Thesis advisor: Thomas Wesner / In the wake of events like the COVID-19 pandemic, the storming of the Capitol, and the Russian invasion of Ukraine, it’s time to start labeling misinformation for what it is: a negative externality to society. The spillover effects from the proliferation of mis- and disinformation have the potential to negatively impact the institution of democracy, civic engagement, and downstream health outcomes. Put simply, to understand the misinformation problem is to understand its complexities, its pitfalls, and its motivations. Taken as a whole, this paper articulates the need for a divergence from conventional economic theory on efficiency to a pro-social, welfare-based approach to internalization efforts. In doing so, this analysis presents a full-scale characterization of misinformation as a negative externality, starting with the reorganization of traditional microeconomic theory, followed by a platform-by-platform evaluation of various internalization strategies and evidence from the literature on the impacts of misinformation, and concluding with a commentary on potential remediation approaches. / Thesis (BA) — Boston College, 2022. / Submitted to: Boston College. College of Arts and Sciences. / Discipline: Departmental Honors. / Discipline: Economics.

Upon interviewing a sample of youths, researchers who study exposure to violence are often faced with a substantial number of reports of exposure to violence and no way to confirm the veracity of these claims. The remarkably high levels of violence exposure reported by young preschool children (Richters & Martinez, 1990), paired with the low concordance between parents and children concerning what violence the child had witnessed and experienced (Howard, Cross, Li & Huang, 1999; Shahinfar, Fox & Leavitt, 2000), suggests that child self-reports of exposure to violence may not be entirely veridical. Since self-reports appear to be the only feasible method of measuring lifelong exposure to violence, determining the possible causes of false reports and investigating possible predictors should be a significant aspect of studying youth exposure to violence. With this thesis, I have investigated the veracity of child self-reports by comparing responses to a life-long exposure to violence questionnaire administered first between 1997 and 1999, and a second time two years later. Inconsistent reports, wherein an initial report of exposure to a particular violent instance was not confirmed by re-reporting at the second interview, were prevalent on a variety of violence-assessment items, including witnessing a murder and being the victim of a physical attack. For instance, when initially interviewed, approximately one in seven Chicago children aged 8 to 17 reported witnessing a shooting at least once during their lives. Alarmingly, 50% of these children did not confirm this instance of violence when interviewed again two years later. In an effort to identify self- or parent-reported characteristics and behaviours predictive of inconsistent responses concerning witnessing a shooting, I conducted several series of Binomial and Multinomial Logistic Regression analyses. Explanatory variables were selected to be representative of two main likely reasons for inconsistent self-reports: misremembering due to forgetting over time and the inaccuracy of children's memories, and misinformation due to an impulsive propensity towards lying or exaggeration. My findings suggest that, in comparison to individuals who confirmed their initial reports of having 'witnessed a shooting', individuals who retracted their initial claims were generally younger, from a higher socio-economic status level neighbourhood, admittedly lacked guilt after misbehaving, desired a lot of attention, and reported less impulsivity as measured by delinquency and behavioural impulsivity scales. / Thesis / Master of Science (MSc)

News consumption is primarily done through online news media outlets and social media. There has been a recent rise in both fake news generation, and consumption. Fake news refers to articles that deliberately contain false information to influence readers. Substantial dissemination of misinformation has been recognized to influence election results. This work focuses on hyperpartisanship in web-searched articles which refers to web searched articles which have polarized views and which represent a sensationalized view of the content. There are many such news websites which cater to propagating biased news for political and/or financial gain. This work uses Natural Language Processing (NLP) techniques on news articles to find out if a web-searched article can be termed as hyperpartisan or not. The methods were developed using a labeled dataset which was released as a part of the SemEval Task 4 - Hyperpartisan News Detection. The model was applied to queries related to U. S. midterm elections in 2018. We found that more than half the articles in web search queries showed hyperpartisanship attributes. / Master of Science / Over the recent years, the World Wide Web (WWW) has become a very important part of society. It has overgrown as a powerful medium not only to communicate with known contacts but also to gather, understand and propagate ideas with the whole world. However, in recent times there has been an increasing generation and consumption of misinformation and disinformation. These type of news, particularly fake and hyperpartisan news are particularly curated so as to hide the actual facts, and to present a biased, made-up view of the issue at hand. This activity can be harmful to the society as greater the spread and/or consumption of such news would be, more would be the negative decisions made by the readers. Thus, it poses a bigger threat to society as it affects the actions of people affected by the news. In this work, we look into a similar genre of misinformation that is hyperpartisan news. Hyperpartisan news follows a hyperpartisan orientation - the news exhibits biased opinions towards a entity (party, people, etc.) In this work, we explore to find how Natural Language Processing (NLP) methods could be used to automate the finding of hyperpartisanship in web searched articles, focusing on extraction of the linguistic features. We extend our work to test our findings in the web-searched articles related to midterm elections 2018.

