Valerie Reyna, PhD - US grants

Children's spontaneous false memories have been the subject of considerable heat and little light recently. This research is testing several hypotheses about false memories for doctor visits that are accompanied by trauma to answer questions that bear on the accuracy of children's witness testimony. Prior research by these investigators has produced a theoretical framework (fuzzy-trace theory) that makes counterintuitive predictions about such important issues as: age variability in spontaneous memory falsifications, age variability in the effects of delay on memory falsification, the persistence of false memories over time, the tendency of mere memory testes to falsify the contents of memory, and the features of experienced events that are most prone to memory falsification. Three experiments will use a false-recognition procedure and will consist of four sessions. Participants will be 180 children between the ages of 23 and 8. Children's experiences to urgent-care clinics will be the focus of subsequent memory interviews to occur 1-3 days after the child sees the physician. Subsequent interviews occur at 6 month and 1 year intervals. Memory interviews will query events that the child did and those that her or she did not experience. False-recognition effect will be used to index spontaneous memory falsification. Differences in spontaneous memory falsification are being tracked as a function of several variables that have forensic relevance: frequency of experience with events, amount of prior interviewing, whether inferences have been made about events, whether inconsistent version of evens have been experienced. These data will resolve tow uncertainties in current scientific knowledge about children's false memories. These are that mere memory tests can create false memories that are stable over time and whether spontaneous false memory increase with age or delay between exposure to events and memory tests. %%% Children's spontaneous false memories have been the subject of considerable heat and little light recently. This research is testing several hypotheses about false memories for doctor visits that are accompanied by trauma to answer questions that bear on the accuracy of children's witness testimony. Prior research by these investigators has produced a theoretical framework (fuzzy-trace theory) that makes counterintuitive predictions about such important issues as: age variability in spontaneous memory falsifications, age variability in the effects of delay on memory falsification, the persistence of false memories over time, the tendency of mere memory testes to falsify the contents of memory, and the features of experienced events that are most prone to memory falsification. Three experiments will use a false-recognition procedure and will consist of four sessions. Participants will be 180 children between the ages of 23 and 8. Children's experiences to urgent-care clinics will be the focus of subsequent memory interviews to occur 1-3 days after the child sees the physician. Subsequent interviews occur at 6 month and 1 year intervals. Memory interviews will query events that the child did and those that her or she did not experience. False-recognition effect will be used to index spontaneous memory falsification. Differences in spontaneous memory falsification are being tracked as a function of several variables that have forensic relevance: frequency of experience with events, amount of prior interviewing, whether inferences have been made about events, whether inconsistent version of evens have been experienced. These data will resolve tow uncertainties in current scientific knowledge about children's false memories. These are that mere memory tests can create false memories that are stable over time and whether spontaneous false memory increase with age or delay between exposure to events and memory tests.

Three million new cases of sexually transmitted diseases (STD) are identified in adolescents every year. The proposed research applies recent advances in cognitive-behavioral theory to increase the effectiveness and durability of multicomponent interventions that promote risk reduction and avoidance in adolescents. In Phase I, we investigate adolescents' psychological representations of decision options (i.e. abstinence vs. risk taking), risk estimates of problem behaviors, and relevant beliefs and attitudes. Specifically, 200 adolescents will be surveyed, presented with realistic sexual decision-making scenarios, and questioned in-depth concerning their perceptions of important risks and consequences (e.g., the cumulative risk of contracting a STD). Hispanics will be oversampled to provide sufficient statistical power to evaluate ethnic differences. In Phase II, 800 adolescents will be randomly assigned to one of three groups: a standard multicomponent intervention (Reducing the Risk), a gist-based enhancement of the standard intervention, or a control group receiving an unrelated intervention. The enhanced intervention will be based on cognitive behavioral theory (e.g., Reyna and Ellis, 1994), as well as on differences identified in Phase 1 between adolescents engaging versus not engaging in risky behaviors. Pre- and post-intervention surveys will assess such outcomes as self-reports of risky behaviors and STD treatment, with follow-up surveys at 3, 6, and 12 months. Multivariate analyses of variance (followed by univariate tests) controlling for background and baseline characteristics, will be used to compare differences among standard, enhanced, and control groups. Logistic regression will be used for categorical outcome measures, such as initiation of intercourse. Thus, the proposed research will provide evidence concerning sexual decision processes in Hispanic and non-Hispanic adolescents, and evaluate a theory-based intervention that focuses on specific behaviors and addresses both social (e.g., refusal skills) and cognitive- developmental factors using a randomized assignment design. We will also compare our survey data to results of a statewide evaluation of abstinence-only programs using the same survey instrument. By emphasizing gist representations, which are well preserved over long time periods and are key memories used in decision making, the enhanced intervention should produce larger and more sustained effects on risk reduction and avoidance.

