Mining Bodily Cues to Deception.

A significant body of research has investigated potential correlates of deception and bodily behavior. The vast majority of these studies consider discrete, subjectively coded bodily movements such as specific hand or head gestures. Such studies fail to consider quantitative aspects of body movement such as the precise movement direction, magnitude and timing. In this paper, we employ an innovative data mining approach to systematically study bodily correlates of deception. We re-analyze motion capture data from a previously published deception study, and experiment with different data coding options. We report how deception detection rates are affected by variables such as body part, the coding of the pose and movement, the length of the observation, and the amount of measurement noise. Our results demonstrate the feasibility of a data mining approach, with detection rates above 65%, significantly outperforming human judgement (52.80%). Owing to the systematic analysis, our analyses allow for an understanding of the importance of various coding factor. Moreover, we can reconcile seemingly discrepant findings in previous research. Our approach highlights the merits of data-driven research to support the validation and development of deception theory.

Multi-Dataset, Multitask Learning of Egocentric Vision Tasks.

For egocentric vision tasks such as action recognition, there is a relative scarcity of labeled data. This increases the risk of overfitting during training. In this paper, we address this issue by introducing a multitask learning scheme that employs related tasks as well as related datasets in the training process. Related tasks are indicative of the performed action, such as the presence of objects and the position of the hands. By including related tasks as additional outputs to be optimized, action recognition performance typically increases because the network focuses on relevant aspects in the video. Still, the training data is limited to a single dataset because the set of action labels usually differs across datasets. To mitigate this issue, we extend the multitask paradigm to include datasets with different label sets. During training, we effectively mix batches with samples from multiple datasets. Our experiments on egocentric action recognition in the EPIC-Kitchens, EGTEA Gaze+, ADL and Charades-EGO datasets demonstrate the improvements of our approach over single-dataset baselines. On EGTEA we surpass the current state-of-the-art by 2.47 percent. We further illustrate the cross-dataset task correlations that emerge automatically with our novel training scheme.

A Personal Model of Trumpery: Linguistic Deception Detection in a Real-World High-Stakes Setting.

Language use differs between truthful and deceptive statements, but not all differences are consistent across people and contexts, complicating the identification of deceit in individuals. By relying on fact-checked tweets, we showed in three studies (Study 1: 469 tweets; Study 2: 484 tweets; Study 3: 24 models) how well personalized linguistic deception detection performs by developing the first deception model tailored to an individual: the 45th U.S. president. First, we found substantial linguistic differences between factually correct and factually incorrect tweets. We developed a quantitative model and achieved 73% overall accuracy. Second, we tested out-of-sample prediction and achieved 74% overall accuracy. Third, we compared our personalized model with linguistic models previously reported in the literature. Our model outperformed existing models by 5 percentage points, demonstrating the added value of personalized linguistic analysis in real-world settings. Our results indicate that factually incorrect tweets by the U.S. president are not random mistakes of the sender.

AdaPool: Exponential Adaptive Pooling for Information-Retaining Downsampling.

Pooling layers are essential building blocks of convolutional neural networks (CNNs), to reduce computational overhead and increase the receptive fields of proceeding convolutional operations. Their goal is to produce downsampled volumes that closely resemble the input volume while, ideally, also being computationally and memory efficient. Meeting both these requirements remains a challenge. To this end, we propose an adaptive and exponentially weighted pooling method: adaPool. Our method learns a regional-specific fusion of two sets of pooling kernels that are based on the exponent of the Dice-Sørensen coefficient and the exponential maximum, respectively. AdaPool improves the preservation of detail on a range of tasks including image and video classification and object detection. A key property of adaPool is its bidirectional nature. In contrast to common pooling methods, the learned weights can also be used to upsample activation maps. We term this method adaUnPool. We evaluate adaUnPool on image and video super-resolution and frame interpolation. For benchmarking, we introduce Inter4K, a novel high-quality, high frame-rate video dataset. Our experiments demonstrate that adaPool systematically achieves better results across tasks and backbones, while introducing a minor additional computational and memory overhead.

Coping with paediatric illness: Child's play? Exploring the effectiveness of a play- and sports-based cognitive behavioural programme for children with chronic health conditions.

