Objective Physiological Measurements but Not Subjective Reports Moderate the Effect of Hunger on Choice Behavior

Abstract

Hunger is a powerful driver of human behavior, and is therefore of great interest to the study of psychology, economics, and consumer behavior. Assessing hunger levels in experiments is often biased, when using self-report methods, or complex, when using blood tests. We propose a novel way of objectively measuring subjects’ levels of hunger by identifying levels of alpha-amylase (AA) enzyme in their saliva samples. We used this measure to uncover the effect of hunger on different types of choice behaviors. We found that hunger increases risk-seeking behavior in a lottery-choice task, modifies levels of vindictiveness in a social decision-making task, but does not have a detectible effect on economic inconsistency in a budget-set choice task. Importantly, these findings were moderated by AA levels and not by self-report measures. We demonstrate the effects hunger has on choice behavior and the problematic nature of subjective measures of physiological states, and propose to use reliable and valid biologically based methods to overcome these problems.

Gambling on an empty stomach: Hunger modulates preferences for learned but not described risks

“…Hunger did not affect description-based risk-taking. This may be surprising given that previous studies reported that hungry individuals were more risk-seeking for food, water and monetary rewards when gambles were explicitly described (Levy et al, 2013;Shabat-Simon et al, 2018;Symmonds et al, 2010). The absence of an effect may be due to differences in task design.…”

“…The effect of decision context is thought to be driven by anticipatory emotions (De Martino et al, 2006) as well as biological needs (Stephens, 1981). Nevertheless, only a handful of studies has investigated the effect of physiological factors, such as hunger, on explicit risk-taking behaviour in humans, and suggest that hunger increases risk-seeking (Levy et al, 2013; Shabat-Simon et al, 2018; Symmonds et al, 2010), but the effect of hunger on experiential risk-taking has not yet been tested in humans. Biological need, which is described as the disparity between the current state and the goal state, has been shown to motivate decision-making in animals that make experiential choices (Aw et al, 2011; Papageorgiou et al, 2016; Pompilio et al, 2006) and has been captured by computational models (van Swieten and Bogacz, 2020).…”

“…Hunger may modulate high-level decision-making systems, with the appetite-stimulating hormone ghrelin activating receptors distributed widely in the cerebral cortex including hippocampus (Zigman et al, 2006) and can enhance memory and performance (Diano et al, 2006). Accordingly, hunger may increase risk-seeking for explicitly described food but also monetary reward (Levy et al, 2013; Shabat-Simon et al, 2018; Symmonds et al, 2010), suggesting that metabolic signals do impact cognitive decisions.…”

Salivary Alpha-Amylase Mediates the Increase in Hunger Levels in Adolescents with Excess Weight after Viewing Food Images

“…20 Furthermore, recent studies have confirmed that in a food intake-related context, higher sAA levels are associated with a physiological measure of perceived satiety. [21][22][23][24] In individuals with lower sAA levels, the digestion of carbohydrates will be slow, which would serve to maintain high levels of hunger over a longer period, leading to greater food consumption before achieving satiety. In the long term, this would increase the propensity toward obesity.…”

“…25 This study aims to evaluate basal sAA differences between excess weight (EW) and normal weight (NW) adolescents, and the association between sAA and subjective hunger levels after viewing food images. Most studies have focused on the role of sAA as a biomarker of hunger after food consumption [21][22][23][24] ; however, to date, very few studies have analyzed the association between sAA levels and feelings of hunger in a food-cue display context. Furthermore, no research has focused on evaluating sAA in adolescents with EW, and most recent studies have been focused on the number of copies of the sAA gene, with few directly comparing sAA levels between groups differing in body weight.…”