Model-Based Explanation of Feedback Effects in Syllogistic Reasoning.

For decades, a significant number of models explaining human syllogistic inference processes were developed. There is profound work fitting the models' parameters and analyzing each model's ability to account for the data in order to support or reject the underlying theories. However, the model parameters are rarely used to extract explanations and hypotheses for phenomena that go beyond the original scope of the models. In this work, we apply three state-of-the-art models,  the probability heuristics model (PHM), mReasoner, and TransSet, to data from reasoning experiments where participants received feedback for their conclusions. We derived hypotheses based on the models' explanations for the feedback effect and put these to the test by conducting an experiment targeting the hypotheses. The work contributes to the field in three ways: (a) the feedback effect could be replicated and was shown to be a robust effect; (b) we demonstrate the use of the model parameters in order to derive new hypotheses; (c) we present possible explanations for the feedback effect based on existing theories.

The Predictive Power of Spatial Relational Reasoning Models: A New Evaluation Approach.

In the last few decades, cognitive theories for explaining human spatial relational reasoning have increased. Few of these theories have been implemented as computational models, however, even fewer have been compared computationally to each other. A computational model comparison requires, among other things, a still missing quantitative benchmark of core spatial relational reasoning problems. By presenting a new evaluation approach, this paper addresses: (1) developing a benchmark including raw data of participants, (2) reimplementation, adaptation, and extension of existing cognitive models to predict individual responses, and (3) a thorough evaluation of the cognitive models on the benchmark data. The paper shifts the research focus of cognitive modeling from reproducing aggregated response patterns toward assessing the predictive power of models for the individual reasoner. It demonstrate that not all psychological effects can discern theories. We discuss implications for modeling spatial relational reasoning.

Generation of an NFκB-Driven Alpharetroviral "All-in-One" Vector Construct as a Potent Tool for CAR NK Cell Therapy.

Immune cell therapeutics are increasingly applied in oncology. Especially chimeric antigen receptor (CAR) T cells are successfully used to treat several B cell malignancies. Efforts to engineer CAR T cells for improved activity against solid tumors include co-delivery of pro-inflammatory cytokines in addition to CARs, via either constitutive cytokine expression or inducible cytokine expression triggered by CAR recognition of its target antigen-so-called "T cells redirected for universal cytokine-mediated killing" (TRUCKs) or fourth-generation CARs. Here, we tested the hypothesis that TRUCK principles could be expanded to improve anticancer functions of NK cells. A comparison of the functionality of inducible promoters responsive to NFAT or NFκB in NK cells showed that, in contrast to T cells, the inclusion of NFκB-responsive elements within the inducible promoter construct was essential for CAR-inducible expression of the transgene. We demonstrated that GD2CAR-specific activation induced a tight NFκB-promoter-driven cytokine release in NK-92 and primary NK cells together with an enhanced cytotoxic capacity against GD2+ target cells, also shown by increased secretion of cytolytic cytokines. The data demonstrate biologically relevant differences between T and NK cells that are important when clinically translating the TRUCK concept to NK cells for the treatment of solid malignancies.

Genetic Correction of IL-10RB Deficiency Reconstitutes Anti-Inflammatory Regulation in iPSC-Derived Macrophages.

Patient material from rare diseases such as very early-onset inflammatory bowel disease (VEO-IBD) is often limited. The use of patient-derived induced pluripotent stem cells (iPSCs) for disease modeling is a promising approach to investigate disease pathomechanisms and therapeutic strategies. We successfully developed VEO-IBD patient-derived iPSC lines harboring a mutation in the IL-10 receptor ß-chain (IL-10RB) associated with defective IL-10 signaling. To characterize the disease phenotype, healthy control and VEO-IBD iPSCs were differentiated into macrophages. IL-10 stimulation induced characteristic signal transducer and activator of transcription 3 (STAT3) and suppressor of cytokine signaling 3 (SOCS3) downstream signaling and anti-inflammatory regulation of lipopolysaccharide (LPS)-mediated cytokine secretion in healthy control iPSC-derived macrophages. In contrast, IL-10 stimulation of macrophages derived from patient iPSCs did not result in STAT3 phosphorylation and subsequent SOCS3 expression, recapitulating the phenotype of cells from patients with IL-10RB deficiency. In line with this, LPS-induced cytokine secretion (e.g., IL-6 and tumor necrosis factor-α (TNF-α)) could not be downregulated by exogenous IL-10 stimulation in VEO-IBD iPSC-derived macrophages. Correction of the IL-10RB defect via lentiviral gene therapy or genome editing in the adeno-associated virus integration site 1 (AAVS1) safe harbor locus led to reconstitution of the anti-inflammatory response. Corrected cells showed IL-10RB expression, IL-10-inducible phosphorylation of STAT3, and subsequent SOCS3 expression. Furthermore, LPS-mediated TNF-α secretion could be modulated by IL-10 stimulation in gene-edited VEO-IBD iPSC-derived macrophages. Our established disease models provide the opportunity to identify and validate new curative molecular therapies and to investigate phenotypes and consequences of additional individual IL-10 signaling pathway-dependent VEO-IBD mutations.

