Highlights of American Psychological Association Conference on Technology, Mind and Society

WHAT:                 Technology, Mind, and Society – an interdisciplinary conference exploring the links between psychology and technology

The conference will focus on efforts to understand and shape the interactions of human beings and technology. Topics to be presented include whether virtual reality is ready for primetime, how fake news persists on social media, the use of games to enhance psychotherapy, how wearable technology may help improve well-being at work, and how technology can help people successfully age.

WHERE:                Grand Hyatt Washington 1000 H St., N.W. Washington, DC 20001

WHEN:                 Thursday, Oct. 3, 5 p.m. to 7:15 p.m.

                             Friday, Oct. 4, 7:30 a.m. to 5:45 p.m.

                            Saturday, Oct. 5, 7:30 a.m. to 5:30 p.m.

Media wishing to attend should register in advance by emailing APA Public Affairs or calling 202-336-5700. Registration is limited to credentialed journalists and journalists who have assignment letters on the letterhead of a media organization. All media personnel must be registered and wear their badges while attending any conference session or activity.

A full program for the conference can be found online here.

Presentations of note include:

“An Experimental Prosocial Intervention to Reduce Cyberbullying in YouTube Comments: Mixed Methods Findings,” Poster Session, Friday, Oct. 4, 7:30 a.m. – 9 a.m.

To help combat cyberbullying in the comments section on YouTube, researchers at the University of Florida are developing a novel strategy employing internet chatbots that identify toxic language online and intervene by automatically posting prosocial comments as if they were real people. Approximately 41% of people in the United States reported experiencing some form of online harassment and 66% believe online harassment is a serious problem, according to a 2017 report by the Pew Research Center. Cyberbullying is consistently associated with various psychological and behavioral problems, including anxiety, depression, substance use and suicide ideation. Supported by a grant from the National Science Foundation, this project has the potential to help build a healthier online environment for internet users.

“Racism and Sexism by Computers Reduce Outrage,” Paper Session, Friday, Oct. 4, 10:45 a.m. – 12 p.m.

In 2014, Amazon created a machine learning bot that screened job applicant resumes. However, the undertaking was scrapped in 2017 after it was discovered that the program was systematically discriminating against women. People tend to express moral outrage when other people discriminate, but how would they react when a computer algorithm does the same thing? Yoachanan Bigman, PhD, of the University of North Carolina Chapel Hill will present results of a series of experiments involving 480 participants that suggest people are less outraged when a computer algorithm discriminates based on race or sex. The findings indicate that people are outraged by discrimination not only because of the harmful consequences, but also the motivation behind the discrimination. When it comes to hiring decisions, automating the process by computer screening can be more efficient and impartial than humans, but caution is needed, he says?. If people’s natural defenses against injustice and discrimination are lowered when the agent is a computer instead of a human, it can make it easier for discrimination to go unnoticed, unobserved and unopposed, according to Bigman.

“The Development of a Virtual Reality Shooter Paradigm and Its Application to a Study on Racial Bias in Police Officers’ Use of Lethal Force,” Paper Session, Friday, Oct. 4, 10:45 a.m. – 12 p.m.

There has been growing media attention on police violence toward unarmed black men and women, and while a growing body of research suggests racial bias may be to blame, critics of the research suggest that the way the data is collected lacks validity. Typically, research subjects sit in front of a computer screen and respond to still images of suspects who appear, either holding a gun or a benign object. These methods are not likely to induce the type of cognitive or affective states experienced by police in high-pressure environments. To better recreate this type of environment and collect more valid data on race and police shootings, researchers at Mount Holyoke College have developed a virtual reality simulator to embed participants in a variety of 360-degree immersive scenarios, including traffic stops and altercations on the street, that unfold unpredictably in real time. For example, suspects may appear in the periphery or even from behind the participants.  Headsets worn by participants can track eye movements and measure pupil dilation as an indicator of physiological arousal. John Tawa, PhD, an assistant professor of psychology at Mount Holyoke College, will present the most recent findings from research using this virtual reality technology to better understand the differences in police reaction times when faced with black or white suspects.

“Using Digital Data to Measure and Understand Mental Health,” Symposium, Friday, Oct. 4, 10:45 a.m. – 12 p.m.

When people post to social media or use wearable devices, the digital data generated through these everyday interactions reveal a great deal about their behaviors in ways that were previously not observable. Participants in this symposium will present research suggesting that these data may help predict relapses in patients suffering from schizophrenia. Existing strategies to detect relapses, including clinical interviews, patient self-reports and family observations are severely limited by reliance on direct, frequent and timely contact with trained professionals as well as accurate and insightful patient and family recall. By analyzing a patient’s historical Facebook postings, researchers at the Georgia Institute of Technology have been able to predict relapses as far as one month in advance with an accuracy of 71%. Using a smartphone app to passively measure mobility and social behavioral patterns  in patients suffering from schizophrenia, researchers from the University of Pennsylvania found a 71% higher rate of behavioral anomalies in the two weeks prior to relapse, suggesting that data collected from smartphones in the background, with no need for active input from subjects, may be a useful tool for predicting relapse.

