Search

Introduction: Elevated body mass index (BMI) may be associated with reduced survival in nonmetastatic colorectal cancer (CRC). Whether this occurs directly, or indirectly through treatment-related mechanisms such as capping of adjuvant chemotherapy (ACT) doses and toxicity, is unclear. This thesis aimed to disentangle the effects of BMI, ACT adherence and toxicity on survival using individual participant data (IPD), causal mediation, and meta-analysis. Methods: Data from four randomised clinical ACT trials (MOSAIC, SCOT, CHRONICLE and PROCTOR-SCRIPT [five datasets – SCOT arms analysed individually]), with derivable BMI (at trial enrolment) cycle-level dosing and toxicity data were utilised from the OCTOPUS consortium. Dose capping was defined as < 95% of the expected (full BSA-based) cycle 1 dose. Two ACT adherence measures were calculated: average cumulative relative dose (ACRD: percentage of actual-to-expected cumulative dose (mg/m2 )) and average relative dose intensity (ARDI: percentage of actual-to-expected dose intensity [DI: cumulative dose/treatment duration (mg/m2/week)]). Directed acyclic graphs pre-defined putative causal pathways/confounders. The primary outcome was overall survival (OS). Trial level chemotherapy and toxicity data were summarised by BMI category (Chapters three and four). Two-stage random effects IPD metaanalyses were performed to assess BMI, adherence, toxicity, and survival relationships (Chapter five). Causal inference mediation analysis methods were explored, followed by metaanalysis of direct, indirect, and total effects from the mediation models (Chapter six). Results I (Chapter 3): A total of 7269 patients from five datasets demonstrated obesity incidence ranging 5.0%-22.8%. Cycle 1 dose capping rates increased with increasing BMI categories (ranging 29.6% to 62.2% of obese patients), with evidence of attrition of dosing differences across administered cycles (excluding MOSAIC). Subsequent cycle dose reductions and early discontinuation tended not to be associated with BMI. Overall, mean ARDI and ACRD were lowest amongst obese patients. Results II (Chapter 4): BMI did not appear to be associated with the occurrence of grade 3+ toxicity across the trials. However, there was a tendency for the incidence of neutropenia to reduce with increasing BMI. Additionally, the proportion of first grade 3+ toxicity episode occurring late increased with increasing BMI. However, results were limited by missing data. Results III (Chapter 5): BMI increments of 5kg/m2 were associated with increased dose capping odds (OR (95%CI): 2.70 (2.00, 3.64)) in addition to reduced ARDI (Coef. -1.08% (-1.44, -0.72)) and ACRD (Coef. -1.14% (-1.91, -0.38)), with no demonstrable BMI-grade 3+ toxicity relationship. Increments of 5% ARDI were significantly associated with reduced OS (HR 1.05 (1.01, 1.09)). Conversely, 5% ACRD increments were associated with improved OS (HR 0.94 (0.91, 0.96)), raising the possibility of a small adverse indirect effect of BMI via reduced ACRD. Grade 3+ toxicity was associated with reduced ACRD (-10.37% (-11.77, -8.97)) and reduced OS (HR 1.37 (1.17, 1.61)). The latter effects attenuated on adjusting for ACRD (HR 1.20 (1.02, 1.41)), suggesting partial mediation via ACRD. BMI 5kg/m2 increments were not associated with OS. Results IV (Chapter 6): Meta-mediation demonstrated no significant total effect (TE) of 5kg/m2 BMI increments on OS. However, a significant adverse natural indirect effect (NIE) was demonstrated via ACRD (1% reduction in mean survival time (MST)), with no natural direct effect (NDE). Furthermore, a significant TE of 5kg/m2 BMI increments on both ARDI and ACRD (1% reduction) was demonstrated, with no NIE mediated via toxicity. Finally, the TE of grade 3+ toxicity on OS was a 19% reduction in MST, partially mediated via ACRD (NIE and NDE demonstrated a 9% and 10% reduction in MST respectively). Conclusion: Elevated BMI did not influence survival from CRC despite modest under-dosing. However, results support full BSA-based dosing for CRC patients with a high BMI, without significant additional toxicity risks. Toxicity may contribute to poorer overall survival via pathways both including and excluding ACRD, and hence dosing decisions should account for other toxicity risk factors.

