Experimental studies of structured memory representations | Sam Gershman (FAS Psychology)
The Computational Cognitive Neuroscience lab aims to understand how richly structured knowledge about the environment is acquired, and how this knowledge aids adaptive behavior. We use a combination of behavioral, neuroimaging and computational techniques to pursue these questions.
One prong of this research focuses on how humans and animals discover the hidden states underlying their observations, and how they represent these states. In some cases, these states correspond to complex data structures, like graphs, grammars or programs. These data structures strongly constrain how agents infer which actions will lead to reward. A second prong of our research is teasing apart the interactions between different learning systems. Evidence suggests the existence of at least two systems: a "goal-directed" system that builds an explicit model of the environment, and a "habitual" system that learns state-action response rules. These two systems are subserved by separate neural pathways that compete for control of behavior, but the systems may also cooperate with one another.
Recent theories of memory suggest that items are more memorable if their “description length” is short. For example, CIAFBINSA is easier to remember than GTFEMNVQO because the first string can be described in terms of 3 “chunks” (CIA, FBI, NSA) instead of 9 unrelated letters. As this example illustrates, in order to predict how memorable an item is, we need to hypothesize a particular description language. Computational models of cognition have developed sophisticated description languages, but the implications for memory are still unexplored. This research project will fill in this gap by developing new variants of classic memory experiments (e.g., change detection, recognition memory, free recall) using structured stimuli, along with computational models of description languages that will quantitatively predict memorability.
To illustrate the generality of this idea, we will examine several cognitive domains that seem on their surface very different but may be organized within a single computational framework. This will include representations of dynamical scenes, words, and noisy functions. The description languages will be derived from probabilistic models that specify distributions over objects in a domain. The description length can be predicted using concepts from information theory.
The primary task for the student working on this project will be to implement the experimental tasks and collect data. Since our lab primarily runs behavioral experiments over the web, the experiments will be implemented as websites and the data collected using Amazon Mechanical Turk. The secondary task for the student will be to using a probabilistic programming language, such as Church (see https://probmods.org/), to implement probabilistic models of the different experimental paradigms and use the models to predict human performance. The student will participate in all lab activities, including lab meeting and informal interactions with graduate students and postdocs, as well as regular meetings with the principal investigator. The student will present their research in lab meeting at the end of the summer.
This research is a unique opportunity for undergraduates to engage in both experimental and theoretical research. For those students interested in pursuing a career in cognitive science, the skillset they will develop (web programming experiments and computational modeling) are in high demand. For those students interested in fields such as machine learning and statistics, they will have the opportunity to do cutting-edge computational research.
Using GIS to study the Russian Empire | Kelly O'Neill (FAS History)
Experience conducting social science research and familiarity with spreadsheet and/or database software are desirable. The fellow must be willing to engage in qualitative and quantitative analysis, and to learn the basics of GIS. Language skills (Russian, French, German or Italian) are a welcome bonus.
My research and teaching interests revolve around the history of the Russian Empire. My first book explores the social, cultural, and economic ramifications of Russia’s first annexation of Crimea under the Catherine the Great (1783). My current projects include studies of the Black Sea slave trade, the development of the Russian wine industry, as well as a book project on the environmental and economic history of Russian rivers. I am particularly interested in exploring the potential of digital history methods and am in the process of developing an historical GIS of the cultural and commercial infrastructure of the tsarist empire.
Space has always been an important theme in Russian history. Many a textbook opens with the declaration that the reader will understand nothing of the tsars, serfs, revolutionaries and artists who inhabit the Russian past if he or she does not appreciate the vast, varied, and generally unforgiving nature of its terrain. Rarely, however, have scholars moved from describing to analyzing imperial space and the ways in which it has shaped human experience.
The innovations of digital scholarship are rapidly removing many of the obstacles that have prevented historians from harnessing the vast quantitative material at our disposal and using it to gain new insights into the human past. Among the many tools that encourage us to annotate, compare, compile, sort, sample, visualize, manipulate, and model, GIS is particularly useful to historians. It allows us, for example, to layer information in a single visual field and identify patterns and relationships that would otherwise have remained obscure. While students working on contemporary issues are able to download rich datasets from the World Health Organization or NASA (to take but two examples), students of the past must generate their own data from scratch. Historical GIS work involves everything from compiling data on trade routes, weather conditions, and population density, to mapping sacred spaces and the circulation of literary texts.
The Imperiia Project (imperiia is the Russian word for ‘empire’) is an attempt to develop a robust historical GIS that documents and analyzes the infrastructure of mobility in the Russian Empire. I initiated the project several years ago and thanks to key contributions from undergraduates and graduate students, along with support from the Center for Geographic Analysis and the Davis Center for Russian and Eurasian Studies, it is beginning to take shape. The project has an online presence (http://worldmap.harvard.edu/maps/russianempire) and we are in the process of developing a suite of interactive tools that will enable users to access and manipulate the geodatabase.
The goal for summer 2016 is to 1) compile data on the migration of peoples within and across Russian borders and 2) examine the impact of migration on relevant rural/urban landscapes. The movement of ethnic and religious minorities, nomadic peoples, peasants, foreign colonists, and others during both war and peace presents a rich array of research possibilities. Fellows will use published primary and secondary sources to research the episodes they have selected, design and build geodatabases, and work with Prof. O’Neill to engage in spatial analysis of their findings. Some will find it necessary to digitize and georeference maps from the Harvard Map Collection. In most cases the result of their work will be a series of annotated map layers that will be incorporated into the Imperiia Project.
This is an excellent opportunity for students interested in conducting original research, learning about historical GIS, and becoming formal contributors to an ongoing piece of digital scholarship.
Building the adult human mind | Susan Carey (FAS Psychology)
We can teach any student what he or she needs to know to do this research, but a background in some aspects of cognitive science is ideal (for example, a course in linguistics, computer science, history of science, analytic philosophy, or cognitive or developmental psychology).
My research seeks to explain the human conceptual repertoire. We are the only animals who can ponder the causes and cures of cancer, or of global warming, wonder about different orders of infinity, or, indeed, think literally billions of thoughts outside of the competence of other animal. This in spite of the fact that our brains are very similar to those of other higher apes. What makes this possible?
I run an internship program for 6 to 8 undergraduates from all over the world; I have had 1 or 2 BLISS fellows ever since the beginning of the BLISS program and they are part of the internship program. A BLISS fellow chooses a project for him/herself from among those we currently have ongoing or planned, collaborates on the design of the experiment to be carried out over the summer, runs the study, collaborates on the data analysis and interpretation of the results. The argument is presented in a poster session at the end of the internship and as a talk to the BLISS program. The week is full—interns are involved in all aspects of bringing a research project to completion.
My current work focuses on two very general cognitive capacities that might distinguish humans from other great apes: executive function and logic. Within each research focus, there are several different research projects the BLISS fellow can choose from.
Each BLISS fellow works in collaboration with a graduate student or postdoc, meeting with him or her several times a week, participates in weekly lab meetings headed by me, participates in a discussion section just for undergraduate interns in the Laboratory for Developmental Studies, and meets with me weekly (together with his or her graduate student/postdoc mentor) throughout all 10 weeks of the placement.
The ultimate goal of the internship is to engage students in bringing empirical data to bear on theoretically important issues, to see the process of making a scientific argument. This experience will inform students’ career choices (i.e., does he or she want to become a research scientist, and if so, are the questions within cognitive science of interest?). Whatever the outcome of these decisions (at this stage of life, it’s important to discover what one does not want to do as part of discovering what one does want to do), this experience make students more critical consumers of popular science reporting in the press, and of course work in the social sciences.
The proximate goal of the internship is to engage students in the fun of the collaborative activity that is science.
Dynamic and multi-period matching markets | Maciej Kotowski (HKS)
• Some training in mathematics beyond college calculus. The student should feel comfortable reading and discussing somewhat technical academic material. (A recent issue of American Economic Review offers a good example of the type of literature relevant for this research project.)
• Familiarity with microeconomics at the intermediate level (Economics 1011a). Completed course work in game theory or market design would be beneficial, but is not necessary.
• A student with little or no background in economics, but with a strong background in mathematics/statistics/computer science would also benefit from this research experience.
• Basic computer programing skills (CS 50) are desirable. (A student interested in developing such skills would be given the chance to do so.)
I am an Associate Professor of Public Policy at Harvard Kennedy School. I am an economist and my research and teaching specialization is microeconomic theory, with particular emphasis on auctions, matching markets, and market design. I have also conducted research on economic networks and on topics in law and economics, including regulation and taxation. Methodologically, my research is primarily theoretical, though firmly motivated by salient real-world features. I am interested in understanding how markets work, what wrinkles compromise their operation, and how they can be best organized. Toward this goal, I develop models of economic phenomena and markets to help guide positive and normative research in market organization and design.
In a matching market agents from two sides must “match” together to realize benefits. Examples are ubiquitous. A firm hiring a worker must find him well-qualified while the worker must find the job offer sufficiently appealing. A student must be keen to attend a college and the college must consider the student admissible. Many well-known internet applications—Uber, AirBnB, and countless dating apps—are also matching markets. In these cases, entrepreneurs have crafted matching mechanisms, sets of tools and rules structuring the interaction among the market’s participants, that help agents coordinate on beneficial outcomes. (If you are reading this, you too are participating in matching market that connects students to research projects.)
