Were the mind any other sort of system that processes information, there would be few objections to the statement that mechanisms (processors) fuelled by resources carry out processes in order to perform tasks. Poor performance may arise because the mechanisms are overloaded, because resources are being depleted faster than they can be renewed, or for both of these reasons.
We can also think of processors as a type of resource; however, they are occupiable rather than depletable. (Both my car and the fuel it uses are travel resources – but the car is an occupiable resource, whereas the fuel is a depletable one.) If we use this sort of terminology, we still need to recognize that both types of resource (depletable and occupiable) are needed to carry out processes to perform tasks. (Having just fuel or just a car will not enable me to give my friend a lift to the airport.)
Kurzban et al. refer to occupiable resources as finite, dynamic, and divisible, and to depletable resources as finite and depletable over time. They say that “mental ‘resources’ are finite, dynamic, and divisible at any given point in time, rather than finite and depletable over time” (sect. 2.4.2, para. 1, emphasis in the original). Here their use of the term “rather than” indicates that they consider occupiable and depletable resources as alternative means for processing information, when, in fact, they are both essential. A system that could process information using mechanisms alone without any energy input would provide us with an example of perpetual motion.
To make their point, the authors do not need to deny the existence of depletable resources. They can accept the existence of such resources but then attribute the effects they discuss to the rational allocation of mechanisms to higher priority tasks.
I turn now to the main substance of the target article. The authors argue that, by adding a mechanism that rationally allocates processors (occupiable resources) to tasks, they can render their processor account superior to the depletable-resources account. Their argument is convincing. However, it is not fair or balanced. They have shown that an occupiable-resource account that incorporates a rational allocation mechanism is superior to a depletable-resources account that does not incorporate such a mechanism. It is perfectly possible to rationally allocate depletable resources – electricity suppliers do it when faced with a large unexpected loss of generating capacity. A mechanism precisely analogous to the one that the authors describe could be added to the depletable-resources account.
If this were done, it is unlikely that the resulting model would be inferior to the one proposed by Kurzban et al. There is no reason to suppose that prioritization of resources using opportunity costs would be any less effective than prioritization of processors using opportunity costs in explaining all the phenomena that the authors discuss. For example, effects of incentives and availability of alternative tasks, such as using a smartphone, could be handled equally well. Furthermore, one could still argue, as the authors do, that “the sensation of ‘mental effort’ is the output of mechanisms designed to measure the opportunity cost of engaging in the current mental task” (sect. 2.3.2, para. 2, italics original).
It might prove difficult to design experiments to distinguish the occupiable-resources account and the depletable-resources account of performance decrements if a rational resource allocation mechanism were added to both types of model. Recovery rates after demanding performance may provide one line of attack.
Unfortunately, there is a third possibility. Both processors and depletable resources may be rationally allocated to tasks. Distinguishing this alternative from the other two is likely to pose further difficulties.
Finally, I consider the authors’ argument that there are no proposals that identify an explicit neural resource beyond Gaillot and Baumeister's (Reference Gailliot and Baumeister2007) argument in favour of glucose. Kurzban et al. say that any such proposals would need to explain: “(1) what the resource is, (2) how that resource is depleted by effortful tasks, (3) how depletion of the resource is sensed and leads to subsequent decrements in task performance, and (4) why some kinds of mental/neural activity, but not others, lead to resource depletion” (sect. 4.1, para. 6).
These seem very stringent conditions for classifying something as a depletable neural resource. There are many neurological problems, such as Parkinsonism, where performance decrements can be attributed to some neural resource (e.g., a neurotransmitter or neurohormone) that cannot be renewed at the rate at which it is depleted. In such cases, the resource has been primarily depleted not by an effortful task but by disease. Effective drug treatments replace the resource. In cases such as this, the depletable resource is not fuelling the processor but acting as a means of signalling for it. However, its depletion still causes performance impairment.
Kurzban et al. appear to exclude depletable resources that serve as signals rather than as fuels from their definition of a depletable resource. For example, though they say they know of no proposals for an explicit neural resource beyond glucose, they still suggest that information about opportunity costs needed for rational allocation may be provided by levels of a neurotransmitter, such as dopamine. This looks like a depletable-resource account: It predicts that chronic depletion of dopamine via disease or experimental manipulation will lead to an inability to regulate its levels in the prefrontal cortex for signalling purposes, and that, as a result, rational allocation would be impaired.
Distinguishing occupiable and depletable resources at the neural level is open to the criticism that all brain constituents are subject to chemical turnover. Ultimately, it is the rate of this turnover that should allow us to distinguish resource types. Occasionally, I need to replace parts of my car when they are broken or worn out. This does not mean that my car is a depletable rather than an occupiable resource: I have to replace parts of my car much less frequently than I have to re-fill it with fuel.
