Christopher Summerfield and Konstantinos Tsetsos, ‘Do Humans Make Good Decisions?’, Trends in Cognitive Sciences, 19 (2015), 27-34.

It is possible to make humans seem like good decision-makers, Summerfield and Tsetsos explain, if you put them in an orderly experimental context and invite them to undertake a perceptual classification task. However, in real-world scenarios, such as when making economic decisions, humans prove to be ‘irrational’, and their choices vary according to context. Drawing together experiments wherein subjects experience ‘variable or volatile’ decisions even in the laboratory, and their decisions are shown to veer away from the ideal, Summerfield and Tsetsos propose that we have evolved an ‘efficient coding of decision-relevant information’.
      The point is that in a rapidly changing environment, and with a remarkable but nevertheless limited capacity to manage information, the human mind is all too ready to respond to expected rather than unexpected information, and to act along an expected path. This is consistent with the cognitive biases – especially confirmation bias – that are widely observed. However, I like the fact that Summerfield and Tsetsos are ready to question whether the resulting decisions are really bad: they don’t conform to ‘statistical optimality’ but they are designed for, and serviceable in, situations where statistical optimality simply cannot be calculated.

*

This made me think about whether literary works are orderly or disorderly environments in which decisions are made. It also made me think about how identifying the particular decision as such, in a complex literary plot or life situation, is not always straightforward. Hamlet has a reputation for indecisiveness: his decision, much delayed, is whether to kill Claudius. Or perhaps his decision is only a decision when he moves to kill Claudius; not killing him is not a decision, it’s an extension of the resting state.
      The hero’s predicament could be seen as a highly complex one, ‘variable and volatile’ in all sorts of ways. Hamlet has to weigh the ghost’s words against the way Claudius and Gertrude are acting, how they are viewed by others, Hamlet’s feelings for his mother, his other emotional entanglements, the problem of whether ghosts can be trusted, and so on. All of this would surely create a situation in which decision-making might revert to the ‘expected inputs’.
      In a revenge tragedy expectation points ruthlessly in one direction: eventually the hero will carry out the necessary act, and then die (or similar). For the audience, then, the ‘expected input’ is manifestly anything that points towards destruction. However, we are often, as in the case of Hamlet, presented with heroes for whom things seem less clear-cut. Hamlet is ill at ease, at odds with his mission in his cast of mind and his tone of voice. The ‘expected inputs’ still point strongly at a decision, but the experience of the individual is out of tune.
      I don’t have a lot more to say about this at this point, not least because the ‘a’ key on my computer (or the ‘’ key as it would rather be called) appears to be in terminal decline. This makes typing very annoying for the time being (until my complete vocabulary refit kicks in). Nonetheless, it seems interesting and productive that literature may configure ‘expected inputs’ and decision points in complex ways that may propose further nuances about these evolved mechanisms.