Dual process theories of mind have been developed over the last decades to posit that humans use heuristics or mental shortcuts (automatic) and analytical (reflective) reasoning while consuming information. Can such theories be used to support users' information consumption in the presence of problematic content in online spaces? To answer, I merge these theories with the idea of affordances from HCI to into the concept of dual process cognitive affordances, consisting of automatic affordance and reflective affordance. Using this concept, I built and tested a set of systems to address two categories of online problematic content: misinformation and filter bubbles. In the first system, NudgeCred, I use cognitive heuristics from the MAIN model to design automatic affordances for better credibility assessment of news tweets from mainstream and misinformative sources. In TransparencyCue, I show the promise of value-centered automatic affordance design inside news articles differentiating content quality. To encourage information consumption outside their ideological filter bubble, in NewsComp, I use comparative annotation to design reflective affordances that enable active engagement with stories from opposing-leaning sources. In OtherTube, I use parasocial interaction, that is, experiencing information feed through the eyes of someone else, to design a reflective affordance that enables recognition of filter bubbles in their YouTube recommendation feeds. Each system shows various degrees of success and outlines considerations in cognitive affordances design. Overall, this thesis showcases the utility of design strategies centered on dual process information cognition model of human mind to combat problematic information space. / Doctor of Philosophy / Over the last several decades, billions of users have moved to the internet for everyday information gathering, allowing information flow around the globe at a massive scale. This flow is managed by algorithms personalized to each users' need, creating a complicated trio of producer-algorithm-consumer. This has resulted in some unforeseen challenges. Bad information producers takes the advantage of system to promote problematic content, such as, false information, termed as misinformation. Personalized algorithms have created filters of what people see oftentimes isolating them from diverse perspectives of information, creating a distorted perception of reality. Augmenting the online technology infrastructure to combat these challenges has become crucial and the overall goal of this thesis. Cognitive psychologists theorize that two cognitive processes are at play when people consume information, also known as dual process theories. Can we design new tools to combat these challenges by tapping into each of these processes? In this thesis, I answer this question through a series of studies. In each of these studies, I combine this theory from psychology with design guides from Human-Computer Interaction to design socio-technical design. I evaluated each of these systems through controlled experimentation. The result of these studies informs ways we can capitalize on users' information processing mechanism to combat various types of problematic information online.

Many methodologies have been used to generate false memories, with the misinformation (MI) paradigm and the Deese-Roediger-McDermott (DRM) paradigm being the most commonly studied. The MI paradigm generates false memories based on retroactive interference across episodes, while the DRM paradigm generates false memories based on semantic similarities across stimuli. Since current research is ambiguous about whether the processes for different types of false memories are similar, the purpose of this project was to compare the neural mechanisms between MI and DRM false memories. We used a novel paradigm to limit methodological differences, while maintaining the defining characteristics of each paradigm. We made ERP predictions for false memories in both paradigms based on four current cognitive theories of false memories: fuzzy-trace theory, spreading activation/monitoring theory, global matching models, and source of activation confusion (SAC) model. We found no LPC, FN400, or N2 neural differences between the two types of false memories. This result is discussed in the context of the theories and the implications about our understanding of false memories. Our results support that there may not be mechanistic differences in false memory recollection when paradigms to produce the false memories are similar. / Ph. D. / Many methodologies have been used to generate false memories (or retrieval of an incorrect detail of an experienced event), with the misinformation (MI) paradigm and the Deese-Roediger-McDermott (DRM) paradigm being the most commonly studied. The MI paradigm generates false memories based on incorrectly described details provided by the experimenter across episodes, while the DRM paradigm generates false memories based on semantic similarities across lists of words. Since current research is ambiguous about whether the processes for different types of false memories are similar, the purpose of this project was to compare the MI and DRM false memories. We used a novel paradigm to limit differences driven by different methods, while maintaining the defining characteristics of each paradigm. The four current memory theories informed our event-related potential (time-locked electroencephalogram) predictions. The four theories are fuzzy-trace theory, spreading activation/monitoring theory, global matching models, and source of activation confusion (SAC) model. We found no late positive component (an ERP component indicating recollective processes), FN400 (an ERP component indicating familiarity processes), or N2 (an ERP component indicating conflict) differences between the two types of false memories. This result is discussed in the context of these theories and the implications about our understanding of false memories. Our results support that there may not be differences in false memory recollection when paradigms to produce the false memories are similar.

