Real-world Reasoning Research Articles

Reliabilist epistemology meets bounded rationality

Epistemic reliabilism holds that a belief is justified if and only if it is produced by a reliable or truth-conducive process. I argue that reliabilism offers an epistemology for bounded rationality. This latter concept refers to normative and descriptive accounts of real-world reasoning instead of some ideal reasoning. However, as initially formulated, reliabilism involves an absolute, context-independent assessment of rationality that does not do justice to the fact that several processes are reliable in some reasoning environments but not in others, as is widely reported in the cognitive sciences literature. I consider possible solutions to this problem. Resorting to ‘normality reliabilism’, a variant of the theory, is one; but I find it insufficient. Therefore, in addition, I propose to relativise the reliability assessment to reasoning environments. This novel version of reliabilism fits bounded rationality better than the original one does.

The Application of Knowledge Engineering via the Use of a Biomimetic Digital Twin Ecosystem, Phenotype-Driven Variant Analysis, and Exome Sequencing to Understand the Molecular Mechanisms of Disease

Applied Artificial Intelligence, particularly Large Language Models, in biomedical research is accelerating, but effective discovery and validation requires a toolset without limitations or bias. On January 30, 2023, the National Academies of Sciences, Engineering, and Medicine (NAS) appointed an ad hoc committee to identify needs and opportunities to advance the mathematical, statistical, and computational foundations of digital twins in applications across science, medicine, engineering, and society. On December 15, 2023, the NAS released a 164 page report, "Foundational Research Gaps and Future Directions for Digital Twins". This report described the importance of using digital twins in biomedical research. We developed an innovative method that incorporated phenotype ranking algorithms with knowledge engineering via a biomimetic digital twin ecosystem. This ecosystem applied real-world reasoning principles to non-normalized, raw data to identify hidden or "dark data". We performed a clinical exome sequencing study on patients with endometriosis and were able to identify four VUSs potentially associated with endometriosis-related disorders in nearly all patients analyzed. One VUS was identified in all patient samples and could be a biomarker for diagnostics. To the best of our knowledge, this is the first study to incorporate the recomandations of the NAS to biomedical research. This method can be used to understand the mechanisms of any disease, for virtual clinical trials, and to identify effective new therapies.

P1462: POPULATION-BASED REAL WORLD RESULTS OF CD19-DIRECTED CAR T-CELL THERAPY FOR PATIENTS WITH RELAPSED OR REFRACTORY LARGE B-CELL LYMPHOMA: A DUTCH CAR T-CELL TUMORBOARD EXPERIENCE

Background: CD19-directed CAR T-cell therapy has become standard of care for patients with relapsed/refractory (R/R) large B-cell lymphoma (LBCL) and durable responses are seen in up to 40% of treated patients. Optimal selection of patients who will benefit from CAR T-cell therapy is important. In the real world reasons for ineligibility of patients who meet the registration label are often not reported. Additionally, data on health-related quality of life (HR-QoL) after infusion are scarce. Real world (in)eligibility and HR-QoL can play an important role in guiding treatment choices with novel therapeutic options becoming available. In the Netherlands, population-based referral and eligibility assessment for CAR T-cell therapy are centralized in a national tumorboard. A multicenter expert team discusses cases and data are collected in a national registry from time of referral. The aims of the tumorboard and registry are to capture population-based data on patients potentially eligible for CAR T-cell therapy and to optimize (cost-) effective, high quality care. Aims: To evaluate real world results of axicabtagene ciloleucel (axi-cel) for R/R LBCL on a population level, including reasons for ineligibility, efficacy, toxicity and (in a subset of patients) HR-QoL. Methods: For this prospective multicenter cohort study, all adult patients referred to the tumorboard for treatment of R/R LBCL with axi-cel were included from May 1st 2020 – October 19th 2021. Descriptive statistics were used for baseline characteristics, Kaplan Meier methods for survival analysis, regression models for univariate/multivariate analysis and linear mixed models for HR-QoL over time. Results: In total, 186 patients were referred to the tumorboard, 57 were not approved for axi-cel during the online tumorboard meeting, mostly (n=23) due to rapid progression (RP). The 129 approved patients were screened at a CAR T center, 32 were rejected (n=21 RP). Ninety-seven patients underwent apheresis of whom 7 did not receive axi-cel, mostly due to RP (n=6). Ninety patients were infused with CAR T-cells. Data of 74 of these patients are currently available for toxicity/efficacy analysis and updated results, including HR-QoL, will be presented during the EHA meeting. Median age was 59-years (min-max 27-79), 81% received bridging therapy. Best overall response rate was 82.4 %, including 66.2% complete remissions (CR). Median time to response was 1 month and all patients achieved their best response within 3 months. Four patients with PR at 1 month achieved a CR at month 3. The PFS and OS curves are shown in Figure 1. All grades CRS occurred in 89%, including ≥grade 3 in 2.7%, median time to onset was 3 days (min-max 0-14) and median duration 6 days (min-max 1-17). All grades ICANS occurred in 58%, including ≥ grade 3 in 27%, with a median time to onset of 7 days (min-max 1-28) and a median duration of 7 days (min-max 1-28). Non-relapse mortality was 1.4% due to one grade 5 ICANS event. Image:Summary/Conclusion: The merit of a national tumorboard is to enable prospective, in-dept evaluation of rejected as well as CART-infused patients. Twenty-eight percent of the referred patients in the Dutch population that fulfilled the indication criteria for R/R LBCL treatment with axi-cel could not receive the therapy due to rapidly progressive disease. A relatively low percentage of patients, 7%, underwent leukapheresis but eventually did not receive axi-cel. Efficacy and toxicity in our cohort are comparable with ZUMA-1 study results and other large real world cohorts.

