Why Do We Experience Mental States?

While mental states are generally thought to be involved in causing behavior, it is not clear why they are useful for this purpose. Would it not be possible for selection to design people so that, like biological automatons, they adaptively responded to stimuli without experiencing mental states (Dennett, 1991)?


All organisms were faced with problems in their evolutionary past in which they had to choose a response. For some problems, the type of response that maximized fitness depended heavily on subtle variations in context; for other problems, the best response was much less dependent upon subtle variations. However, there is no way to design a nervous system that simply calculates the fitness consequences of different actions to determine the option that maximizes reproduction (Symons, 1992; Tooby & Cosmides, 1990).

Still, the organism must have a nervous system that allows it to make good choices. Selection could imbue the nervous system with a suite of hard-wired response rules, each tailored to a particular context. If the organism is merely a collection of hard-wired rules, it will respond to environmental stimuli without necessarily experiencing mental states. However, hard-wired solutions are less tenable for problems in which the optimal response varies dramatically with subtle changes in context. To deal with a nearly infinite number of subtly varying contexts that could be encountered, the biological automaton must come equipped with a nearly infinite number of hard-wired rules, each of which is invoked in a slightly different context.

Alternatively, selection could design a nervous system that allows the organism, before
making a choice, to internally simulate the likely outcomes of behavioral options and predict which ones best satisfy internal goals (Alexander, 1989). Presumably, internal simulation would be less cumbersome than storing a large number of hard-wired rules and more efficient than post hoc learning. It would require the ability to internally represent the self and the external environment, including other actors if their behaviors must be simulated. Since actors have goals, internal simulation would also require representation of the motivational systems of the actors in the simulation. The organism must then make behavioral decisions on the basis of the outcomes of the internal simulation, which will require some internal standard of utility for identifying and comparing desirable and undesirable outcomes (e.g., aesthetic experience). In short, mental states (such as perceptions, beliefs, emotions, intentions, etc.) may allow an organism to identify a behavioral option (from a large suite of options) that approximates an optimal solution to a problem posed by the environment.

That’s Paul Andrews in an old Evolution and Human Behavior paper.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s