A neural cognitive model of argumentation with application to legal inference and decision making

Formal models of argumentation have been investigated in several areas, from multi-agent systems and artificial intelligence (AI) to decision making, philosophy and law. In artificial intelligence, logic-based models have been the standard for the representation of argumentative reasoning. More recently, the standard logic-based models have been shown equivalent to standard connectionist models. This has created a new line of research where (i) neural networks can be used as a parallel computational model for argumentation and (ii) neural networks can be used to combine argumentation, quantitative reasoning and statistical learning. At the same time, non-standard logic models of argumentation started to emerge. In this paper, we propose a connectionist cognitive model of argumentation that accounts for both standard and non-standard forms of argumentation. The model is shown to be an adequate framework for dealing with standard and non-standard argumentation, including joint-attacks, argument support, ordered attacks, disjunctive attacks, meta-level attacks, self-defeating attacks, argument accrual and uncertainty. We show that the neural cognitive approach offers an adequate way of modelling all of these different aspects of argumentation. We have applied the framework to the modelling of a public prosecution charging decision as part of a real legal decision making case study containing many of the above aspects of argumentation. The results show that the model can be a useful tool in the analysis of legal decision making, including the analysis of what-if questions and the analysis of alternative conclusions. The approach opens up two new perspectives in the short-term: the use of neural networks for computing prevailing arguments efficiently through the propagation in parallel of neuronal activations, and the use of the same networks to evolve the structure of the argumentation network through learning (e.g. to learn the strength of arguments from data).     

判断与决策研究纵观:基于模糊痕迹理论的视角(英文)

模糊痕迹理论是用于解释记忆、判断与决策的综合性理论,该理论的提出和发展主要基于对信息存储、表征、提取和加工过程的研究。本文首先介绍了模糊痕迹理论的基本原则,在此基础上重点讨论了其要义(gist)如何发挥核心作用,使得模糊痕迹理论有别于其他传统的决策模型。该理论将高级直觉与原始冲动性进行了区分,并且预测决策误差来源于判断与决策的各种不同成分,如背景知识、信息表征、提取和加工过程等。模糊痕迹理论不仅可以解释诸如框架效应、合取谬误等传统决策与判断文献中常讨论的误差现象,同时基于该理论的研究还得到了一些与传统决策理论相悖的新发现。此外,对脑与行为如何发育性变化的研究为我们了解成人的认知过程提供了至关重要的新视角,这些对脑与行为的发育性研究和对特殊人群的研究结果也都支持了模糊痕迹理论对要义加工依赖的预测。     

Counter‐intuitive moral judgement following traumatic brain injury

Several neurological patient populations, including traumatic brain injury (TBI), appear to produce an abnormally ‘utilitarian’ pattern of judgements to moral dilemmas; they tend to make judgements that maximize the welfare of the majority, rather than deontological judgements based on the following of moral rules (e.g., do not harm others). However, this patient research has always used extreme dilemmas with highly valued moral rules (e.g., do not kill). Data from healthy participants, however, suggest that when a wider range of dilemmas are employed, involving less valued moral rules (e.g., do not lie), moral judgements demonstrate sensitivity to the psychological intuitiveness of the judgements, rather than their deontological or utilitarian content (Kahane et al., Social Cognitive and Affective Neuroscience, 7, 2011, 393). We sought the moral judgements of 30 TBI participants and 30 controls on moral dilemmas where content (utilitarian/deontological) and intuition (intuitive/counter‐intuitive) were measured concurrently. Overall TBI participants made utilitarian judgements in equal proportions to controls; disproportionately favouring utilitarian judgements only when they were counter‐intuitive, and deontological judgements only when they were counter‐intuitive. These results speak against the view that TBI causes a specific utilitarian bias, suggesting instead that moral intuition is broadly disrupted following TBI.     

Evaluating performance over time: Is improving better than being consistently good?

In many decision contexts, people evaluate others based on intertemporal performance records and commonly face a choice between two distinct profiles: performance that is consistently high versus performance that improves over time to that high level. We proposed that these two profiles could be appealing for different reasons, and thus evaluators’ preferences will differ across decision contexts. In three studies, participants were presented with candidates (e.g., students, employees) displaying the two profiles, and evaluated each candidate in terms of performance, future expectations, and deservingness. The consistent candidate was rated higher on performance, but lower on future expectations, than the improved candidate. Consequently, in achievement-based decisions (e.g., selecting a student for a scholarship), the consistent candidate was viewed as most deserving, whereas in potential-based decisions (e.g., selecting an employee for promotion), the improved candidate was preferred. These effects were mediated by the relative weight that evaluators placed on performance and future expectations.     

Do tests facilitate decision‐making performance on a transfer task?

Testing can improve later recall of information. However, much less is known about the potential use of testing in promoting the transfer of learning. In this study, we investigated whether testing improves decision‐making performance on a transfer task in a sample of 98 university students using a between‐subjects design. After studying several statements about a fictional disease under different learning conditions (restudy, free recall, and multiple‐choice), participants were asked to recall this information and subsequently make medical decisions concerning the fictional disease (i.e., the transfer task). The present study found no advantage of testing conditions over restudy condition on the 30‐min delayed memory task. However, participants in the active retrieval practice (i.e., free recall) group performed significantly better on the transfer task over those in both restudy and multiple‐choice groups. These results suggest free‐recall tests promote the transfer of learning.