"This is largest human cognitive performance dataset," P. Murali Doraiswamy told InformationWeek Education in a phone interview. Doraiswamy is professor of psychiatry at Duke University Medical Center, member of the Duke Institute for Brain Sciences, and co-author of the research.
Most cognitive experiments involve a handful of undergraduate volunteers or patients, Doraiswamy said, noting that even large-scale disease studies might involve just a few thousand individuals.
"This is an order of magnitude bigger," he said. For example, the first study, which examined the effects of sleep and alcohol consumption on cognitive abilities, used data from 162,462; 161,717; and 127,048 people.
Still, Doraiswamy said he was fully expecting pushback in some academic circles, given the nature of the data.
"But [the] tech-savvy will say this [crowd-sourced, patient-based data] is the wave of the future," he said. Besides, these studies will complement, not replace, controlled experiments in a lab, he said.
A legitimate methodological complaint, Doraiswamy continued, is that the study participants were self-selecting -- that is, coming exclusively from people who downloaded and played Lumosity games. For this reason, Doraiswamy is interested in building a randomized, large-scale study, such as comparing the performance of students at 500 schools that use Lumosity apps against students at 500 schools that don't.
Ramping Up Research
"Right now we have about 36 projects, involving 45 different academic researchers at 31 universities around the world," Daniel Sternberg, Ph.D, data scientist at Lumosity and lead author of the study told InformationWeek Education.
To manage these research partnerships, Lumosity announced the Human Cognition Project, which it describes as containing the "world's largest and continuously growing dataset of human cognitive performance."
Like Doraiswamy, Sternberg emphasized the sheer size of the dataset.
For instance, the second study examined how learning ability changes over the lifespan and how aging might affect learning across distinct cognitive abilities. The study included adults ages 18-74, and looked at how age influences improvement over the course of the first 25 sessions of a cognitive task.
But instead of comparing a handful of 25-year-olds to a handful of 74-year-olds, the size of the dataset -- the smallest segment was 22,718 and the largest was 107,478 -- allowed the study to examine cognitive ability at every year between 25 and 74, Sternberg said.
Lumosity has no plans to push its dataset into the public domain or create an API for academics, citing both privacy and competitive concerns. However, Sternberg said Lumosity will make public the specific anonymized dataset used in any published research. None of the shared data will make use of unique identifiers, such as IP addresses or, in most cases, geography.
Although it's obvious that one of the reasons Lumosity is engaged in this research is to inform its game design and market intelligence, that isn't the only reason. "Our science team, our neuroscientists, are intrinsically interested in this, even if it's not necessarily core to our business," Sternberg said.
Mobile technology is changing the way we live, work and play. Attend Mobile Commerce World , June 24-26 in San Francisco, to learn how to develop and deploy your company’s mobile commerce strategy. Spend three days learning from key players at Walmart, Amazon, Zappos, 1-800-FLOWERS, Advance Auto Parts, Alex and Ani, Groupon, REI, Vegas.com and more who are harnessing the power of mobile commerce. Register with code MP_BTMEDIWKAXE and save $200 on an All Access pass.
6 Tools to Protect Big DataMost IT teams have their conventional databases covered in terms of security and business continuity. But as we enter the era of big data, Hadoop, and NoSQL, protection schemes need to evolve. In fact, big data could drive the next big security strategy shift.
Big Data Brings Big Security ProblemsWhy should big data be more difficult to secure? In a word, variety. But the business won’t wait to use it to predict customer behavior, find correlations across disparate data sources, predict fraud or financial risk, and more.