The Social Cognition of Intentional Action*

Malingering is defined as an intentional social act of simulating or exaggerating illness. However, determinations of intentionality are often met with suspicion, and among experts in malingering nobody quite likes to make intentionality judgements. A serious medical, legal, and political management of malingering, however, must confront the problem of intentionality. Both malingerers (in planning their deception) and examiners (in trying to detect it) rely on a folk concept of intentionality that has evolved over a million years and has been refined in everyday social practice. Though folk judgements of intentionality are not perfect, they are well grounded in a systematic conceptual framework and thus provide the starting point for a comprehensive analysis of malingering as a social act. This chapter discusses the folk-conceptual and cognitive underpinnings of intentionality judgements and offers implications for a better understanding and handling of malingering. Important distinctions are introduced, such as between intentions (the mental state of being committed to act) and intentionality (the skilful and conscious performance of an intended action), and factors are identified that increase or decrease the validity of intentionality judgements. The picture of malingering painted here is one of a complex but tractable type of intentional action that is, in principle, no more elusive than other intentional actions people perceive and manage every day. Malle Intentional Action 3 Malingering is typically defined as the intentional attempt to simulate or exaggerate illness symptoms in order to reach a consciously desired goal (e.g., health benefits, release from military duty). Because of its intentional and conscious quality, malingering is more blameworthy in our social and legal system than hypochondria, hysteria, or related disorders, which count as unintentional because their illness symptoms are caused by nonconscious factors (e.g., anxiety). However, the distinction between intentional and unintentional behaviour is often treated with suspicion. In particular, within the complex web of individuals and institutions who deal with the problem of malingering, nobody wants to make intentionality judgements. Investigators for the insurance industry or the government are only interested in “facts,” not judgements (even though, naturally, brute facts do not have the causal power to deny benefits or demand repayment; they have causal power only insofar as individuals judge them as relevant and decisive for the question at issue). Physicians focus on describing illness symptoms and organic deficits (or their absence) and claim that judgements about intentional deception lie beyond their professional expertise—they are the lawyers’ province. Lawyers, on their part, argue that they are no experts at distinguishing malingering from hysteria and would rather have the physicians make that call. This avoidance of intentionality is misguided, and for two reasons. First, if malingering is distinct from hypochondria and if one of the distinguishing features is intentionality, then no parties involved can ignore the issue of intentionality. Second, intentionality judgements are made every day, by all social agents, and that has been the case for probably about a million years (Malle, in press-b; McWhinney, in press). Judging the intentions and goals of other people is a common social practice, and as such it is neither metaphysically impossible (as some philosophers have it) nor completely reliable. But if it weren’t the case that most people are correct in their intentionality judgements a Malle Intentional Action 4 good part of the time, social interaction would be impossible, and the evolution of humankind would have been halted long ago. An adequate foundation for social and professional dealings with malingering must involve a clear concept of intentionality. However, scholars of various disciplines have long argued about the meaning intentionality without reaching a consensual understanding of this phenomenon. Rather than add to the numerous expert definitions, I suggest a different approach, namely to focus on the social cognition of intentionality. Social cognition comprises the tasks and tools involved in the human interpretation behaviour, and judgements of intentionality play a central role therein. These judgements have been rendered reliable and socially shareable (see Freyd 1983) by the evolution of a folk concept of intentionality. A folk concept operates like a filter that classifies certain perceptual input into significant categories and thus frames or interprets the perceptual input in ways that facilitate subsequent processing, including prediction, explanation, evaluation, and action (Malle, in press-a). At least some folk concepts are historically and cross-culturally stable, and there is good reason to believe that intentionality is one of them (Ames 2001; Malle & Knobe 1997; Malle & Nelson, in press). Three reasons make this approach compelling. First, malingering agents act in accordance with their folk concept of intentionality—i.e., they generate action that is intentional in their minds. Second, malingering agents conceal the true motives of their actions in light of what they consider other people’s perceptions of intentionality—i.e., they try to behave in ways that do not appear intentional to others. And third, social perceivers of malingering behaviour, such as physicians, insurance agents, jury members or judges, cannot help but use their folk concept of intentionality to distinguish malingering from genuine illness. Malle Intentional Action 5 In this chapter, I thus introduce a definition of intentionality that is grounded in people’s social practice of judging intentionality, and I examine some of the features on which people base such judgements. This contribution does not solve the problem of judging intentionality in any particular case, but it lays the groundwork for an adequate approach to malingering in general by providing a systematic conceptual language that can be used in social as well as legal contexts. 