APS Minds and Technology Article Collection

Artificial Intelligence and Persuasion: A Construal-Level Account

Tae Woo Kim and Adam Duhachek

Kim and Duhachek found that messages by nonhuman artificial agents (AAs) are more persuasive when they highlight how an action is performed (e.g., “Apply sunscreen before going out”) rather than why an action is performed (e.g., “Using sunscreen means healthy skin”). Participants likely judged an AA’s actions as more appropriate when they represented how (i.e., low-level construals) as opposed to why (i.e., high-level construals) because they perceived the AA as lacking goals. However, when the AA showed learning capabilities, participants were more open to persuasion from high-level construal messages than low-level messages. This indicates that perceptions of learning capabilities may change people’s assumptions about AAs.

Concerns About Automation and Negative Sentiment Toward Immigration

Monica Gamez-Djokic and Adam Waytz

The rise of sophisticated technology that can automate certain tasks not only threatens jobs but also appears to have social and psychological consequences, such as increasing negative attitudes toward immigration. In 12 studies, Gamez-Djokic and Waytz found that, both in the United States and in Europe, people who perceive automation as a greater threat to employment also tend to hold negative perceptions about immigrants. Automation concerns are also linked to support for restrictive immigration policies and, in the context of layoffs, an increase in discrimination against immigrants.

People Reject Algorithms in Uncertain Decision Domains Because They Have Diminishing Sensitivity to Forecasting Error

Berkeley J. Dietvorst and Soaham Bharti

People may be unwilling to use algorithmic decision-makers (e.g., virtual doctors, self-driving cars) in inherently uncertain domains, such as financial investing or medical decision-making. In nine studies, Dietvorst and Bharti showed that people have diminishing sensitivity to forecasting errors—they perceive “relatively large subjective differences between different magnitudes of near-perfect forecasts (the best possible forecasts that produce little to no error) and relatively small subjective differences between forecasts with greater amounts of error.” As a result, they are less likely to choose the best decision-makers in domains that are more unpredictable (e.g., with random outcomes vs. with outcomes determined by an equation) and instead tend to prefer decision-makers based on their perceived likelihood of producing a near-perfect choice and with high variance in performance. This leads people to favor riskier and often worse-performing decision-makers, such as human judgment, in uncertain domains.

Detecting the Trustworthiness of Novel Partners in Economic Exchange

David DeSteno, Cynthia Breazeal, Robert H. Frank, David Pizarro, Jolie Baumann, Leah Dickens, and Jin Joo Lee

Can humans detect the trustworthiness of others by observing their nonverbal signals? Participants were partnered with a robot called Nexi. Each participant spent 10 minutes conversing with Nexi while she displayed several behavioral cues found to be predicative of untrustworthiness (target-cue condition) or while she refrained from displaying any behavioral target cues (control condition). Participants then played the Give-Some game — an economic game based on trust — with Nexi. Participants in the target-cue condition reported trusting Nexi less and gave less to, and expected to receive less from, Nexi in the economic game. This suggests that humans can detect trustworthiness in others (even in robots) by examining their nonverbal signals.

Digital Chameleons: Automatic Assimilation of Nonverbal Gestures in Immersive Virtual Environments

Jeremy N. Bailenson and Nick Yee

A nonhuman, nonverbal mimicker might exert social influence like a human would, this research suggests. Participants interacted with an embodied artificial intelligence agent in immersive virtual reality. The agent either mimicked a participant’s head movements at a 4-s delay or utilized prerecorded movements of another participant as it verbally presented an argument. Mimicking agents were more persuasive and received more positive trait ratings than nonmimickers, despite participants’ inability to explicitly detect the mimicry. These data indicate the ability to use automatic, indiscriminate mimicking (i.e., a computer algorithm blindly applied to all movements) to gain social influence. Furthermore, this is the first study to demonstrate social influence effects with a nonhuman, nonverbal mimicker.