“Rethinking Anthropomorphism: The Antecedents, Unexpected Consequences, and Potential Remedy for Perceiving Machines as Humanlike,” Symposium, Friday, Oct. 4, 1:15 – 2:30 p.m.

With the rapid rise of robots and artificial intelligence in the public sphere, anthropomorphism, or the tendency for people to attribute human-like characteristics to nonhuman entities, has become an important topic for psychologists and other professionals who design, develop and even regulate these emerging technologies. What leads people to perceive machines and algorithms as humanlike, and what are the potential consequences of the anthropomorphism of technology? Participants in this symposium will present research suggesting that how human a robot appears systematically influences people’s beliefs regarding the robot’s humanlike mental capacities on a number dimensions; that chatbots that intentionally make an error (like a typo) and subsequently correct it appear more humanlike; and that as people perceive technological products as more humanlike, they also perceive and treat actual people more as if they were objects. 

“News Consumption on Social Media: Uninformed but Unaware,” Paper Session, Saturday, Oct. 5, 11 a.m. – 12:15 p.m.

What makes people more likely to like or share fake news on social media without reviewing it thoroughly first? Researchers from the University of Zurich present findings from two online experiments suggesting that when people perceive news items to be relevant to the public but not relevant to them personally, they are more likely to think they know more about the topic without reading the article. Social media users use social signals, like the number of likes and shares an article receives as well as comments from others, to decide if news might be important for others to know. In order to alleviate their fear of missing out, users will fool themselves into thinking they are better informed about content they consider publicly relevant and often will not click through and read the new item thoroughly before sharing unless they perceive that it is personally relevant to them. Until now, most research on social media content focused on engagement without considering the effects of social media on information processing. This research suggests that social signals on news content can lead individuals to feel better informed on a news article without reading it.

“Measuring Psychological Variables using Mobile Sensing Technologies: Modeling Big Data and Implications for Research and Designing Intelligent Support for Well-Being and Productivity at Work,” Symposium, Saturday, Oct. 5, 11 a.m. – 12:15 p.m.

Over the past century, computers and more advanced education of the workforce have led to increased productivity, but continuous connections to mobile devices and expectations of immediate communication through email or instant messaging adds more stress to the workplace. Poor workplace well-being can lead to negative impacts on employee physical and psychological health, resulting in reduced job performance and increased absenteeism. Researchers in this symposium present findings from the mPerf project, a multimillion-dollar effort to measure employee well-being using a variety of unobtrusive, wearable mobile sensors to gather psychological data from hundreds of employees of several organizations collected over 10 weeks. Presentations will provide an overview of the sensor technologies used and data analytics approaches, as well as preliminary findings from the project.

“Employing the Virtual Reality Buffet to Measure and Understand Parent Feeding Behavior,” Paper Session, Saturday, Oct. 5, 11 a.m. – 12:15 p.m.

Researchers from the National Institutes of Health are employing virtual reality to better understand how parents’ food choices can affect their children’s weight. One important predictor of adult obesity is eating behavior during childhood, a process in which parents play a significant role. One of the difficulties in researching parents’ food choices for their children is that the use of real food can be costly and wasteful. Another limitation is sterile laboratory conditions can change participants’ behaviors. To overcome these problems, researchers have developed an immersive virtual reality buffet simulation where parents can create a meal for their children in a realistic environment that can still be controlled. Using this technology, researchers have found that providing parents information on how family risk of obesity for their children later in life is associated with parents’ choosing fewer calories for their children. They also found that providing that risk information induces guilt among parents and choosing healthier foods from the virtual reality buffet can reduce those feelings of guilt.

“Transforming Access to Mental Health Care for Rural and Undeserved Populations,” Symposium, Saturday, Oct. 5, 3 – 4:15 p.m.

Approximately one out of five Americans is suffering from some type of mental illness, according to the most recent data from the National Survey on Drug Use and Health, and approximately 4.5% have a serious mental illness. One out of five adults with mental illness report that they are not able to receive the treatment they would like. One of the largest barriers to receiving treatment is a lack of access. More than 115 million Americans are living in areas lacking readily accessible mental health care, according to the U.S. Department of Health and Human Services. One strategy for addressing this shortage is telehealth, which uses modern telecommunications, including telephone and video conferencing, to allow individuals to  access mental health treatment remotely, either from home or a local primary care facility. Participants in this symposium will present data outlining the scope of the unmet need for mental health care services in rural areas and how the application of telehealth strategies holds promise for meeting that unmet need. They will also present on how one clinic, the Texas A&M Telehealth Counseling Clinic, has  deployed this strategy to deliver mental health services to seven rural counties in central Texas and how the organization’s efforts could serve as model for other counseling centers to deliver mental health services where there are few or none available.

Original post https://alertarticles.info