You wake up, uncomfortable, to beeping sounds and urgent voices. “He’s awake!” There are voices. Hurried footsteps. You notice the uncomfortable tubes. You’re in hospital. Over the next hour you learn you’ve been suffering from a mysterious it-could-only-happen-in-Hollywood illness. You’re recovering now but you have completely forgotten who you are. You have simply no recollection of your previous life. You can still speak English. People arrive who say they’re your friends and relatives. Some of them do indeed bear a physical resemblance. They take you to an apartment, bring you groceries, and help you remember your former life. After a couple of weeks, your friends are getting worried that you’re sitting in the apartment all the time. One suggests you try ‘riding’ a metal contraption. You know it’s called a bike but you have no idea how to ride it. Your friend shows you photos of yourself, on this very bike, and says you’ve always been a very accomplished ‘cyclist’. You put it off but you’re intrigued. Later the same day, and presumably motivated by the same desire to see you getting active again, another friend comes round with some rollerblades. You think you know the drill and so you ask whether you used to be an expert rollerblader. She says no but that you’ve always wanted to have a go. Now, which one of these activities, in your strangely amnesiac state, do you think you’re more likely to be able to do? If pushed by your well-meaning friends to try one or the other, which would you feel safer having a go at? Why? I’m guessing you said bike riding. Because apparently you’ve done it before. So now tell me. Which comes first, self-efficacy beliefs or the behavioural repertoire they’re about? In other words, do you get confident at something and then do it, or do it until you can do it confidently? In my humble opinion, a great many psychological theories get this wrong. Most of them in fact. A new paper by Professor Jane Ogden questions the usefulness of a large number of the models used by health psychologists to predict health behaviours. Amongst the issues she considers is how ‘common sense’ it is for the models to predict that self-efficacy causes behaviour, for example that people who are confident in cooking will cook more. She says we don’t need the model partly because it’s just codified common sense. As it happens I agree with Prof Ogden that these models tend to be fairly useless but for quite different reasons. As far as I can tell, she doesn’t have a problem with the central tenet of the information processing model of psychology, that thoughts cause behaviours. If you accept that general principle then it stands to reason that self-efficacy beliefs are causal with respect to their behavioural domains. I.e. If you believe that thoughts cause overt behaviours, then your model is bound to predict that self-efficacy beliefs about cooking, say, are causal with respect to the act of cooking itself. And indeed there are correlational studies that show a link between self efficacy and other behaviours but a first year undergraduate will tell you correlation does not imply causation. Where are the experiments where self-efficacy beliefs were enhanced without the participant actually practicing the behaviour, and where the result was an increased proclivity to engage in the behaviour? I can’t find them. Prof Ogden comes at this another way and declares these theories unfalsifiable. Yes indeed. And what’s worse, all these pretty boxes and arrows give us a false sense of progress. As scientists, we find a nice fat correlation coefficient and a significant p value and we raise our arms in triumph, without pausing to ask what the model tells us about practice. Does it give us demonstrably effective ways to intervene? If not, you can keep your theory. I’m not interested. This is the central problem with the information processing account in psychology. Far too often it posits the cause of behaviour as something internal to the organism — something not readily amenable to outside influence. And thus it leaves practiticing psychologists groping in the dark trying to find a technology that will affect the changes clients so ardently desire.