UNDERSTANDING COVID-19 VACCINE HESITANCY: REASONS FOR VACCINE REFUSAL FROM PATIENTS EVALUATED IN A FLORIDA ICU

TYPE: Abstract TOPIC: Chest Infections PURPOSE: To better understand reasons for COVID-19 vaccine hesitancy so as to better target education efforts. METHODS: Consecutive unvaccinated patients evaluated for admission to the Intensive Care Unit due to COVID-19 infection were asked their reason(s) for not being vaccinated. Reasons were categorized and presented along with basic demographic data. RESULTS: CONCLUSIONS: The most common reasons for not getting vaccinated included fear of potential side effects, and the perception that the vaccine was too new. Severity of COVID-19 and access to vaccination were less frequently reported. CLINICAL IMPLICATIONS: The unvaccinated population of COVID-19 patients is by far the number one driver of morbidity and mortality from this disease. Vaccination remains the most effective way to prevent severe illness yet vaccine refussal remains at unacceptably high rates. This study provides a glimpse into the real world reasons why patients in a single community who end up very ill have not gotten vaccinated. Providing education highlighting the safety of the vaccine, including its exceptional track record and wide use, appears to be essential to vaccine acceptance in this cohort. Understanding the reasons individuals decline vaccination on a local level is crucial in targeting education and additional resources. DISCLOSURE: Nothing to declare. KEYWORD: COVID-19

Novice Non-Native English Language Teachers’ Imaginary and Actual Decision Making and Pedagogical Reasoning: Student and Personal Features (Research Article)

Two of the central concepts in teaching skills are decision making and pedagogical reasoning. Taking benefit from the dearth of studies on teachers’ actual or real-world decisions, this study aimed to respond to this invitation by keeping track of novice Iranian English as a foreign language (EFL) teachers’ decisions in two different times using six research-oriented teaching scenarios reflecting the student and personal features. Furthermore, their pedagogical reasoning was also attended to once through their responses to imaginary teaching scenarios and once through their actual classroom decisions. The participants comprised of ten novice Iranian EFL (six female and four male) teachers with an age range of 19 to 25 and a male experienced teacher, aged 30, who acted as a researcher-as-participant and was only accountable for the novices’ real-world reasoning. The data were collected through utilizing a total of six teaching scenarios, classroom observation, and video stimulated recalls. The findings, obtained through conversation analysis and pertinent vignettes and excerpts, revealed that the participants underwent a change in their decisions in two of the three scenarios reflecting the student features, while an approximate conformity could be observed in all scenarios mirroring personal features. It was revealed that whenever the teachers’ reasoning changed, their decisions underwent some changes as well. In addition, the findings showed that the flow of conversation in the classroom could be strongly influenced by the teachers’ decisions. A number of implications and recommendations for further research are also pinpointed.

Real-World Reasons for Steroid Switching and Associated Clinical Outcomes Among Patients with Dystrophinopathies in the US: a Nationwide Review of Medical Records (2378)

Real-World Reasons for Steroid Switching and Associated Clinical Outcomes Among Patients with Dystrophinopathies in the US: a Nationwide Review of Medical Records (2378)

Propensities and Second Order Uncertainty: A Modified Taxi Cab Problem.

