Kurzban et al.'s opportunity cost model (OCM) is proposed to explain the origins and adaptive nature of mental effort. The authors argue that if current and competing tasks both require executive functions, these tasks will be compared on their value. If the value of a competing task exceeds that of the current task, mental effort is experienced. This experience of effort is adaptive in nature, as it signals that executive functions should not be used for the current task but are better applied to the competing task. Kurzban et al. argue that their model can explain a wide variety of phenomena, including the finding that performance degrades if executive functions are used repeatedly, a phenomenon known as “resource depletion.” We argue that the OCM account of resource depletion requires three assumptions, two of which are likely not to be satisfied. We therefore discuss an alternative model that does not require the two unlikely assumptions.

However, before doing so, it is necessary to specify our interpretation of two key OCM concepts: “task value” and “effort.” In general, Kurzban et al. seem to define task value in terms of the positive aspects of a task (cf. sect. 2.4.1), yet in some instances they seem to allude to negative aspects as well (cf. Fig. 1). In addition, effort is generally defined as the discrepancy between current and alternative task values (cf. Abstract), yet occasionally the term seems to refer to a property of a single task (e.g., “might explain why subjects in self-control conditions exert less effort”; sect. 3.1, para. 10). In the following we adhere to the authors' general interpretations: Task value is defined only in terms of positive aspects, and effort is an index of the discrepancy between current and alternative task values.

The effects of repeated usage of executive functions are often taken to suggest that resources for executive function become depleted, hence the name “resource depletion” (Muraven & Baumeister Reference Muraven and Baumeister2000). Yet, this interpretation is subject to debate, as it has been suggested that the effects of repeated use of executive functions are better explained in terms of a depletion of motivation, rather than by a depletion of resources (Hagger et al. Reference Hagger, Wood, Stiff and Chatzisarantis2010a). Accordingly, Kurzban et al. provide an OCM account of the effects of repeated usage of executive functions, an account in which task value, a concept related to motivation, plays a key role. Below we argue that this OCM account relies on three assumptions, two of which are unlikely.

The OCM's first assumption is that a preceding executive function task reduces the value of a current executive function task. Kurzban et al. suggest one potential mechanism for this reduction: In the beginning of an experiment participants may feel obliged towards the experimenter, and therefore task value is high. But as the experiment proceeds, obligations are gradually fulfilled, and therefore task value decreases. The second assumption of the OCM account is that task value is compared to the value of a competing task also requiring executive functions; the authors focus specifically on the competing “task” of daydreaming. However, to our knowledge, there is no convincing evidence that daydreaming requires executive functions, and therefore it is not likely that this assumption is satisfied. A third assumption of the OCM account is that task value is defined only in terms of its positive aspects, and not by its negative ones, as, for example, task difficulty. This would imply that tasks differing in difficulty can have equal value and, thus, should lead to an equal experience of effort. As this corollary of assumption 3 is to our knowledge not supported by empirical evidence (Morsella et al. Reference Morsella, Wilson, Berger, Honhongva, Gazzaley and Bargh2009), we conclude that assumption 3 is not likely to be satisfied.

We therefore argue that a model of the effects of sequential usage of executive functions is needed that does include motivation, yet does not rely on the aforementioned unlikely assumptions. A recently proposed simple formal model satisfies these requirements (Huizenga et al. Reference Huizenga, van der Molen, Bexkens, Bos and van den Wildenberg2012). In this model, motivation determines the fraction of required resources that will be allocated to tasks, in which required resources depend on task difficulty. It is assumed that motivation decreases with repeated usage of executive functions, and as a result, performance will decrease also. This model does not require the unlikely second assumption, as there is no comparison of motivation (“value”) associated with current and alternative tasks. In addition, it does not require the unlikely third assumption, as task difficulty is explicitly incorporated into the model.

The model, however, does require the first assumption, as it is assumed that motivation decreases with repeated usage of executive functions. This assumption certainly needs further investigation, at a behavioral as well as at a neurophysiological level. At a behavioral level, it needs to be investigated whether indices of experienced motivation (e.g., Carlson & Tamm Reference Carlson and Tamm2000) mediate the effects of sequential use of executive functions. At the neurophysiological level, the effect of repeated use of executive functions on dopamine, a “motivational” neurotransmitter (Salamone & Correa Reference Salamone and Correa2012) that improves executive functions (Pessoa Reference Pessoa2009), needs further consideration. For example, in simple learning tasks, phasic dopamine releases decrease with repeated exposure to stimuli that are associated with expected reward (Schultz et al. Reference Schultz, Apicella and Ljungberg1993). An intriguing possibility is that these dopamine levels would also decrease with repeated performance on executive function tasks (Boksem & Tops Reference Boksem and Tops2008; Lorist et al. Reference Lorist, Boksem and Ridderinkhof2005).

To conclude, an advantage of the OCM account of “resource depletion” is that it includes motivation (value). A disadvantage, however, is that the OCM account relies on two unlikely assumptions. Therefore, an alternative model, relying only on the assumption that motivation decreases with repeated usage of executive functions, requires further investigation, both at a behavioral and at a neurophysiological level.