What is the suspicion that emerged in the research that 'cash benefits have increased the power of children's brain waves'?



In February 2022, a

study was published stating that 'financial support for poor families improved the brain activity of newborns,' which received a great deal of attention and praise from various media. However, 'whether or not income distribution is right or wrong, there are doubts about the experimental method for drawing that conclusion,' said the American think tank CSPI Center.

About Those Baby Brainwaves: Why “Policy Relevant” Social Science is Mostly a Fraud --Executive Summary --CSPI Center
https://cspicenter.org/blog/cspi/about-those-baby-brainwaves-why-policy-relevant-social-science-is-mostly-a-fraud-executive-summary/

This time, the CSPI Center is focusing on ' The impact of a poverty reduction intervention on infant brain activity ' announced by Sonya V Troller-Renfree, a psychologist at Columbia University Teachers College. About the effect on the brain activity of infants) '. In this paper, the researchers said, 'In an experiment in which 100 pairs of mothers and children in low-income households with newborn children were given cash of $ 333 (about 40,000 yen) per month, the group that was given cash was given. Infants showed strong brain wave power in the high frequency band. '

The study was heavily featured in major media outlets such as The New York Times and Vox , followed by policy think tanks and medical news aggregators.



However, the CSPI Center said of this study, 'As is often the case in policy-related social science studies, the content of the claim is highly exaggerated and the experimental methodology is questionable.' The evidence for the relationship between brain waves and cognitive ability is very weak, and most interventions to improve a child's cognitive ability are ineffective or zero, 'he said, arguing that support for mothers and children in poor families should be supported. It is said that it was a research with conclusions to support it.

One of the problems pointed out by the CSPI Center is the process of selecting data that is the basis for the increased power of brain waves. Troller-Renfree and colleagues initially started their research with plans to conduct experiments to measure theta waves (4-8Hz), alpha waves (8-13Hz), and gamma waves (35Hz and above) in the brain. These three brain waves were selected based on the fact that another study suggested that 'improvement in economic conditions would have a significant effect.'

However, when we actually measured it, the biggest result was the beta wave (13-35Hz), which was not originally planned. So, Troller-Renfree et al. Cited two papers as the latest research results showing the relationship between income and beta waves, and concluded that 'paying money increased the power of brain waves such as beta waves.' That. However, one of the two papers targeted only 60 children, and the other, contrary to the purpose of the study by Troller-Renfree et al., 'Between income and beta waves. There is a negative correlation. '



In addition, ' (PDF file) Multiple test problem ', which increases the risk of finding a significant difference even though there is not really a significant difference by repeating the test, has not been corrected. ' There are many inappropriate points in the paper, such as points, 'the number of participants in the experiment is small and the sample size is small', and 'the data targeted for analysis are only the data of studies targeting extreme households'. Can be seen.

Due to these problems, Troller-Renfree's dissertation was initially welcomed by politicians and others who said that welfare should be expanded, but it has since been exposed to many criticisms. In response to this, the UBI Center, which studies basic income policy, deleted all articles related to this research, and other research institutes took measures such as adding disclaimers. In addition, various news media such as Vox have also taken measures such as updating articles and posting conflicting opinions.

For this reason, the CSPI Center said, 'Cases like this one rethink the role of social sciences in public policy debates.' 'The flashy research that supports popular policies is usually unreliable. Yes, the dramatic effect is usually an exaggeration, a p-hacking of favorable data selection, or a coincidence. '

in Science, Posted by log1l_ks