The study of people’s ability to engage in causal probabilistic reasoning has typically used fixed-point estimates for key figures. For example, in the classic taxi-cab problem, where a witness provides evidence on which of two cab companies (the more common ‘green’/less common ‘blue’) were responsible for a hit and run incident, solvers are told the witness’s ability to judge cab color is 80%. In reality, there is likely to be some uncertainty around this estimate (perhaps we tested the witness and they were correct 4/5 times), known as second-order uncertainty, producing a distribution rather than a fixed probability. While generally more closely matching real world reasoning, a further important ramification of this is that our best estimate of the witness’ accuracy can and should change when the witness makes the claim that the cab was blue. We present a Bayesian Network model of this problem, and show that, while the witness’s report does increase our probability of the cab being blue, it simultaneously decreases our estimate of their future accuracy (because blue cabs are less common). We presented this version of the problem to 131 participants, requiring them to update their estimates of both the probability the cab involved was blue, as well as the witness’s accuracy, after they claim it was blue. We also required participants to explain their reasoning process and provided follow up questions to probe various aspects of their reasoning. While some participants responded normatively, the majority self-reported ‘assuming’ one of the probabilities was a certainty. Around a quarter assumed the cab was green, and thus the witness was wrong, decreasing their estimate of their accuracy. Another quarter assumed the witness was correct and actually increased their estimate of their accuracy, showing a circular logic similar to that seen in the confirmation bias/belief polarization literature. Around half of participants refused to make any change, with convergent evidence suggesting that these participants do not see the relevance of the witness’s report to their accuracy before we know for certain whether they are correct or incorrect.

P2794Real-world reasons and outcomes for 1-month versus longer dual antiplatelet therapy strategies with a polymer-free biolimus A9-coated stent: insights from the all-comers FREEDOM registry

Abstract Background The LEADERS FREE trial established the favourable clinical profile of a new polymer-free biolimus A9-eluting stent (PF-BES) in patients at high bleeding risk when used with a 1-month dual anti-platelet therapy (DAPT) strategy. Purpose This is the first study evaluating real-world reasons and outcomes for a 1-month versus longer DAPT strategies following PF-BES percutaneous coronary intervention (PCI). Methods FREEDOM is an all-comers registry including all patients who underwent PCI with at least one PF-BES at 10 sites, between January 2016 and July 2018. Patients were stratified according to DAPT strategy at discharge (1-month vs >1-month). Baseline features, reasons for PF-BES as reported by the treating physician, and outcomes were compared between groups. Primary outcomes were the 390-day estimates of a patient-oriented composite endpoint (POCE: death, myocardial infarction (MI) or target vessel revascularization) and of a device-oriented composite endpoint (DOCE: cardiac death, target vessel-MI or ischemia-driven target lesion revascularization). Incidence rates were adjusted for clinically relevant factors and outcome predictors. To avoid survival bias, landmark analyses starting from 1-month post-PCI were further carried. Results Following PF-BES PCI, 328 (40.3%) patients were discharged with 1-month DAPT and 485 (59.6%) patients with longer DAPT (median 12 months, IQR 6–12 months). Patients with hypertension or on oral anticoagulation (OAT) were more likely and patients with a previous or an index MI were less likely to be discharged on 1-month DAPT. Patients prescribed with 1-month DAPT were more likely to have had a PF-BES for a LEADERS FREE high bleeding risk criterium than those with longer DAPT (90.2% vs 69.9%, p=0.001). The same association was observed when the reason for PF-BES was a planned major surgery (13.1% vs 6.2%; p=0.001) or OAC to be continued after PCI (38.7% vs 16.5%, p<0.001). Conversely, patients with planned longer DAPT were more likely to have had a PF-BES following the operator preference (2.4% vs 15.5%, p<0.001) or for a reason other than a LEADERS FREE criterium or operator preference (5.2% vs 11.3%, p<0.001). No between-groups differences in the occurrence of the primary outcomes (1-month vs >1-month DAPT: POCE 11.9% vs 13.2%, p=0.747; adj-HR 1.29 [95% CI 0.78–2.10]; DOCE: 4.8% vs 8.1%, p=0.500; adj-HR 1.01 [95% CI 0.50–2.07]) and of bleedings (any: 11.3% vs 9.4%, p=0.472; adj-HR 0.87 [95% CI 0.51–1-50]; BARC 3–5: 4.2% vs 2.2%; p=0.108; adj-HR 1.50 [95% CI 0.58–3.87]) were observed. Landmark analyses showed similar results. Conclusions In a large contemporary all-comers registry, factors reflecting the operator-perceived patient high bleeding risk were the main drivers of a very-short DAPT strategy following PF-BES PCI. We found no interaction of DAPT duration with outcomes following PF-BES PCI, an observation warranting investigation in adequately powered randomised studies.