Do Dr. Google and Health Apps Have (Comparable) Side Effects? An Experimental Study

Stefanie M. Jungmann, Sebastian Brand, Johanna Kolb, and Michael Witthöft

Using Google or health-related apps to diagnose symptoms may increase health anxiety and have other negative effects. Participants did a fast-breathing exercise to induce hyperventilation and then either Googled possible causes for their symptoms, used a diagnostic app, or simply waited. Googling or using the app led to increased anxiety, stronger negative affect, and increased perceived need to consult a physician. These findings support the idea that health-related Internet use may contribute to emotional distress, leading to so-called cyberchondria.

Massive Open Online Interventions: A Novel Model for Delivering Behavioral-Health Services Worldwide

Ricardo F. Muñoz, Eduardo L. Bunge, Ken Chen, Stephen M. Schueller, Julia I. Bravin, Elizabeth A. Shaughnessy, and Eliseo J. Pérez-Stable

In recent years there has been a call to expand access to behavioral-health services. In this article, the authors describe their work providing a massive open online intervention (MOOI) aimed at helping people stop smoking. Although the multi-component intervention was open for 30 months, the authors, in this article, specifically focus on data from the last 18 months of recruitment. The researchers found that quit rates were 39.2% at 1 month, 43.5% at 3 months, 45.7% at 6 months, and 50.3% at 12 months. These quit rates matched closely to previously published quit rates seen in participants recruited in the first year of the intervention. This consistency provides added support for the usefulness of MOOIs as an effective way of disseminating large-scale treatments.  

The Promise of Neurotechnology in Clinical Translational Science

Susan W. White, J. Anthony Richey, Denis Gracanin, Martha Ann Bell, Stephen LaConte, Marika Coffman, Andrea Trubanova, and Inyoung Kim

The use of neurotechnology — devices and applications that can be used to assess and change neural function — has the potential to augment and improve the way we diagnose and treat a variety of different clinical disorders. White and colleagues provide a seven-principle framework that can be used to guide the development, evaluation, and application of these new technologies. Although neurotechnology has the potential to provide care with few side effects to a wide audience who might not have access to traditional in-person care, there are still some hurdles to be overcome. An integrative approach will be needed to combine measurements and insights gleaned from use of this technology to identify and target the underlying mechanisms of specific disorders.

Mental Health on the Go: Effects of a Gamified Attention-Bias Modification Mobile Application in Trait-Anxious Adults

Tracy A. Dennis and Laura J. O’Toole

Can mobile-application interventions be effective at reducing anxiety and stress reactivity? Participants who scored relatively high on a survey for anxiety played a gamified attention-bias modification training (ABMT) mobile application or a placebo training (PT) application for 25 or 45 minutes. Compared with the PT application, both the long and short versions of the ABMT game reduced participants’ levels of anxiety and stress reactivity. Participants paying the long version of the ABMT game also showed reduced attention bias to threat and easier disengagement from threat when compared with participants playing the short ABMT application or the PT application. These findings indicate that even a single session of gamified ABMT may be useful in reducing anxiety.

Robots as Mirrors of the Human Mind

Agnieszka Wykowska

Robots can increase our knowledge about human cognition and serve as tools for research in psychological science. Wykowska gives examples in which robots have been used to study mechanisms of social cognition that require reciprocal interaction between two people (e.g., joint attention, when one person directs their attention to a location and their partner attends there in response). The author also discusses whether and when robots are perceived as possessing human characteristics and how robots have been used to implement computational models of human cognition.

Augmented Reality as a Tool for Studying Visual Plasticity: 2009 to 2018

Min Bao and Stephen A. Engel

Augmented reality (AR) is a technology in which virtual objects are superimposed on or composited with the real world so the user can see or interact with them. Bao and Engel review the use of AR in basic psychological research. They focus on studies of visual plasticity, in which the visual system changes its function in response to environmental demands. AR has shed new light on mechanisms of long-term contrast adaptation and ocular-dominance plasticity. For example, viewing an altered reality depriving the visual system of vertical information for 4 hr increased observers’ sensitivity to the deprived orientation, and the effects of adaptation seemed to gradually increase in strength over a range of durations from 1 min to 8 hr. These findings indicate that distinct mechanisms may control visual adaptation over different time scales. AR has also been shown to shift ocular dominance (i.e., change which eye is stronger and more dominant than the other in certain tasks). Bao and Engel suggest that shifting ocular dominance through the use of AR might be an effective treatment of certain visual disorders, such as amblyopia (lazy eye). They further propose that AR could be used to study plasticity of other visual properties, such as color or motion.

