Cook et al. argue that the proper entry-point to decipher mirror neurons (MNs) is the associative-learning hypothesis. They argue that the standard “genetic” hypothesis implicitly suggests that MNs are adaptive, that is, favored by natural selection. Instead, they argue that MNs exist as a by-product of associative learning – where the genetic component is rather learning as domain-general capability. Put differently, there are no genetic blueprints specific for the MNs phenotype. MNs are simply the by-product of the broader learning mechanism: When the spectator's sensory and motor neurons are excited as the spectator watches another organism doing action X, it is because the same neurons got excited when the spectator, in the past, has undertaken a similar action, say X+. That is, MNs fire as a result of learning, when X+ is judged to be close enough to X. So, to put it plainly, the spectator's MNs fire mainly because of “remembrance”: One can relate to the observed action because it reminds one of one's own experience, whether it is joy or suffering. So, MNs are ultimately the outcome of a self-centered mechanism, where one can reach out to others only if one has undergone the same situation.
Despite its many insightful payoffs, the associative-learning hypothesis leaves two blind spots. First, it cannot, at first approximation, explain how the spectator could understand the action of another, where the spectator has never been in a similar situation. That is, how could there be “action understanding” in cases where X and X+ are far apart? For instance, men can normally understand the pain of women in labor, although they could not have entered such an experience. Adam Smith (Reference Smith, Raphael and Macfie1976b, p.317) mentions this example to repudiate Hobbes' “selfish” theory of sympathy. For Smith, when one sympathizes with another person, it is not because of “remembrance” of his or her joy or suffering – but rather because of the ability to view matters from the station of the other.
Second, if MNs are about learned response, how could MNs fire in some cases and fail to fire in other cases – in cases where the observed actions, X and X+, are very close? For instance, while a spectator's MNs fire when the spectator sees a hand reaching for a cup, they do not fire if the spectator witnesses the same motion, but no cup in sight. The reason for this is MNs usually fire when the action has meaning, that is, there is a goal to reach. As also noted by the Cook et al., the spectator's MNs do not fire when the spectator watches object-absent or pantomimed motion of the hand, even when the motion is very similar to the motion of trying to grasp a cup (see Gallese et al. Reference Gallese, Fadiga, Fogassi and Rizzolatti1996).
Actually, these two cases raise the same question: What is the link between MNs and “meaning” or “action understanding”? To see the link, we have to move beyond the associative-learning approach. To introduce the issue of context, á la Cook et al., is merely ad hoc. A more promising way to decipher MNs is through the anatomy of decision making (Khalil Reference Khalil2011). It seems that MNs fire when the spectator intends to “understand” the intention of the observed actor and, in fact, does understand it. As such, the spectator places himself or herself in the station of the observed actor and imagines what it would be like. Such switching of stations, emphasized long ago by Smith (Reference Smith, Raphael and Macfie1976a), amounts to the spectator looking at the matter not from his or her own self-centered station, but rather from the station of the observed. In such case, the objective of the spectator is knowledge or understanding. If the spectator cannot make sense of the observed action, MNs would not fire. However, if the organism can make sense of the observed action, MNs would fire.
The situation is different if the organism's objective is not knowledge but rather self-enjoyment (utility). In this case, the neurons would fire or not fire irrespective of understanding. Actually, in this case, the neurons in question differ from what is technically called MNs (Khalil Reference Khalil2011). The spectator here uses the observed action as a stimulus – similar to what Cook et al. argue – to remember his or her own experience. In this case, the organism here simply wants to indulge itself in the observed sensation. It does not care to find out “why” or “how come” the observed sensation has risen. It simply takes the observed reaction, without any regards to the observed reaction in relation to its cause, as a way to indulge in self-enjoyment. In this case, the organism does not place itself in the station of the observed, but rather stays in its own station.
To decipher this minute, but intricate difference between the two objectives – understanding as opposed to utility – we need to identify at the entry point what is the objective of the organism (see Khalil Reference Khalil2013). But such a set-up is the hallmark of rational-decision hypothesis. Namely, to understand the function of any trait, we must first identify the actor and the actor's objective.