Fake news, misinformation och desinformation är begrepp som synts i den allmänna debatten de senaste åren och har använts i såväl politiska sammanhang för att påverka opinion liksom för ekonomiska syften. En särskilt central aktör i spridningen av sådan information är även sociala medier. I kampen mot fake news, misinformation och desinformation utvecklas olika lösningar för att göra det enklare för användare att urskilja sant från falskt. Tidigare forskning påvisar emellertid kontraproduktiva effekter utav flera av de sätt som vilseledande information idag motarbetas på. Ett mer tillförlitligt sätt är därför att främja människors egna psykologiska försvar och därför underbygga den källkritiska förmågan. Källkritik hos människor har också tidigare undersökts, men främst utefter människors förmåga. Föreliggande studies syfte var därför att istället undersöka sociala medieanvändares upplevelser av ett källkritiskt förhållningssätt. För att besvara studiens frågeställningar genomfördes 9 semistrukturerade intervjuer med sociala medieanvändare. Datan analyserades därefter utifrån studiens teoretiska ramverk Theory of planner behavior (TPB), varpå deltagarnas attityder, subjektiva normer, upplevda egna förmåga och intentioner skulle ge en förklaring för de befintliga beteenden som uppvisades. Resultaten visar bland annat att deltagarna besitter en god förståelse för hur källkritik bör tillämpas men att tillvägagångssättet uppfattas som kognitivt belastande och tidskrävande vilket i förlängningen gör att förfarandet främst uppges ske när information anses vara viktig. Framtida lösningar bör således sänka resurströskeln utan att för den delen kompromissa användares analytiska förmåga. / Fake news, Misinformation and disinformation are terms that have been prevalent in the public debate in recent years and have been used in political contexts to influence political opinion as well as for economical purposes. A particularly central actor in the dissemination of such information is social media. To combat fake news, misinformation and disinformation, certain solutions are developed to make it easier for users to distinguish true from false. However, previous research showscounterproductive effects in many of the ways that deceptive information is being fought today. A more reliable way is therefore to promote people's psychological defence and instead underpin critical skills in evaluation of sources. People's ability to critically evaluate the reliability of sources have also been a subject for research, although mainly focused on people's capabilities. The present studies aim was therefore instead to explore social media users' subjective experiences of practicing source evaluation. To answer the studies research questions, 9 semi-structured interviews were conducted with social media users. The data was then analysed with the help of the studies theoretical framework Theory of planned behaviour (TPB), from which the participants' attitudes, subjective norms, perceived behaviouralcontrol and intentions would explain the existing behaviours. Inter alia, the results show that the participants have a good knowledge of how sources should be evaluated although the behaviour is viewed as cognitively cumbersome and time consuming which in the long run purports the behaviour to mainly happen when the information is viewed as important. Future solutions should therefore lower the resource requirements without compromising the users’ analytical capabilities.