1. The Concept of Intentionality In court and in daily interactions, people regularly make judgements of intentionality. However, surprisingly little research has been devoted to the concept that underlies these judgements. A number of researchers have offered theoretical discussions (Heider 1958; Jones & Davis 1965; Fiske 1989; Shaver 1985), but their respective models disagree on the specific components that make up intentionality. Malle and Knobe (1997) therefore relied on an empirical approach to reconstruct the folk concept of intentionality, using qualitative and experimental methods. In a first study, participants read descriptions of 20 behaviours and rated them for their intentionality, using an 8-point scale ranging from “not at all” (0) to “completely” (7) intentional. About one half of the participants received a working definition of intentionality before they rated the 20 behaviours. The definition read: “What do we mean by intentional? This means that the person had a reason to do what she did and that she chose to do so.” The assumption was that if people used their own folk concept to rate the behaviours, then there should be high agreement among participants with and without an experimenter-provided definition. Agreement was high in the whole sample. Any two people’s intentionality ratings showed an average intercorrelation of r (20) = .64, and any one person showed an average correlation of r (20) = .80 with the remaining group, Malle Intentional Action 6 resulting in an inter-rater reliability of α = .99. More important, the experimenterprovided definition had absolutely no effect on average agreement, so it appears that people share a folk concept of intentionality that they spontaneously use to judge behaviours. The question now becomes what specific components, or “necessary conditions,” this folk concept has. Malle and Knobe (1997) answered this question in two steps. The first was to examine people’s direct and explicit definitions of intentionality; the second was to experimentally manipulate components of intentionality and thus demonstrate their reliable effect on judgements of intentionality. A sample of 159 undergraduate students provided explicit definitions in response to the question “When you say that somebody performed an action intentionally, what does this mean?” Twenty participants (13%) provided only synonyms of the term intentionally (e.g., “on purpose,” “purposefully,” “deliberately”). Of the remaining 139 participants, 54% mentioned exactly one component, 31% mentioned two or more. After initial inspection of the definitions, two coders classified them into various categories, of which four reached substantial frequencies accounting for 96% of the meaningful definitions. These four categories were desire, belief, intention, and awareness. To qualify for the desire category, a definition had to mention the desire for an outcome or the outcome itself as a goal, purpose, or aim (e.g., “He did it in hopes of getting some result”). To qualify for the belief category, a definition had to mention thoughts about the consequences of the act or about the act itself (e.g., “She thought about the act and its effect”). To qualify for the intention category, a definition had to mention the intention to perform the act, or states of intending, meaning, deciding, choosing, or planning to perform the act (“She made a decision to perform the action”). Finally, to qualify for the awareness category, a definition had to mention awareness of the act while the person is Malle Intentional Action 7 performing it (e.g., “He knows what he is doing”). Each of these four categories was mentioned by a quarter to a half of participants, but none of the participants mentioned all four components, presumably because the instructions to this study (“What does it mean that...”) did not encourage exhaustive definitions. Significantly, however, those who mentioned two or more components drew careful distinctions between them. They distinguished, for example, between intention and desire: “The person meant to act that way and was motivated to do so”; between belief and intention: “Someone gave thought to the action beforehand and chose to do it”; between belief and awareness: “This person thought about the action before he did it and was fully aware of performing the action while he was doing it”; and between intention and awareness: “They decided to do something and then did it with full awareness of what they were doing” The folk concept of intentionality, as reconstructed from explicit definitions, thus encompasses four components. For