Despite their prevalence, many questions remain concerning the operation and design of matching markets. A feature of many matching markets, such as those cited above, is the important role played by time and the dynamic nature of market’s population. For example, a worker’s choice of a job today often affects his (possible) jobs in the future. Similarly, employers worry about employee retention. A related problem arises with organ transplantation. As new patients arrive, new donor organs become available only sporadically. Determining who matches with whom and when are important questions.
This project is part of the principal investigator’s ongoing research program studying dynamic and multi-period matching markets, such as the situations described above. At a high level, the program aims to develop a unified theoretical framework for the analysis of such markets. At a more fine-grained level, the project involves creating succinct and tractable models of dynamic matching markets that will inform market design and policy. Some of the key questions considered include:
- What is the appropriate “equilibrium concept” for a model of a dynamic matching economy? Under what circumstances do insights from models of static economies generalize?
- Which markets are particularly impacted by their dynamic characteristics? When has ignoring the dynamic nature of agents’ interaction led to suboptimal policy prescriptions or fragile market designs? Can simple “patches” address potential shortcomings or are fundamental redesigns warranted?
- What features of a matching mechanism ensure its good operation in a market with a dynamic population or when agents interact over many periods?
- What role do commitment, incentives, and reputation play in the successful operation of dynamic matching economies?
A research assistant engaged in this project will help propose answers to these and other questions. Within the project’s scope, the assistant’s tasks can be tailored to meet his/her skills and interest. These range on a continuum depending on the assistant’s prior familiarity with matching markets, his/her technical abilities, and his/her demonstrated interest in research in microeconomic theory. A student who is new to the research questions considered would first hone his or her understanding of matching markets and market design more broadly. To do so, the assistant may contribute to a review of the literature by developing new typologies of prior contributions or by consolidating and linking existing results. The assistant may be tasked with identifying, studying, and documenting the operation of real-world matching markets and preparing case studies to serve as examples. The opportunity to develop additional technical skills, such as computer simulation or data analysis, would also be possible. At the other extreme, a student who is already familiar with the topic (perhaps through coursework or prior experience), very mathematically adept, or with extensive technical/computer skills may prefer to develop his or her own models or market simulations. Beyond acquiring a familiarity with recent research on matching markets, the research assistant will strengthen skills that are useful in both academic and non-academic settings, such as computer programing, data analysis, academic writing, and editing.
While there is considerable flexibility in structuring the research assistant’s day-to-day tasks, the project’s broader methodology imposes some constraints. In particular, an interest in microeconomic theory, a degree of mathematical sophistication and maturity, an enthusiastic engagement with technical social-science research, and a genuine curiosity about the world are prerequisites to benefit fully from this research experience.
Further reading on matching markets and market design:
- David Gale and Lloyd S. Shapley. College admissions and the stability of marriage. The American Mathematical Monthly, 69(1):9–15, January 1962.
- Alvin E. Roth. The economist as engineer: Game theory, experimentation, and computation as tools for design economics. Econometrica, 70(4):1341–1378, July 2002.
- Alvin E. Roth. Who Gets What—And Why: The New Economics of Matchmaking and Market Design. Eamon Dolan/Houghton Mifflin Harcourt, 2015.
Computational measures of engagement across differences | Dustin Tingley (FAS Government)
The ideal fellow will have some experience or coursework in computational social science, computer science, statistics or a related field. The fellow should have an interest in learning more about social network analysis, text analysis, and online education.
The BLISS fellow will assist in developing new methods for evaluating how people with diverse beliefs engage with one another in online classes with the ultimately goal of encouraging greater attention to the development of civic competencies in online learning. The fellow will have the opportunity to interact with a multi-university team including Justin Reich, Fellow at the Harvard’s Berkman Center for Internet and Society and Director of the Teaching Systems Lab at MIT; Dustin Tingley, Professor of Government at Harvard; Brandon Stewart, Assistant Professor of Sociology at Princeton; and a postdoctoral fellow who is in the process of being hired.
The BLISS fellow will work over the summer to support three lines of research and development in the project. First, she or he will learn about and make use of social network analysis and natural language processing tools, including Structural Topic Models, to develop new measures of the degree to which students in online courses engage across political differences. As the project develops a set of informative measures, the fellow will assist in the design and building of software tools that take standardized data formats (primarily from the edX learning management system) from online discussion forums and generate these measures. Second, she or he will participate in a small qualitative study in coding and evaluating discussion forum threads to develop a “ground truth” of discussion quality. Third, she or he will aid in developing a set of experiments to improve the quantity and quality of engagement across different in politically themed online courses and evaluate how our measures are responsive to those experiments.
Why have we been dying less from heart disease and stroke? | Ankur Pandya (HSPH)
The student should have taken at least 1-3 of the following courses:
• Probability theory
• Statistics (introduction and one intermediate course, ideally)
The student must be comfortable using SAS or STATA to analyze data. The student should know that this project is quantitative in nature, only involves secondary data analysis (i.e., the student will not be collecting any primary data), and mostly involves “computer work” as opposed to being “on the ground” or in the community. There is considerable autonomy on the part of the student, but I will guide and direct the research as much as they need/prefer.
Dr. Pandya’s research focuses on developing and evaluating quantitative models that have the potential to improve cardiovascular disease policies in developed and developing countries. His research has been covered in The New York Times, National Public Radio, NBC Nightly News, and other media outlets.
The goal of this project is to re-examine the key causes for the reduction of cardiovascular disease (heart disease and stroke, which is the leading cause of death in the United States and worldwide) using large nationally representative datasets. We believe that previous (widely-cited) estimates for these causes relied on flawed assumptions regarding the role of diet and exercise on blood pressure and cholesterol reductions, and that re-analyzing these data can shed light on important public health questions related to a major disease area.
Specifically, the student will:
- Participate in the conceptual study design process, which involves statistics, epidemiology, and cardiovascular disease health disciplines
- Clean and analyze large nationally representative (NHANES) datasets using SAS and/or STATA software
- Prepare results tables and figures
- Write a scientific abstract and (potentially) manuscript that describes the background, methods, findings, policy implications of this work
The student will work with a real, large, nationally representative dataset to help answer an important public health question using quantitative methods. The student will have an opportunity to “own” some of this work by being co-author on any scientific abstracts or manuscripts that they contribute to. This should be a nice, defined, applied public health project that gives the student exposure to a real research project while sitting at an academic research center (Center for Health Decision Science at the Harvard T.H. Chan School of Public Health).
God, the Grand Canyon, and your 401(k)... | Jennifer Lerner (HKS) and Charlie Dorison (GSAS)
... Experiments Investigating how Religion and Awe influence Financial Patience
The ideal student will have completed a course in introductory psychology or religion. Some economics coursework would also be desirable, although not necessary. Previous research experience is a plus. Students with different backgrounds will still be considered if the student has a strong interest in the topic and a strong academic record.
Dr. Jennifer Lerner is Professor of Public Policy and Management at the Harvard Kennedy School of Government and Co-founder of the Harvard Decision Science Laboratory. She is the first psychologist in the history of the Harvard Kennedy School to receive tenure. Drawing insights from psychology, economics, and neuroscience, her research examines exciting new areas of human judgment and decision making.
One specific area of research examines how different specific emotions influence economic decisions over two points in time. For example, Professor Lerner has found that sadness exacerbates financial impatience, whereas gratitude reduces impatience. Across all areas, her work aims to expand the evidentiary base for designing public policies that maximize human wellbeing.
A recent revolution in emotion science has uncovered exciting new opportunities for research examining the influence of specific emotions on economic decision making. Within this quickly growing field of emotion and decision making, it has been only in the last decade that researchers have begun to examine a powerful, spiritual emotion: awe. Deeply tied to both religious experience and spirituality, awe has the power to shift how people think about time, money, and themselves. Drawing primarily from psychology, but also on the fields of religious studies and behavioral economics, the current project will use experiments in a laboratory setting to examine how religion and awe causally influence economic decision making. The project will focus both on the mechanisms through which religion and awe influence financial decision making (e.g., through diminishment of the self) and on important practical and policy applications of this research (e.g., retirement savings).
An undergraduate will have the opportunity to help with this project at an early stage, allowing for greater ownership of the research and a role in the project from the ground up. The selected research assistant will work on such things as: finding journal articles, summarizing readings, discussing ideas, graphing data, and running experiments in a laboratory. In addition to acquiring these skills and gaining knowledge, the research assistant will receive feedback on how to organize ideas, review scientific papers, and generate potential practical applications.
The interdisciplinary nature of this research will allow the chosen student to develop analytical skills that cut across the methodologies and theories from a single field. The chosen undergraduate will receive feedback from me on a number of areas that are relevant in any professional setting. These include how to organize ideas, how to review and analyze scientific evidence, and how to generate and communicate practical applications from data, among others. Additionally, learning skills relevant to the research process – such as reading papers and identifying the most important information, analyzing data and drawing conclusions, clearly communicating arguments – will be valuable in any academic or professional setting.
