- Award ID(s):
- NSF-PAR ID:
- Date Published:
- Journal Name:
- Page Range / eLocation ID:
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
While EXplainable Artificial Intelligence (XAI) approaches aim to improve human-AI collaborative decision-making by improving model transparency and mental model formations, experiential factors associated with human users can cause challenges in ways system designers do not anticipate. In this paper, we first showcase a user study on how anchoring bias can potentially affect mental model formations when users initially interact with an intelligent system and the role of explanations in addressing this bias. Using a video activity recognition tool in cooking domain, we asked participants to verify whether a set of kitchen policies are being followed, with each policy focusing on a weakness or a strength. We controlled the order of the policies and the presence of explanations to test our hypotheses. Our main finding shows that those who observed system strengths early-on were more prone to automation bias and made significantly more errors due to positive first impressions of the system, while they built a more accurate mental model of the system competencies. On the other hand, those who encountered weaknesses earlier made significantly fewer errors since they tended to rely more on themselves, while they also underestimated model competencies due to having a more negative first impression of the model. Motivated by these findings and similar existing work, we formalize and present a conceptual model of user’s past experiences that examine the relations between user’s backgrounds, experiences, and human factors in XAI systems based on usage time. Our work presents strong findings and implications, aiming to raise the awareness of AI designers towards biases associated with user impressions and backgrounds.more » « less
Research has long sought to identify which individuals are best at accurately perceiving others' personalities or are
good judges, yet consistent predictors of this ability have been difficult to find. In the current studies, we revisit this question by examining a novel physiological correlate of social sensitivity, cardiac vagal flexibility, which reflects dynamic modulation of cardiac vagal control. Method
We examined whether greater cardiac vagal flexibility was associated with forming more accurate personality impressions, defined as viewing targets more in line with their distinctive self‐reported profile of traits, in two studies, including a thin‐slice video perceptions study (
N= 109) and a dyadic interaction study ( N= 175). Results
Across studies, we found that individuals higher in vagal flexibility formed significantly more accurate first impressions of others' more observable personality traits (e.g., extraversion, creativity, warmth). These associations held while including a range of relevant covariates, including cardiac vagal tone, sympathetic activation, and gender.
In sum, social sensitivity as indexed by cardiac vagal flexibility is linked to forming more accurate impressions of others' observable traits, shedding light on a characteristic that may help to identify the elusive good judge and providing insight into its neurobiological underpinnings.
null (Ed.)Abstract People spontaneously infer other people’s psychology from faces, encompassing inferences of their affective states, cognitive states, and stable traits such as personality. These judgments are known to be often invalid, but nonetheless bias many social decisions. Their importance and ubiquity have made them popular targets for automated prediction using deep convolutional neural networks (DCNNs). Here, we investigated the applicability of this approach: how well does it generalize, and what biases does it introduce? We compared three distinct sets of features (from a face identification DCNN, an object recognition DCNN, and using facial geometry), and tested their prediction across multiple out-of-sample datasets. Across judgments and datasets, features from both pre-trained DCNNs provided better predictions than did facial geometry. However, predictions using object recognition DCNN features were not robust to superficial cues (e.g., color and hair style). Importantly, predictions using face identification DCNN features were not specific: models trained to predict one social judgment (e.g., trustworthiness) also significantly predicted other social judgments (e.g., femininity and criminal), and at an even higher accuracy in some cases than predicting the judgment of interest (e.g., trustworthiness). Models trained to predict affective states (e.g., happy) also significantly predicted judgments of stable traits (e.g., sociable), and vice versa. Our analysis pipeline not only provides a flexible and efficient framework for predicting affective and social judgments from faces but also highlights the dangers of such automated predictions: correlated but unintended judgments can drive the predictions of the intended judgments.more » « less
Multi-modal bio-sensing has recently been used as effective research tools in affective computing, autism, clinical disorders, and virtual reality among other areas. However, none of the existing bio-sensing systems support multi-modality in a wearable manner outside well-controlled laboratory environments with research-grade measurements. This work attempts to bridge this gap by developing a wearable multi-modal biosensing system capable of collecting, synchronizing, recording and transmitting data from multiple bio-sensors: PPG, EEG, eye-gaze headset, body motion capture, GSR, etc. while also providing task modulation features including visual-stimulus tagging. This study describes the development and integration of the various components of our system. We evaluate the developed sensors by comparing their measurements to those obtained by a standard research-grade bio-sensors. We first evaluate different sensor modalities of our headset, namely earlobe-based PPG module with motion-noise canceling for ECG during heart-beat calculation. We also compare the steady-state visually evoked potentials (SSVEP) measured by our shielded dry EEG sensors with the potentials obtained by commercially available dry EEG sensors. We also investigate the effect of head movements on the accuracy and precision of our wearable eyegaze system. Furthermore, we carry out two practical tasks to demonstrate the applications of using multiple sensor modalities for exploring previously unanswerable questions in bio-sensing. Specifically, utilizing bio-sensing we show which strategy works best for playing Where is Waldo? visual-search game, changes in EEG corresponding to true versus false target fixations in this game, and predicting the loss/draw/win states through biosensing modalities while learning their limitations in a Rock-Paper-Scissors game.more » « less
To date, many AI initiatives (eg, AI4K12, CS for All) developed standards and frameworks as guidance for educators to create accessible and engaging Artificial Intelligence (AI) learning experiences for K‐12 students. These efforts revealed a significant need to prepare youth to gain a fundamental understanding of how intelligence is created, applied, and its potential to perpetuate bias and unfairness. This study contributes to the growing interest in K‐12 AI education by examining student learning of modelling real‐world text data. Four students from an Advanced Placement computer science classroom at a public high school participated in this study. Our qualitative analysis reveals that the students developed nuanced and in‐depth understandings of how text classification models—a type of AI application—are trained. Specifically, we found that in modelling texts, students: (1) drew on their social experiences and cultural knowledge to create predictive features, (2) engineered predictive features to address model errors, (3) described model learning patterns from training data and (4) reasoned about noisy features when comparing models. This study contributes to an initial understanding of student learning of modelling unstructured data and offers implications for scaffolding in‐depth reasoning about model decision making.
What is already known about this topic
Scholarly attention has turned to examining Artificial Intelligence (AI) literacy in K‐12 to help students understand the working mechanism of AI technologies and critically evaluate automated decisions made by computer models.
While efforts have been made to engage students in understanding AI through building machine learning models with data, few of them go in‐depth into teaching and learning of feature engineering, a critical concept in modelling data.
There is a need for research to examine students' data modelling processes, particularly in the little‐researched realm of unstructured data.
What this paper adds
Results show that students developed nuanced understandings of models learning patterns in data for automated decision making.
Results demonstrate that students drew on prior experience and knowledge in creating features from unstructured data in the learning task of building text classification models.
Students needed support in performing feature engineering practices, reasoning about noisy features and exploring features in rich social contexts that the data set is situated in.
Implications for practice and/or policy
It is important for schools to provide hands‐on model building experiences for students to understand and evaluate automated decisions from AI technologies.
Students should be empowered to draw on their cultural and social backgrounds as they create models and evaluate data sources.
To extend this work, educators should consider opportunities to integrate AI learning in other disciplinary subjects (ie, outside of computer science classes).