We surely suffer from a kind of mental myopia, an inordinate focus on things close at hand and a corresponding difficulty taking in what's further afield. This fact underlies many foibles of human reasoning, including, no doubt, our tendency to “explain many observed patterns in terms of the inherent features of the things that instantiate these patterns” (sect. 1, para. 1). Cimpian & Salomon (C&S) artfully demonstrate just how general this tendency is, and how plausibly it is linked to a diverse range of phenomena, from system justification to essentialism and beyond. So far, so good. But need all this inherence be the result of an actual inherence heuristic?
Let us consider two accounts. Story 1 is C&S's, in which the bias towards inherent explanations emerges because we employ an intuitive heuristic that reflexively generates inherent explanations. The heuristic can be overridden by more deliberative processing, but because overriding requires effort, we often end up with inherence even when we shouldn't. On story 2, the same regularities are explained by something more general – let's call it the explanatory heuristic. The explanatory heuristic is very much like the inherence heuristic. In fact, it follows the very same steps pictured in C&S's Figure 2. When faced with something to explain, the explanatory heuristic also conducts a fast, shallow search for relevant information (the “mental shotgun” [sect. 2.2, para. 2]). It, too, passes what it's gathered to a “storyteller” (sect. 2.2, para. 7) that looks for a satisficing explanation, and when it finds one, it too passes that explanation to consciousness, producing the same intuitive sense of having explained. So how is it different? Well, notice that we haven't said anything about inherence. That's because the explanatory heuristic doesn't care about inherence. It just takes what's at its fingertips and stitches it together. As C&S review, what's at its fingertips will often be the sort of stuff that leads to inherent explanations. (We too often attend to focal entities instead of their history; to enduring rather than transient properties; and so on.) In those cases, inherence is what we'll get. But if different stuff is available (the situational sort of stuff, say), it will happily output different kinds of explanations, and they'll feel just as intuitively satisfying. In short, the explanatory heuristic can also explain the bias towards inherent explanations, but it does so it via bias in the input rather than bias in the heuristic.
If the two accounts are so similar, it is fair to ask if we're splitting hairs. I don't think so. I take C&S to be proposing a genuine psychological mechanism, the kind of thing that is in the head, such that it can be blocked or revised, for example. If that's right, the two accounts are proposing different mechanisms and therefore different answers to the question of why so much inherence. Crucially, they also make different predictions regarding what will happen when our explanatory system is stressed – predictions that will be familiar from other work within dual-process frameworks. For the inherence heuristic, manipulations that tax conscious reasoning should lead to more inherence and fewer other forms of explanation, because it should disrupt the more effortful blocking process. For the explanatory heuristic, such manipulations should simply increase reliance on whatever is readily accessible to the mental shotgun, irrespective of whether that information favors dispositional or situational explanations. As far as I'm aware, the work most relevant to this issue favors the latter, more general picture. Most critically, while it's true that cognitive load sometimes leads to greater reliance on inherent or dispositional inferences (e.g., Gilbert et al. Reference Gilbert, Pelham and Krull1988), in other cases it leads to greater reliance on noninherent, situational inferences. The crucial point is not that people are capable of noninherent explanations; obviously they are, and this can be explained as the effortful blocking of the inherence heuristic. But the data suggest that such explanations often do not look like effortful blocking; in fact, they often look just as fast and automatic, just as much like an intuitive heuristic, as do inherent explanations! This happens, for example, when you are directly asked to make situational inferences (Krull & Dill Reference Krull and Dill1996; Krull & Erickson Reference Krull and Erickson1995), or have been thinking about the dynamic nature of human attributes (Molden et al. Reference Molden, Plaks and Dweck2006), or even when incidentally exposed to relevant information (Todd et al. Reference Todd, Molden, Ham and Vonk2011). All this is difficult for the inherence heuristic to account for, but is consistent with the more general explanatory heuristic: If the information within the shotgun's range includes things consistent with a noninherent explanation (“I'm supposed to be thinking about situations!” “Traits are sometimes malleable!”), then those things become, rapidly and automatically, grist for the explanatory mill.
The question I hope to have raised with these comments is whether our reliance (and overreliance) on inherent explanations requires postulating a heuristic devoted to that end, or whether it could emerge from the same dynamics of our bounded and situated rationality that also produce noninherent explanations. On this alternative account, the reason we often fail to perceive “complex chains of historical causes” (sect. 1, para. 1) is simply that they're complex and historical, and therefore difficult for our explanatory apparatus to sink its teeth into. But whatever we think of these alternatives, C&S have done the field a service by carefully tracing the strands of inherence winding their way through so many of the phenomena we study and, in so doing, generating many promising avenues for future investigation.