Counterfactual thought describes the process of thinking about alternatives to reality. Counterfactual thought can result in various counterfactual emotions, one of which - regret- has received particular research attention. Regret is the emotion we feel when things would have turned out better than they are now, if only we had acted differently in some way. Thus, regret involves a comparison between reality and a counterfactual alternative. Functional theories of regret argue that experiencing or anticipating regret can promote adaptive behavior, as aversive emotional feedback signals that a sub-optimal decision has been made. Furthermore, thinking counterfactually about how poor outcomes could have been avoided also draws attention to ameliorative actions, which can then be enacted upon when faced with similar decisions in the future. Indeed, many studies have evidenced that experiencing or anticipating regret can promote adaptive behavior, in both adults and children. However, studies of regret have focused primarily on the context of self-interested decision making, i.e. when an individual makes a decision which results in a poor outcome for themselves. I refer to this as intrapersonal regret. However, it is also possible to experience regret about decision that have negative consequences for someone other than ourselves: I call this interpersonal regret. The primary aim of this thesis was to investigate the development of interpersonal regret. Motivating this aim was the suggestion that, if intrapersonal regret informs self-interested decision making process, perhaps interpersonal regret might inform social and moral decision making. In Studies 1 through 3, I describe efforts to design an interpersonal regret task which could overcome some limitations which had been identified in existing regret tasks, while maintaining a similar structure overall. In Studies 1 and 2, 8-9-year-olds reported feeling sadder when they learned that making a different choice would have prevented the poor outcome of a peer, and did so significantly more often in regret trials than control trials (where making a different choice would not have prevented the poor outcome). Additionally, 8-9-year-olds reported feeling sadder significantly more often than 4-5-year-olds in regret trials, as is consistent with developmental changes in counterfactual thinking abilities. However, while this evidence suggested that sadder responses were underpinned by consideration of the counterfactual alternative - what would have happened if they had made a different choice - it was not until Study 3 that 8-9-year-olds reported feeling sadder in such circumstances significantly more often than chance. Having piloted and refined the interpersonal regret measure, its associations with theoretically relevant constructs were examined in Study 4; resource donation and sympathy. In addition to examining interpersonal regret, this thesis also investigated prosocial risk taking (PSRT). PSRT was defined as taking a risk with the primary intention of benefitting a person other than oneself. Importantly, PSRT required a potential cost to one's own outcomes, making it more likely that the intentions underlying such a choice were genuinely altruistic. Few 8-9-year-olds were willing to take a prosocial risk to benefit a peer; those who were did so consistently across trials and were more likely to experience interpersonal regret. Engaging in PSRT was also associated with donating a larger amount of resources to another child, and paying a cost to create equality with another child. These findings indicated our tasks had successfully operationalized a novel indicator of prosociality. Although recent work has discussed PSRT as a theoretical construct, we are the first to propose a measure of PSRT, and to examine its occurrence in children.

This thesis examines the counterfactuals, both syntactic and structural, of Virgil's Aeneid. These are alternative stories presented by speakers (primary narrator and characters) as conceivable but only partly or not at all materialized; they clash with the more visible story of the fated advent of Rome and Augustus. The purpose of the study is to envisage some alternatives to the Aeneid as readable from the actualization, in some cases already under way, of the counterfactuals scrutinized; these include the universe and the text not starting or collapsing, Troy surviving, the Greeks losing the Trojan war, Aeneas repeatedly failing to carry out his mission, and Turnus defeating him. Virgil's counterfactual language and images are examined in relation to a number of obvious sources as well as developments contemporary to Virgil, as relevant to the individual case: Homer (particularly if not constructions in the Iliad: A would have happened, if not B), pre-Stoic and Stoic studies of conditionals, linguistic changes in Latin (uses of the indicative / subjunctive and coordination / subordination), the practice of comparison between pairs of differently actualized entities in Livy and earlier historians (synkrisis) and political language in Horace. The main strands of modern thought on counterfactuality from logical, linguistic, narratological and psychological viewpoints are also taken into consideration. An attempt has been made to examine all syntactic counterfactuals spoken in the Aeneid, including some but not all cases of possible counterfactuals, such as wishes, and two types of structural counterfactuals: those which become such when characters are saved through the diversion of weapons, and those which depict partial Troys. The different types of analyses have been as far as possible integrated. My conclusion is that Virgil was revealing but also deflecting alternative stories to that of the destiny of Rome and Augustus as well as those of Aeneas' career circulating at the time.

