Our reward is in merely anticipating what might await us on the other side of that click. Our past disappointments only heighten the anticipation that, this time, we will be rewarded for our trust.
David A.Tomar, editor-in-chief of The Quad
Positive reinforcement has an even stronger effect when we do not know exactly what we are getting. Therefore, the variability of the reward is particularly attractive and fuels curiosity. “It’s as if your brain – a born pattern recognition apparatus – can’t resist the temptation”. Any form of user feedback is a variable reward, because human reactions are unpredictable.
Social media algorithms are usually “adaptive”, which means that they are constantly changing themselves a bit in order to achieve better results.
Jaron Lanier, in: “Ten reasons why you need to delete your social media accounts immediately”, p.23
Algorithms thus contain a kind of “jump component” that makes them briefly jump out of their sequence – similar to mutation in natural evolution. Such events open up new possibilities for change and thus opportunities for improvement.
Feedback generated by social bots, news feed and video recommendations are subject to this principle. Even if user behaviour and preferences are known, algorithms repeatedly provide surprises in the form of deviating recommendations.
So-called “clickbaits” that are widespread in the networks also entice users with variable rewards.
rewards. These are polarising headlines in the form of links that invite clicks and promise sensational insights. Mostly, however, there is nothing behind them and the links only serve the purpose of generating as much traffic as possible on third-party websites.