Adapting Perception, Action, and Technology for Mathematical Reasoning

Robert L. Goldstone, Tyler Marghetis, Erik Weitnauer, Erin R. Ottmar, and David Landy

Although mathematical reasoning has existed for less than 6,000 years, specific regions in the brain have been associated with these cognitive functions. How has a relatively new activity in human history made such an evolutionary impact on the brain? Goldstone and colleagues suggest that people engage in mathematical reasoning using three strategies: by reusing mental processes that evolved for other purposes, by adapting those routines to the new requirements of mathematics, and by inventing tools to improve our mental processes. They propose and present evidence for a hypothesis—the Rigged Up Perception-Action Systems—informed by these strategies

Learning by Communicating in Natural Language With Conversational Agents

Arthur C. Graesser, Haiying Li, and Carol Forsyth

Tutoring is effective not simply because tutors are highly knowledgeable and can lecture students, but because tutors encourage students to generate answers to problems. Recently developed computer-based tutoring programs can simulate the conversation patterns used by human tutors and engage students in natural-language discussions. Graesser, Li, and Forsyth describe the conversation patterns used in tutoring and how these have been implemented in intelligent tutoring systems. 

Realizing the Potential of Behavioral Intervention Technologies

Stephen M. Schueller, Ricardo F. Muñoz, and David C. Mohr

Technology is continuously evolving, and psychologists have begun to take advantage of these advances. One of the main areas this is evident is in the rise of behavioral intervention technologies (BITs). Schueller, Muñoz, and Mohr describe the advantages of BITs over traditional therapies and the challenges inherent in making them work. Although these technologies are promising, the authors caution that they will only live up to their potential if designers are deliberate and thoughtful in their design, evaluation, and implementation. 

Neuroergonomics: Brain, Cognition, and Performance at Work

Raja Parasuraman

Neuroergonomics is the study of the human brain in relation to performance at work and other everyday settings. Parasuraman describes research using neuroimaging and molecular genetics in four areas: (a) multitasking and mental workload, (b) human error, (c) neuroadaptive interfaces, and (d) individual differences in cognition. The author’s analyses indicate that theoretical and practical understanding of how humans interact with and use technology might be considerably enriched if researchers also consider the human brain that makes such activities possible.

The Human Relation With Nature and Technological Nature

Peter H. Kahn, Jr., Rachel L. Severson, and Jolina H. Ruckert

Technological nature includes technologies that in various ways mediate, augment, or simulate the natural world. Current examples of technological nature include videos and live webcams of nature, robot animals, and immersive virtual environments. But does it matter for the physical and psychological well-being of the human species that actual nature is being replaced with technological nature? The answer might be “yes,” Kahn and colleagues suggest and draw on evolutionary and cross-cultural developmental accounts of the human relation with nature and psychological research on the effects of technological nature to support their argument. The authors also discuss the issue—and area for future research—of “environmental generational amnesia.” The concern is that, by adapting gradually to the loss of actual nature and to the increase of technological nature, humans will lower the baseline across generations for what counts as a full measure of the human experience and of human flourishing.

Technition: When Tools Come Out of the Closet

François Osiurak, Mathieu Lesourd, Jordan Navarro, and Emanuelle Reynaud

Osiurak and colleagues review recent psychological and neuroimaging studies to propose that the human technical mind originated because of neurocognitive skills unique to humans, such as technical reasoning skills involving the area PF within the left inferior parietal lobe. The researchers thus propose a new field in the cognitive sciences—technition. Technition focuses on human technical skills and might allow researchers to better understand how humans transform their environment by making and using tools and by building constructions.