Little is known about how play affects the development of children with a chronic condition. Studying play poses major methodological challenges in measuring differences in play behaviour, which results in a relative scarcity of research on this subject. This pilot study seeks to provide novel directions for research in this area. The effectiveness of a play- and sports-based cognitive behavioural programme for children (8-12 years) with a chronic condition was studied. The children and parents completed a battery of measurement tools before and after the programme. Moreover, the application of automated computer analyses of behaviour was piloted. Behaviour (Child Behavior Checklist) seemed to be positively affected by the programme. An increase in psychological well-being was observed (KIDSCREEN). Perceived competence (Self-Perception Profile for Children) and actual motor competence (Canadian Agility and Movement Skill Assessment) did not show any positive trends. These results of 13 participants suggest that children might learn to better cope with their illness by stimulating play behaviour. For the analysis of the effectiveness of programmes like this, we therefore propose to focus on measuring behaviour and quality of life. In addition, pilot measurements showed that automated analysis of play can provide important insights into the participation of children.

Light Field Saliency Detection with Deep Convolutional Networks.

Light field imaging presents an attractive alternative to RGB imaging because of the recording of the direction of the incoming light. The detection of salient regions in a light field image benefits from the additional modeling of angular patterns. For RGB imaging, methods using CNNs have achieved excellent results on a range of tasks, including saliency detection. However, it is not trivial to use CNN-based methods for saliency detection on light field images because these methods are not specifically designed for processing light field inputs. In addition, current light field datasets are not sufficiently large to train CNNs. To overcome these issues, we present a new Lytro Illum dataset, which contains 640 light fields and their corresponding ground-truth saliency maps. Compared to current publicly available light field saliency datasets [1], [2], our new dataset is larger, of higher quality, contains more variation and more types of light field inputs. This makes our dataset suitable for training deeper networks and benchmarking. Furthermore, we propose a novel end-to-end CNN-based framework for light field saliency detection. Specifically, we propose three novel MAC (Model Angular Changes) blocks to process light field micro-lens images. We systematically study the impact of different architecture variants and compare light field saliency with regular 2D saliency. Our extensive comparisons indicate that our novel network significantly outperforms state-of-the-art methods on the proposed dataset and has desired generalization abilities on other existing datasets.

To freeze or not to freeze: A culture-sensitive motion capture approach to detecting deceit.

We present a new signal for detecting deception: full body motion. Previous work on detecting deception from body movement has relied either on human judges or on specific gestures (such as fidgeting or gaze aversion) that are coded by humans. While this research has helped to build the foundation of the field, results are often characterized by inconsistent and contradictory findings, with small-stakes lies under lab conditions detected at rates little better than guessing. We examine whether a full body motion capture suit, which records the position, velocity, and orientation of 23 points in the subject's body, could yield a better signal of deception. Interviewees of South Asian (n = 60) or White British culture (n = 30) were required to either tell the truth or lie about two experienced tasks while being interviewed by somebody from their own (n = 60) or different culture (n = 30). We discovered that full body motion-the sum of joint displacements-was indicative of lying 74.4% of the time. Further analyses indicated that including individual limb data in our full body motion measurements can increase its discriminatory power to 82.2%. Furthermore, movement was guilt- and penitential-related, and occurred independently of anxiety, cognitive load, and cultural background. It appears that full body motion can be an objective nonverbal indicator of deceit, showing that lying does not cause people to freeze.

Learning to recognize rat social behavior: Novel dataset and cross-dataset application.

BACKGROUND: Social behavior is an important aspect of rodent models. Automated measuring tools that make use of video analysis and machine learning are an increasingly attractive alternative to manual annotation. Because machine learning-based methods need to be trained, it is important that they are validated using data from different experiment settings. NEW METHOD: To develop and validate automated measuring tools, there is a need for annotated rodent interaction datasets. Currently, the availability of such datasets is limited to two mouse datasets. We introduce the first, publicly available rat social interaction dataset, RatSI. RESULTS: We demonstrate the practical value of the novel dataset by using it as the training set for a rat interaction recognition method. We show that behavior variations induced by the experiment setting can lead to reduced performance, which illustrates the importance of cross-dataset validation. Consequently, we add a simple adaptation step to our method and improve the recognition performance. COMPARISON WITH EXISTING METHODS: Most existing methods are trained and evaluated in one experimental setting, which limits the predictive power of the evaluation to that particular setting. We demonstrate that cross-dataset experiments provide more insight in the performance of classifiers. CONCLUSIONS: With our novel, public dataset we encourage the development and validation of automated recognition methods. We are convinced that cross-dataset validation enhances our understanding of rodent interactions and facilitates the development of more sophisticated recognition methods. Combining them with adaptation techniques may enable us to apply automated recognition methods to a variety of animals and experiment settings.

