuc - 0236764
Why do families, colleagues, and even nations come to remember the past in the same way? How do initially discrepant memories come to converge on a single rendering of the past? Will everyday conversations change people's memories so that a collective memory emerges? And why do some memories spread across a group, while others remain localized or disappear? With NSF support, under the supervision of Dr. William Hirst, Alex Cuc addresses these critical issues for the study of collective memory, focusing on the formation of collective memories in families.
To a large extent, researchers have failed to examine the effect of social structure on memory, concentrating on universal mental mechanisms that serve as the foundation for mnemonic processing. Mr. Cuc innovatively explores the interaction between these well-understood mental mechanisms and the less understood dynamics of conversation as he articulates the conditions under which collective memories are formed through conversation. The intellectual merits of the project stem primarily from the fresh insights Mr. Cuc offers about this interaction and the rigorous methodology he develops to chart this interaction.
Broader impacts of the project include a better understanding of the formation of collective memories, not just of families, but of any group, including nations. This understanding serves as a foundation for exploring in the future the way collective memories support group identity, be it family identity or national identity.

Memory does not merely record what occurred at a particular time in the past, but also reflects what intervenes between past and present. As a result, the frequent conversations people have with each other about a past event can radically influence the way they subsequently remember. Conversations cannot only impose new memories onto listeners, they can also reshape memories by, paradoxically, promoting forgetting, as the investigator's research has recently established. This socially induced forgetting is quite specific: The inevitable selective recounting emerging in a conversation will strengthen some memories, but will also weaken or induce forgetting, in both speaker and listener, for unmentioned memories related to what was mentioned, without having a similar effect on unmentioned, unrelated memories. The proposed research explores factors that might influence this socially induced forgetting, specifically, the position of power and expertise of the speaker, the trust the listener has in the speaker, and the goals of the conversation. By examining these factors, the investigator will specify critical conditions under which socially induced forgetting may occur and thereby further our understanding of how a memory becomes, to a substantial degree, a product of the conversations that mediate the initial experience and a subsequent act of remembrance.

The expected findings will have a broad impact because they will underscore the danger of depending on the accuracy of memory and will articulate how something as commonplace as a conversation can reshape memory, not just through errors of commission, but also through errors of omission. Specifically, the expected findings will (1) add to growing reservations of the legal community about the reliability of eyewitness testimony, (2) identify how the conversations community members have about their community's past can shape its collective memory, and in turn, collective identity and collective actions, and (3) caution educators to limit their enthusiasm for distributed or collaborative learning. In each instance, the research will emphasize that the seemingly positive act of collaborative remembering will carry with it not just benefits, but also costs. In doing so, the proposed research will unravel the complex ways in which both individual and collective memories are held captive by social interactions, and help both institutions and individuals calibrate their assumptions about memory accuracy to better deal with learning and the effect that the past has on the present.

Juries are the bedrock of the American judicial system. Six or twelve people join together in a collaborative effort to arrive at a verdict. During deliberation, jurors must not only come to a verdict based on the evidence presented during the trial, but they must also remember the evidence. To a large extent, jurors must rely on their own memory. Here, the assumption of the Courts is that the collective memory of a jury is likely to lead to a better, more complete recollection than any individual effort. The difficulty with this position is that there is burgeoning psychological evidence indicating that groups often remember less than the sum of what individuals are capable of remembering. Moreover, extant research demonstrates that what one person says in a conversation can alter the memory of other participants - by imposing misleading information or inducing forgetting, for example. Although both types of conversational effects can affect subsequent decision-making, it is possible that, even more critically, both types can be mitigated in certain circumstances. To this end, the present research explores the way jury deliberation might reshape the memories jurors have of the evidence presented during the trial, the effect any change in memory might have on the verdict of the jury, and the means of mitigating these deliberation effects.

The research involves seven experiments employing a mock jury methodology. They will examine whether (1) research concerning collaborative remembering applies to jury deliberations; (2) the emergent consensus story of a jury reflects the conversational dynamics consistent with what is known about conversational remembering; (3) the verdict of a jury can be traced to these conversational influences on memory; (4) the conversational influences on memory are a function of the composition of the jury; and (5) ways exist to mitigate these potentially negative conversational effects through instruction or use of mnemonic technology.

The present research explores how collective memories are formed, maintained, changed, and forgotten. How do collective memories form? To a large extent, the formation of collective memories occurs through acts of communication. One person might talk to another person about a past event, or one person might watch a video about an event on the Internet. Eventually, some of these memories are propagated throughout the group. The present research examines how acts of person-to-person communication can establish, strengthen, change and weaken what people remember of their own and others' collective past. It tests the mechanisms by which this interpersonal influence can spread across a network, affect the memories of those with whom there is no direct connection, and in so doing, have a marked effect on the collective memory formed across the network.

Three sets of experiments are proposed. Using a modified version of the serial reproduction task, the first set examines the extent to which the three conversational influences of interest - reinforcement, induced forgetting, and the implantation of memories - spread across transmission chains of various sizes. It tests whether the extent of this propagation is a function of the strength of these conversational influences and whether these strengths, in turn, are a function of social relationships and the medium of communication. The second phase explores mnemonic convergence, that is, the extent to which a chain of individuals converges onto the same shared memories. Specifically, it investigates whether the extent of propagation can not exceed the length of the transmission chain if convergence is to occur. The third phase moves beyond transmission chains to complex social networks. It now probes for the degree to which mnemonic convergence is a function of both the extent of propagation, or in turn, the strength of conversational influences, and the average length between all pairings of individuals in the network. It also introduces cluster coefficients as an additional variable of interest. The proposed research brings together for the first time two strands of relevant research: the burgeoning experimentally-based psychological literature on the effect of communication on memory in dyads and the widely discussed work on the propagation of influence through social networks. This synergistic approach produces a model with great general applicability, particularly in the domain of national security. Its interdisciplinarity will link psychological research on individual memory with work on networks and social connectedness. As a result, graduate students working on the proposed research will receive a unique training. Overall, the proposed research provides an empirical foundation for a model of collective memory formation that will impact on societally important issues.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.