Machine Learning and Psychological Research: The Unexplored Effect of Measurement

Ross Jacobucci and Kevin J. Grimm

Machine learning (i.e., artificial intelligence, big data, data mining) can benefit many areas of psychological science, such as those that use biological or genetic variables. However, more traditional areas of research have not benefited from machine learning. Jacobucci and Grimm suggest that this might be because of measurement errors that prevent machine-learning algorithms from accurately modeling the data. They provide simulated examples that show that measurement quality is very important for model selection in machine learning, and they advance recommendations for better integration of machine learning with statistics in traditional psychological science. 

Extending the Passive-Sensing Toolbox: Using Smart-Home Technology in Psychological Science

Benjamin W. Nelson and Nicholas B. Allen

Smart-home technology offers an unprecedented opportunity for psychological scientists to directly observe behavior in an important natural setting: the home. Smart-home devices, which are connected to the Internet, can continuously and unobtrusively capture individuals’ social and interpersonal behaviors, their affect, and their physical activity in an ecologically valid context. Nelson and Allen argue that this technology surpasses the capabilities of traditional methods, such as questionnaires or laboratory experiments, and newer experience-sampling methods that use digital devices to collect data at various time points. Smart-home technology has potential applications in research, assessment, and intervention, but it comes with technical challenges (e.g., poor audio or video quality) and analytical challenges (e.g., processing large volumes of data). Given the highly identifiable nature of the data collected by smart-home devices, researchers must carefully consider the ethical issues that may arise with respect to participants’ privacy and confidentiality. Nelson and Allen conclude that smart-home technology, if used with care and consideration, can advance psychological theory and practice, producing new insights into individuals and their relationships.

No Appointment Necessary: Treating Mental Illness Outside the Therapist’s Office

Bethany A. Teachman

Twenty-five years ago, it was assumed that face-to-face interaction was necessary for successful treatment of mental illness, but this view has recently begun to change. The burgeoning field of ehealth has provided new ways to reach people in need of mental health care. Early studies of internet-based treatments suggest that they can be just as effective for some people as traditional methods of care. Although there are still many questions to be answered about edelivery systems for mental health care, this approach is a promising way to meet growing mental health needs.

Improving Transparency, Falsifiability, and Rigor by Making Hypothesis Tests Machine-Readable

Daniël Lakens and Lisa M. DeBruine

Lakens and DeBruine propose an approach to make hypothesis tests machine-readable. Specifying hypothesis tests in ways that a computer can read and evaluate might increase the rigor and transparency of hypothesis testing as well as facilitate finding and reusing these tests and their results (e.g., in meta-analyses). The authors describe what a machine-readable hypothesis test should look like and demonstrate its feasibility in a real-life example (DeBruine’s 2002 study on facial resemblance and trust), using the prototype R package scienceverse.

Citizens Versus the Internet: Confronting Digital Challenges With Cognitive Tools

Anastasia Kozyreva, Stephan Lewandowsky, and Ralph Hertwig

The Internet, despite its many advantages, is also responsible for some negative outcomes, such as the spread of misinformation, rising incivility in online interactions, the facilitation of ideological extremisms, and a decline in decision autonomy. Kozyreva and colleagues identify the challenges that online environments pose to their users and to the functioning of society, and they explain how psychological science can inform interventions that allow users to gain more control over their digital environments.

The authors describe three approaches informed by psychological science that may help to create a digital world with fewer negative consequences: Nudges aim to guide people’s behavior through the design of choice architectures (e.g., default privacy-respecting settings); technocognition aims to design technological solutions informed by psychological principles (e.g., making it more difficult to share offensive material); and boosts aim to improve people’s cognitive and motivational competencies.

Focusing on boosts, Kozyreva and colleagues describe four tools that the designers of digital environments can implement to improve Internet users’ competencies. Self-nudging aims to enhance users’ control over their digital environments. Deliberate ignorance consists of deliberately not seeking or using information. Simple decision aids (e.g., the use of linguistic cues to distinguish between authentic and fictitious online product reviews) can help people accurately evaluate the content they encounter online. And inoculation, much like a vaccine, uses small and controlled exposure to misinformation to boost people’s resilience to online misinformation and manipulation.

Kozyreva and colleagues suggest that these tools can become “simple rules for mindful Internet behavior that could become as routine as washing one’s hands” and can be adopted by Internet users and policymakers to design and foster “Internet for citizens.”