This project continues the researchers' program of research on children's false memories. That program of research is theory-driven, and to date, it has produced a number of new findings about false memories that were originally predicted on theoretical grounds. This project consists of three large-scale developmental experiments. The experiments focus on two cognitive operations that are believed to be key controlling factors in children's false-memory reports: recollection rejection and phantom recollection. The existence of both operations is predicted by the researchers' theoretical account of false memory (i.e., fuzzy-trace theory), and evidence of the use of both operations by adults has been reported in previous experiments. Recollection rejection is an operation that suppresses the reporting of false events that are consistent with the gist of children's experience. Suppression is accomplished via the detection of mismatches between false-but-gist-consistent events and verbatim traces of actual events. For instance, after hearing a narrative containing the statement "The bird is inside the cage" and "The cage is under the table," children could reject the unpresented statement "The bird is under the table" by retrieving verbatim traces of either presented sentence and noticing that the words "bird" and "table" do not appear together in the sentence. Phantom recollection, on the other hand, is an operation that supports the reporting of false events that are consistent with the gist of children's experience by generating illusory vivid phenomenologies for those events. Fuzzy-trace theory posits that gist memories sometimes cause false events to be accompanied by illusory vivid phenomenologies that are difficult to distinguish from the vivid phenomenologies that accompany true events. These phantom recollections cause things that were not experienced to be remembered as physical "occurrences" (e.g, they are seen in the mind's eye or heard in the mind's ear). Fuzzy-trace theory predicts that phantom recollection can occur when two conditions are met: (a) Experienced events repeatedly cue some familiar meaning, so that gist memories of that meaning are very strong, and (b) the false events that provoke phantom recollection are especially good retrieval cues for strong gist memories.

The experiments implement a new paradigm, conjoint recognition. This paradigm provides a reliable behavioral method of measuring recollection rejection and phantom recollection in children. Children respond to memory tests under three types of instructions (accept only experienced events, accept only nonexperienced events that are consistent with the gist of experience, and accept experienced events + nonexperienced events that are consistent with the gist of experience). Recollection rejection and phantom recollection are identified with distinct patterns of responses across these different types of instructions. A mathematical model that is defined over the paradigm corrects raw response patterns for the influence of extraneous variables, which allows researchers to separate and quantify the effects of recollection rejection and phantom recollection. Implementation of this paradigm generates detailed findings on how recollection rejection and phantom recollection affect false-memory reports, on how these operations change with age, and on how they react to forensically-significant task variables. In Experiment 1, the effects of recollection rejection and phantom recollection on spontaneous false memories of narratives are studied in younger children (4- and 5-year-olds), older children (10- and 11-year-olds), and adults. Experiment 1 also determines how the two operations react to five forensically-significant task variables (repetition of events, recency of events, order of memory tests, timing of memory tests, and repetition of memory tests). Fuzzy-trace theory makes predictions about each of these variable's effects on recollection rejection and phantom recollection. In Experiment 2, the effects of recollection rejection and phantom recollection on implanted false memories of narratives are studied at the same three age levels by interpolating a memory-suggestion procedure following narrative presentation. Experiment 2 includes the same five task variables as Experiment 1. The aim of Experiment 3 is to determine how well the results of Experiments 1 and 2 generalize to more naturalistic eyewitness situations. The narrative false-memory procedure is replaced by a standardized quasi-forensic task involving a misdemeanor theft. The effects of recollection rejection and phantom recollection on both spontaneous and implanted false memories are assessed with this eyewitness procedure. The same three age levels are studied, and the design includes three of the five task variables that figure in Experiments 1 and 2 (repetition of events, timing of memory tests, and repetition of memory tests).

