Jaswal & Akhtar (J&A) assert that there is no social disinterest in autistic individuals. Social motivation is a complex construct and autism a highly heterogeneous condition, so the question of social motivation in autism might not have a global yes-or-no answer. Hence, defining what is different and what is not in social interest in autism remains open. As a contribution to the discussion, we would like to examine another domain that might also challenge the social motivation account in autism and furthermore, point to alternative explanations of autism: mutual social influence.
The idea that autistic individuals might be less influenced by others, and that they might try to influence others to a lesser extent, almost seems a tautology: “autism” derives from a Greek word, whose literal meaning is “self-centered.” Indeed, many aspects of autistic behaviour might be (and often have been) interpreted as resulting from a reduced social influence: As an example, atypical interests could be indicative of a tendency to orient one's preferences independently from any influence from others, or even from any interest in others. Experimental results also suggest that autistic individuals are less influenced by others: When sharing an amount of money (Izuma et al. Reference Izuma, Matsumoto, Camerer and Adolphs2011) or completing a cognitive task (Chevallier et al. Reference Chevallier, Parish-Morris, Tonge, Le, Miller and Schultz2014; de C Hamilton & Krendl Reference de C. Hamilton and Krendl2007), they are less likely than non-autistics to modify their behaviour in the presence of an observer. Similarly, autistic individuals would be less prone to be influenced by their peers’ opinions (Bowler & Worley Reference Bowler and Worley1994) or to attune their speech to their audience (Chevallier et al. Reference Chevallier, Molesworth and Happé2012) when making a statement. Reduced social interest/motivation is typically interpreted as resulting from an insensitivity to one's social environment. This is also the case for signs of reduced social influence.
However, there are numerous situations in which the influence of others is clearly observable in autistics’ behaviour: A majority of studies have found an intact orienting of attention by gaze cueing (for a review, see Nation & Penny Reference Nation and Penny2008). Both automatic and voluntary imitation have been found intact (Bird et al. Reference Bird, Leighton, Press and Heyes2007) or even enhanced (Spengler et al. Reference Spengler, Bird and Brass2010) in autistic individuals. Mimetic desire, the spontaneous propensity to pursue goals that others pursue, is a basic case of social influence that is believed to implicitly shape preferences based on tacit observation of others’ behaviours (Lebreton et al. Reference Lebreton, Kawa, Forgeot d'Arc, Daunizeau and Pessiglione2012). Mimetic desire has been found intact in autistic adults (Forgeot d'Arc et al. Reference Forgeot d'Arc, Vinckier, Lebreton, Soulières, Mottron and Pessiglione2016), suggesting that autistic atypical interests are likely to have other determinants than a lack of influence by one's peers’ interests. Autistics are also influenced by stereotypes shared in their social environment (de C Hamilton & Krendl Reference de C. Hamilton and Krendl2007; Hirschfeld et al. Reference Hirschfeld, Bartmess, White and Frith2007). However, a recent study (Birmingham et al. Reference Birmingham, Stanley, Nair and Adolphs2015) reported an intriguing contrast: an implicit measure revealed stereotypes in both autistic and non-autistic participants but, when asked more explicitly, non-autistic participants reported fewer of them, suggesting that they tended to minimize them when they could, while autistic participants reported them more transparently.
As a whole, basic mechanisms of social influence such as imitation, attentional cueing, mimetic desire, or sensitivity to stereotypes are present in autistics. In these situations, social environment is used in a one-sided way, as a source of information. Conversely, in most situations in which autistic individuals might appear immune to social influence, social environment is involved in a bidirectional relationship: In other words, autistics are less likely to adjust their behaviour in front of an audience to appear more generous (Izuma et al. Reference Izuma, Matsumoto, Camerer and Adolphs2011), more performant (Chevallier et al. Reference Chevallier, Parish-Morris, Tonge, Le, Miller and Schultz2014) or more friendly (Chevallier et al. Reference Chevallier, Molesworth and Happé2012), less confronting (Bowler & Worley Reference Bowler and Worley1994) or less prone to shameful stereotypes (Birmingham et al. Reference Birmingham, Stanley, Nair and Adolphs2015). Thus, the difference between autistics and non-autistics in adjusting their behaviour in the presence of peers might be more about influencing, than being influenced.
