by University of Amsterdam

Credit: Pixabay/CC0 Public Domain

Young people today are growing up in a social media-saturated world where technology plays a central role in shaping most of their experiences. The rapid rise of social media use has consequently created parental and societal fears about young people's social and psychological well-being.

Now, for the first time, a team of researchers led by the University of Amsterdam has used real social media data to show that young people may indeed be more sensitive to social media feedback (likes) than adults, and that this directly impacts their engagement and their mood. The results were published on 23 October in the journal Science Advances.

One of the fears surrounding social media is that it could provoke anxiety in young people by driving them to keep using the apps more than they want to so they can gather more and more likes. Team member Wouter van den Bos explains, "Adolescence is a developmental period during which both reward and rejection sensitivity are particularly strong, and these have, respectively, been linked to increased impulsive behavior and depressive symptoms."

Crucial period in our lives

The researchers used a three-pronged approach to examine the issue. First, they looked at a large dataset of real-life Instagram posts, and used a computational model to capture the sensitivity to likes. Second came an experimental study that mimicked the features of social media platforms and could be used to track changes in mood.

Finally, an exploratory neuroimaging study showed that sensitivity to social media feedback is related to individual differences in amygdala volume. Taken together, the three studies showed converging evidence that young people may indeed be more sensitive to social media feedback than adults.

Computational approach. (A and B) The RL model states that the agent will select the latency until it will post again, τPostt, after it receives feedback on the current post. This posting latency is drawn from an exponential distribution determined by the current policy (Policyt). It posits that the agent’s posting latency is influenced by the number of likes received (e.g., 23 likes represented by the heart). The model states that the agent maximizes the reward rate by adjusting the policy (Policy update) after receiving a certain number of likes for a particular post. The model policy is adjusted on the basis of the learning rate (α), the change in posting latency (ΔτPostt), and the net reward prediction error (δ). The learning rate parameter indicates the sensitivity to social media feedback, and it is our prime variable of interest. The δ consists of the difference between the reward received and the average net reward rate (R) which considers both the effort cost associated with quick responses and the opportunity cost (or missed opportunities) of slow responding. The RL model was fit to individual Instagram trace data (see Fig. 2, A to C, for frequency distributions), where we estimated individual values for three free parameters: learning rate, α; effort cost sensitivity, C; and initial policy, ρ, based on maximum likelihood estimation. We focused on the two first parameters to test age differences in sensitivity to likes (α) and effort associated with posting. Credit: Science Advances (2024). DOI: 10.1126/sciadv.adp8775

Adolescence is a crucial period in our lives, marked by heightened sensitivity to peer approval and rejection. In the context of this research, this heightened sensitivity leads to an interesting paradox: While receiving likes seems to generate a feeling of connectedness and can enhance young people's mood, this positive outcome could also create such a pull toward the apps that it could lead to problematic overuse.

On the other hand, given their sensitivity, young people would stop using the platforms sooner than adults if they weren't receiving the likes, but this could also lead to increasingly negative moods.

Intervention for negative impacts

With the researchers' results suggesting that the current design of social media platforms may have both positive and negative impacts on young people, it seems interventions to address the negative side effects may be needed.

The researchers propose that firstly, platforms should change incentive structures, shifting the emphasis away from likes and towards more meaningful engagement. Secondly, they suggest that we should not only focus on strengthening the digital literacy of young people—since they probably know more about this topic than any other generation—but instead focus on developing skillful emotional regulation in online environments.

Study first author Ana da Silva Pinho notes, "While social media plays an important role in fostering some aspects of youth development, such as identity formation and social connection, our study reveals that it may also present challenges, particularly relating to young people's moods. Given the growing concerns about the impact of social media on mental health, it is crucial that we further understand how young people engage with and respond to social media, while also addressing the unique aspects of their developmental stages."

More information: Ana da Silva Pinho et al, Youths' sensitivity to social media feedback: A computational account, Science Advances (2024). DOI: 10.1126/sciadv.adp8775

Journal information: Science Advances 

Provided by University of Amsterdam