New review prompts a re-think on what low sugar levels do to our thinking

Glucose. Fuel for our cells, vital for life. But how fundamental is it to how we think?

According to dual-systems theory (best known from Nobel laureate Daniel Kahneman's work), low blood glucose favours the use of fast and dirty System One thinking over the deliberative, effortful System Two. Similarly, the ego depletion theory of Roy Baumeister sees glucose as a resource that gets used up whenever we resist a temptation.

But the authors of a new meta-analysis published in Psychological Bulletin find these claims hard to swallow. Their review suggests that glucose levels may change our decisions about food, but little else.

Jacob Orquin at Aarhus University and Robert Kurzban at the University of Pennsylvania searched the decision-making literature, finding 36 articles that directly investigated glucose by measuring blood concentration, providing participants with sugar solution, or via interventions such as wafting food smells, which triggers some amount of glucose production.

The authors pored through the articles and tabulated every effect, its direction as well as its size. They found the effects were very variable, often operating in different directions from study to study. But when the data was organised according to a key factor, a consistent pattern began to emerge. That factor? Food.

In payment tasks – involving hypothetical purchases ("how much would you pay for …") and actual purchases while shopping – low blood glucose did increase people's willingness to overspend … on food. But it actually made them less willing to spend money on non-food products. When it came to persistence on tasks (such as time spent trying to complete a puzzle), low glucose decreased willingness to work for non-food rewards, but led to more tenacious work towards food-related goals. And when people were given the choice between receiving a small amount now or a large one later, low glucose led to a large bias towards immediate gratification when food was the payoff, compared to a much smaller bias for non-food.

This pattern of results doesn't fit the notion of glucose as willpower-fuel. It suggests instead that low glucose is a signal that, to ensure future wellbeing, food should be prioritised – by paying more for it, working harder for it, and grabbing a little now rather than taking the promise of more in the future. This signaling account also explains the recent discovery that you don't need to consume glucose to produce some cognitive effects, simply tasting it is enough (by swishing around the mouth); no fuel has been received, but presumably the signaling system is temporarily fooled by the taste receptors.

Kahneman can sleep easy – the findings from this meta-analysis aren't a blow to his dual process theory as a whole, merely the specific claim that glucose has a role in switching between thinking smart and slow. The meta-analysis is a more substantial problem for the claims of ego depletion, which are intimately related to the idea that willpower is a finite resource that depends on glucose.

Based on the prior glucose research and theory, some publications have recommended strategies like eating chocolate before tense marital discussions or stacking emergency Jelly Belly's in the office desk drawer. But according to this meta-analysis, these strategies will yield little benefit; the main implication of being low on glucose is a greater preoccupation with finding something to eat. There's a lot of strong psychological science out there to help with building everyday habits and making better decisions, so if you're looking for a dose of something, we recommend you check those out instead.

Further reading

Orquin, J., & Kurzban, R. (2015). A Meta-Analysis of Blood Glucose Effects on Human Decision Making. Psychological Bulletin DOI: 10.1037/bul0000035