Is it then the case that autistics do not influence others? A study by Scheeren et al. (Reference Scheeren, Begeer, Banerjee, Meerum Terwogt and Koot2010) adds two important elements to the discussion: (1) In a situation where the goal was to convince an audience, autistics and non-autistics both made attempts to present themselves favourably. This suggests that, while only non-autistics seem to manage their reputation by default, as a background process in many situations, autistics might at least do it when it is set as an explicit goal. (2) In addition, when the participants in the Scheeren et al. study were instructed about specific expectations of their audience, only non-autistics appeared to take them into account in their presentation. Hence, even when both autistics and non-autistics try to influence others, non-autistics appear more strategic in how they do it.
Far from reflecting a general immunity to social influence, a reduced mutual influence in autism might relate to two specific components: First, autistics might have a higher threshold for attempting to influence others, the threshold being related to either the motivation or the detection of a possibility to influence in a given situation. Second, the skill to flexibly adjust to specific expectations from the audience seems to be less developed in autistics. Contrary to other components of social influence, this skill necessarily involves strategies based on sophisticated recursive representations (i.e., theory of mind; de C Hamilton & Lind Reference de C Hamilton and Lind2016). Both the threshold and the skill components of mutual social influence require further investigation in autism. In particular, although theory of mind has been extensively studied as a decoding tool in social environment (Senju et al. Reference Senju, Southgate, White and Frith2009), its strategic use during reciprocal social interaction, and more specifically its role in mutual social influence, remains virtually unexplored in autism.
There is no doubt to us that autistic persons are influenced by their social environment. Hence, we will have to consider whether the term “autism” [self-centered] is appropriate – both from an ethical and a descriptive point of view – to label this domain of human diversity and the persons who relate to it.
Jaswal & Akhtar (J&A) assert that there is no social disinterest in autistic individuals. Social motivation is a complex construct and autism a highly heterogeneous condition, so the question of social motivation in autism might not have a global yes-or-no answer. Hence, defining what is different and what is not in social interest in autism remains open. As a contribution to the discussion, we would like to examine another domain that might also challenge the social motivation account in autism and furthermore, point to alternative explanations of autism: mutual social influence.
The idea that autistic individuals might be less influenced by others, and that they might try to influence others to a lesser extent, almost seems a tautology: “autism” derives from a Greek word, whose literal meaning is “self-centered.” Indeed, many aspects of autistic behaviour might be (and often have been) interpreted as resulting from a reduced social influence: As an example, atypical interests could be indicative of a tendency to orient one's preferences independently from any influence from others, or even from any interest in others. Experimental results also suggest that autistic individuals are less influenced by others: When sharing an amount of money (Izuma et al. Reference Izuma, Matsumoto, Camerer and Adolphs2011) or completing a cognitive task (Chevallier et al. Reference Chevallier, Parish-Morris, Tonge, Le, Miller and Schultz2014; de C Hamilton & Krendl Reference de C. Hamilton and Krendl2007), they are less likely than non-autistics to modify their behaviour in the presence of an observer. Similarly, autistic individuals would be less prone to be influenced by their peers’ opinions (Bowler & Worley Reference Bowler and Worley1994) or to attune their speech to their audience (Chevallier et al. Reference Chevallier, Molesworth and Happé2012) when making a statement. Reduced social interest/motivation is typically interpreted as resulting from an insensitivity to one's social environment. This is also the case for signs of reduced social influence.