Killer Apps: Developing Novel Applications That Enhance Team Coordination, Communication, and Effectiveness.

As part of the Lorentz workshop, "Interdisciplinary Insights into Group and Team Dynamics," held in Leiden, Netherlands, this article describes how Geeks and Groupies (computer and social scientists) may benefit from interdisciplinary collaboration toward the development of killer apps in team contexts that are meaningful and challenging for both. First, we discuss interaction processes during team meetings as a research topic for both Groupies and Geeks. Second, we highlight teamwork in health care settings as an interdisciplinary research challenge. Third, we discuss how an automated solution for optimal team design could benefit team effectiveness and feed into team-based interventions. Fourth, we discuss team collaboration in massive open online courses as a challenge for both Geeks and Groupies. We argue for the necessary integration of social and computational research insights and approaches. In the hope of inspiring future interdisciplinary collaborations, we develop criteria for evaluating killer apps-including the four proposed here-and discuss future research challenges and opportunities that potentially derive from these developments.

When Lying Feels the Right Thing to Do.

Fraud is a pervasive and challenging problem that costs society large amounts of money. By no means all fraud is committed by 'professional criminals': much is done by ordinary people who indulge in small-scale opportunistic deception. In this paper, we set out to investigate when people behave dishonestly, for example by committing fraud, in an online context. We conducted three studies to investigate how the rejection of one's efforts, operationalized in different ways, affected the amount of cheating and information falsification. Study 1 demonstrated that people behave more dishonestly when rejected. Studies 2 and 3 were conducted in order to disentangle the confounding factors of the nature of the rejection and the financial rewards that are usually associated with dishonest behavior. It was demonstrated that rejection in general, rather than the nature of a rejection, caused people to behave more dishonestly. When a rejection was based on subjective grounds, dishonest behavior increased with approximately 10%, but this difference was not statistically significant. We subsequently measured whether dishonesty was driven by the financial loss associated with rejection, or emotional factors such as a desire for revenge. We found that rejected participants were just as dishonest when their cheating did not led to financial gain. However, they felt stronger emotions when there was no money involved. This seems to suggest that upon rejection, emotional involvement, especially a reduction in happiness, drives dishonest behavior more strongly than a rational cost-benefit analysis. These results indicate that rejection causes people to behave more dishonestly, specifically in online settings. Firms wishing to deter customers and employees from committing fraud may therefore benefit from transparency and clear policy guidelines, discouraging people to submit claims that are likely to be rejected.

AMAB: automated measurement and analysis of body motion.

Technologies that measure human nonverbal behavior have existed for some time, and their use in the analysis of social behavior has become more popular following the development of sensor technologies that record full-body movement. However, a standardized methodology to efficiently represent and analyze full-body motion is absent. In this article, we present automated measurement and analysis of body motion (AMAB), a methodology for examining individual and interpersonal nonverbal behavior from the output of full-body motion tracking systems. We address the recording, screening, and normalization of the data, providing methods for standardizing the data across recording condition and across subject body sizes. We then propose a series of dependent measures to operationalize common research questions in psychological research. We present practical examples from several application areas to demonstrate the efficacy of our proposed method for full-body measurements and comparisons across time, space, body parts, and subjects.

Accuracy of head orientation perception in triadic situations: experiment in a virtual environment.

Research has revealed high accuracy in the perception of gaze in dyadic (sender- receiver) situations. Triadic situations differ from these in that an observer has to report where a sender is looking, not relative to himself. This is more difficult owing to the less favourable position of the observer. The effect of the position of the observer on the accuracy of the identification of the sender's looking direction is relatively unexplored. Here, we investigate this, focusing exclusively on head orientation. We used a virtual environment to ensure good stimulus control. We found a mean angular error close to 5 degrees. A higher observer viewpoint results in more accurate identification. Similarly, a viewpoint with a smaller angle to the sender's midsagittal plane leads to an improvement in identification performance. Also, we found an effect of underestimation of the error in horizontal direction, similar to findings for dyadic situations.