You have
Access
We surely suffer from a kind of mental myopia, an inordinate focus on things close at hand and a corresponding difficulty taking in what's further afield. This fact underlies many foibles of human reasoning, including, no doubt, our tendency to “explain many observed patterns in terms of the inherent features of the things that instantiate these patterns” (sect. 1, para. 1). Cimpian & Salomon (C&S) artfully demonstrate just how general this tendency is, and how plausibly it is linked to a diverse range of phenomena, from system justification to essentialism and beyond. So far, so good. But need all this inherence be the result of an actual inherence heuristic?
Let us consider two accounts. Story 1 is C&S's, in which the bias towards inherent explanations emerges because we employ an intuitive heuristic that reflexively generates inherent explanations. The heuristic can be overridden by more deliberative processing, but because overriding requires effort, we often end up with inherence even when we shouldn't. On story 2, the same regularities are explained by something more general – let's call it the explanatory heuristic. The explanatory heuristic is very much like the inherence heuristic. In fact, it follows the very same steps pictured in C&S's Figure 2. When faced with something to explain, the explanatory heuristic also conducts a fast, shallow search for relevant information (the “mental shotgun” [sect. 2.2, para. 2]). It, too, passes what it's gathered to a “storyteller” (sect. 2.2, para. 7) that looks for a satisficing explanation, and when it finds one, it too passes that explanation to consciousness, producing the same intuitive sense of having explained. So how is it different? Well, notice that we haven't said anything about inherence. That's because the explanatory heuristic doesn't care about inherence. It just takes what's at its fingertips and stitches it together. As C&S review, what's at its fingertips will often be the sort of stuff that leads to inherent explanations. (We too often attend to focal entities instead of their history; to enduring rather than transient properties; and so on.) In those cases, inherence is what we'll get. But if different stuff is available (the situational sort of stuff, say), it will happily output different kinds of explanations, and they'll feel just as intuitively satisfying. In short, the explanatory heuristic can also explain the bias towards inherent explanations, but it does so it via bias in the input rather than bias in the heuristic.
If the two accounts are so similar, it is fair to ask if we're splitting hairs. I don't think so. I take C&S to be proposing a genuine psychological mechanism, the kind of thing that is in the head, such that it can be blocked or revised, for example. If that's right, the two accounts are proposing different mechanisms and therefore different answers to the question of why so much inherence. Crucially, they also make different predictions regarding what will happen when our explanatory system is stressed – predictions that will be familiar from other work within dual-process frameworks. For the inherence heuristic, manipulations that tax conscious reasoning should lead to more inherence and fewer other forms of explanation, because it should disrupt the more effortful blocking process. For the explanatory heuristic, such manipulations should simply increase reliance on whatever is readily accessible to the mental shotgun, irrespective of whether that information favors dispositional or situational explanations. As far as I'm aware, the work most relevant to this issue favors the latter, more general picture. Most critically, while it's true that cognitive load sometimes leads to greater reliance on inherent or dispositional inferences (e.g., Gilbert et al. Reference Gilbert, Pelham and Krull1988), in other cases it leads to greater reliance on noninherent, situational inferences. The crucial point is not that people are capable of noninherent explanations; obviously they are, and this can be explained as the effortful blocking of the inherence heuristic. But the data suggest that such explanations often do not look like effortful blocking; in fact, they often look just as fast and automatic, just as much like an intuitive heuristic, as do inherent explanations! This happens, for example, when you are directly asked to make situational inferences (Krull & Dill Reference Krull and Dill1996; Krull & Erickson Reference Krull and Erickson1995), or have been thinking about the dynamic nature of human attributes (Molden et al. Reference Molden, Plaks and Dweck2006), or even when incidentally exposed to relevant information (Todd et al. Reference Todd, Molden, Ham and Vonk2011). All this is difficult for the inherence heuristic to account for, but is consistent with the more general explanatory heuristic: If the information within the shotgun's range includes things consistent with a noninherent explanation (“I'm supposed to be thinking about situations!” “Traits are sometimes malleable!”), then those things become, rapidly and automatically, grist for the explanatory mill.
The question I hope to have raised with these comments is whether our reliance (and overreliance) on inherent explanations requires postulating a heuristic devoted to that end, or whether it could emerge from the same dynamics of our bounded and situated rationality that also produce noninherent explanations. On this alternative account, the reason we often fail to perceive “complex chains of historical causes” (sect. 1, para. 1) is simply that they're complex and historical, and therefore difficult for our explanatory apparatus to sink its teeth into. But whatever we think of these alternatives, C&S have done the field a service by carefully tracing the strands of inherence winding their way through so many of the phenomena we study and, in so doing, generating many promising avenues for future investigation.