Real-world reasons for discontinuation of biologics and apremilast in patients with psoriasis: The University of South Florida Dermatology clinic experience

Real-world reasons for discontinuation of biologics and apremilast in patients with psoriasis: The University of South Florida Dermatology clinic experience

Syllogisms delivered in an angry voice lead to improved performance and engagement of a different neural system compared to neutral voice.

Despite the fact that most real-world reasoning occurs in some emotional context, very little is known about the underlying behavioral and neural implications of such context. To further understand the role of emotional context in logical reasoning we scanned 15 participants with fMRI while they engaged in logical reasoning about neutral syllogisms presented through the auditory channel in a sad, angry, or neutral tone of voice. Exposure to angry voice led to improved reasoning performance compared to exposure to sad and neutral voice. A likely explanation for this effect is that exposure to expressions of anger increases selective attention toward the relevant features of target stimuli, in this case the reasoning task. Supporting this interpretation, reasoning in the context of angry voice was accompanied by activation in the superior frontal gyrus—a region known to be associated with selective attention. Our findings contribute to a greater understanding of the neural processes that underlie reasoning in an emotional context by demonstrating that two emotional contexts, despite being of the same (negative) valence, have different effects on reasoning.

The Singularity and the State of the Art in Artificial Intelligence

The state of the art in automating basic cognitive tasks, including vision and natural language understanding, is far below human abilities. Real-world reasoning, which is an unavoidable part of many advanced forms of computer vision and natural language understanding, is particularly difficult---suggesting the advent of computers with superhuman general intelligence is not imminent. The possibility of attaining a singularity by computers that lack these abilities is discussed briefly.

Notes towards an Anthropology of Political Revolutions

AbstractWhile resistance and rebellion have remained core themes in anthropology at least since the 1960s, anthropologists have paid much less attention to the study of political revolutions as real historical events. Yet there are compelling real-world reasons why they should orient their analytical apparatus and ethnographic efforts towards revolutionary events. This article advances a series of reasons why anthropology can enrich and supplement existing political science and history traditions in the study of political revolutions. Anthropology can do so via key concepts developed by Victor Turner: “liminality,” “social drama,” “communitas,” “frame,” and “play.” Turner's ritual approach gains further relevance when linked to another series of concepts developed by Marcel Mauss, Gabriel Tarde, Georg Simmel, and Gregory Bateson, such as “imitation,” “trickster,” “schismogenesis,” and “crowd behavior.” To study revolutions implies not only a focus on political behavior “from below,” but also recognition of moments where “high and low” are relativized or subverted, and where the micro- and macro-levels fuse in critical conjunctions.

An exploratory investigation of real-world reasoning in paranoia

Paranoid thinking has been linked to greater availability in memory of past threats to the self. However, remembered experiences may not always closely resemble events that trigger paranoia, so novel explanations must be elaborated for the likelihood of threat to be determined. We investigated the ability of paranoid individuals to construct explanations for everyday situations and whether these modulate their emotional impact. Twenty-one participants experiencing paranoia and 21 healthy controls completed a mental simulation task that yields a measure of the coherence of reasoning in everyday situations. When responses featured positive content, clinical participants produced less coherent narratives in response to paranoid themed scenarios than healthy controls. There was no significant difference between the groups when responses featured negative content. The current study suggests that difficulty in scenario construction may exacerbate paranoia by reducing access to non-threatening explanations for everyday events, and this consequently increases distress.