This thesis involves the submission of published academic work with a critical commentary, in accordance with the regulations of the University of the West of England on Supervised DPhil degrees. Nine papers are submitted, published from 1998 to 2013. Although their subject matter is diverse, I argue in the critical commentary that the work is concerned with critical organizational history and historiography, counterfactuals and modality, and connections, boundaries and identity. The works submitted are theoretical rather than empirical in nature (hence 'theory' in the thesis title) and are thematically connected either by a focus on practice (academic practice, or in and around organizations more generally; hence 'practice'), or by a concern with connecting separate bodies of theory or disciplinary areas (hence 'boundary work'), or both. There are also common themes connecting the papers, in as much as organizational history - broadly conceived as the connection between the organizational past and present, and its interpretation, representation, and so on - and strategy (traditionally and historically concerned with organizational action which connects the present to the future) remain a major focus. The second clause of the thesis title is intended to reflect these concerns. In the introduction to the commentary I briefly comment on the title of the thesis and its relation to the selected papers, list the papers selected for the thesis, and outline the structure of the commentary. I then discuss in the second major section of the commentary, the background to the studies, their themes, and their originality and significance. In a sub-section, I briefly reflect upon their influence and impact, referring among other things to citation data presented as part of the commentary. I then provide a methodologically informed account of the papers, describing and assessing the extent of research competence displayed, as well as discussing the approaches to theory and theorising in the different papers presented. Finally, I clarify, on a paper by paper basis, my personal research contribution to each of the studies, before concluding the commentary with a final reflection on the work submitted.

This study aimed to examine the psychological effects of considering a move into residential care. It sought to explore the wistful ‘prefactual’ and ‘counterfactual’ evaluation of ‘what if’ and ‘if only’ scenarios. Sanna, Carter, and Small’s (2006) ‘Time, Environment, Motivation, Personality, and Outcome’ (TEMPO) model was applied to investigate whether individuals moving closer in time to a prefactual scenario (a hypothetical vignette about two older adults facing a move towards residential care) express increased prefactual/counterfactual statements. Additional hypotheses explored the impact of personality and outcome (mood as input) factors on prefactual/counterfactual statement generation. This study employed a naturalistic experimental design. The main grouping variable was each participant’s life stage (working-age adults or older adults). These two groups were selected to represent two distinct stages along the TEMPO timeline. The dependent variable involved the number of written prefactual/counterfactual statements. In total, 33 working-age adults and 33 older adults completed the study. Each wrote what they thought could be better or worse about each character’s position in the prefactual scenario. They also completed relevant demographic information and information about personal circumstances, along with a range of personality measures. Independent-Samples T Tests revealed a significantly higher number of prefactual/counterfactual statements generated by the older adult group for the prefactual scenario. Effect sizes were medium to large. However, tests of personal scenarios, and the effects of personality and outcome did not reach significance. The implications of these findings, in relation to research and clinical work, were discussed. This was particularly in relation to furthering the investigation of prefactual and counterfactual generation and in relation to the significant emotional implications of considering a move away from independent living. The limitations of this research were discussed and related to future research possibilities, particularly concerning the potential impact of prefactual and counterfactual thinking on behaviour.

