As we study the factors that enhance intrinsic motivation and undermine intrinsic motivation as we were talking about in the last lecture, one has become the most controversial at all and it's the issue of rewards. In fact, some of the studies in self-determination theory will show that rewards actually can have a negative effect on intrinsic motivation, a phenomenon that's been just a mystery to so many people. But it's also a phenomenon that has a lot of applied implications. So, I want to spend this short lecture just discussing what the meaning of this is. This phenomena really popped out in an early classic experiment by my colleague at DC that was published in 1971. In that experiment, he brought college students into a laboratory setting and he gave them a puzzle that he had pre-tested and showed that was typically a lot of fun for college students. It's called a soma puzzle. It's a spatial puzzle that we try and match up patterns using a block set of configurations. Most people found it fun. For some of the subjects he brought into his laboratory setting, he just had them engage in the activity, for others he told them that he would give them a financial reward for every problem they solve, and so, he gave them some dollars for each of their successful solutions and then this was a three session paradigm. So, they first were engaged in the activity, and the second session they either got rewarded or were not rewarded for the activity, and in a third session he brought them back in for the free choice paradigm which we just explained in the last lecture. So, they were free to either engage in the activity or not engage in the activity during that third period. What he found is that the participants who had been rewarded for their engagement in the activity with financial rewards, were less likely to persist in the activity during free choice than the ones who had not been rewarded at all. This was an explosive phenomenon that time because of course many apparent theorists thought or rewards should enhance intrinsic motivation, they certainly shouldn't be worse than no rewards at all, but at the same time this particular study started to get picked up by other people, was replicated, extended in different kinds of settings and has become a real phenomena. I'm going to give you a more current example of it just to really illustrate it in a more concrete way, and this is an experiment that was done at the Max Planck Institute by Felix Warneken and Michael Tomasello. They were asserting and have interest in the idea that humans are intrinsically motivated to be helpful to one another, that there's a kind of intrinsic motivation to our altruism. In part of showing that, they showed that toddlers as early as 16 months old and certainly by 20 months, are spontaneously oriented to help adults who were in need. In laboratory settings, they would have adults drop something on the floor, be unable to reach or open a door and they found that toddlers at 20 months of age, close to 90 percent of the time would come over to the adults aid. So, spontaneously and seemingly just out of interest toddlers want to help others. But they thought, if this is really intrinsically motivated, then we should be able to show the undermining effect by rewarding children for engaging in this altruistic behavior. So, they then did a study with 20-month-old toddlers which was the following and in laboratory settings, the adult would do something like drop something, leave a door open etc. and then if a child came over to help them in one condition which they called the neutral condition, the adults simply did nothing. So, if the adult picked up or if the child picked up the materials on the floor of the adult said nothing, did nothing, just accepted the materials from the child. The second condition was a what we call a non-controlling praise condition. Here the adults all said thank you for doing that. So, non-controlling acknowledgment of the good deed done. A third condition the adult did something which many of us might think would be a motivating event. They gave the childhood reward for helping. There was a special toy in the corner of the room that required a ball to operate and when the child did a helpful thing the adult turned and gave them one of the balls to operate that fun machine and said, "Here you get this for having done that" They made it clear that the reward was for the behavior that had just occurred. So, children would finish this session and then they were brought back for a second session and, again they got opportunities to help or not help and one thing Tomasello found is that, when children were in the neutral condition or in that non-controlling praise condition, they continue to help at higher rates. Whereas before they were helping around 90 percent of the time they continued to help around 90 percent of the time in this neutral and this praise condition. But when rewarded for helping, children's rate of helping went down dramatically. Now, that helping rate was around 54 percent of the time. A significant drop off in helping. So here, when children were rewarded with a positive event for helping, they were now much less likely to help in the future, and we can ask ourselves the question, well why is that? Well self-determination theory argues that before when the children were helping, they had what we called an internal perceived locus of causality. They were helping because they wanted to and they were doing it just out of their own initiative. But once they were rewarded for it, something shifted in their psyche about this behavior. Now they're helping because of what we call an external perceived locus of causality. Because some external event is now the driver of their behavior. When that external event is no longer obvious in the environment, then neither is the motivation that it was supporting. So, we think that the undermining effect occurs because the reward shifts the perceived locus