Psychophysiological anticipation of positive outcomes promotes advantageous decision-making in normal older persons

Previous studies have demonstrated that aging has an adverse effect on laboratory decision-making in some older adults, and such findings have important implications for real-world reasoning and judgment. Emotion, and its accompanying somatic responses, is thought to contribute significantly to decision-making. In the present study, we had two objectives: 1) to investigate decision-making in a new sample of elderly participants, using the Iowa Gambling Task (IGT); and 2) to investigate psychophysiological correlates of decision-making, focusing on anticipatory skin conductance responses (SCRs) that participants produce immediately prior to their behavioral response. We hypothesized that the previous behavioral findings would be replicated, and further, that the older adult participants with IGT impairments would show a lack of discriminatory anticipatory SCRs during the IGT. The results supported both predictions. First, a subgroup of the new elderly sample demonstrated impaired decision-making on the IGT, replicating our previous findings. Second, the participants with impaired IGT performance failed to demonstrate discriminatory anticipatory SCRs for advantageous versus disadvantageous choices, whereas participants with normal IGT performance did demonstrate such discrimination; in the latter case, however, SCR magnitude was higher for advantageous decisions (unlike the pattern in young normal adults). Our data lead to the suggestion that strong decision-making abilities among older adults may be a function of positive somatic markers, whereas poor decision-making abilities may arise from an abnormal somatic response generated in anticipation of a future event.

Enhancing disjunctive logic programming systems by SAT checkers

Disjunctive logic programming (DLP) with stable model semantics is a powerful nonmonotonic formalism for knowledge representation and reasoning. Reasoning with DLP is harder than with normal (∨-free) logic programs, because stable model checking—deciding whether a given model is a stable model of a propositional DLP program—is co-NP-complete, while it is polynomial for normal logic programs. This paper proposes a new transformation Γ M( P) , which reduces stable model checking to UNSAT—i.e., to deciding whether a given CNF formula is unsatisfiable. The stability of a model M of a program P thus can be verified by calling a Satisfiability Checker on the CNF formula Γ M( P) . The transformation is parsimonious (i.e., no new symbol is added), and efficiently computable, as it runs in logarithmic space (and therefore in polynomial time). Moreover, the size of the generated CNF formula never exceeds the size of the input (and is usually much smaller). We complement this transformation with modular evaluation results, which allow for efficient handling of large real-world reasoning problems. The proposed approach to stable model checking has been implemented in DLV—a state-of-the-art implementation of DLP. A number of experiments and benchmarks have been run using SATZ as Satisfiability checker. The results of the experiments are very positive and confirm the usefulness of our techniques.

Planning the project management way: Efficient planning by effective integration of causal and resource reasoning in RealPlan

In most real-world reasoning problems, planning and scheduling phases are loosely coupled. For example, in project planning, the user comes up with a task list and schedules it with a scheduling tool like Microsoft Project. One can view automated planning in a similar way in which there is an action selection phase where actions are selected and ordered to reach the desired goals, and a resource allocation phase where enough resources are assigned to ensure the successful execution of the chosen actions. On the other hand, most existing automated planners studied in Artificial Intelligence do not exploit this loose-coupling and perform both action selection and resource assignment employing the same algorithm. The current work shows that the above strategy severely curtails the scale-up potential of existing state of the art planners which can be overcome by leveraging the loose coupling. Specifically, a novel planning framework called RealPlan is developed in which resource allocation is de-coupled from planning and is handled in a separate scheduling phase. The scheduling problem with discrete resources is represented as a Constraint Satisfaction Problem (CSP) problem, and the planner and scheduler interact either in a master-slave manner or in a peer-peer relationship. In the former, the scheduler simply tries to assign resources to the abstract causal plan passed to it by the planner and returns success. In the latter, a more sophisticated “multi-module dependency directed backtracking” approach is used where the failure explanation in the scheduler is translated back to the planner and serves as a nogood to direct planner search. RealPlan not only preserves both the correctness as well as the quality (measured in length) of the plan but also improves efficiency. Moreover, the failure-driven learning of constraints can serve as an elegant and effective approach for integrating planning and scheduling systems. Beyond the context of planner efficiency, the current work can be viewed as an important step towards merging planning with real-world problem solving where plan failure during execution can be resolved by undertaking only necessary resource re-allocation and not complete re-planning.

Reasoning with Bounded Rationality

Oaksford and Chater (1992) have criticized an earlier discussion of theories of human reasoning (Evans, 1991) for omission of the concept of `bounded rationality'. They argue that all existing theories of deductive competence would become computationally intractable when applied to complex real-world problems. In reply, it is accepted that we cannot use explicit deductive reasoning based on either inference rules or mental models in the face of complex problems. However, it is suggested first of all that we do engage in explicit reasoning with relatively simple but important real-world problems, and that the mental models theory provides a plausible account of how this is achieved. Complex real-world problem see are knowledge-based and are thus probably dealt with by the induction of schemas or by connectionist mechanisms which we know to be very efficient in the human brain. Thus much real-world reasoning will be implicit and inductive, rather than explicit and deductive.