Student social support R&D | Todd Rogers (HKS)
The student should be interested in psychology, education, decision-making, influence, and/or behavioral economics. The student should also be motivated, punctual, hardworking, and flexible.
The Student Social Support Research & Development Lab (S3 Lab) uses data and behavioral science to develop and prove scalable, high Return-on-Investment interventions that mobilize and empower students’ social support systems to improve achievement. S3 Lab is conducting over a dozen projects in nearly 400 schools and universities involving more than 135,000 students.
A sampling of projects we are working on:
• Reducing absences among at-risk K-12 students by providing parents with regular messages targeting specific attendance beliefs.
• Turning friends and family in college students’ social networks into “Study Supporters” by regularly empowering them to get more involved with updates about the students’ courses, upcoming deadlines, and campus resources.
• “Inviting the village” that surrounds K-12 students (e.g. grandparents, mentors, coaches, family friends) by empowering these supporters with actionable information they can use to encourage and support students (e.g., report cards, class schedules, attendance reports, etc.).
The BLISS fellow will work on randomized field experiments that test educational interventions focused on mobilizing and empowering social systems (e.g., friends and family) to support student achievement. These interventions will be based on behavioral science insights, and will be designed to “nudge” students toward better educational outcomes. The lab will be conducting some of these during the summer at online colleges, community colleges, and in advance of school-year projects in large K-12 schools. The BLISS fellow may continue to work on these after the summer, if interested. In addition, highly motivated students may have the opportunity to develop these projects into a senior thesis.
The BLISS Fellow will join the S3 Lab’s eight Research Fellows and Lab Coordinator in their office space at the Harvard Kennedy School. Each Research Fellow manages a portfolio of projects run at multiple sites. The BLISS Fellow will have opportunities to work on most, if not all, of the projects, which will be at various stages of the research process. Tasks may include entering data into excel, analyzing data, maintaining a citations database, using Qualtrics to create surveys, sending out information to participants and partners, and creating new treatment materials. Students will also spend time reading scholarly articles and writing literature reviews. The summer months also provide a unique opportunity because this is the period for us to brainstorm, create and design new studies, and meet with current and potential partners with proposals. The lab is committed to providing a comprehensive and challenging research experience for undergraduates with a healthy helping of fun (including daily lab lunches)!
There are several possible research projects that the BLISS undergraduate student may help with. Two are described below:
Turning Pre-Existing Relationships into Academic Support (College): College students often have strong outside-of-college relationships, and these friends and family often report feeling explicitly uninvited by colleges to be a part of helping students succeed in college. In this intervention we develop an automated system that enlists and coaches outside-of-college friends and family to turn their pre-existing social interactions with college students into opportunities to actively support and encourage the students. These “study supporters” receive weekly SMS messages, Facebook updates, and/or emails (cc’ing the students) that reflect course content and discussion tips. This work will help us understand what information most influences the behavior of supporters and which supporters within students’ networks are the most influential.
Inviting the Village (K-12): Students often have several adult figures in their communities whom they admire and who exert influence on them. These may be grandparents, aunts and uncles, older siblings, ministers or pastors, mentors, family friends, or even work supervisors. These “positive community members” care about the students and actively take an interest in and encourage the students, yet are often shut out of communications from students’ schools. Schools are not permitted to send them progress reports, absence notifications, or other relevant school information. This project involves guardians authorizing schools to communicate directly with these adults (in addition to with guardians) to support students’ academic success.
This opportunity will both teach the basics of working in a behavioral science lab and allow students to delve into designing protocols and implementing studies. The student will learn how to design viable surveys as well as structure papers and citations. Additionally, they will learn to quickly analyze academic works and summarize their important findings. The BLISS Fellow will also have the opportunity to help plan for the upcoming school year studies. The BLISS Fellow will also learn about the nitty gritty details of running research studies at a university. Most importantly, the fellow will get to work closely with a fantastic group of researchers dedicated to creating education interventions at all academic levels with varying past experiences, providing many different perspectives for future guidance for undergraduate and post graduate life.
Economic decision making and health behavior | Jennifer Lerner & Joowon Kim (HKS)
An ideal candidate would demonstrate a strong interest in key drivers of human behavior and how people make decisions involving economic or health outcomes. Introductory courses in economics, psychology, or statistics would be helpful, but not necessary.
Dr. Jennifer Lerner is Professor of Public Policy and Management at the Harvard Kennedy School of Government and Co-founder of the Harvard Decision Science Laboratory. She is the first psychologist in the history of the Harvard Kennedy School to receive tenure. Drawing insights from psychology, economics, and neuroscience, her research examines human judgment and decision making. Together with colleagues, she has developed a theoretical framework that successfully predicts the effects of specific emotions on specific judgment and choice outcomes. Applied widely, the framework has been especially useful in predicting emotion effects on perceptions of risk, economic decisions, and attributions of responsibility.
As a BLISS fellow, you will participate in a research project on economic decision making and health behavior, especially examining the drivers of addictive behaviors like smoking. Your responsibilities will include:
- Leading scientific experiments – occasionally with elite government and military leaders at the Harvard Decision Science Laboratory
- Conducting background research and literature reviews
- Coding data, summarizing results, and preparing it for presentation
- Participating in managing all other aspects of scientific research processes
- Finding ways to apply behavioral insights of our research to inform public policy
Our BLISS fellow will be working with full-time researchers (often with an advanced degree) on an interdisciplinary project in a professional setting. In addition to basic training in research methods and software, the fellow will gain hands-on experience in conducting scientific research, synthesizing information, and communicating ideas effectively. Members of my research team will be available to provide valuable feedback. The skills and experience gained from this position will be applicable to a variety of academic disciplines and career settings, furthering the fellow’s education and professional development.
Development of cooperative behavior in children | Felix Warneken (FAS Psychology)
An interest in psychology and working with children is essential. Previous experience in either area (or both) is a plus.
We are interested in the origins of human social behaviors, especially the origins of cooperation. What motivates children to share? When do children develop a sense of fairness? How do they judge other people’s actions as nice or mean? Most of our studies focus on children from early childhood into school age to elucidate the interplay of biological predispositions and cultural norms across development. In addition, we take an evolutionary perspective and compare the findings from child studies with that of other animals, especially chimpanzees.
Our main expertise is in developing child studies that target the most important questions about human social life, often drawing on insights from various disciplines, including economics, primatology, and anthropology. By studying children in this way, we can gain important insights into the origins of human nature.
In the summer of 2016, we will conduct several studies on children’s cooperative behavior. Some of our studies focus on children’s sense of fairness, and how this develops over time. Previous work has shown that while young children tend to be more selfish and keep valuable resources to themselves, older children value fairness more and are willing to share with others. Do children learn that when they share with someone now, this person might reciprocate the favor in the future? Do children share because others are watching and they want to appear fair? Or do children develop a general sense of fairness and apply equality norms to guide their behavior? We investigate these questions by developing child-appropriate experimental paradigms that can be used across different age groups. In some cases, we even use these tasks cross-culturally.
We also investigate children’s sharing and reciprocity behaviors using various paradigms. Are children more willing to give up a valued resource in order to gain a future reward? We also pair children with different partners, one cooperative and one uncooperative, and ask whether children of different ages will differentiate their sharing behavior based on the behavior of their partner. In all cases, children have a lot of fun participating in these studies – and so do our experimenters!
A new project that will begin in the summer of 2016 will use an interactive video game to assess third-party punishment. With a graduate student and a computer science student, research assistants will have the opportunity to develop and implement a study in which children play a computer game and get to decide whether to compensate the characters who have been treated unfairly or to punish the characters who have been unfair to others.
Studies will be conducted in the testing rooms of the Social Cognitive Development Group at the Department of Psychology, as well as in public parks and museums in the Greater Boston area. We conduct studies typically in teams of three researchers, one graduate student or senior lab member together with two undergraduate research assistants. The summer is the most productive time for data collection, and it is thus the most rewarding time for everyone involved, enabling students to participate in research projects from the planning phase all the way to the data analysis. BLISS fellows will gain experience in designing and conducting experimental studies with children ranging in age from toddlers to young teenagers. We video-record all studies, and undergraduate students will be trained in video coding as well as basic statistical analyses. In addition to experimental data, some studies will include questionnaires and short interviews to learn more about the children’s behavior outside of the testing situation.
Students will be supervised by the PI, as well as a postdoc and graduate students. We hold weekly lab meetings, where undergraduate students will discuss background papers from psychology, behavioral economics, and anthropology, and summarize their research project. At the end of the program, BLISS fellows give presentations about their research project.
BLISS students will have the opportunity to participate in all aspects of experimental research, including help running the studies and interact with children, video coding, and data analyses. This includes learning how to work with statistical and behavioral coding software. In addition to hands-on experience, our summer program gives students the opportunity to learn about the ongoing research by reading and discussing the relevant background literature. This provides students with hands-on experience on how to conduct experimental psychological studies.
In addition to working on the child studies, students participate in weekly lab meetings, discuss psychological publications, present ongoing studies, and have the option to write a literature review or study proposal at the end of the summer. Students will be paired up with a graduate student or postdoctoral fellow to closely work on a particular project, and meet with the faculty sponsor to discuss the project they are working on to gain a deeper understanding of the research topic. Combining hands-on experience of conducting research and discussing the relevant literature will provide students with insight into science that no regular undergraduate course can offer.
