Towards an incentive salience model of intertemporal choice

Abstract

This theoretical paper presents an incentive salience model of intertemporal choice. The model is a variation of the quasi-hyperbolic discounting model. Based on the distinction between ‘wanting' and ‘liking', the paper presents one possible explanation of impulsive choices of smaller sooner rewards instead of larger later ones. These impulsive choices are induced by cues that trigger strong motivational ‘wanting' to obtain smaller sooner rewards, but do not necessarily influence the degree to which the rewards are ‘liked'. Cue-triggered ‘wanting' can occur when an individual is in a specific need deprivation state, perceives a cue previously associated with an immediately obtainable reward, knows that the cued reward can reduce the current deprivation state, and lacks self-control. Attributable to the integration of cue-triggered ‘wanting' into an intertemporal choice model, the incentive salience can account for anomalies in intertemporal choice such as present-biased preferences and the domain effect

The evolution of consumption and its welfare effects

“…a deprivation of their need for a positive self-image – by engaging in consumption activities symbolic of the ideal individual they would like to be. This is particularly true for consumers who tend to define their ideal self-image in terms of material possessions (Lades 2012a, b, Chap. 5).…”

Impulsive consumption and reflexive thought: Nudging ethical consumer behavior

“…Consequently, cues can casue individuals to impulsively 'want' identity-relevant goods that they turn out not to 'like' after, or during, the purchase. 8 For example, when an individual's actual self-image falls short in a social comparison with an ideal self-image adopted from the mass media, the individual 7 Adapting the incentive salience model presented in Lades (2011) to the case of identity-related consumption leads to the following formulation of the individual's motivation V (st) to consume an identityrelevant good ct that is rewarding (r(ct) > 1):…”

Autonomy-Enhancing Paternalism

“…Present-biased preferences are described by the parameter β. While β is below 1 in terms of decision utility, in terms of experienced utility, β is equal to 1 (see Lades, 2012 for a micro-foundation of β based on the dissociation of utility into components of wanting and liking). Since we assume that also the perceived benefits and costs from period-t consumption are additively separable from the perceived benefits and costs from consumption in any other period, at any time period the decision utility û(x, z) corresponding to 26.…”

“…Individual preferences are synonymous to "decision utility" (Kahneman et al, 1997) and "wanting" (Berridge and Aldridge, 2008). Preferences, but not actual happiness, can be subject to faulty affective forecasts or other biases (see more extensively Lades, 2012;Witt and Binder, 2013). 27.…”

“…The parameter β depicts a timeinconsistent preference for immediate gratification that reflects impulsivity. While β < 1 in terms of preferences or wanting, in terms of happiness or liking β = 1 (see Lades, 2012 for a micro-foundation of β based on the dissociation of wanting and liking). Since we assume that also the perceived benefits and costs from period-t consumption are additively separable from the perceived benefits and costs from consumption in any other period, the preferences corresponding to the consumption of x and z with the potential for distorted perceptions (α ≠ 1) and present-biased preferences (β < 1) can be written as…”