[unreadable] DESCRIPTION (provided by applicant): The Society for Medical Decision Making (SMDM) is dedicated to promoting rigorous, scientific research on health care decisions and policy initiatives. For over a quarter century, it has been on the cutting edge in studying decision making and developing support tools to improve the quality of patient care. While these clinical approaches and tools have been reported in medical journals and employed in numerous health care settings, several barriers have limited their more widespread adoption and use for real-time decision making. A major impediment has been the lack of coherent and evidence-based predictive scientific theory. In addition, the recent proliferation of patient decision aids and their use by a wide range of consumers, researchers and policy-makers makes standardized evaluation of their quality more crucial. To address these issues, the theme of this year's annual meeting is Better Decisions, Better Care: Advancing Decision Support to Improve Health Care. A key feature of this year's meeting will be a symposium devoted to presenting leading, empirically supported theories of medical decision making and health. [unreadable] [unreadable] The Society's Annual Meeting (October 14 -18, 2006) features current research in a rich cross-section of disciplines including decision analysis, shared decision making, evidence based medicine, cost effectiveness analysis, health economics, meta-analysis, computational methods, decision psychology, and medical ethics. We seek support for one symposium, a workshop, and 9 short courses. Symposium #2, Theories of Medical Decision Making and Health, will present scientifically rigorous explanatory and predictive theories in health and medical decision making that have been tested empirically. The symposium will bring together leading proponents of influential and diverse theories in medical decision making, and will emphasize the need for theoretically motivated research in domains of practical importance, including behavior change in health, cancer prevention and control, and physician decision making and decision support. Our dinner symposium, Improving Patient Decision-aids by Establishing State-of-the-art Standards for Content: The IPDAS Project, will feature experts in decision aid development and evaluation. The Short Courses will help attendees acquire skills related to the conference theme. [unreadable] [unreadable] Summaries of the symposia will be published in the Society's newsletter and disseminated through the Society's website. Furthermore, the Editor of the journal Medical Decision Making has enthusiastically agreed to publish the proceedings from the conference as a special issue (symposium participants will contribute articles), if funding can be obtained. We anticipate that the proposed program can make major strides toward disseminating research illustrating evidence-based approaches to real time decision-making by patients, physicians, and members of the health care team. This is a venue in which strong intellectual challenges to the status quo can engage the best researchers in the field. [unreadable] [unreadable] [unreadable] [unreadable]

The Workshop on Development of Higher-Order Cognitive Processes in Adolescence and Young Adulthood: Social, Behavioral, and Biological Influences on Learning will advance basic science in higher-order cognition and its development in adolescence and young adulthood by focusing leading scientists on key problems that are ripe for groundbreaking discoveries. The workshop will bring together separate but overlapping research communities. These communities include those with basic research expertise in learning (e.g., cognition, development, neuroscience) with those who have expertise in more applied research, especially education. Themes of the workshop will focus around the following areas: 1) Metrics of Meaning (devising valid measures of meaning, a construct that is at the heart of research on comprehension, reasoning, and information processing generally), 2) Reasoning and Judgment, and 3) Neuroscience of Reasoning, Representation, and Retrieval. Special attention will be given to acquisition of mathematical knowledge and skills, and their application in reasoning (e.g., scientific reasoning) and judgment (e.g., economic judgments). The workshop is also expected to facilitate development of new metrics and methods for studying higher-order cognition.

The workshop will enhance mutual understanding among research communities and stimulate fruitful collaborations. It will foster the translation of research on the basic science of higher-order cognition to solve pressing problems in education, economics, and workforce development, especially the development of mathematical knowledge and reasoning skills essential for competitiveness in the 21st century in science, technology, engineering, and mathematics.

