More Like this

1. Robots as Moral Advisors: The Effects of Deontological, Virtue, and Confucian Role Ethics on Encouraging Honest Behavior

   https://doi.org/10.1145/3434074.3446908  

   Kim, Boyoung; Wen, Ruchen; Zhu, Qin; Williams, Tom; Phillips, Elizabeth (March 2021, ACM/IEEE International Conference on Human-Robot Interaction)

   null (Ed.)

   We examined how robots can successfully serve as moral advisors for humans. We evaluated the effectiveness of moral advice grounded in deontological, virtue, and Confucian role ethics frameworks in encouraging humans to make honest decisions. Participants were introduced to a tempting situation where extra monetary gain could be earned by choosing to cheat (i.e., violating the norm of honesty). Prior to their decision, a robot encouraged honest choices by offering a piece of moral advice grounded in one of the three ethics frameworks. While the robot’s advice was overall not effective at discouraging dishonest choices, there was preliminary evidence indicating the relative effectiveness of moral advice drawn from deontology. We also explored how different cultural orientations (i.e., vertical and horizontal collectivism and individualism) influence honest decisions across differentially-framed moral advice. We found that individuals with a strong cultural orientation of establishing their own power and status through competition (i.e., high vertical individualism) were more likely to make dishonest choices, especially when moral advice was drawn from virtue ethics. Our findings suggest the importance of considering different ethical frameworks and cultural differences to design robots that can guide humans to comply with the norm of honesty. 

   more » « less
2. Combining forecasts from advisors: The impact of advice independence and verbal versus numeric format.

   https://doi.org/10.1037/xge0001611  

   Strueder, Jeremy D; Windschitl, Paul D (August 2024, Journal of Experimental Psychology: General)

   Past research on advice-taking has suggested that people are often insensitive to the level of advice independence when combining forecasts from advisors. However, this has primarily been tested for cases in which people receive numeric forecasts. Recent work by Mislavsky and Gaertig (2022) shows that people sometimes employ different strategies when combining verbal versus numeric forecasts about the likelihood of future events. Specifically, likelihood judgments based on two verbal forecasts (e.g., "rather likely") are more often extreme (relative to the forecasts) than are likelihood judgments based on two numeric forecasts (e.g., "70% probability"). The goal of the present research was to investigate whether advice-takers' use of combination strategies can be sensitive to advice independence when differences in independence are highly salient and whether sensitivity to advice independence depends on the format in which advice is given. In two studies, we found that advice-takers became more extreme with their own likelihood estimate when combining forecasts from advisors who use separate evidence, as opposed to the same evidence. We also found that two verbal forecasts generally resulted in more extreme combined likelihood estimates than two numeric forecasts. However, the results did not suggest that sensitivity to advice independence depends on the format of advice. 

   more » « less
3. TuringAdvice: A Generative and Dynamic Evaluation of Language Use

   https://doi.org/10.18653/v1/2021.naacl-main.386  

   Zellers, Rowan; Holtzman, Ari; Clark, Elizabeth; Qin, Lianhui; Farhadi, Ali; Choi, Yejin (June 2021, Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume)

   We propose TuringAdvice, a new challenge task and dataset for language understanding models. Given a written situation that a real person is currently facing, a model must generate helpful advice in natural language. Our evaluation framework tests a fundamental aspect of human language understanding: our ability to use language to resolve open-ended situations by communicating with each other. Empirical results show that today’s models struggle at TuringAdvice, even multibillion parameter models finetuned on 600k in-domain training examples. The best model, T5, writes advice that is at least as helpful as human-written advice in only 14% of cases; a much larger non-finetunable GPT3 model does even worse at 4%. This low performance reveals language understanding errors that are hard to spot outside of a generative setting, showing much room for progress. 

   more » « less
4. Emotion understanding and maternal sensitivity as protective factors against hostile attribution bias in anger‐prone children

   https://doi.org/10.1111/sode.12336  

   Wong, Maria S.; Chen, Xi; McElwain, Nancy L. (October 2018, Social Development)

   Abstract The present research examined children’s anger proneness, emotion understanding, and maternal sensitivity during toddlerhood as predictors of children’s hostile attribution bias (HAB) during the later preschool years. At 2.8 years (N = 128), maternal sensitivity (e.g., child‐centered behavior) was observed during mother–child play and snack, and parents reported on children’s anger proneness. At 3.3 years, emotion understanding (i.e., ability to identify emotional expressions accurately) was measured via an interactive puppet interview. At 4.8 and 5.4 years, children's HAB was assessed via child responses to hypothetical vignettes of ambiguous peer provocations. Path models revealed that maternal sensitivity predicted fewer hostile attributions. In addition, emotion understanding and maternal sensitivity emerged as buffers against the negative effect of anger proneness on HAB. Specifically, greater anger proneness was associated with more frequent hostile attributions, but only when children had lower emotion understanding or had mothers who were less sensitive. The findings highlight the interplay between intrapersonal and interpersonal factors in early childhood that contribute to a hostile attribution bias during the preschool period. 

   more » « less
5. The untold impact of learning approaches on software fault-proneness predictions: an analysis of temporal aspects

   https://doi.org/10.1007/s10664-024-10454-8  

   Ahmad, Mohammad Jamil; Goseva-Popstojanova, Katerina; Lutz, Robyn R (July 2024, Empirical Software Engineering)

   This paper aims to improve software fault-proneness prediction by investigating the unex- plored effects on classification performance of the temporal decisions made by practitioners and researchers regarding (i) the interval for which they will collect longitudinal features (soft- ware metrics data), and (ii) the interval for which they will predict software bugs (the target variable). We call these specifics of the data used for training and of the target variable being predicted the learning approach, and explore the impact of the two most common learning approaches on the performance of software fault-proneness prediction, both within a single release of a software product and across releases. The paper presents empirical results from a study based on data extracted from 64 releases of twelve open-source projects. Results show that the learning approach has a substantial, and typically unacknowledged, impact on classi- fication performance. Specifically, we show that one learning approach leads to significantly better performance than the other, both within-release and across-releases. Furthermore, this paper uncovers that, for within-release predictions, the difference in classification perfor- mance is due to different levels of class imbalance in the two learning approaches. Our findings show that improved specification of the learning approach is essential to under- standing and explaining the performance of fault-proneness prediction models, as well as to avoiding misleading comparisons among them. The paper concludes with some practical recommendations and research directions based on our findings toward improved software fault-proneness prediction. This preprint has not undergone any post-submission improvements or corrections. The Version of Record of this article is published in Empirical Software Engineering, and is available online at https://doi.org/10.1007/s10664-024-10454-8. 

   more » « less