of causality. Whereas before I was doing it because I wanted to, now I'm doing it because you've rewarded me and that means that my behavior has become dependent on your rewards and I've lost the intrinsic motivation I already had. Now, there's been a lot, a lot of studies on the rewards effects on intrinsic motivation. In fact in 1999, we presented a meta-analysis of over a 101 studies that had been done on this phenomenon, and I'm going to give you a tree diagram that represents the results of this meta-analysis. Don't get overwhelmed by the results, I just use it to illustrate a couple of things here. One is that, if you look at the top of this tree diagram, there's a box that's called all rewards. In general, rewards have a mild negative effect on intrinsic motivation. Here's a d of minus 24, which means it's a weak effect, and that's because we think there are some kinds of rewards that actually enhance intrinsic motivation and some that decrease it. When you give people a tangible reward and say, "If you do this I will give you this, that conveys an external perceived locus of causality and that tends to undermine and you can see that in a tangible reward box. On the other hand, sometimes people apply what we call a verbal reward which is you say, "Oh you did really well at that activity" Which is otherwise called praise, and praise typically would enhance competence and therefore enhance intrinsic motivation. So, verbal rewards typically would enhance tangible rewards would typically decrease. But if you take a tangible reward and you say, "Do this and I will give you this" that will definitely undermine intrinsic motivation. But, if you gave that reward differently, if the person did the activity, finished the activity, and then went up to him and you said, "Oh you did such a good job at that. I wanted to give you this reward as an acknowledgment of that" This would be an unexpected reward and it wouldn't undermine because it was never controlling the person's behavior in the first place. My point is that it not all rewards undermine intrinsic motivation only those that are experienced as controlling and there's a lot of nuances in both how to give out rewards and how to design incentive strategies, so as to preserve the person's sense of autonomy and not undermine their intrinsic motivation. But oftentimes people use rewards as a clumsy or a blunt instrument and it has these negative effects. Now, we can show the undermining effect of rewards on intrinsic motivation in lots of different ways, but I want to show you one other illustration of it that shows it in a neuropsychological format. This was an experiment that was done by Kou Murayama and his colleagues. What he did is he brought people into a laboratory setting where there was an MRI so that he could do FMRI analyses on their brain activity while they were engaged in activities. He designed the activity that was fun, that you can do inside of a magnet that was basically a reaction time game and in one condition he would have people play the game and when they had an accurate response in the game, they got a financial reward. In another condition, he just gave no financial reward and had people simply play the game. In the figure that I show you now, we're looking at a brain activation diagram here of the bilateral striatum. This is the reward center of our brain or what some people call the reward center of the brain. It's in the area of our brain that's very sensitive to reward effects that are going on in our environment or internally. What you can see in this diagram, is that in this first session where people are getting financial rewards for accurate responses is in this game, there's a lot of activity in this bilateral striatum area. There's a lot of firing going on. But noteworthy, people who are not getting rewards at all or also showing activation in the bilateral striatum and that's because the activity itself is rewarding, it's fun and it's interesting. In a second session Murayama and colleagues bring people back into the laboratory setting but this time no participants are getting any financial rewards and they have a good plausible cover story as to why that's the case. But now you see that when people re-engage in this activity, the people who had been previously rewarded are now showing very little or no activation in this area of the bilateral striatum, meaning they're not finding the task rewarding anymore. But people who had never been given financial rewards are still showing activation in this area of the brain. They're still finding this task rewarding. This is a one way of demonstrating that undermining effect. Similarly, you can look in another area, the lateral prefrontal cortical areas where we would typically look for cognitive engagement, and we see the same pattern here which is people who are very engaged when they were getting rewards but now in a period where they're not getting rewards they're disengaged in this task, where people who would never receive rewards remain as engaged before. If this were just an extinction paradigm as behaviorists would talk about it, you'd expect people who had been previously rewarded just to drop to the level of activity of people who had never been rewarded. But they dropped below that activity which is why we call this the undermining effect. Now, this impact of rewards has lots of practical implications because rewards can sometimes have a backfire effect when we're trying to use them to motivate people, we need to understand that in a nuanced way when we're using rewards in classrooms and in organizational settings and on sport fields. As we go on in this course we're going to be talking about just these kinds of practical implications and how we can use rewards to either support or undermine people's autonomy. But hopefully this little lecture has given you some sense of how sometimes even a positive thing can end up to be a negative for a certain kind of behavior.