DESCRIPTION (provided by applicant): This application addresses the broad Challenge Area: 15 Translational Science and the specific Challenge area Topic: 15-RR-101* Applied Translational Technology Development Memory declines, especially in recall, are hallmarks of healthy aging and conversion to cognitive impairment. Our goal is to use highly sensitive mathematical modeling techniques to improve the ability of clinical recall tests to predict future cognitive impairment and to diagnose current impairment. Our research will focus on one of the most widely used clinical tests of such declines, the Rey Auditory Verbal Learning Test (RAVLT). Our specific aims are to apply mathematical models to RAVLT data in order to: (a) substantially improve the ability of the RAVLT and similar clinical recall tests to predict future impairment and to diagnose current impairment;(b) separate different clinically important components of memory from one another in accordance with current theories of the memory processes that underlie performance on the RAVLT and similar tests;(c) identify the components of memory that differentiate cognitive changes that are associated with normal aging from changes that are associated with conversion to impairment;and (d) provide separate scores for different memory components of RAVLT data, which can be used to better predict behavioral and biological markers of future impairment and to identify current impairment. The research will consist of 2 phases, spanning 2 years. Both phases will rely on mathematical modeling tools and software that we have already developed. Our preliminary studies have shown that RAVLT-type tests are inherently noisy measures of impairment because 3 different memory processes are responsible for performance, but only 1 of them (gist-based reconstruction) is responsible for conversion to impairment. Therefore, in both phases of research, we will investigate how predictive and diagnostic power are improved when our modeling tools are used to remove this noise. Noise will be removed by computing separate scores for the reconstruction component of performance and for the other 2 components (direct access of verbatim traces and meta-cognitive confidence). During Phase I, this question will be investigated using a very large sample of subjects who participated in the Aging, Demographics, and Memory Study (ADAMS) portion of NIA's Healthy Retirement Study. The first phase will establish whether noise-free scores greatly improve our ability to separate groups of subjects that differ on biological markers of impairment (e.g., the ApoE genotype), behavioral markers of impairment (e.g., neuropsychological tests), and clinical diagnoses of impairment. During Phase II, this question will be investigated in a longitudinal study of 200 adults (aged 70 and above) who will be administered a neuropsychological test battery, and who will also be administered 3 versions of the RAVLT, spaced at 6-month intervals. The second phase will establish whether noise-free scores greatly improve our ability to differentiate individual people who differ in biological markers of impairment, behavioral markers of impairment, and clinical diagnoses of impairment. PUBLIC HEALTH RELEVANCE: This research will apply state-of-the-art mathematical models to clinical tests of memory to dramatically improve such tests'ability to predict future cognitive impairment in older adults and to diagnose current impairment. Findings will be used to develop low-burden tools that remove the noise for such tests and provide scores for the component memory process that are associated with conversion to impairment.

DESCRIPTION (provided by applicant): Decisions about whether to be tested for genetic risk of breast cancer are difficult. There are qualitative and quantitative dimensions of this decision. Quantitative dimensions include understanding conditional probabilities, relative and absolute risk, and the logic of statistical risk models. Qualitative dimensions include understanding what is breast cancer, what does genetic risk for breast cancer mean, what people should do in the event of positive and negative test results, and deciding under what circumstances a person should consider being tested. Aims. The goals of this project are to understand how women who have never had cancer themselves decide about whether to undergo predictive testing for genetic risk of breast cancer, and to develop and test a web-based computerized Intelligent Tutoring System (ITS) to help women make this decision using information already vetted, approved, and available on the National Cancer Institute web site. The first aim is better understand decision-making processes. The second aim is to develop a web- based AutoTutor, a sophisticated ITS with an animated conversational agent. Innovation. This is, we believe, the first use of an ITS to improve patients' medical decision making. These tutorials will teach women about the qualitative and quantitative concepts related to predictive testing. The ultimate goal is helping women make better decisions about genetic testing for breast cancer risk. Methods. Dimensions of this research and development project are developing the web-based AutoTutor; conducting randomized controlled experiments; and carrying out fine-grained cognitive analyses. The fine-grained analysis will integrate detailed process data with outcomes and posttest responses from 120 participants. The AutoTutor will be developed and tested in three phases corresponding to two tutor modules emphasizing qualitative and quantitative content, and a post-production phase. This will be accomplished through an iterative process with cycles of (1) preliminary research, (2) tutor development, (3) empirical research, and (4) tutor revision. New dependent measurers will be developed in a study with 60 participants. Three controlled experiments will empirically test the AutoTutor and assess decision-making. Two experiments of 120 participants each will address each module and a third web-based experiment with 80 participants will test the complete tutor. Participants will be randomly assigned to the AutoTutor, the National Cancer Institute web site or a control group receiving unrelated information. We will work from the beginning to lay the foundations for the next, more sophisticated generation of the AutoTutor. Personnel. PIs Christopher Wolfe at Miami University and Valerie Reyna at Cornell University have considerable experience with research on medical decision-making, learning technologies and web-based interventions, web-based psychology experiments, quantitative decision making, and verbal reasoning. Expert consultants are Nananda Col MD, breast cancer expert and director of the Center for Outcomes Research and Evaluation, Maine Medical Center, and genetic counselor Sara Knapke. PUBLIC HEALTH RELEVANCE: The goal of this project is to develop a web-based Intelligent Tutor about qualitative and quantitative dimensions of the decision to undergo predictive testing for genetic risk of breast cancer. The purpose is to understand how women make this decision and help improve decision making. Research methods include randomized controlled experiments and fine-grained cognitive analysis.