an agent to perform an action intentionally, the agent must have (a) a desire for an outcome; (b) beliefs about an action that leads to that outcome; (c) an intention to perform the action; and (d) awareness of fulfilling the intention while performing the action. To illustrate these components with a malingering behaviour, suppose person P wants to receive disability benefits [desire]. He learns that those benefits are given, for example, to workers who have done hard labour of at least five years (which is true for P) and display severe immobility in their joints or spine [beliefs]. He therefore plans to display exactly those symptoms to the company’s physician [intention]. During the exam, P executes his plan with specific attention to displaying immobility [awareness]. Some theoretical models of intentionality postulate a fifth component of skill or ability (e.g., Mele & Moser 1994), but people did not mention this component in their own Malle Intentional Action 8 definitions. Malle and Knobe (1997) therefore conducted an initial study to explore whether skill may be implicitly used in people’s intentionality judgements, even if it was not explicitly mentioned. In a vignette presented to 141 undergraduate students, a novice at darts surprisingly hits triple 20 (a very difficult throw) on his first try. His partner dismisses the throw as a fluke, so the novice tries again, this time missing badly. Surely, he wanted and tried to hit the triple 20 each time? Most participants (77%) agreed. But would people infer that he hit the triple 20 intentionally the first time? This was not the case, as only 16% said that he hit it intentionally. So most people felt that the novice tried or intended to hit the target, but, without any evidence of skill, they did not feel that he hit it intentionally. Instead, he got lucky. When the scenario was altered to suggest that the novice did have skill—he hit the triple 20 twice in a row—a significantly greater number of participants (55%) were willing to grant that he hit it intentionally even on his first try (p < .001). These initial results suggested that people were sensitive to skill information when making judgements of intentionality. The skill component may have been omitted from explicit definitions because people focused on social behaviours, for which skill can be assumed, in contrast to, say, artistic or athletic behaviours, for which skill cannot be assumed. A more systematic study explored this possible fifth component of intentionality (Malle & Knobe 1997, Study 3). If skill indeed plays a role, it could only be a necessary condition of intentionally performing an action, not a necessary condition of forming an intention. Forming an intention requires only a desire for an outcome and beliefs about an action leading to that outcome (and of course a process of reasoning to combine desires and beliefs; see Malle & Knobe 2001). Once the agent tries to execute that intention, however, skill will be Malle Intentional Action 9 necessary for successfully acting as intended—to perform the action intentionally (not just out of luck or by accident). Thus, the prediction was that a skill component should be necessary for judgements of intentionality (whether the agent truly performed the action intentionally) but not for judgements of intention (whether the agent merely tried or planned to act a certain way). A sample of 132 undergraduate students read a vignette that described a person named David flipping a coin to land on tails, which settled a debate among David and his friends over whether they should go to a movie or not. Additional information was experimentally manipulated to provide information about the presence or absence of David’s skill of making the coin land on the side he wants (“he has not been able to do better than chance” vs. “by now, he almost always succeeds”); desire (“he wants to see the movie” vs. “he does not want to see the movie”); and belief (David hears the suggestion that “tails” stands for going to the movie vs. he does not hear it). The awareness component was always implied to be present. Participants then answered two questions: “Do you think that David tried to make the coin land on tails?” and “Do you think that David made the coin land on tails intentionally?” (Some people were asked only one questions, others both, but the results were identical.) As predicted, the presence of both belief and desire was necessary for an ascription of intention (81% for belief & desire vs. 21% for desire only and 31% for belief only), and the presence of skill was necessary for an ascription of intentionality (76% for belief & desire & skill vs. 3% for belief & desire only). This finding not only skill as a genuine component of intentionality but also highlights that people distinguish between judgements of intention (a mental state of planning or trying) and judgements of intentionality (the quality of an action performed intentionally). To return to the earlier malingering example of a person P trying to gain disability benefits, P must be capable of Malle Intentional Action 10 faking an immobile spine or joint restrictions. If he happens to actually have an immobile spine (without knowing it and even without trying to pretend immobility), then he was not malingering intentionally, even though he had the intention to malinger in the first place.