Syftet med den här studien är att undersöka vilka konspirationsteorier som förekommer bland medlemmar i tre Facebookgrupper utifrån sex inlägg om Covid-19. Studien jämför även inläggen om Covid-19 i Facebook-grupperna med information från de officiella källorna World Health Organization och Folkhälsomyndigheten, hur informationen från de olika källorna framställs och skillnaderna i informationen mellan dem. Detta bidrar till en bättre förståelse om användningen och olika typer av vilseledande information. Studien utgår från en kvalitativ språklig analys eftersom det sällan förekommer studier om textens innehåll i vilseledande information.  Studien uppmärksammar tidigare forskning om sambandet mellan sociala medier och konspirationsteorier. De teoretiska utgångspunkterna är tre typer av vilseledande information och begrepp såsom social kontroll och selektiv exponering förekommer. Resultatet visar att det förekommer fyra huvudsakliga konspirationsteorier och sex tillhörande konspirationsteorier i diskussioner om Covid-19 i de undersökta grupperna på Facebook. I dessa diskussioner är desinformation mest förekommande. Resultatet diskuteras i jämförelse med information från World Health Organization och Folkhälsomyndigheten. Den vilseledande informationen bildar budskap och tolkningar genom tecken och språk i form av texter. Diskussionerna belyser slutligen att den tidigare forskningen, till viss del, stämmer överens med vår studie samt ger förslag på eventuell framtida forskning om konspirationsteorier. / The purpose of this study is to investigate the conspiracy theories that exist among members of three Facebook groups based on six posts about Covid-19. The study also compares the posts about Covid-19 in the Facebook groups with information from the official sources World Health Organization and the Swedish Public Health Authority, how the information from the different sources is presented and the differences in the information between them. This contributes to a better understanding of the use and different types of misleading information. The study is based on a qualitative linguistic analysis since there are only a few studies on the text's content in misleading information. The study draws attention to previous research on the connection between social media and conspiracy theories. The theoretical starting points are three different types of misinformation and concepts such as social control and selective exposure occur. The results show that there are four main conspiracy theories and six related conspiracy theories in discussions about Covid-19 in the analyzed groups on Facebook. In these discussions, disinformation is most common. The result is discussed in comparison with information from the World Health Organization and the Swedish Public Health Authority. The misleading information forms messages and interpretations through signs and language in the form of texts. Finally, the discussions highlight that the previous research, to some extent, is consistent with our study and provide suggestions for possible future research on conspiracy theories.

It is hypothesized that useful diagnostic information can reside in the wrong answers of multiple-choice tests, and that properly designed distractors can yield indications of misinformation and missing information in algorithmic reasoning on the part of the test taker. In addition to summarizing the literature regarding diagnostic research as opposed to scoring research, this study proposes a methodology for analyzing test results and compares the findings with those from the research of Birenbaum and Tatsuoka and others. The proposed method identifies the conditions of misinformation and missing information, and it contains a statistical compensation for careless errors. Strengths and weaknesses of the method are explored, and suggestions for further research are offered.

This paper explores research on the misinformation effect and hypothesizes a new explanation for the occurrence misinformation effect. Current psychological theories states the misinformation effect occurs when memories are skewed by the presentation of new information after an initial event. This effect has been tested in a multitude of ways, including testing words lists, pictures, colors, and change blindness. Socially, the misinformation effect has been used to explain the inaccuracies of eyewitness testimony. The current judicial system relies on the fallible memory of people and has wrongfully imprisoned numerous innocents. The purpose of this research is to show the misinformation effect is not a problem with memory storage and retrieval, but rather a product of selective stimulus control.

This thesis examines how traditional media could reinforce its position as a reliable source of information, overcoming the huge amount of misinformation in public space. First, the thesis analyzes the concept of fake news, examining different concepts of this expression across American and Anglo-Saxon environments. In the following chapters, the thesis presents a link between trust in media and the spread of fake news, outlines psychological factors that allow for fake news to be spread easily and highlights the role of media literacy. Next chapters list the specific ways in which traditional media can fight against fake news, for example through slow journalism or factual verification. That is how the thesis gets to the three analyzed fact-checking departments, which were established as a part of foreign traditional media: ARD Faktenfinder, BBC Reality Check and RFE / RL / VOA Polygraph.info. These departments were founded between 2015 and 2017 and perceive verification of information as an independent journalistic output. Using semi-structured interviews, the thesis examines how these departments work, how they define fake news, what are their future plans or whether they actively contribute to the development of media literacy. Comparative analysis has shown that none of the departments has...