Journalists love to write headlines about how something "affects the brain." CNN's recent splash, "Teens, this is how social media affects your brain" is a good example. Can the study they report really support the causal claim that social media affects your brain? Let's check it out. Here's one of the journalist's ominous opening sentences
Social media has become second nature -- but what impact is this having on our brain?
This line implies that social media might even damage the teenage brain...but does it? Let's read on. The journalist is reviewing several aspects of the study in this article, but I will focus only on the main claim. Here's the journalist's description:
In a recent study, researchers at the UCLA brain mapping center used an fMRI scanner to image the brains of 32 teenagers as they used a bespoke social media app resembling Instagram. By watching the activity inside different regions of the brain as the teens used the app, the team found certain regions became activated by "likes", with the brain's reward center becoming especially active.
"When teens learn that their own pictures have supposedly received a lot of likes, they show significantly greater activation in parts of the brain's reward circuitry," says lead author Lauren Sherman. "This is the same group of regions responding when we see pictures of a person we love or when we win money."
(In the text's Chapter 5, fMRI is described as a physiological measure. As participants engage in tasks in an MRI scanner, the machine detects changes in brain activity, which are then analyzed by a computer, resulting in an estimate of which brain regions are more active.)
- Like many fMRI studies, this one uses a repeated measures experimental design. See if you can guess what the study's independent (IV) and dependent variables (DV) would have been.
- Assuming this was a repeated measures design, how might the researchers have controlled for order effects?
- The journalist makes a broad causal claim here, "Social media affects your brain." What is a more specific causal claim that the experiment in this study might be able to support?
- Assuming the study did follow the repeated-measures design we assumed in Question 1, can the study support this causal claim? Apply the three causal criteria.
1. We know that the study's participants viewed 140 Instagram photos while lying in an MRI scanner. Some of the photos received a lot of likes (assigned by the experimenter), and some received fewer. Therefore, the independent variable was probably the number of "likes" received by a photo (high vs. low), and the dependent variable can be described as "level of activity in the nucleus accumbens"
2. The journalist indicates that people saw 140 photos that varied on the number of likes they had received. To control for order effects, we'd want to be sure that the researchers presented the photos in different orders for each person. In addition, we'd also want to be sure that the number of likes assigned to each photo was not confounded with the contents of the photos. It would be a problem, for example, if photos with people in them had been assigned more "likes" than photos of landscapes. The researchers should be sure that likes were assigned regardless of photo content.
3. Instead of the general statement, "social media affects your brain," it is more specific to say, "viewing Instagram photos that have received more likes causes more activity in the nucleus accumbens."
4. Yes, the study probably can support this causal claim. The study reported covariance, because the nucleus accumbens was shown to be more active, on average, when viewing photos that had received more "likes" compared to photos that had received fewer "likes." Because the researchers manipulated how many likes the photos received, the study establishes temporal precedence--people saw the likes first, and then their nucleus accumbens activity was measured directly after. Finally, if we assume that the researchers controlled for order effects and potential confounds (see question 2, above), internal validity would have been established, as well.
It is certainly exciting to read that researchers have identified a specific region that is active when people like a teen's Instagram photos. But keep in mind that Instagram is not necessarily special. Neuroscientists would assert that every interaction we have with the world--talking to friends, listening to music, drinking a coffee--"affects" our brain. Would a well-liked Instagram photo get more nucleus accumbens activity than these pleasant activities? Future research might be able to tell us.
You can read the original journal article on which this study is based by visiting the journal, Psychological Science.