Intergenerational mobility, inequality, taxation and redistribution | Stefanie Stantcheva (FAS Economics)
The ideal student should have strong skills in Stata or R. Matlab or Python experience would be very welcomed, but is not required.
My research focuses on the design of the tax system, taking into account important labor market features, more complex social preferences, and long-term effects such as human capital acquisition. For instance, my recent empirical work looks into the effect of the top tax rate on international mobility as well as the role of interactions in knowledge creation and diffusion, using data on inventors from the United States and Europe.
Inequality and redistributive policies are at the heart of my research interests. Previous work includes research on the optimal design of redistributive policies and several empirical findings on the topic. For instance, in joint work with Thomas Piketty and Emmanuel Saez, we show that top tax rate cuts are associated with an increase in top one percent income shares but not higher economic growth. I am also currently working on social preferences for redistribution and, more specifically, on the impact of perceptions of social and intergenerational mobility on support for redistribution.
The selected student would work on several projects. The first is called “Social mobility and preferences for redistribution”. In this project, in collaboration with Prof. Alberto Alesina, we study how perceptions of social mobility affect support for redistribution. While income and wealth inequality in the United States have been growing over the past decades, support for redistribution has remained flat. One conjecture is that Americans believe in a high social mobility, or in the “American Dream”, namely that everybody gets a fair chance to make it in life and that one’s income is mostly the result of one’s own effort, rather than of one’s background. Are these perceptions of social mobility consistent with reality? Do they evolve with reality or are they sticky?
In order to answer these questions we design a survey asking respondents a series of detailed background questions related to their socio-economic characteristics, political preferences, and own experience of social mobility relative to their parents, we will then be able to compile a large, representative and comprehensive new dataset that can be used for future research on social mobility. We then perform online randomized experiments that change people’s perceptions of social mobility, by showing respondents in treatment groups various sets of (true) information regarding social mobility. We consider the effect of social mobility perceptions on the support for various redistributive policies: “equality of opportunity” policies, which level the playing field (such as education policies and inheritance taxation), and “equality of outcome” policies (such as income taxation and direct transfer policies). We study whether the effects of the treatments are amplified or dampened by a respondents’ own subjective experience of social mobility, by whether they have children, and by how risk averse they are. The survey will not only be distributed in the United States but also in France, Italy, Sweden and the United Kingdom
The other project, joint with Emmanuel Saez (UC Berkeley) will consider the question of the optimal taxation of capital. Here, the student could help us with analyzing data on capital and wealth holdings across countries.
The third project relates to the effects of taxes on innovation by firms and entrepreneurs.
The student will be involved in several stages of the process in all these projects.
For all projects, the student will participate in the search of and analysis of the data (some of which will be directly collected by us), using programming softwares such as Stata or R. First, s/he will help us generate some important descriptive statistics and key facts based on the data, and then will help us evaluate the causal linkages between different variables in the data. There will typically also be cross-countries comparisons of the results. Second, the student will help us produce presentation materials for academic conferences that we will attend and will help us editing the research paper.
At every stage, the students will interact with me to discuss his intermediate findings, potential concerns and ideas for future explorations.
The project includes a variety of tasks that provide preparation for graduate school, such as analyzing data, monitoring a survey, writing and creating presentations, and editing research papers. It is therefore a unique opportunity to strengthen one’s research skills, particularly for students interested in questions regarding inequality, taxation, redistribution, and innovation. The student will evolve in a team-oriented environment where he will be able to observe first-hand how a researcher proceeds. Rigor and creativity will be encouraged in the analysis of the data collected and the student will be able to discuss his intermediary findings with me. Interacting with faculty members regularly will help the student build connections and make sure that research is the career path he/she wants to take.
Gauging young elites’ affinity with America’s middle class | John Donahue (HKS)
The most important attributes are curiosity, resourcefulness, and creativity. Facility with Stata or another statistical package is a plus, but good Excel skills should probably suffice.
Based on a few systematic data sources as well as informal observations while co-founding Harvard’s joint degree program in business and government and chairing the Masters in Public Policy degree, I suspect that young American elites are becoming no less idealistic but more global—and less narrowly national—in their affinities. This trend, if true, would be in many ways commendable. But it has implications for the prospects of many imaginable options for shoring up America’s middle class. The preliminary plan for my next book, which would engage much of my energy in 2017 and 2018, is to explore how to square the restoration of shared American prosperity with realism about the cosmopolitan sentiments of many of the most talented young Americans.
Before leaning into a major effort to deal with the implications of the trend, I need to make sure that it is a trend and that I am not making unwarranted generalizations from a few hundred Harvard graduate students. If there is indeed a meaningful evolution in the sentiments and ambitions of young American elites there will be evidence to be found. I propose to devote the summer of 2016 to seeking such evidence. This will require approximately half of my time—to brainstorm over potential data sources, provide detailed research direction, and assess and interpret the evidence as it accumulated—and most of the summer for a resourceful research assistant. Likely sources include survey research, Bureau of Labor Statistics and Center for Education Statistics compilations, literature reviews, higher-education institutions’ data on choices of majors and first jobs, and other obvious hunting grounds. I expect a creative student will be able to identify data sources that have not yet occurred to me.
I anticipate three kinds of activities—with the third category very broad. For all of them I would be working closely with the student researcher, but expect that my time involvement will be considerably less as I will have some other obligations during the summer.
- Accessing and processing known data sources that may offer some purchase on trends in the beliefs, values, and behaviors of young American elites. These will likely include official statistical assemblages such as the Department of Education’s National Center for Education Statistics as well as survey research including the Harvard Public Opinion Project.
- Reviewing relevant literatures, including public opinion, economics, sociology, and political science both to identify existing work relevant to the broader research project and to flush out potentially useful data sources employed by other scholars.
More generally, exploiting what we learn in the first two categories to find reliable evidence of shifting elite affinities—or to be able to declare with reasonable confidence that such evidence does not exist. By a few weeks into the summer the researcher and I will know a great deal more than I do now about prior work in related areas and about where to look for high-grade empirical ore. Throughout the effort creativity and good judgment will be required for making valid inferences, since “elite young Americans” is no kind of standard category.
The researcher would learn about framing an empirical strategy to validate (or not) a big, broad supposition. This would be useful in most thesis projects. She or he would also likely gain some familiarity with a number of standard statistical sources (including the Bureau of Labor Statistics, National Center for Education Statistics, and Current Population Survey) as well as the major repositories of opinion-research data.
Suicide and NIMH Research Domain Criteria | Matthew Nock (FAS Psychology)
Ideally, students should have classroom-based course experience that has provided an introduction to the theories and methods that guide research in clinical psychology, such as Introduction to Psychology and Abnormal Psychology. We expect to build on the foundation students received in those courses by applying what was learned to the conduct of research on psychopathology and other behavior problems. Becoming familiar with the RDoC matrix is strongly encouraged.
Suicide is a leading cause of death in the US and around the world; yet, we do not yet understand who is most at-risk, and what factors cause suicidal behavior. We have just begun working with the National Institute of Mental Health to perform a massive literature search and categorization of research into the RDoC (Research Domain Criteria) matrix. This is a new framework for studying mental disorders, looking at different levels of information (Genes, Molecules, Cells, Circuits, Physiology, Behavior, Self-Reports, and Paradigms), with the goal of better understanding “basic dimensions of functioning” (nimh.nih.gov/research-priorities/rdoc).
Our goal is to better understand which key constructs, and then what combinations/interactions are most important for predicting the different parts of the pathway to suicide. This is will tailor the framework specifically for thinking about and studying suicide moving forward and provide guidance as we engage in a comprehensive and iterative program of research to test which constructs are involved.
As a BLISS student, you will participate in various aspects of the clinical research process through your active involvement in the research of our psychology laboratory. All students will first complete comprehensive human subjects training to learn about the history of research ethics as well as appropriate ways to interact with human subjects and the protection and careful use of research data. Through attending weekly lab meetings, you will learn how: research ideas are generated and specific hypotheses formed, research studies are designed to test these hypotheses, various research methods are used depending on the question addressed, data are collected and analyzed, and results are presented at meetings and in scientific publications. By participating in the RDoC project, you will gain an appreciation for how the field of psychological research is being re-shaped, a better understanding of research findings, and a greater ability to synthesize and describe scientific results.
The work that you would do on the RDoC project specifically will involve:
- Assisting with literature searches and retrieving PDF files of relevant papers
- Evaluating papers for inclusion in our RDoC database
- Learning to code and enter findings from these papers into the RDoC database
- Participating in the reporting of our process and the resulting matrix
The experience would be especially useful for students planning on pursuing a career in clinical psychology and/or psychological research. The program is designed to provide a well-rounded, hands-on introduction to all aspects of research in the realm of clinical psychology. The skills learned (e.g., critical thinking, research design, statistical analyses, science writing) will be useful more generally, and so students not planning on a career in clinical psychology research would also benefit from participating in the BLISS experience with our lab.