DESCRIPTION (provided by applicant): Risky behavior contributes to substantial morbidity and mortality during adolescence, and unhealthy patterns of behavior that debut during this period have consequences that play out over a lifetime. For example, in a representative sample of 5,547 12- to 19-year-olds, none met criteria for cardiovascular health. Sensitivity to rewards (e.g., tasty foods) contributes to obesity and other unhealthy behaviors, and recent reviews of research on risky decision making in adolescence have focused on this topic. Although there is evidence that adolescents are more sensitive than adults are to rewards, not all studies find this pattern: Some studies find less sensitivity to rewards among adolescents, which cannot be explained simply by reward stages (anticipation vs. receipt). Other studies were not designed to isolate reward sensitivity, and so confound it with known developmental differences in risk attitudes, memory for outcomes, or feedback-induced strategies. Also, definitions of reward sensitivity vary across fields, and research on adolescent decision making does not distinguish among 4 different hypotheses tested here. These hypotheses make starkly different predictions about adolescent risk taking and effects of incentives on their behavior. Moreover, we examine the interplay between such factors as sensitivity to reward and risk, on the one hand, with emotions and drive states on the other hand. We test surprising, but theoretically motivated predictions, for example: (1) Drive states will induce reverse framing (taking greater risks for greater rewards and accepting larger sure losses) among adolescents even for objectively low rewards. (2) Inducing gist processing will have a protective effect on ris taking for rewards, reducing vulnerability to drive states. (3) Although most theories anticipate that adolescents will be more vulnerable to strong emotion than adults, and less able to accurately forecast their risky decisions, there is theoretical justification for the prediction tht adolescents will approach such risks more coldly than adults. Adolescents and adults will provide reward ratings and make decisions involving these same rewards (in counterbalanced order) using consequential and motivating incentive-compatible procedures. We examine common currency and domain-specific effects for candy bars and money, and use neuroimaging to test hypotheses about neural circuitry of risk taking. The Principal Investigator and other investigators are highly proficient data analysts and mathematical modelers. Analyses will include standard ANOVA for factorial designs (Table 1) with either decision choices or reward ratings as dependent variables. Using multiple regressions, measures of individual differences (principal components analysis will be used to reduce the number of predictors; see Reyna, Estrada et al., 2011), plus laboratory decisions and reward ratings, will be used to predict real-life risk taking on the Adolescent Risk Questionnaire. Therefore, we both manipulate levels of reward and measure sensitivity to reward as an individual difference, as well as manipulate challenges to cognitive control (e.g., drive states) and measure cognitive control, including inhibition (Behavioral Inhibition Scale and go/no-go task).