Predictive Modeling of Individual Human Cognition: Upper Bounds and a New Perspective on Performance.

Model evaluation is commonly performed by relying on aggregated data as well as relative metrics for model comparison and selection. In light of recent criticism about the prevailing perspectives on cognitive modeling, we investigate models for human syllogistic reasoning in terms of predictive accuracy on individual responses. By contrasting cognitive models with statistical baselines such as random guessing or the most frequently selected response option as well as data-driven neural networks, we obtain information about the progress cognitive modeling could achieve for syllogistic reasoning to date, its remaining potential, and upper bounds of performance future models should strive to exceed. The methods presented in this article are not restricted to the domains of reasoning but generalize to other fields of behavioral research and can serve as useful additions to the modern modeler's toolbox.

Modeling Human Syllogistic Reasoning: The Role of "No Valid Conclusion".

Syllogistic reasoning, that is the drawing of inferences for categorical-quantified assertions, is one of the oldest branches of deductive reasoning research with a history exceeding 100 years. In syllogistic reasoning experiments, "No Valid Conclusion" (NVC) is one of the most frequently selected responses and corresponds to the logically correct conclusion for 58% of the syllogistic problem domain. To date, NVC is often neglected in computational models or just treated as a by-product of the underlying inferential mechanisms such as a last resort when the search for alternatives is exhausted. We illustrate that NVC represents a major shortcoming of current models for human syllogistic reasoning. By introducing heuristic rules for predicting NVC, we demonstrate that simple extensions of the existing models result in substantial improvements in their predictive performances. Our results emphasize the need for better NVC handling in cognitive modeling of human reasoning and provide directions for modelers on how to enhance their approaches.

Hematopoietic stem cell gene therapy for IFNγR1 deficiency protects mice from mycobacterial infections.

Mendelian susceptibility to mycobacterial disease is a rare primary immunodeficiency characterized by severe infections caused by weakly virulent mycobacteria. Biallelic null mutations in genes encoding interferon gamma receptor 1 or 2 (IFNGR1 or IFNGR2) result in a life-threatening disease phenotype in early childhood. Recombinant interferon γ (IFN-γ) therapy is inefficient, and hematopoietic stem cell transplantation has a poor prognosis. Thus, we developed a hematopoietic stem cell (HSC) gene therapy approach using lentiviral vectors that express Ifnγr1 either constitutively or myeloid specifically. Transduction of mouse Ifnγr1-/- HSCs led to stable IFNγR1 expression on macrophages, which rescued their cellular responses to IFN-γ. As a consequence, genetically corrected HSC-derived macrophages were able to suppress T-cell activation and showed restored antimycobacterial activity against Mycobacterium avium and Mycobacterium bovis Bacille Calmette-Guérin (BCG) in vitro. Transplantation of genetically corrected HSCs into Ifnγr1-/- mice before BCG infection prevented manifestations of severe BCG disease and maintained lung and spleen organ integrity, which was accompanied by a reduced mycobacterial burden in lung and spleen and a prolonged overall survival in animals that received a transplant. In summary, we demonstrate an HSC-based gene therapy approach for IFNγR1 deficiency, which protects mice from severe mycobacterial infections, thereby laying the foundation for a new therapeutic intervention in corresponding human patients.