Bombing the way to state-building? Lessons from the Vietnam War | Melissa Dell (FAS Economics)
Some experience with programming – in a language such as Python, R, or C – is required. This could be acquired by successful completion of CS50, or through past work experience. Completion of an introductory statistics or econometrics course is preferred but not required. Vietnamese language skills are not required.
My research examines long-run economic development, seeking to better understand the massive divergence in global incomes that has occurred over the past two centuries. While average incomes in some regions have grown by more than 20-fold, other areas have stagnated, with poverty and insecurity remaining pervasive. In particular, I focus on the role of political and institutional variables in generating persistent economic disparities. Methodologically, my work relies on econometric and statistical modeling, the collection of detailed archival data, and in depth knowledge about the local and historical context.
Interventions in weakly institutionalized societies, such as Vietnam and Afghanistan, have been central to U.S. foreign policy over the past half-century. These have been amongst the most costly expenditures in the U.S. federal budget and may have important national security consequences. The United States has employed a variety of strategies to defeat insurgents and strengthen states, ranging from military force to development aid to information and propaganda campaigns. Yet rigorous causal evidence about how different strategies influence security, governance, civic engagement, and economic outcomes remains limited. The project identifies the causal impacts of key interventions employed during the Vietnam War. It examines the bombing of civilian population centers by exploiting discontinuities in an algorithm used to target air strikes. The impacts of persuasion campaigns are identified using a similar approach that exploits discontinuities in the algorithm used to target different propaganda themes. Finally, the study exploits a spatial discontinuity to identify the effects of the war's largest development aid program.
The Vietnam War provides an excellent laboratory in which to examine questions of central importance to state building and counterinsurgency. More than twice as many tons of explosives were dropped during the conflict as during World War II, the psychological operations campaign was the largest in history, dropping over a billion leaflets in 1969 alone; and 12 billion USD of development aid were deployed, one of the largest aid allocations to a single country in U.S. history (Butterfield, 2004). Spending on the Vietnam War during the Lyndon Johnson administration was seventeen times greater than on the War on Poverty, and the Department of Defense (DoD) estimated in 2003 that the Vietnam War has cost US taxpayers over one trillion USD (Appy, 2015). Importantly, the United States utilized quantitative metrics for resource allocation to an unprecedented extent, spurred by the systems analysis perspective that Secretary of Defense Robert McNamara brought to the DoD, resulting in rich data and unique opportunities for causal identification.
Last summer, two BLISS fellows worked on identifying the impacts of bombing on state building, with a focus on air strikes that hit civilian population centers. Estimates document that the bombing of South Vietnamese villages led more Vietnamese to join the communist insurgency; worsened security conditions; and weakened participatory governance, public goods provision, and civic engagement. The results suggest that the bombing of civilian population centers facilitated Viet Cong recruitment, in turn increasing insurgent attacks on government sympathizers. This plausibly led to a general decline in law and order, public goods provision, and civic engagement. In contrast, the development aid program improved access to health care and increased primary school completion rates. While economic outcomes and civic engagement were not affected, the program plausibly improved security and reduced participation in the communist insurgency. These results cast considerable doubt on the common Vietnam War Era adage ``get the people by the balls and their hearts and minds will follow'' (Kodosky, 2007). Instead, results are more consistent with the views of James Scott (1998, 2009) that citizens have many ways in which to undermine a state that they do not trust, which can lead top down approaches to state building that employ force to backfire.
This summer, the BLISS fellow will work on estimating the impacts of persuasion campaigns employed during the Vietnam War. These questions are highly policy relevant, as the U.S. seeks to revamp its persuasion campaign to counter ISIS and other insurgent groups. Through persuasion, the state aims to change citizens' perceptions about the costs and benefits of supporting state and non-state actors. Modernization theory hypothesizes that a critical step in state building is to change the way citizens think about the state, relative to more traditional authorities (Weber, 1976), and psychological operations were central to the Vietnam War. The majority of efforts were targeted towards encouraging Viet Cong to defect, through leaflets, loud speakers, mobile information teams, and other methods of persuasion.
A natural experiment will exploit discontinuities in the scoring used to target psychological operations to identify causal effects. The Joint U.S. Public Affairs Office identified different propaganda themes, such as ``anti-VC'', ``Chieu Hoi [defect]'', ``people's self defense'', ``local administration'', and ``prosperity for all''. Priority rankings for 14 themes were calculated, using a different algorithm for each theme to combine different district level characteristics. Districts with the highest rankings on each of the priority indicators were then treated with that theme during the following quarter, and the discontinuity in whether the district's theme ranking was above or below the treatment threshold can be used to identify the causal effects of propaganda. For example, if the top 100 districts on the priority ranking for theme i were treated with propaganda for i, the district ranked 100th would be treated whereas the district ranked 101th would not, creating as-if random variation that can be used to identify causal effects.
If time permits, the BLISS fellow will also examine the bombing of insurgent supply lines. In these cases, the air strikes did not typically hit population centers, and the project will examine whether the impacts of this type of bombing are different from those of air strikes that hit populated areas.
The BLISS fellow will focus on data analysis and modeling. Depending on the skills and interests of the student, analyses may including data plotting and regression in R or Stata, programming of statistical models and simulation exercises in R, geospatial modeling and mapping using Python and ArcGIS, data parsing using Python, or batch optical character recognition using Python and ABBYY.
At the end of each day, the BLISS fellow will be expected to send the PI a summary of the work completed that day. At the beginning of each day, the BLISS fellow will discuss the work for that day and any difficulties met during the previous day with someone from the project team. The PI or an experienced research assistant will be available at all times to answer questions.
During the summer, my team of research assistants works together in a collaborative environment that includes weekly group meetings for all team members, in which insights are shared across projects. There are also project-specific meetings between research assistants, graduate students, and myself on a near-daily basis. Last year, we had 20 full and part-time research assistants and interns from top universities in the U.S., Asia, and Latin America who regularly attended these meetings. Overall, this fosters a highly collaborative environment where students learn both from each other and from interactions with myself and other faculty collaborators.
This research experience will provide extensive opportunities for the BLISS fellow to hone his or her problem-solving skills. The student will gain experience developing simple prototypes to test various data analysis approaches. There will be an emphasis on understanding the underlying mechanics of how the programs that the student writes function. This understanding can then be used to efficiently and accurately debug and improve the programs. The student will also gain experience manipulating large datasets. These are all skills that will be useful for writing research papers or a senior thesis using empirical data, regardless of the student’s field of study. These skills will be particularly useful for students interested in pursuing a career in data science, analytics, or academic research and are more generally applicable to approaching problems in a systematic way.
Petitioning in American history | Daniel Carpenter (FAS Government)
Use of spreadsheets and willingness to learn new software will be both essential. Knowledge of U.S. and/or North American politics and history will be helpful. Although it is not necessary, knowledge of either French and/or Spanish may also be helpful.
A research team headed by Daniel Carpenter, Freed Professor of Government, is examining the diffusion of petitioning in American history. Our focus is both “micro” (how individual petitions were put together) and “macro” (how petitions in greater numbers were sent to Congress and state legislatures). We will collaborate with undergraduate researchers to examine both of these processes, focusing on slavery-related petitions at first and broadening our scope to include petitions on a range of subjects and petitions up to the present day.
The petition stands as one of the most vital institutions and expressive patterns of American culture and American democracy. Petitions comprised a critical vehicle for the expression of resistance during the American Revolution; they were common means of communicating across cultural boundaries; they were central modes of expression and organization in dozens of social movements ranging from temperance and anti-Sabbatarianism, to women’s suffrage and minority rights, to anti-slavery and anti-segregation campaigns.
Reflecting this history, the amount of petitioning in American history appears to be vast. Historical records suggest that, for the first 150 years of the American republic, petitioning was an incredibly widespread and common practice in everyday life for millions of Americans, a practice that welcomed the energies of African-Americans, Native Americans and women whose liberties in voting, property-owning, mobility and other forms of expression were severely circumscribed for most of American history. Among many repositories, the petitions and memorials collection of the National Archives and Records Administration for the House of Representatives (from 1789 to 1945 alone, Record Group 233) holds approximately 6,100 cubic feet of petitioning materials. Plausible calculations suggest that this collection alone may contain more than one million petitions, with a billion or more signatures. Taking into account the wealth of petition collections in state archives and other locations, it is plausible that much more petitioning material is available in other venues.
We have three organizing aims for the coming year:
1. Create General Databases of Petitions and Spatial Data on Their Origins and Diffusion. A first goal of the Petitions Project Archive is to create a publicly accessible database of thousands of anti-slavery, finance-related and Native American petitions from the 18th and 19th centuries, so that citizens, teachers and students can access digital images with the click of a mouse. We will create searchable text files of each petition (including its signatory list), so that citizens and scholars can query our database about individual signers of a petition as well as the particular language to which they affixed their names. At least one access interface for the petitions will appear in the form of a map (both of the United States and including state-specific maps for states from which petitions were sent). Within these maps, the petitions will be linked to towns and countries (and statistically geo-coded) so that public users and researchers can examine petitions by the point of their origin.