Lewis's counterfactual analysis of causation starts with the claim that c causes e if ~ C > ~ E, where c and e are events, C and E are the propositions that c and e respectively occur, ~ is negation and > is the counterfactual conditional. The purpose of my project is to provide a counterfactual analysis of causation which departs signigicantly from Lewis's starting point, and thus can hope to solve several stubborn problems for that approach. Whereas Lewis starts with a sufficiency claim, my analysis claims that a certain counterfactual is necessary for causation. I say that, if c causes e, then ~ E > ~ C - I call the latter the Reverse Counterfactual. This will often, perhaps always, be a backtracking counterfactual, so two chapters are devoted to defending a conception of counterfactuals which allows backtracking. Thus prepared, I argue that the Reverse Counterfactual is true of causes, but not of mere conditions for an effect. This provides a neat analysis of the principles governing causal selection, which is extended in a discussion of causal transitivity. Standard counterfactual accounts suffer counterexamples from preemption, but I argue that the Reverse Counterfactual has resources to deal neatly with those too. Finally I argue that the Reverse counterfactual, as a necessary condition oncausation, is the most we can hope for: in principle, there can be no counterfactual sufficient condition for causation.

Counterfactuals are alternative outcomes to past events. Curiosity for the counterfactual acts as an important driver of learning under uncertainty and helps to improve on past behaviors. This experiment observed counterfactual information collection rates during times of uncertainty and also assessed the willingness of nonhuman primates to gain information when there was a cost or benefit associated. Humans (Experiment 1), capuchins, and rhesus macaques (Experiments 1 and 2) were presented with a three-choice gambling task that consisted of hidden and visible reward values. When choosing visible reward values, participants could view the counterfactual information associated with the gamble. Only humans collected counterfactual information at levels different than chance. Humans also collected counterfactual information significantly more than capuchins. When monkeys saw trials in which the counterfactual information had a greater or lesser reward value compared to the noninformative option, the difference between the reward values was the sole predictor of choice.

The present study provides an example of implementing the difference in differences (DD) estimator for a two-group, pretest-posttest design with K-12 educational intervention data. The goal is to explore the basis for causal inference via Rubin's potential outcomes framework. The DD method is introduced to educational researchers, as it is seldom implemented in educational research. DD analytic methods' mathematical formulae and assumptions are explored to understand the opportunity and the challenges of using the DD estimator for causal inference in educational research. For this example, the teacher intervention effect is estimated with multi-cohort student outcome data. First, the DD method is used to detect the average treatment effect (ATE) with linear regression as a baseline model. Second, the analysis is repeated using linear regression with cluster robust standard errors. Finally, a linear mixed effects analysis is provided with a random intercept model. Resulting standard errors, parameter estimates, and inferential statistics are compared among these three analyses to explore the best holistic analytic method for this context.

Dopamine in the nucleus accumbens is an important neural substrate for valuation and decision-making. Dominant theories generally discretize and homogenize decision-making, when it is in fact a continuous process, with evaluation and re-evaluation components that extend beyond simple outcome prediction into consideration of past and future value. Furthermore, individual animals use distinct strategies to achieve their goals, requiring different computational processes. Extensive work has examined mesolimbic dopamine in the context of reward prediction error, but major gaps persist in our understanding of how dopamine tracks imagined past and future rewards to influence decision confidence. Moreover, there is little consideration of strategy-dependent differences in value processing that may shape dopaminergic encoding. In the studies presented in this dissertation we used an economic foraging task in mice, and found that strategy-specific dopamine dynamics reflected decision confidence during evaluation, as well as both past and future counterfactual value during re-evaluation. We found that inhibition of dopamine terminals altered counterfactual processing during re-evaluation. Individually-tailored optogenetic stimulation of mesolimbic dopamine terminals altered decision confidence during evaluation and carried over to counterfactual re-evaluation, in a strategy-specific manner. We provide evidence that mesolimbic dopamine is tightly linked to decision confidence and counterfactual information, through signals that go beyond reward prediction errors to more complex encoding of imagined past and future value.