To have a well-functioning society, businesses, insurance companies, and members of the public depend on a reliable and competent jury system. In a civil case in which a defendant injures a plaintiff and a damage award must be determined, civil jurors must translate their subjective, qualitative sense about the severity of the plaintiff's injury into dollars. Yet many injuries are difficult to value, and civil juries receive limited guidance. Some jurors, too, have difficulty working with numbers. This research project examines how to improve jury damage award decision making so that damage awards are well-matched to the harm done. The project tests predictions from a new model of damage award decision making that focuses on the gist or underlying meaning that people derive from trial evidence. These predictions are (a) most jurors have a coherent sense of the gist of damage awards; (b) a gist-based understanding of numerical magnitude is essential for coherent judgments; and (c) this gist-based understanding can be facilitated with simple interventions to improve jury decision making. Three experiments are proposed to identify the jury damage award decision process and to test easily implemented and cost-effective solutions to reduce unwarranted inconsistency. The question of fair and consistent evaluation of injury arises in many business and insurance contexts, so effective solutions discovered through this research could prove useful in other domains beyond the courtroom.

The legal system regularly requires fact finders to translate their qualitative impressions into quantitative judgments such as criminal sentences, fines, and damage awards. However, the relationship between qualitative and quantitative judgments in law is under-studied and under-theorized; no comprehensive theory yet explains it. The proposed research promises to move theory in the field forward by testing specific predictions derived from a new model of damage award decision making. The experiments extend fuzzy-trace theory to damage award decision making, testing a number of specific predictions about meaningfulness and relative gist of anchors (or benchmarks), relevant and irrelevant emotional evidence, and the utility of instructions about how to scale damages. Specifically, the first experiment investigates how potential jurors interpret the gist of case facts, how that influences damage awards, and whether providing meaningful anchor amounts reduces variability in award amounts. The second experiment focuses on effects of relevant and irrelevant emotion, and on the mapping process--how simple questions can help orient potential jurors to the magnitude of damages--with the aim of reducing variability in award judgments. The third experiment tests our new model in a realistic mock civil jury setting, including jury deliberations. Data collection will include multiple measures of numeracy (facility with numbers), allowing the researchers to determine how numeracy influences decision making about dollars in the context of civil damage awards.

While the novel coronavirus sweeps the globe, democratic societies face a quandary, namely, how to encourage sacrifices to reduce risk, even when the threat is invisible and seems to be receding. When risk is high, individuals may need to engage in extreme forms of social distancing for long periods, but this choice comes with economic and personal costs. Not engaging in these risk-reduction activities, however, could cause millions of deaths. It is therefore crucial for public health communicators to understand rationales for refusing these actions and to base risk communication on empirically supported principles, including the need to convey the gist (bottom-line meaning) of these actions to individuals making these challenging choices.

This project consists of three studies, each with two samples at different time points, with the ultimate goal in subsequent studies of converting the gists of social media messages into effective risk-communication interventions. The approach is motivated by Fuzzy Trace Theory?an evidence-based account of health decision making under risk?which posits that decisions are based on qualitative gist representations of stimuli that encode basic meaning in context. The scholars test time-sensitive hypotheses using bottom-up unsupervised machine-learning algorithms to characterize the topics in social media messages about COVID-19 and top-down human judgments about theoretically predictable mental representations of the gist of perceived risks and benefits of risk-reduction behaviors. Study 1 quantifies the prevalence of gists pertaining to social distancing and other risk-reduction behaviors (e.g., hand washing) by extracting millions of social media messages from Twitter and public Facebook posts. This involves automatic coding an ongoing corpus of millions of social media messages using probabilistic topic models. Study 2 provides a systematic approach with human judges to interpreting the topics extracted by a topic model at different time points. Study 3 involves a different group of human judges assessing topics for their consistency with a targeted set of gist representations of risks and benefits, along with gist principles that express values. The researchers adapt items used successfully in prior research. Similar gist representations have predicted self-reported risk-reduction behaviors for numerous health conditions. Thus, for Study 3, the researchers validate topic gists by fielding a theoretically motivated survey analyzed for reliability and construct validity. The focus is on perceptions of the gist of risks and benefits, especially gists that can promote the health of individuals and society. Overall, this project introduces novel techniques for eliciting gists from social media that can be used to generate meaningful public health communications. The project directly informs existing attempts by public health communicators to express the risks, benefits, and actions that members of the population should take to mitigate the spread of the COVID-19 pandemic. The measures are validated against human users, enabling the team to achieve both accuracy and scale. This project serves as the basis for a larger effort that can increase the extent to which gist elicitation may be automated, helping public health communicators to quickly understand which gists should be communicated to which communities during outbreaks.

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.