2. Create Signatory Lists and Map – in Network and Geographical Data – Sequences of Signatures. One of the biggest tasks that we face is transcribing the signatory lists from these hundreds of images of petitions. Once these signatory lists are transcribed the names will then be looked up in ancestry.com. This task will enable us to examine the exact location of the signatories and hopefully gain an understanding of the issues that specific families deemed relevant. Once the background is obtained on the signatory the data will be geocoded and a map will be created.
We will also, in the course of the coming year, be analyzing many petitions with geocoded signatory data. These include some particular antislavery petitions sent to Congress from New York City and Rochester, New York, as well as some Iraq-War petitions from Wisconsin in 2006, where each signatory was identified with her or his address. This petition and others like it will allow us to map the spatial creation of the petition, going from ‘top” to “bottom” of the signatory list (or where papers were pasted together, multiple signatory lists). Using other biographical and city historical data, we can examine which persons signed (and which did not) and consider the role played by networks and civic space in petition canvassing.
3. Examine Contemporary Petitions, including Electronic Petitions. We will supplement these historical petitions with data on contemporary petitions (one of our team members has assembled a database of 2006 petitions related to the Iraq War, also with signatories geocoded). Through partnerships we are developing with political organizations, we are also gaining access to electronic petitions. Modern petitions – whether paper or electronic – also have a significant role to play in politics. In the past few years we have witnessed business policy reversals by Bank of America and Verizon because young people have started online petitions contesting their practices.
Work with petitions like these provides a unique opportunity to work closely with a highly skilled and thoughtful research team. The BLISS fellow-collaborator will be able to see instant results, and Professor Carpenter is known for his outstanding teaching and undergraduate mentorship. For example: The first step is the transcription of the signatory lists of specified petitions. Once completed, the BLISS fellow-collaborator will move onto the phase of conducting genealogical research utilizing the cutting edge tools that are becoming widely available through Ancestry.com. Once completed, the BLISS fellow-collaborator will develop the knowledge to geocode the data so that it will accurately reflect the location of the said movement on the specified petition. The BLISS fellow-collaborator will then search newspapers from the specified time period and location that may provide more insight into the reasoning behind the signatories desire to participate in the movement.
The cognitive and neural underpinnings of punishment | Fiery Cushman & Justin Martin (FAS Psychology)
The only requirement is an interest in psychology and/or cognitive neuroscience. Many interns have a background in cognitive science, psychology or cognitive neuroscience, though this is not necessary. Experience with data analysis and/or computer programming is a plus, though not required. Interns will be taught all skills needed to be successful in this opportunity.
The Moral Psychology Research Lab aims to organize the astonishing complexity of moral judgment around basic functional principles. Much of it is motivated by a simple idea: Because we use punishments and rewards to modify others' behavior, one function of morality is to teach others how to behave, while another complementary function is to learn appropriate patterns of behavior. In general, our research asks questions like: Why do we sometimes punish accidents? And, when do we think it is OK to sacrifice one person for the good of many? We are interested in understanding the cognitive processes that give rise to moral judgment, their development, and their evolutionary history. In answering these questions, we employ behavioral experiments—both in the lab and online—as well as functional neuroimaging (fMRI) and psychophysiology.
Students will have the opportunity to participate in the Moral Psychology Research Lab’s summer internship program. This program is focused on giving students a rewarding, full-time research experience. Principally, students will have an opportunity to be involved in all aspects of research: Conceptualizing relevant research questions, designing experiments targeted toward specific research questions, conducting behavioral and neuroimaging experiments, analyzing data (both behavioral and neuroimaging) and/or data write-up. Because projects are at various stages of development, students can participate in both beginning and ending stages of research projects simultaneously. In addition to conducting research, students will participate in weekly discussions with mentors and other interns motivated by current papers on moral psychology designed toward developing a sophisticated and critical eye for evaluating research.
Specific projects that students may be involved with include:
Pedagogy and punishment: From an evolutionary standpoint, why do we punish others? What causes me to want to harm a stranger who pushes into me on the street, even though I’ll never meet them again? One reason why evolution might have favored a psychology of punishment is to teach others: That is, we punish others so they treat us better. This project will focus on whether or not pedagogy plays a role in punishment. Using both in-lab and online methods, we will construct situations in which participants either have an opportunity to learn or teach through punishment.
The effortful nature of moral judgments: Moral decisions largely depend on two sources of information: Intentions (e.g. What an agent is trying to do?) and outcomes (e.g. The good or bad consequences that an agent brings about). A large, on-going project within the lab is to understand the cognitive and neural processes that operate over representations of these two factors. That is, we seek to understand how people process information about intentions and outcomes and incorporate that information into a moral judgment. One aspect of this large project focuses on the degree to which processing information about intentions versus outcomes requires different levels of deliberative, effortful processing. Understanding this will potentially shed light on the different functional goals that these processes are targeted towards.
The neural representation of causality: When deciding whether someone who causes harm is worthy of punishment, one factor that influences our decision is how causally responsible they are for harm. Is there a clear connection between them and a bad outcome? This study will investigate how people process sequences of causal events. That is, we are interested in understanding how the brain makes sense of the ways in which people cause events to occur. We will use fMRI to investigate what areas of the brain process causality, how this is accomplished and how these regions interact with regions responsible for making punishment decisions.
Punitive encouragement of cooperation: At the broadest level, punishment has the potential to motivate cooperation among unrelated and anonymous strangers. Specifically, one reason why evolution might have favored a psychology that supports punishment is because punishment can encourage others to be helpful and cooperative, benefitting our entire species. Here, using behavioral economics and game theory, we investigate the extent to which punishment can lead to increased cooperation and what factors influence whether punishment will be a successful strategy or not.
Across all of these projects, daily activities may involve:
- Conduct literature searches under the direction of their mentor
- Create experimental stimuli
- Recruit study participants and run behavioral and online studies
- Assist in running functional neuroimaging (fMRI) studies
- Participate in a weekly Reading Group to talk about 1-2 journal articles with other interns, while 1-2 research mentors moderate the discussion.
- Participate in weekly Lab Meetings where they will watch presentations, hear in-depth discussions of theory, methods, etc., and see part of a typical day-in-the-life of a grad student.
- Participate in regular meetings with their mentors.
- Make a final presentation to the Moral Psychology Research Lab, detailing the research they conducted over the summer and providing an interim summer of the findings as well as potential future directions.
Getting involved in research as an undergrad is an invaluable experience, one that all members of the Moral Psychology Research Lab had. Understanding the scientific process and seeing one’s own ideas tested and brought to life is exciting and important, both for those interested in science as a career and those who do not envision such a path. Our goal with the summer internship program is to instill an appreciation for how science is conducted, the benefits of a scientific mindset and the joys of research. The entire lab has benefited from exceptional experiences getting deeply involved in science early on in our careers and we seek to give that opportunity to others.
Harvard Decision Science Laboratory internship | Nina Cohodes and Gabe Mansur (HKS)
Undergraduates from all disciplines are welcome to apply as we support a diverse pool of investigators. We do recommend that students have taken an introductory course in economics and/or psychology. Interest in decision science in general as well as willingness to jump into the variety of different projects that are likely to be running in the summer is important. The great opportunity of working for the Decision Science Laboratory is that students here can learn skills they want to acquire. We provide structured training in the development of experimental methods and experience in conducting experiments.
The Harvard Decision Science Laboratory is a world-class, university-wide biobehavioral research facility. Our lab supports faculty investigators from across Harvard. Our researchers have come to us from the departments of Psychology, Economics, Government, and Linguistics; the Business School, the Kennedy School, the Law School, the Medical School, the School of Public Health, the School of Engineering and Applied Sciences, the Design School, and the Radcliffe Institute. We are an emerging hub for the conduct of experimental research in the behavioral and decision sciences.
If you are interested in experimental research but haven’t yet had a chance to work on an experimental team… If you are curious to know how behavioral researchers are exploring cognitive biases and other factors that predictably—and unconsciously—shape our decisions… If you want to work with faculty and other researchers from a variety of different disciplines and schools across Harvard in order to explore different topics and research questions in behavioral science… Then you should be a BLISS Fellow at the Harvard Decision Science Laboratory (HSDL).
Who we are
Researchers from all across the university, and at all levels of the scholarly community — from senior faculty to undergraduates — use our facility to investigate how emotion, neuroscience, and cognitive processes combine to shape human judgment and decision-making. Researchers who do work here include such folks as Amy Cuddy and Francesca Gino at Harvard Business School, Joshua Greene in the department of Psychology, Dustin Tingley in the department of Government, and Günther Fink in the School of Public Health.
Our laboratory provides experimenters with the means to monitor specific physiological signals (heart rate and heart-rate variability, thoracic blood flow, blood pressure, peripheral temperature, respiration rate, skin conductance, electromyography) as well as neuroendocrine processes (through salivary assays); to relate these signals to specifically induced affective states; and to pursue specific research hypotheses as to how these relate to risk perception and risk tolerance, evaluation of alternatives, and choices of action.
HDSL also maintains a large subject pool as a resource to experimenters conducting work here. Participants in the pool can take part in both in-lab and online studies conducted through HDSL. We are constantly seeking ways of improving both the size of, and variety within, our subject pool.