However, there are numerous situations in which the influence of others is clearly observable in autistics’ behaviour: A majority of studies have found an intact orienting of attention by gaze cueing (for a review, see Nation & Penny Reference Nation and Penny2008). Both automatic and voluntary imitation have been found intact (Bird et al. Reference Bird, Leighton, Press and Heyes2007) or even enhanced (Spengler et al. Reference Spengler, Bird and Brass2010) in autistic individuals. Mimetic desire, the spontaneous propensity to pursue goals that others pursue, is a basic case of social influence that is believed to implicitly shape preferences based on tacit observation of others’ behaviours (Lebreton et al. Reference Lebreton, Kawa, Forgeot d'Arc, Daunizeau and Pessiglione2012). Mimetic desire has been found intact in autistic adults (Forgeot d'Arc et al. Reference Forgeot d'Arc, Vinckier, Lebreton, Soulières, Mottron and Pessiglione2016), suggesting that autistic atypical interests are likely to have other determinants than a lack of influence by one's peers’ interests. Autistics are also influenced by stereotypes shared in their social environment (de C Hamilton & Krendl Reference de C. Hamilton and Krendl2007; Hirschfeld et al. Reference Hirschfeld, Bartmess, White and Frith2007). However, a recent study (Birmingham et al. Reference Birmingham, Stanley, Nair and Adolphs2015) reported an intriguing contrast: an implicit measure revealed stereotypes in both autistic and non-autistic participants but, when asked more explicitly, non-autistic participants reported fewer of them, suggesting that they tended to minimize them when they could, while autistic participants reported them more transparently.
As a whole, basic mechanisms of social influence such as imitation, attentional cueing, mimetic desire, or sensitivity to stereotypes are present in autistics. In these situations, social environment is used in a one-sided way, as a source of information. Conversely, in most situations in which autistic individuals might appear immune to social influence, social environment is involved in a bidirectional relationship: In other words, autistics are less likely to adjust their behaviour in front of an audience to appear more generous (Izuma et al. Reference Izuma, Matsumoto, Camerer and Adolphs2011), more performant (Chevallier et al. Reference Chevallier, Parish-Morris, Tonge, Le, Miller and Schultz2014) or more friendly (Chevallier et al. Reference Chevallier, Molesworth and Happé2012), less confronting (Bowler & Worley Reference Bowler and Worley1994) or less prone to shameful stereotypes (Birmingham et al. Reference Birmingham, Stanley, Nair and Adolphs2015). Thus, the difference between autistics and non-autistics in adjusting their behaviour in the presence of peers might be more about influencing, than being influenced.
Is it then the case that autistics do not influence others? A study by Scheeren et al. (Reference Scheeren, Begeer, Banerjee, Meerum Terwogt and Koot2010) adds two important elements to the discussion: (1) In a situation where the goal was to convince an audience, autistics and non-autistics both made attempts to present themselves favourably. This suggests that, while only non-autistics seem to manage their reputation by default, as a background process in many situations, autistics might at least do it when it is set as an explicit goal. (2) In addition, when the participants in the Scheeren et al. study were instructed about specific expectations of their audience, only non-autistics appeared to take them into account in their presentation. Hence, even when both autistics and non-autistics try to influence others, non-autistics appear more strategic in how they do it.
Far from reflecting a general immunity to social influence, a reduced mutual influence in autism might relate to two specific components: First, autistics might have a higher threshold for attempting to influence others, the threshold being related to either the motivation or the detection of a possibility to influence in a given situation. Second, the skill to flexibly adjust to specific expectations from the audience seems to be less developed in autistics. Contrary to other components of social influence, this skill necessarily involves strategies based on sophisticated recursive representations (i.e., theory of mind; de C Hamilton & Lind Reference de C Hamilton and Lind2016). Both the threshold and the skill components of mutual social influence require further investigation in autism. In particular, although theory of mind has been extensively studied as a decoding tool in social environment (Senju et al. Reference Senju, Southgate, White and Frith2009), its strategic use during reciprocal social interaction, and more specifically its role in mutual social influence, remains virtually unexplored in autism.
There is no doubt to us that autistic persons are influenced by their social environment. Hence, we will have to consider whether the term “autism” [self-centered] is appropriate – both from an ethical and a descriptive point of view – to label this domain of human diversity and the persons who relate to it.