For U.S. foreign policy, the long decade of 1897-1909 was one of the most consequential eras in U.S. history. The United States was vastly expanding its involvement in the world. It acquired several overseas territories, established a large presence in the Pacific, brought down a European power, deployed thousands of troops to fight wars in Asia for the first time, began building an interoceanic canal, consolidated its power over the Caribbean, tripled the size of its military, and deepened its diplomacy outside of the Western hemisphere. This dissertation explains why by analyzing six major presidential decisions of William McKinley and Theodore Roosevelt: the annexation of Hawaii, the pursuit of war against Spain, the annexation of the Philippines, intervention in Panama, hosting the Portsmouth Peace Conference, and supporting and negotiating the Algeciras Conference. In doing so this dissertation proposes and applies a new interdisciplinary method for explaining historical decisions, drawing from history, international relations, and decision science. It categorizes explanatory factors into two groups: “decision-generating factors” (the forces that enabled and put the decision on the policy agenda) and “decision-making factors” (the forces that swayed the president’s mind). Within those categories, factors are further labeled “first order” and “second order” to signify the degree of influence. Those are then ranked. This method allows one to compare how various factors mattered in a decision, and to what degree. The primary findings are that systemic political forces, like U.S. power and geography, tended to be the most important decision-generating factors. Events were mostly of secondary importance, affecting the timing and expression of the decision point. Once the decision point arose, McKinley and Roosevelt mattered greatly as leaders. Their choices are explained as much by their interpretation of broader social and political forces as by their peculiar psychologies, like their morality. McKinley emerges as an overlooked leader of U.S. foreign policy. His decisions were more consequential than Roosevelt’s, though they each left their own significant mark in history. This study is the first pairwise, volume-length analysis of the foreign policies of McKinley and Roosevelt.

To estimate the effect of an exogenous intervention on a treated unit, a control unit is necessary. Control time-series are chosen to render threats to internal validity implausible. While a literature review can often reveal the most plausible threats to the internal validity for a particular research design or substantive outcome, literature reviews rarely suggest an appropriate control time-series. Sometimes, appropriate control time series simply do not exist. When an appropriate control time-series is not available, it may be possible to construct a close approximation of an ideal control series from a set of control units that are less than appropriate individually. I describe the statistical theory underlying the synthetic control group design, outline the modeling procedure to construct a synthetic control group, and identify aspects of the model construction process that are in need of formalization. The importance of choosing an appropriate scaling method for the outcome of interest, and the inclusion of pre-intervention observations versus covariates as predictors are discussed at length in Chapters 4 and 5. Beyond those prescriptive conclusions, the early chapters trace the development of causal reasoning, contextualize synthetic control methods within the econometric wave, and serve as an algebraic reference. The dissertation concludes with a discussion of some of some remaining unknowns regarding synthetic control applications, including predictor “trimming”, fuzzy onset treatments, and restricting the analysis timeframe. Guidelines for social science applications of the synthetic control method are outlined, and readers are cautioned to avoid applying the method in a few specific circumstances.

Fake news overwhelmed social media platforms, like Twitter and Facebook, during the 2016 U.S. presidential election. This new, digital brand of fake news that can be spread much more rapidly than older forms, is coupled with a lack of academic research into its effects, the reasons that certain individuals trust its veracity, and methods of decreasing overall belief in fake news. Confirmation bias is one of several reasons why individuals fall victim to fake news; although there are a few strategies that can be used to combat the negative effects of confirmation bias; counterfactual reasoning is one that has demonstrated success. In the current study, participants were assigned to either a counterfactual priming scenario or a non-counterfactual scenario, before assessing the accuracy of a series of images that all contained verified fake news. Results showed little evidence that the counterfactual prime lowered accuracy judgments about politically congruent images for both left- and right-leaning individuals, but it seemed to be effective with respect to judgments about politically incongruent images. Individual political leaning was the biggest predictor of overall belief in fake news for politically congruent images. The findings overall provide an understanding about how decisions are made about the veracity of politically congruent or politically incongruent fake news.

Two studies were conducted to examine the role of implicit theories of intelligence in the motivational consequences of performance-related counterfactual thinking. Overall, upward and downward counterfactuals generated in response to performance feedback differentially influenced incremental and entity theorists' propensity to engage in remedial action. In Study 1 it was found that incremental theorists reported greater motivation for remedial action when generating upward compared to downward counterfactuals for imagined poor performance, whereas entity theorists reported less motivation for remedial action when generating upward as compared to downward counterfactuals. In study 2 it was found that incremental theorists perceived remedial action to be more useful and engaged in more remedial action following upward compared to downward counterfactuals, whereas counterfactual direction did not influence entity theorists' perceptions of the usefulness of remedial action or actual engagement in remedial action. The results of these studies suggest that the perceived attainability of future improvement is an important moderator of the counterfactual direction and motivation link.