Were the mind any other sort of system that processes information, there would be few objections to the statement that mechanisms (processors) fuelled by resources carry out processes in order to perform tasks. Poor performance may arise because the mechanisms are overloaded, because resources are being depleted faster than they can be renewed, or for both of these reasons.
We can also think of processors as a type of resource; however, they are occupiable rather than depletable. (Both my car and the fuel it uses are travel resources – but the car is an occupiable resource, whereas the fuel is a depletable one.) If we use this sort of terminology, we still need to recognize that both types of resource (depletable and occupiable) are needed to carry out processes to perform tasks. (Having just fuel or just a car will not enable me to give my friend a lift to the airport.)
Kurzban et al. refer to occupiable resources as finite, dynamic, and divisible, and to depletable resources as finite and depletable over time. They say that “mental ‘resources’ are finite, dynamic, and divisible at any given point in time, rather than finite and depletable over time” (sect. 2.4.2, para. 1, emphasis in the original). Here their use of the term “rather than” indicates that they consider occupiable and depletable resources as alternative means for processing information, when, in fact, they are both essential. A system that could process information using mechanisms alone without any energy input would provide us with an example of perpetual motion.
To make their point, the authors do not need to deny the existence of depletable resources. They can accept the existence of such resources but then attribute the effects they discuss to the rational allocation of mechanisms to higher priority tasks.
I turn now to the main substance of the target article. The authors argue that, by adding a mechanism that rationally allocates processors (occupiable resources) to tasks, they can render their processor account superior to the depletable-resources account. Their argument is convincing. However, it is not fair or balanced. They have shown that an occupiable-resource account that incorporates a rational allocation mechanism is superior to a depletable-resources account that does not incorporate such a mechanism. It is perfectly possible to rationally allocate depletable resources – electricity suppliers do it when faced with a large unexpected loss of generating capacity. A mechanism precisely analogous to the one that the authors describe could be added to the depletable-resources account.
If this were done, it is unlikely that the resulting model would be inferior to the one proposed by Kurzban et al. There is no reason to suppose that prioritization of resources using opportunity costs would be any less effective than prioritization of processors using opportunity costs in explaining all the phenomena that the authors discuss. For example, effects of incentives and availability of alternative tasks, such as using a smartphone, could be handled equally well. Furthermore, one could still argue, as the authors do, that “the sensation of ‘mental effort’ is the output of mechanisms designed to measure the opportunity cost of engaging in the current mental task” (sect. 2.3.2, para. 2, italics original).
It might prove difficult to design experiments to distinguish the occupiable-resources account and the depletable-resources account of performance decrements if a rational resource allocation mechanism were added to both types of model. Recovery rates after demanding performance may provide one line of attack.
Unfortunately, there is a third possibility. Both processors and depletable resources may be rationally allocated to tasks. Distinguishing this alternative from the other two is likely to pose further difficulties.
Finally, I consider the authors’ argument that there are no proposals that identify an explicit neural resource beyond Gaillot and Baumeister's (Reference Gailliot and Baumeister2007) argument in favour of glucose. Kurzban et al. say that any such proposals would need to explain: “(1) what the resource is, (2) how that resource is depleted by effortful tasks, (3) how depletion of the resource is sensed and leads to subsequent decrements in task performance, and (4) why some kinds of mental/neural activity, but not others, lead to resource depletion” (sect. 4.1, para. 6).
These seem very stringent conditions for classifying something as a depletable neural resource. There are many neurological problems, such as Parkinsonism, where performance decrements can be attributed to some neural resource (e.g., a neurotransmitter or neurohormone) that cannot be renewed at the rate at which it is depleted. In such cases, the resource has been primarily depleted not by an effortful task but by disease. Effective drug treatments replace the resource. In cases such as this, the depletable resource is not fuelling the processor but acting as a means of signalling for it. However, its depletion still causes performance impairment.
Kurzban et al. appear to exclude depletable resources that serve as signals rather than as fuels from their definition of a depletable resource. For example, though they say they know of no proposals for an explicit neural resource beyond glucose, they still suggest that information about opportunity costs needed for rational allocation may be provided by levels of a neurotransmitter, such as dopamine. This looks like a depletable-resource account: It predicts that chronic depletion of dopamine via disease or experimental manipulation will lead to an inability to regulate its levels in the prefrontal cortex for signalling purposes, and that, as a result, rational allocation would be impaired.
Distinguishing occupiable and depletable resources at the neural level is open to the criticism that all brain constituents are subject to chemical turnover. Ultimately, it is the rate of this turnover that should allow us to distinguish resource types. Occasionally, I need to replace parts of my car when they are broken or worn out. This does not mean that my car is a depletable rather than an occupiable resource: I have to replace parts of my car much less frequently than I have to re-fill it with fuel.