What you’ll do
BLISS Fellows at HDSL will be assigned to work alongside investigators undertaking research in the lab during the summer. The months of the summer can be busy in the lab, as they offer researchers an opportunity to conduct experiments free of the obligations and distractions of the academic year.
BLISS Fellows will have an opportunity to work with experimenters in the conceptualization and design of their experiments; in programming these designs using standard tools for subject interaction (for example, MediaLab, z-Tree, E-Prime, MatLab, Qualtrics); in conducting the experiments and interacting with subjects in the lab; and in compiling and analyzing experimental data.
In addition, as part of their basic training in the tools of this laboratory, as an HDSL BLISS Fellow you’ll be trained on the use of physiological monitoring systems for behavioral experiments. This “Physio Boot Camp” involves training in the correct placement of sensors, assuring the clarity of a signal for data collection, skills for respectful interactions with subjects, and use of software tools (specifically BioLab) for the analysis of physiological data. Core Competency Training. Through online and in-person trainings and work experience in the laboratory, the HDSL training program for BLISS Fellows is designed to help develop some or all of the following core competencies:
- Ability to read, critically engage, and coherently discuss scientific journal articles
- Ability to synthesize findings from published studies and develop an annotated bibliography
- Understanding the purpose of, function of, and recruiting for a subject pool
- Ability to use psychophysiological sensing systems, reduction of physiological data into analyzable form
- Ability to design experiments using software tools (e.g., Qualtrics, MediaLab)
- Experience in submitting experimental design for review and approval by IRB
- Demonstrated capability to perform experimental studies
- Ability to perform basic data analysis using statistical modeling and analysis software
- Ability to translate experimental data into graphical representations
- Ability to construct and deliver oral presentations of experimental science using such tools as Excel and PowerPoint
Politics, policy, and public opinion | Jennifer Hochschild (FAS Government and African & African American Studies)
Ability to identify, summarize, and evaluate academic literatures (after guidance on how to find relevant materials). The mental discipline to develop a coding scheme (with guidance) and use it to develop systematic analyses of media and other reports. Ideally, quantitative and software skills to develop a Qualtric survey, implement it on MTurk, do basic analyses of the results, and write up the findings.
Politics and ideology of genomic science. Based on a national survey, elite interviews, coded media reports, and other evidence, my co-author (Maya Sen, of HKS) and I are writing a book about the ways in which Americans are coming to understand, evaluate, and use genomic science in the public arena. We focus on three arenas: (a) use of genomics to explore the meaning of “race,” group categories, and individual ancestry; (b) use of genomics in the criminal justice system through forensic biobanks, familial searching, and DNA tests aimed at exoneration; and (3) use of genomics in the medical and science arena to understand hereditary traits and diseases, and to diagnose or prescribe treatments where relevant. The medical arena includes pre-natal and post-natal testing, biobanks, and research on genetic contributors to everything from aggression to mental illness.
We are studying the views and preferences of the American public, governmental actors, the media, and scientific experts. The book is organized around the political and policy implications of the intersection of two dimensions: technology optimism vs. pessimism, and genetic determinism vs. environmental causes or free will. There are three projects to complete over the summer:
- The BLISS student will help us with elite interviews. We have identified most of the subjects; the tasks are to help me set up, do background research on, and participate in the interviews themselves, at least those (the majority) in the Boston area. The RA will also transcribe the interviews, and help me develop a coding scheme for evaluating the 75 or so completed or new interviews. This may involve use of qualitative coding software.
- The BLISS student will do one of two literature reviews (we have not decided which—it could depend on the student’s preference). One is to organize and make sense of various disciplines’ use of the concept of “technology optimism/pessimism,” a.k.a. “risk tolerance/aversion.” A variant of this concept appears in psychology, sociology, economics, cultural studies, even political science – but each set of scholars understands something slightly different and uses a different evidence base. We need a clear map of the various uses and empirical studies. Alternatively, the second literature search would be on various disciplines and media treatment of the issue of heritability of traits and behaviors. Without going too far into the details of the science itself, we need a clear mapping of research on whether such phenomena as aggression, violence, sexual orientation, intelligence, musical ability, athletic ability, obesity, drug addiction, and alcoholism are inherited, or responses to social conditions, or in some sense freely chosen. In this lit review, we also are interested in the ideological valence of various authors or positions on these sensitive questions.
- The BLISS student will help to develop a new survey focused around the topic of genetic determinism/social construction/free will. We will apply for funding to do the survey once we have a basic design – so the student will search iPoll and other sources for extant questions, help us to write new questions, test question wording using Qualtrics software and MTurk samples, and do basic analyses of the resulting pretest.
Political and policy disputes within racial or ethnic groups. This will involve a survey of ten metropolitan areas in the United States (if funded – application is pending), and case studies in four large metro areas. It is co-authored with Vesla Weaver (Yale University). The issue is how intra-group differences, especially but not only involving class-based inequality, are playing out in specific policy and political contexts. We are studying policing in New York City, pension reform in Chicago, gentrification in Atlanta, and treatment of undocumented immigrants in Los Angeles. The survey will examine perceptions of discrimination, views about social and tax policies, views on local political disputes, and perceptions of other groups and classes; it will be evenly divided among non-Hispanic whites, blacks, and Hispanics of all races (English and Spanish speaking). The case studies will involve several weeks in each metro area after preliminary research, with interviews, attendance at meetings and other events, and examination of pertinent documents.
The goal of this project is to understand and explain when, how, and why disputes within racial or ethnic groups become important in the context of traditional patterns of inter-group conflict. We will also seek to explain electoral outcomes, policy patterns, and institutional developments as affected by class and other conflicts within racial or ethnic groups.
The central task will be to help us prepare for, and to participate in (if desired), site visits to the four metro areas of study. That is, the BLISS student will evaluate media reports, academic studies, and organizational data on the four policy arenas described above (policing, pension reform, gentrification, treatment of undocumented immigrants) in the four relevant locations. The student will help to identify appropriate subjects for interviews, and do background research on those subjects. The student will also try to identify events that we should attend while in the case study metro areas, and other forms of documentation or evidence that we should examine. If it is feasible and desirable, the student will accompany Prof. Weaver or myself on some or all of one or more site visits – to help with interviews, make detailed observations and field notes, and otherwise share in the research task.
If the student has quantitative analytic skills, as we hope, he or she can help to analyze several surveys that we have already conducted on this topic, and/or help us finish setting up and beginning to analyze the new survey for which we are now requesting funding.
Uses of history for policymakers | Graham Allison (HKS)
A researcher should have an active interest in history, and academic background in history, political science (more qualitatively focused than quantitative), or philosophy.
Applied History is the explicit attempt to illuminate current challenges and choices by analyzing historical precedents and analogues. Mainstream historians begin with an event or era, and attempt to provide an account of ‘what happened’ and why. Applied historians begin with a current choice or predicament and then analyze the historical record to provide perspective on the issue, find clues about what is likely to happen, suggest possible interventions, and assess probable consequences.
In the earliest days of historical writing in ancient Athens, Thucydides and other writers considered history a vital teacher for statesmen. Into the nineteenth and twentieth centuries, historical knowledge maintained a central place in strategy and statesmanship, inspiring such leaders as Bismarck, Churchill, and Kissinger. In recent years, however, this tight history-policy relationship has begun to deteriorate. Policymakers have retreated to increasingly facile analogizing (think repeated references to Vietnam and the Appeasement at Munich), while historians have become distrustful of the policy community’s misuse of their craft.
Housed at the Belfer Center for Science and International Affairs, our project on Applied History is building on the work of the late Harvard professors Ernest May and Richard Neustadt to help rebuild this critical history-policy relationship. The first stage of our research is to develop a methodology that analysts can use to apply historical analogies to current challenges.
A key component of developing a rigorous methodology for an applied history approach to today’s policy issues is a literature review on the subject of how thinkers in the past have conceived of history’s usefulness. Research assistants at the Belfer Center have begun to survey the literature on uses of history, but a proper review of such a broad topic will require more intensive work involving significant reading and research. The BLISS fellow will identify sources (from Ancient Greeks to moderns) who have spoken on the topic of how—and how much—historical knowledge should inform our treatment of current challenges. This research will require a creative approach. Since researchers will find useful intellectual strands in sources as diverse as Herodotus and Tolstoy, and across disciplines, the project will draw on literature, philosophy, historical writing, and even government documents and memos produced by policymakers of decades past.
The BLISS fellow will summarize each source concisely, listing its main contributions to the literature. At the end of the summer, these takeaways will be compiled into an extended research memorandum which can be used by our Applied History team in writing a publication on Applied History methodology. A condensed version of this memo will also be crafted tracing the evolution of thought on “uses of history” over time.
Though much of this project will involve self-directed library research, Prof. Allison and the Belfer Center’s full-time research staff will work with the BLISS fellow to formulate a research plan, find sources, analyze takeaways, and package notes into a useful research product.
The BLISS fellow will also assist Belfer’s researchers with the Center’s Applied History portfolio, and may also assist with work summarizing articles or editing as needed. However, the primary focus will be on library research on uses of history.
