What are the underlying processes that enable human beings to recognize a happy face? Clearly, featural and configural cues will help to identify the distinctive smile. In addition, the motivational state of the observer will influence the interpretation of emotional expressions. Therefore, a model accounting for emotion recognition is only complete if bottom-up and top-down aspects are integrated.

We address how the motor system can contribute to the “meaning” component of smile perception. A smile perceiver can ground the meaning of non-prototypical smiles by interacting with the presenter to maintain the presenter's type of smile. In this case, the meaning of that smile is congruent with the motor behavior that elicits that smile (such as a funny gesture).

Niedenthal et al's classification of smiles erroneously conflates psychological mechanisms and adaptive functions. This confusion weakens the rationale behind the types of smiles they chose to individuate, and it obfuscates the distinction between the communicative versus denotative nature of smiles and the role of perceived-gaze direction in emotion recognition.

If embodied models no longer address the symbol grounding problem and a “disembodied” conceptual system can step in and resolve categorizations when embodied simulations fail, then perhaps the next step in theory-building is to isolate the unique contributions of embodied simulation. What is a disembodied conceptual system incapable of doing with respect to semantic processing or the categorization of smiles?

Many species' non-threat facial expression involves an open mouth and retracted lip corners – the smile. This served to make an accompanying vocalization sound like it originated from a smaller vocalizer. That such signals are deceptive and benefit primarily the signaler undermines the notion that the perception of the smile employs embodied simulation of the smiler's state.

Differentiation of the forms and functions of different smiles is needed, but they should be based on the empirical data on distinctions that senders and receivers make and on the physical cues that are employed. Such data would allow for a test of whether smiles can be differentiated using perceptual cues alone or whether mimicry or simulation are necessary.

Eye contact plays a critical role in many aspects of face processing, including the processing of smiles. We propose that this is achieved by a subcortical route, which is activated by eye contact and modulates the cortical areas involve in social cognition, including the processing of facial expression. This mechanism could be impaired in individuals with autism spectrum disorders.

Gender differences in face-based emotion recognition, notably differential use of mimicry, may compromise the extent to which the Simulation of Smiles (SIMS) model can be generalized to populations besides the adult females on which it has been tested. Much work indicates sex differences in face-based emotion recognition, including smile recognition.

The parent-infant dyad, characterized by contingent social interactions that develop over the first three months postpartum, may depend heavily on parental brain responses to the infant, including the capacity to smile. A range of brain regions may subserve this social key function in parents and contribute to similar capacities in normal infants, capacities that may go awry in circumstances of reduced care.

Embodied simulation and the epistemic motivation to search for the “meaning” of other people's behaviors are not necessary for specific and functional responding to, and hence processing of, human facial expressions. Rather, facial expression processing can be achieved through lower-cognitive, heuristical perceptual processing and expression of prototypical morphological musculature movement patterns that communicate discrete trustworthiness and capacity cues to conspecifics.

Processing of facial expressions goes beyond simple pattern recognition. To elucidate this problem, Niedenthal et al. offer a model that identifies multiple embodied and disembodied routes for expression processing, and spell out conditions triggering use of different routes. I elaborate on this model by discussing recent research on emotional recognition in individuals with autism, who can use multiple routes of emotion processing, and consequently can show atypical and typical patterns of embodied simulation and mimicry.

The set of 30 stimulating commentaries on our target article helps to define the areas of our initial position that should be reiterated or else made clearer and, more importantly, the ways in which moderators of and extensions to the SIMS can be imagined. In our response, we divide the areas of discussion into (1) a clarification of our meaning of “functional,” (2) a consideration of our proposed categories of smiles, (3) a reminder about the role of top-down processes in the interpretation of smile meaning in SIMS, (4) an evaluation of the role of eye contact in the interpretation of facial expression of emotion, and (5) an assessment of the possible moderators of the core SIMS model. We end with an appreciation of the proposed extensions to the model, and note that the future of research on the problem of the smile appears to us to be assured.