School of Education faculty and students are presenting at the 59th Annual Meeting of the Psychonomic Society in New Orleans, November 15-18.
The society, formed in 1960 in Madison, Wisconsin, has emerged as the preeminent society for the experimental study of cognition. Over 4,100 members conduct research on questions concerning memory, learning, problem solving, decision making, language, attention, and perception. Members also connect with research in biology, chemistry, statistics, computer science, medicine, law, and business.
School of Education Presentations (organized alphabetically by title)
Title: Improving the Effects of Working Memory Training in Young and Older Adults: The Effects of Gamification and Spacing
Presenter: Susanne M. Jaeggi, University of California, Irvine
Co-authors: Martin Buschkuehl, MIND Research Institute; Anja Pahor and Aaron Seitz, University of California, Riverside; Priti Shah, Patricia A. Reuter-Lorenz, and John Jonides, University of Michigan, Ann Arbor
Abstract: The vast majority of our nation’s population will experience some cognitive decline as a function of age. Therefore, the development of effective interventions to mitigate age-related cognitive decline is of critical importance in that those interventions might not only impact older adults’ cognitive functioning, but ultimately, contribute to their health and quality of life. There is accumulating evidence that cognitive interventions targeting working memory are beneficial in that they show generalizing effects that go beyond specific training effects. Despite the promising results, more research is needed to make cognitive interventions more robust, and to uncover their underlying mechanisms. I will be presenting the results of several randomized controlled multi-site trials targeting older and young adults in which we focus on the interventions’ optimal scheduling (i.e., spacing of training sessions) as well as the question of whether adding certain gaming elements might improve engagement, and ultimately, performance. Our interventions indicate promising effects that last up to six months after training completion. I will be discussing the implications of our findings for learning and plasticity across the lifespan.
Title: Predicting the Quality of Learning on a Working Memory Training Task
Presenter: Shafee Mohammed, University of California, Irvine,
Co-authors: Benjamin Katz, University of Michigan; Chelsea Parlett, University of California, Irvine; Martin Buschkuehl, MIND Research Institute; Mark Steyvers, University of California, Irvine; John Jonides and Priti Shah, University of Michigan; Susanne M. Jaeggi, University of California, Irvine
Abstract: The idea of training, in a generalized form (education) or specific skills (such as working memory) to enable learning, is not new. Despite our extraordinary ability to learn, the quality of learning often differs among individuals. However, studies seeking to understand the interactions of the past and future events of human learning on a single task to understand cognitive flexibility are rare. Here, we predict final performance occurred on two WM training tasks from initial performance and demographics while establishing a stable ‘baseline performance’ that is sufficient to accurately predict final performance (a small dataset with 450+ participants and a large dataset with 8000+ participants completing 15 sessions of training.) Our predictive model allows classification of final performance into above or below median with 86% accuracy. After the first 6-8 sessions, nontraining features (demographic information & training parameters) carried diminishing value as the information gained from training behavior is more valuable for prediction. This stable baseline performance might indicate the window of cognitive flexibility and/or the peak performance. Our work may allow early identification of high/low performers in learning.
Title: Variations in Working Memory Ability in Sighted and Unsighted Individuals
Presenter: Karen Arcos, University of California, Irvine
Co-authors: Susanne M. Jaeggi, and Emily D. Grossman, University of California, Irvine
Abstract: We investigate whether and how different senses impact short-term memory (STM) and working memory (WM) in blind and sighted humans. People who are blind have superior STM relative to the sighted. Braille’s influence on memory in blind individuals is less clear. Braille is a written code for the blind, in which raised dot patterns felt with the fingertips represent characters. Sighted participants and legally blind participants were recruited. Participants completed a demographic questionnaire. They also completed STM and WM tasks: visual, auditory, and haptic digit span tasks and a verbal n-back task. Dependent variables were task accuracy. On the digit span, sighted participants remembered significantly more items visually compared to auditorily. They also showed the expected load-dependent effect on the n-back task. Blind participants scored significantly higher than the sighted on the auditory digit span. The difference was even higher among blind braille readers compared to the sighted on the same version. Preliminary data indicate modality’s role in memory is important. Findings will better explain how encoding modality differentially influences memory performance in both groups.
Title: Working Memory Training Reduces Neural Recruitment in Older Adults: Support for the Compensation Related Utilization of Neural Circuits Hypothesis
Presenters: Alexandru D. Iordan and Katherine A. Cooke, University of Michigan
Co-authors: Kyle D. Moored, Johns Hopkins University; Benjamin Katz, Virginia Tech; Martin Buschkuehl, MIND Research Institute; Susanne M. Jaeggi, University of California, Irvine; Thad A. Polk, Scott J. Peltier, John Jonides, and Patricia A. Reuter-Lorenz, University of Michigan
Abstract: Functional neuroimaging studies have documented greater, more widespread activation in older (OA) than young adults (YA) at the same level of task demand. According to the Compensation Related Utilization of Neural Circuits Hypothesis (CRUNCH), additional neural resources are recruited with increasing demand regardless of age, however OA over recruit at lower levels of demand than YA, and reach capacity sooner causing activation decline at higher loads (quadratic CRUNCH curve). We tested whether adaptive training on a verbal working memory (WM) task would improve performance and reduce age-related over-activation. Pre-training, OA showed greater WM network recruitment than YA across all loads with decline at the highest load. Ten days of training improved WM performance in both groups. At the neural level, OA showed a CRUNCH curve shift indicating reduced recruitment at medium loads and greater recruitment at the highest load. Training may improve the dynamic range of neural recruitment in OA.
2018 Meeting Program