because somewhere on Earth it’s probably still Wednesday…
“Citizen science” projects have successfully gotten people to classify galaxy shapes, find optimal protein folding structures or decipher manuscripts. But is it possible to use online participation not just to harness users problem-solving abilities but for actual experiments and data gathering in cognitive research? There has been much furore recently about a study on the spread of emotion that used data not only gathered through Facebook without users’ consent, but for which users’ newsfeed was actively manipulated. In contrast, researchers from the University of Oxford and the University of Birmingham now show that data for cognitive science experiments can be crowdsourced with informed consent, and results from the mass data collection replicate results which researchers got when they carried out experiments with the same aim in a carefully controlled lab setting.
The researchers got an app produced for smartphones, which allows users to play four short games that are based on classic experimental paradigms of cognitive science. These tested decision-making, short-term memory, perception and action inhibition but were packaged as competitive games. In one of them, users tapped their screen when fruit fell from a tree, but not when the fruit turned brown while falling. When the users tapped correctly and quickly enough, they progressed to the next level, where fruits turned brown later in their fall. Before playing, users filled out a questionnaire about themselves and gave informed consent for the use of their data. Interestingly, when analyzing the data, the researchers got the same results that other researchers got when they carried out experiments with the same goal in a lab setting, with only a few participants. One advantage of mass data collection over the internet is that the participants come from a wider demographic than the usual lab rats for cognitive science, cash-strapped university students. Soon, there will be two billion people using smartphones worldwide. That’s quite a wide pool of potential participants for online experiments, if done correctly. And, Facebook – no, informed consent is not optional.
If you are a football fan, you probably remember Götze’s title-scoring goal for Germany in Sunday’s world-championship final. If you are not a football fan, you probably remember the pleasant evenings you spent before the football craze set in. The memories are clear before your inner eye, but how do you store them in your brain?
Neuroscientists know that one brain region, called the hippocampus, is our memory storage. However, they have three different theories of how the brain cells, or neurons, in this region can store the memory of Götze’s goal. According to the first theory, one neuron encodes this memory. So this neuron, and only this neuron, sends a signal when you think of Götze scoring – you have a “goal neuron”. The second theory states that many neurons together send signals in a pattern, and this pattern is typical only of your goal memory. But each neuron also contributes to many other memories, like that of the cool beer you drank alongside. Basically, you have a “goal pattern”. The third theory falls in the middle: only a few neurons signal when you think of Götze’s goal, and each neuron also stores a few other memories. But which theory is true?
Psychologists in the US tested this by looking at the brain activity of people recognizing familiar versus new words. The participants in the study were epilepsy patients who wore wire electrodes in preparation for possible surgery, with the aim to find out where in the brain seizures took place. To test how their brains store memory, the researchers gave participants a list of 32 words, which they studied. In the memory test, they were shown 64 words – the 32 from the original list, and 32 new words. The participants were asked to say which ones were “old” words and which ones “new”. Using the electrodes, the research could see the areas of the hippocampus in which neurons sent signals when a new or old word was shown. They found that neurons in some areas signal more often when the participant sees an “old” word rather than a “new” word. For each area, neurons only signal when the participant sees a few “old” words.
This supports the third theory: it is likely that in the hippocampus our memory of Götze’s goal is stored by a group of a few neurons. Each of these neurons, together with a different group of neurons, also stores a few other memories – maybe that of Argentina’s very near misses.
Orginal research paper in PNAS: www.pnas.org/content/early/2014/06/11/1408365111.abstract