The BLISS fellow will gain deep and substantive knowledge about how history has influenced and guided decision-makers over time. The BLISS fellow will also be exposed to the skills needed to conduct thorough research. The fellow’s home base of the Belfer Center is shared by leading thinkers in international policy, and researchers regularly interact with a spectrum of professors, visiting fellows, postdoctoral fellows, and distinguished visitors. The BLISS fellow will receive direct feedback on their work skills, writing, and analysis, and will also be included in activities related to the Applied History project.
The origins of state capacity in East Asia | Yuhua Wang (FAS Government)
The project requires the student to be able to read Chinese. And some quantitative data analysis skills are also helpful.
This project examines the historical origins of state capacity at the sub-national and cross-national levels in East Asia with a focus on China. Using historical, quantitative data, I investigate the impacts of war, crops, and the imperial civil service exam system on contemporary state capacity in East Asian countries and sub-national China.
Responsibilities primarily include collecting data from Chinese sources, reviewing relevant literature, and conducting data analysis. At the early stage, the student might be asked to collect information from Chinese texts, such as yearbooks or gazetteers. The student will also be asked to read relevant literature and write a summary. I will also involve the students in the data analysis in the later stage of the project.
The student will be exposed to every stage of a new research project from research design and data collection to data analysis and writing. I will have regular meetings with the student to communicate about the project, share my ideas, and seek suggestions. I will also involve the student in the data analysis process, so he/she can learn some practical quantitative skills.
New dynamics in urban politics | Ryan Enos (FAS Government)
None, other than being organized and comfortable reaching out to people. Statistical training is plus, but not necessary.
Professor Enos and his team are studying new and changing phenomenon in urban politics. Cities are the centers of innovation in the United States and, in many respects, the center of democratic life - but, despite this, cities are understudied. We are undertaking a large-scale data collection on cities to better understand their politics and policies. This includes a large-scale survey of mayors, collection of administrative data, such as 311 calls, and infrastructure data, such as quality of public transportation.
Research assistants will work with Professor Enos and his team to develop strategies for data collection and then will be responsible for the collection and creation of original data sets. This includes the creation, distribution, and analysis of surveys, the collection and analysis of city council meeting records, and the geolocating of constituent service calls.
Students will learn skills in data management and analysis and will be exposed to a growing field of academic research.
Memory, creativity, and imagination in young and old adults | Daniel Schacter (FAS Psychology)
It would be helpful though not necessary to have taken an introductory psychology or related course. Specific skills required or the position will be taught during the summer.
Research in my lab is generally concerned with the nature of memory and how it affects related processes such as imagination, problem solving, and creativity. We are also focused on studying the effects of aging on memory and cognition. To accomplish these goals, we use a combination of cognitive/behavioral and neuroimaging (i.e.,fMRI) techniques.
One of our main summer projects will focus on the relationship between memory and creativity in young and old adults. We have recently carried out experiments showing that brief training in recalling details of previous experiences can boost performance on a subsequent test that measures creative thinking by requiring people to generate unusual uses for common objects. We will be exploring various aspects of this effect in both young and old participants. We also plan to carry out related studies focused on the effects of brief training in recalling details of previous experiences on how young and old participants think about personally worrisome future experiences, based on recent work with young adults indicating that such training can be helpful for processing worrisome future events.
The student’s main role would be to assist in scheduling and running experimental participants and to score and analyze experimental data. The student will also likely assist with fMRI scanning, either in their main project or helping out with an additional project in the lab concerning neural correlates of creativity and memory. To provide context for the research, a reading list of relevant articles and books will be created and assigned so that the student can appreciate the issues at stake in the research. The student will also attend our weekly lab meetings and meet regularly with me and the graduate student and/or post-doctoral fellow who is running the project in which they are participating.
The student will acquire a good deal of knowledge about the study of memory, creativity, and aging, and become familiar with the day-to-day operations of a research laboratory. The student will acquire basic knowledge and skills pertaining to the design of memory/creativity experiments, and will also learn new analysis techniques. Exposure to fMRI scanning should help to broaden the student’s research experience. Previous undergraduate participants in our lab’s summer research have found the experience to be extremely beneficial.
Dynamics of national attachment in the United States | Bart Bonikowski (FAS Sociology)
Ideally, the BLISS fellow would have experience with data cleaning and basic statistical analyses in Stata or R (running descriptive statistics and OLS regressions would be a good baseline skill set). Knowledge of more advanced statistical methods and Python programming skills would be an asset. Finally, familiarity with conducting literature searches and producing annotated bibliographies and literature reviews would be helpful.
My work examines the role of meaning-making in politics. Among my primary research interests is the question of how different segments of a given national population vary in their understanding of the nation and how these competing perceptions affect political attitudes and behaviors. I have previously explored this question using cross-national survey data, but am now pursuing three new projects on the topic: a study of the affective dimensions of nationalist beliefs using online survey experiments and physiological data collected in the laboratory setting; a project tracing the daily fluctuations in levels of U.S. national identification using all public tweets posted since 2010; and a mixed-methods study comparing national sentiments across ethnic minority groups in the U.S.
The BLISS participant will have the opportunity to join the project on the dynamics of national attachment. The project uses Twitter data to systematically track daily fluctuations in American collective identification over the period of four years, from 2010 to 2014. The primary descriptive task is to understand what kinds of social, political, and economic events produce spikes in national consciousness over time. Such spikes, however, are likely to vary in their durability and symmetry: some periods of increased national attachment may be more fleeting than others and some may come on more suddenly than others. Understanding this variation and explaining it with reference to different types of nation-relevant events is the second objective of the project. Finally, the degree to which any event produces a surge in national self-consciousness depends in part on how people understand their nation, which in turn is likely to differ across different regions of the United States. Our third objective, then, is to understand the geographic variation in the dynamics of American national attachment.
The internship opportunity is designed to give the BLISS fellow a comprehensive overview of social scientific research. The fellow will have the opportunity to participate in all stages of the research process, including preparing literature reviews, developing a research design, collecting social media data, cleaning and analyzing data, and producing conference presentations and scholarly manuscripts. The fellow will also be given access to past grant proposals and IRB applications used in the development of the project. Extensive mentorship will be provided by me and other students working on the project; at the same time, however, the BLISS fellow will be granted considerable autonomy in his or her work. The research team will meet weekly to discuss the progress of the study. Upon the completion of the BLISS program, the fellow will have the option of applying for a paid research position.
Diversity, justice, and democracy... | Danielle Allen (FAS Government; HGSE)
...Past and present at the Edmond J. Safra Center for Ethics
Any student with a background in ethics, history, political science, political theory, sociology, or journalism would be well suited to this project. Students from other disciplines are also welcome to apply, particularly if they have an interest in questions of justice and diversity. Previous experience working with archives, managing large data sets, multi-media projects, writing for the web, and/or research and analysis of primary documents is preferred but not required.
My research interests are primarily located in democratic theory, political sociology, and the history of political thought. I am currently running a number of projects: a historical and interpretive reevaluation of the Declaration of Independence; a research project on new assessment methodologies for evaluating the humanistic craft; a youth in participatory politics project; and a multi-faceted research program that tackles normative questions of diversity, justice and democracy.
We have a terrific undergraduate research opportunity available at the Edmond J. Safra Center for Ethics, the primary activities for which would include archival research, evaluation, and writing. This project would offer the student a chance to work closely with the Center’s research staff under Danielle Allen’s supervision. The Center is entering its 30th anniversary and is seeking to connect its rich history of research in the areas of philosophical and practical ethics with its current focus on the theme “Diversity, Justice, and Democracy”.
This summer we will be digging into archival materials (written, audio, and video) that have been collected over the past three decades. Among these materials are interviews, discussions, and lectures from the likes of Peter Singer, Richard Dawkins, Elizabeth Warren, Atul Gawande, Jack Abramoff, Elliott Spitzer, and Larissa Macfarquhar. Our student researcher(s) would take the lead on cataloguing the archival materials and on evaluating their quality as potential contributions to current-day debates on topics such as immigration, citizenship, inequality, policing, economic justice, and effective altruism. The student researcher(s) will then produce a research paper (and possibly several short written works, audio-visual pieces, or case studies) that critically engage with the work done at the Center on the broad theme of diversity, justice and democracy. These projects could take several forms, depending on student interest and the research findings. We would help the researcher(s) tie their projects to the current work being conducted at the Center (e.g. workshops on inequalities in cities and randomized control trials, a national project on ethics pedagogy, a large-scale study on policing) and would provide opportunities for them to engage with the Center’s activities after the summer work has been completed.
Successful completion of student projects will be fostered by the Center’s research staff through dedicated weekly meetings and other consultations as needed. The student(s) will be provided with office space at the Center, where they will enjoy easy access to the archival materials, and frequent interaction with fellow researchers and their mentors. Although the student(s) will be supervised closely, there will be considerable freedom to direct their research toward their own interests.
We expect the student(s) will learn a tremendous amount from evaluating the rich primary source materials housed in the Center’s archives. The student(s) will also benefit from having the chance to develop their own research questions, gain insight into ethical analysis of pressing societal dilemmas, research and write articles for public and academic consumption, work as a part of a collaborative research team with a common purpose, and receive close mentorship and guidance throughout the research process.