Cook et al. argue that the proper entry-point to decipher mirror neurons (MNs) is the associative-learning hypothesis. They argue that the standard “genetic” hypothesis implicitly suggests that MNs are adaptive, that is, favored by natural selection. Instead, they argue that MNs exist as a by-product of associative learning – where the genetic component is rather learning as domain-general capability. Put differently, there are no genetic blueprints specific for the MNs phenotype. MNs are simply the by-product of the broader learning mechanism: When the spectator's sensory and motor neurons are excited as the spectator watches another organism doing action X, it is because the same neurons got excited when the spectator, in the past, has undertaken a similar action, say X+. That is, MNs fire as a result of learning, when X+ is judged to be close enough to X. So, to put it plainly, the spectator's MNs fire mainly because of “remembrance”: One can relate to the observed action because it reminds one of one's own experience, whether it is joy or suffering. So, MNs are ultimately the outcome of a self-centered mechanism, where one can reach out to others only if one has undergone the same situation.
Despite its many insightful payoffs, the associative-learning hypothesis leaves two blind spots. First, it cannot, at first approximation, explain how the spectator could understand the action of another, where the spectator has never been in a similar situation. That is, how could there be “action understanding” in cases where X and X+ are far apart? For instance, men can normally understand the pain of women in labor, although they could not have entered such an experience. Adam Smith (Reference Smith, Raphael and Macfie1976b, p.317) mentions this example to repudiate Hobbes' “selfish” theory of sympathy. For Smith, when one sympathizes with another person, it is not because of “remembrance” of his or her joy or suffering – but rather because of the ability to view matters from the station of the other.
Second, if MNs are about learned response, how could MNs fire in some cases and fail to fire in other cases – in cases where the observed actions, X and X+, are very close? For instance, while a spectator's MNs fire when the spectator sees a hand reaching for a cup, they do not fire if the spectator witnesses the same motion, but no cup in sight. The reason for this is MNs usually fire when the action has meaning, that is, there is a goal to reach. As also noted by the Cook et al., the spectator's MNs do not fire when the spectator watches object-absent or pantomimed motion of the hand, even when the motion is very similar to the motion of trying to grasp a cup (see Gallese et al. Reference Gallese, Fadiga, Fogassi and Rizzolatti1996).
Actually, these two cases raise the same question: What is the link between MNs and “meaning” or “action understanding”? To see the link, we have to move beyond the associative-learning approach. To introduce the issue of context, á la Cook et al., is merely ad hoc. A more promising way to decipher MNs is through the anatomy of decision making (Khalil Reference Khalil2011). It seems that MNs fire when the spectator intends to “understand” the intention of the observed actor and, in fact, does understand it. As such, the spectator places himself or herself in the station of the observed actor and imagines what it would be like. Such switching of stations, emphasized long ago by Smith (Reference Smith, Raphael and Macfie1976a), amounts to the spectator looking at the matter not from his or her own self-centered station, but rather from the station of the observed. In such case, the objective of the spectator is knowledge or understanding. If the spectator cannot make sense of the observed action, MNs would not fire. However, if the organism can make sense of the observed action, MNs would fire.
The situation is different if the organism's objective is not knowledge but rather self-enjoyment (utility). In this case, the neurons would fire or not fire irrespective of understanding. Actually, in this case, the neurons in question differ from what is technically called MNs (Khalil Reference Khalil2011). The spectator here uses the observed action as a stimulus – similar to what Cook et al. argue – to remember his or her own experience. In this case, the organism here simply wants to indulge itself in the observed sensation. It does not care to find out “why” or “how come” the observed sensation has risen. It simply takes the observed reaction, without any regards to the observed reaction in relation to its cause, as a way to indulge in self-enjoyment. In this case, the organism does not place itself in the station of the observed, but rather stays in its own station.
To decipher this minute, but intricate difference between the two objectives – understanding as opposed to utility – we need to identify at the entry point what is the objective of the organism (see Khalil Reference Khalil2013). But such a set-up is the hallmark of rational-decision hypothesis. Namely, to understand the function of any trait, we must first identify the actor and the actor's objective.